Comparative Mammalian Immunology. The Evolution and Diversity of the Immune systems of Mammals 9780323952194

Citation preview

Comparative Mammalian Immunology The Evolution and Diversity of the Immune systems of Mammals

Developments in Immunology

Comparative Mammalian Immunology The Evolution and Diversity of the Immune systems of Mammals

Ian R. Tizard BVMS, BSc, PhD, ScD (Hons), ACVM (Hons) Department of Veterinary Pathobiology, Texas A&M University, TX, United States

Academic Press is an imprint of Elsevier 125 London Wall, London EC2Y 5AS, United Kingdom 525 B Street, Suite 1650, San Diego, CA 92101, United States 50 Hampshire Street, 5th Floor, Cambridge, MA 02139, United States The Boulevard, Langford Lane, Kidlington, Oxford OX5 1GB, United Kingdom Copyright © 2023 Elsevier Inc. All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher. Details on how to seek permission, further information about the Publisher’s permissions policies and our arrangements with organizations such as the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www. elsevier.com/permissions. This book and the individual contributions contained in it are protected under copyright by the Publisher (other than as may be noted herein). Notices Knowledge and best practice in this field are constantly changing. As new research and experience broaden our understanding, changes in research methods, professional practices, or medical treatment may become necessary. Practitioners and researchers must always rely on their own experience and knowledge in evaluating and using any information, methods, compounds, or experiments described herein. In using such information or methods they should be mindful of their own safety and the safety of others, including parties for whom they have a professional responsibility. To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors, assume any liability for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions, or ideas contained in the material herein. ISBN: 978-0-323-95219-4 For Information on all Academic Press publications visit our website at https://www.elsevier.com/books-and-journals

Publisher: Stacy Masucci Acquisitions Editor: Wendi Baker Editorial Project Manager: Matthew Mapes Production Project Manager: Omer Mukthar Cover Designer: Mark Rogers Typeset by MPS Limited, Chennai, India

Dedication To Claire

This page intentionally left blank

Contents Preface Acknowledgments Acronyms

xvii xxi xxiii

Section 1 Mammalian immunology 1. The evolution of the mammals and their immune systems

3

1.1 Amniotes 1.2 The origins of the mammals 1.3 Mammalian phylogeny 1.3.1 Modern mammals 1.3.2 Eutherians 1.3.3 The initial branching events 1.3.4 The Xenarthra 1.3.5 The Afrotheria 1.3.6 The Laurasiatheria 1.3.7 The Euarchontoglires 1.3.8 After the K-Pg event 1.4 The evolution of mammalian immunity References

3 4 5 6 6 7 8 8 8 8 8 9 12

2. The evolution of viviparity 2.1 The evolution of the placenta 2.1.1 Monotremes 2.1.2 Marsupials 2.1.3 Eutherians 2.1.4 Epitheliochorial placentas 2.1.5 Endotheliochorial placentas 2.1.6 Hemochorial placentas 2.2 Transfer of immunoglobulins 2.3 Maternal-fetal tolerance 2.4 Mechanisms of tolerance 2.4.1 Anatomical adaptations 2.4.2 Localized immunosuppression 2.4.3 Inhibition of complement activation 2.4.4 Suppression of adaptive immunity 2.5 Regulatory cells 2.5.1 Natural killer cells

15 15 15 16 17 17 18 19 19 20 21 21 22 23 23 23 23

2.5.2 Regulatory T cells 2.6 Other immunosuppressive mechanisms 2.6.1 Glycoproteins 2.6.2 Cytokines 2.6.3 Blocking antibodies 2.6.4 Microchimerism 2.6.5 Adaptive immunity 2.6.6 Sperm References

24 25 25 25 26 26 27 27 27

3. The evolution and role of lactation

29

3.1 The origins of lactation 3.2 The functions of milk 3.2.1 Nutritional functions 3.2.2 Intestinal development 3.2.3 Protective functions 3.3 Lactation and the microbiota 3.3.1 The gut-mammary axis 3.4 Adaptive immunity 3.5 Colostrum 3.5.1 Production 3.5.2 Immunoglobulin transfer 3.5.3 Composition 3.5.4 Colostral lymphocytes 3.6 Milk 3.6.1 Milk immunoglobulins References

4. Endothermy and immunity 4.1 The evolution of endothermy and homeothermy 4.2 The benefits of endothermy 4.3 The role of brown adipose tissue 4.4 Fevers 4.4.1 Fevers and endothermy 4.4.2 Hibernation 4.4.3 The costs 4.4.4 Fevers and innate immunity 4.4.5 Fevers and T cell functions 4.4.6 Fevers and B cell functions

29 31 31 31 32 32 33 33 34 34 34 34 35 36 36 38

41 41 43 43 45 46 46 46 47 48 48 vii

viii

Contents

4.4.7 Fevers and bacterial diseases 4.4.8 Fevers and viral diseases 4.5 Fevers, fungi, and the rise of the mammals 4.5.1 Fungi and endothermy References

5. The microbiota
immune system relationship 5.1 5.2 5.3 5.4

Herbivores Carnivores The microbiota
immune relationship The location of the microbiota 5.4.1 Skin 5.4.2 The respiratory tract 5.4.3 The genitourinary system 5.4.4 The gastrointestinal tract 5.5 The functions of the microbiota 5.5.1 Nutritional efficiency 5.5.2 Intestinal protection 5.5.3 Development of the immune system 5.5.4 Regulation of immunity 5.6 Dysbiosis 5.7 Behaviors 5.7.1 Odors 5.7.2 Hibernation 5.7.3 The aryl hydrocarbon receptor 5.8 Environmental microbiota References

6. Innate immunity: basic features 6.1 Constitutive innate immunity 6.1.1 Antibacterial peptides 6.1.2 The complement system 6.1.3 Ribonucleic acid interference 6.2 Induced innate immunity 6.2.1 Pattern recognition receptors 6.2.2 Toll-like receptors 6.2.3 Cell surface toll-like receptors 6.2.4 Intracellular toll-like receptors 6.2.5 Toll-like receptor signaling 6.2.6 Retinoic acid-inducible gene-1-like receptors 6.2.7 Nucleotide-binding oligomerization domain-like receptors 6.2.8 AIM2 receptors 6.2.9 Cyclic GMP-AMP synthase 6.2.10 Pathogen-associated molecular patterns 6.2.11 Bacterial DNA

48 49 49 49 50

53 54 54 54 56 56 57 58 58 59 59 59 59 60 65 65 65 66 66 66 66

69 70 70 71 72 72 72 72 74 74 75 76 76 77 77 77 77

6.2.12 Viral nucleic acids 6.2.13 Damage-associated molecular patterns 6.2.14 Soluble pattern-recognition molecules 6.3 Inflammasomes 6.4 Inflammatory cytokines 6.4.1 Tumor necrosis factor-alpha 6.4.2 Interleukin-1 6.4.3 Interleukin-6 6.4.4 Chemokines 6.4.5 Interferons 6.5 Leukocytes 6.5.1 Sentinel cells 6.5.2 Blood cells 6.5.3 Neutrophils 6.5.4 Eosinophils 6.5.5 Macrophages 6.5.6 Dendritic cells 6.5.7 Mast cells 6.6 The costs of innate immunity References

7. The mammalian major histocompatibility complex 7.1 Major histocompatibility complex structure 7.2 Major histocompatibility complex class Ia molecules 7.2.1 Structure 7.2.2 Gene arrangement 7.2.3 Evolution 7.2.4 Polymorphism 7.2.5 Nonpolymorphic major histocompatibility complex class I molecules 7.3 Major histocompatibility complex class II molecules 7.3.1 Structure 7.3.2 Gene arrangement 7.3.3 Polymorphism 7.3.4 Evolution 7.4 Major histocompatibility complex class III molecules 7.5 Mammalian variations 7.5.1 Major histocompatibility complex molecules and disease 7.6 Major histocompatibility complex and body odors 7.6.1 Odorant receptor-major histocompatibility complex linkage References

78 78 78 80 80 80 80 81 81 82 83 83 83 83 84 84 84 85 85 85

89 89 90 90 91 91 93

93 94 94 94 95 95 95 96 96 97 98 98

Contents

8. T Cells and their receptors 8.1 Flexible immunity 8.2 T cell evolution 8.3 T cell antigen receptors 8.3.1 The antigen-binding chains 8.3.2 The signal transduction components 8.4 T cell antigen receptor functions 8.4.1 Receptor-antigen binding 8.5 Antigen receptor diversity 8.5.1 Gene rearrangement 8.5.2 Base insertion and deletion 8.5.3 Somatic mutation 8.6 T cell receptor diversity 8.6.1 Gene structure 8.6.2 Possible combinations 8.6.3 TRC genes and habitat 8.7 γ/δ T cells 8.7.1 γ/δ-high species 8.7.2 γ/δ-low species 8.7.3 Invariant T cells 8.8 Memory T cells References

9. Mammalian B cells 9.1 Before the mammals 9.1.1 Fish 9.1.2 Amphibians 9.1.3 Reptiles 9.1.4 Antibodies in mammals 9.2 B cell antigen receptor structure 9.2.1 Light chains 9.2.2 Heavy chains 9.3 B cell antigen receptor diversity 9.3.1 IGH locus 9.3.2 IGL locus 9.3.3 IGK locus 9.4 Evolution 9.4.1 V region clans 9.4.2 Immunoglobulin D 9.4.3 Immunoglobulin E 9.5 Generation of immunoglobulin diversity 9.5.1 Recombination signal sequences 9.5.2 Gene rearrangement 9.5.3 Base deletion and insertion 9.5.4 Receptor editing 9.5.5 Somatic hypermutation 9.5.6 Gene conversion 9.5.7 Receptor assembly 9.5.8 Intestinal bacteria and the B cell repertoire 9.5.9 Epigenetics

101 101 101 102 103 104 106 106 107 107 107 107 107 107 111 111 111 111 113 114 114 115

117 117 117 118 118 118 118 119 119 121 122 122 122 122 124 125 125 126 126 127 127 129 129 130 130 130 131

ix

9.5.10 Fc receptors 9.5.11 Fc receptor-like molecules References

131 131 132

10. Mammalian innate lymphoid cells

135

10.1 Innate helper cells 10.1.1 Group 1 innate lymphoid cells 10.1.2 Group 2 innate lymphoid cells 10.1.3 Group 3 innate lymphoid cells 10.2 Natural killer cells 10.3 Nature killer cell receptors 10.3.1 The leukocyte receptor complex 10.3.2 The natural killer complex 10.3.3 Other natural killer cell receptors 10.4 “Trained” immunity 10.5 Natural killer cell subsets 10.6 Natural killer T cells 10.6.1 The CD1 system 10.6.2 The MR1 system References

11. The mammalian lymphoid system 11.1 Sources of lymphocytes 11.1.1 Lymphoid tissue inducer cells 11.2 Primary lymphoid organs 11.3 Thymus 11.3.1 Structure 11.3.2 Function 11.3.3 Thymic hormones 11.3.4 Thymic involution 11.3.5 Species differences 11.4 Peyer’s patches 11.4.1 Structure 11.5 Bone marrow 11.6 Secondary lymphoid organs 11.7 Spleen 11.7.1 Red pulp 11.7.2 White pulp 11.7.3 Function 11.8 Lymph nodes 11.8.1 Structure 11.8.2 Function 11.8.3 Lymphocyte circulation 11.8.4 Species differences 11.8.5 Hemolymph nodes 11.8.6 Other secondary lymphoid organs 11.8.7 Mucosal-associated lymphoid tissues 11.8.8 Tonsils

135 135 135 136 136 137 138 140 141 143 144 144 145 146 146

149 149 149 149 150 150 151 151 151 152 152 152 154 154 154 155 155 156 158 158 159 160 161 162 163 163 163

x

Contents

11.8.9 Bronchus-associated lymphoid tissue 11.8.10 Peyer’s patches 11.8.11 Lymphoglandular complexes 11.8.12 Cecal appendix 11.8.13 Cryptopatches 11.8.14 Anal tonsils 11.8.15 Tertiary lymphoid organs References

Section 2 Mammalian orders 12. The monotremes: echidnas and platypus 12.1 Reproduction 12.1.1 Lactation 12.1.2 Venom 12.2 Hematology 12.3 Innate immunity 12.4 Lymphoid tissues 12.4.1 Thymus 12.4.2 Spleen 12.4.3 Lymphoid nodules 12.4.4 Gut-associated lymphoid tissues 12.4.5 Monotreme major histocompatibility complex 12.4.6 Natural killer cell receptors 12.5 B cells and immunoglobulins 12.5.1 Immunoglobulin M 12.5.2 Immunoglobulin D 12.5.3 Immunoglobulin O 12.5.4 Immunoglobulin A 12.5.5 Immunoglobulin G 12.5.6 Immunoglobulin E 12.5.7 Light chains 12.5.8 V region genes 12.5.9 Immunoglobulin receptors 12.6 T cells and cell-mediated immunity 12.6.1 TRA and TRB genes 12.6.2 TRG 12.6.3 TRD 12.6.4 TRM References

164 165 165 166 166 166 166 167

169 171 171 172 173 173 173 174 174 174 174 175 175 176 177 177 178 178 178 178 178 179 179 179 180 180 181 181 181 182

13. Marsupials: Opossums to Kangaroos 185 13.1 Reproduction and lactation 13.1.1 Protection in the pouch 13.1.2 Lactation

186 187 188

13.1.3 Ameridelphia 13.1.4 Australidelphia 13.2 Hematology 13.3 Innate immunity 13.3.1 Cytokines 13.4 Lymphoid organs 13.4.1 Thymus 13.4.2 Bone marrow 13.4.3 Spleen 13.4.4 Lymph nodes 13.4.5 Gut-associated lymphoid tissues 13.5 The marsupial MHC 13.5.1 Opossum 13.5.2 Australidelphia 13.5.3 The natural killer complex 13.5.4 The leukocyte receptor complex 13.6 B cells and immunoglobulins 13.6.1 Opossum 13.6.2 Heavy chains 13.6.3 Light chains 13.6.4 Fc receptors 13.7 T cells and cell-mediated immunity 13.7.1 The T cell antigen receptors 13.7.2 Other T cell receptors References

14. Tylopoda: Camels and llamas 14.1 Reproduction and lactation 14.2 Hematology 14.3 Innate immunity 14.3.1 Leukocytes 14.3.2 Complement 14.3.3 Cytokines 14.4 Lymphoid organs 14.4.1 Thymus 14.4.2 Spleen 14.4.3 Lymph nodes 14.4.4 Hemal nodes 14.4.5 Mucosal lymphoid tissues 14.5 The major histocompatibility complex 14.5.1 Major histocompatibility complex class I 14.5.2 Major histocompatibility complex class II 14.5.3 Major histocompatibility complex class III 14.5.4 The Natural Killer receptor complexes 14.6 B cells and Immunoglobulins 14.6.1 Old-world camels 14.6.2 Light chains 14.7 Heavy-chain-only antibodies

189 189 190 190 192 192 192 193 193 193 194 194 194 195 196 196 196 197 197 198 198 198 198 200 201

205 206 206 207 207 207 207 207 207 208 208 208 208 209 210 210 210 210 211 211 212 213

Contents

14.7.1 V domain structure 14.7.2 VHH gene segments 14.7.3 Heavy Chain-only Antibody functions 14.8 New-world camels 14.8.1 Constant domains 14.9 T cells and cell-mediated immunity 14.9.1 TRA/D 14.9.2 TRB 14.9.3 TRG 14.9.4 Somatic hypermutation References

15. Suiformes: Pigs and Peccaries 15.1 Reproduction and lactation 15.1.1 Cell-mediated immunity and colostrum 15.1.2 Antibody-mediated immunity 15.2 Hematology 15.2.1 Blood leukocytes 15.3 Innate immunity 15.3.1 Pattern recognition receptors 15.3.2 Acute-phase proteins 15.3.3 Antimicrobial peptides 15.3.4 Cytokines 15.4 Lymphoid organs 15.4.1 Thymus 15.4.2 Spleen 15.4.3 Lymph nodes 15.4.4 Mucosa-associated lymphoid tissues 15.4.5 Dendritic cells 15.5 Major histocompatibility complex 15.5.1 Major histocompatibility complex class Ia 15.5.2 Class II 15.5.3 The natural killer receptor complex 15.6 B cells and immunoglobulins 15.6.1 Immunoglobulin heavy chains 15.7 B cell receptor development 15.8 T cells and cell-mediated immunity 15.8.1 Workshop cluster1 cells 15.8.2 T cell receptors 15.8.3 Natural killer T cells References

16. The cetaceans: whales and dolphins 16.1 Reproduction and lactation 16.2 Hematology

213 213 214 214 214 215 215 216 216 216 217

219 220 220 220 221 221 221 221 222 222 222 223 223 223 223 224 225 226 226 227 227 227 228 230 231 231 232 233 234

237 238 238

16.3 Innate immunity 16.3.1 Neutrophils 16.3.2 Cytokines 16.4 Lymphoid organs 16.4.1 Thymus 16.4.2 Spleen 16.4.3 Lymph nodes 16.4.4 Mucosal associated lymphoid tissues 16.4.5 Anal tonsils 16.5 The major histocompatibility complex 16.5.1 The MHC Class I region 16.5.2 The MHC Class II region 16.5.3 DR loci 16.5.4 DQ loci 16.5.5 Other MHC class II loci 16.5.6 The natural killer receptor complex 16.6 B cells and immunoglobulins 16.6.1 IGH genes 16.6.2 IgM 16.6.3 IgA 16.6.4 IgD 16.6.5 IgG 16.6.6 IGV genes 16.6.7 Light chains 16.7 T cells and cell-mediated immunity 16.7.1 Pressure adaptation References

xi

239 239 239 239 239 240 240 241 242 242 243 243 244 244 244 244 244 244 244 245 245 245 246 246 246 246 247

17. Ruminants: cattle, sheep, and goats

251

17.1 Reproduction and lactation 17.1.1 Secretion and composition of colostrum and milk 17.1.2 Absorption of colostrum 17.1.3 Cell-mediated immunity and colostrum 17.2 Hematology 17.3 Innate immunity 17.3.1 Toll-like receptors 17.3.2 Defensins 17.3.3 Complement 17.3.4 Conglutinin 17.3.5 Cytokines 17.4 Lymphoid organs 17.4.1 Thymus 17.4.2 Spleen 17.4.3 Lymph nodes 17.4.4 Hemal nodes 17.4.5 Mucosal associated lymphoid tissues 17.4.6 The genital lymphoid ring

252 253 253 254 254 255 255 256 256 256 256 257 257 257 257 258 258 260

xii

Contents

17.5 The major histocompatibility complex 17.5.1 The MHC class I region 17.5.2 The MHC class II region 17.5.3 The MHC Class IIb region 17.5.4 The MHC class III region 17.5.5 The natural killer receptor complex 17.5.6 Leukocyte receptor complex 17.5.7 Natural killer complex 17.5.8 Natural killer receptor ligands 17.5.9 Natural killer cell functions 17.5.10 Dendritic cells 17.6 B cells and immunoglobulins 17.6.1 Immunoglobulin heavy chains 17.6.2 IGHM 17.6.3 IGHD 17.6.4 IGHG 17.6.5 IGHV 17.6.6 Ultralong VH CR3 17.6.7 Light chains 17.6.8 Receptor assembly 17.7 T cells and cell-mediated immunity 17.7.1 T cell antigen receptors 17.7.2 T cell receptor genes 17.7.3 TRA/D 17.7.4 TRB 17.7.5 TRG 17.7.6 γ/δ T Cell functions 17.7.7 γ/δ T cells as innate cells 17.7.8 γ/δ T cells as Th1 cells 17.7.9 γ/δ T cells as Treg cells 17.7.10 Workshop cluster 1 proteins 17.7.11 The role of CD163 17.8 Sheep (Ovis aires) and goats (Capra hircus) 17.9 Reproduction and lactation 17.9.1 Sheep 17.10 Innate immunity 17.10.1 Sheep 17.11 Lymphoid organs 17.11.1 Tonsils 17.11.2 Peyer’s patches 17.12 Major histocompatibility complex 17.12.1 Sheep 17.12.2 Natural killer cell receptors 17.13 B cells and immunoglobulins 17.13.1 Sheep 17.13.2 Goats 17.14 T cells and cell-mediated immunity 17.14.1 Sheep 17.14.2 TRA/D 17.14.3 TRB 17.14.4 TRG 17.14.5 Goats

260 260 261 261 262 262 262 263 263 263 264 264 264 264 265 265 265 265 266 267 267 267 267 268 268 268 268 269 269 269 270 271 271 271 271 272 272 272 272 272 272 272 273 273 273 273 274 274 274 274 275 275

17.14.6 WC1 17.15 Other species 17.15.1 Water Buffalo (Bulbalis bulbalis) 17.15.2 Domestic Yaks. (Bos grunniens) References

18. Chiropterans: the bats 18.1 Reproduction and lactation 18.2 Hematology 18.3 Innate immunity 18.3.1 Pattern recognition receptors 18.3.2 Inflammatory responses 18.3.3 Interferon pathways 18.3.4 MicroRNA 18.3.5 Body temperature and hibernation 18.3.6 Immune reconstitution inflammatory syndrome! 18.4 Lymphoid organs 18.4.1 Thymus 18.4.2 Spleen 18.4.3 Peyer’s patches 18.5 The major histocompatibility complex 18.5.1 The MHC class I region 18.5.2 The MHC class II region 18.5.3 The natural killer cell receptor complex 18.6 B cells and immunoglobulins 18.6.1 IGH 18.6.2 IgM 18.6.3 IgD 18.6.4 IgG 18.6.5 IgA 18.6.6 IgE 18.6.7 IGHV 18.6.8 Light chains 18.7 T cells and cell-mediated immunity 18.7.1 TRA/D 18.7.2 TRB 18.7.3 TRG References

19. Feliformes: The cats and their relatives 19.1 The evolution of carnivory 19.1.1 Reproduction and lactation 19.1.2 Hematology 19.1.3 Innate immunity

275 275 275 275 276

281 282 283 283 283 283 284 286 286 287 287 287 287 287 288 288 289 289 290 290 291 291 291 291 291 292 292 292 292 293 293 293

297 297 299 299 301

Contents

19.2 Dendritic cells 19.3 Cytokines 19.3.1 Lymphoid organs 19.4 Bronchus-associated lymphoid tissue 19.5 Mucosal lymphoid tissues 19.5.1 The major histocompatibility complex 19.6 The natural killer cell receptor complex 19.6.1 B cells and immunoglobulins 19.6.2 T cells and cell-mediated immunity 19.7 Other cats 19.8 Hyenas References

20. Caniforms: Dogs, bears, and their relatives 20.1 The domestic dog (Canis lupus familiaris) 20.1.1 Reproduction and lactation 20.2 Hematology 20.2.1 Gray eosinophils 20.3 Innate immunity 20.3.1 Acute-phase proteins 20.3.2 Natural killer cells 20.3.3 Dendritic cells 20.4 Lymphoid organs 20.4.1 Thymus 20.4.2 Spleen 20.4.3 Mucosal lymphoid tissues 20.5 Major histocompatibility complex 20.5.1 The MHC class I region 20.5.2 The MHC class II region 20.5.3 The MHC class III region 20.6 B cells and immunoglobulins 20.6.1 Immunoglobulin heavy chains 20.6.2 Canine IgD 20.6.3 The IgG subclasses 20.6.4 The IgE subclasses 20.6.5 Canine IgA 20.6.6 Immunoglobulin light chains 20.7 T cells and cell-mediated immunity 20.7.1 T cell antigen receptor genes 20.8 MUSTELIDS 20.8.1 Major histocompatibility complex 20.8.2 Immunoglobulins 20.8.3 TCRs 20.9 PROCYONIDS 20.9.1 Lymphoid tissues 20.10 URSIDS 20.10.1 Hibernation

302 302 302 302 303 303 304 304 305 307 307 308

311 311 311 312 313 313 313 314 315 315 315 315 316 316 316 318 318 318 318 319 319 319 320 320 320 321 322 322 322 322 322 322 322 323

20.10.2 Climate change and immunity 20.10.3 Major histocompatibility complex 20.11 PINNIPEDS 20.11.1 Lymphoid tissues 20.11.2 Major histocompatibility complex 20.11.3 Natural killer cells and receptors References

21. The perissodactyls: horses and their relatives 21.1 Reproduction and lactation 21.2 Hematology 21.3 Innate immunity 21.3.1 Toll-like receptors 21.3.2 Antimicrobial peptides 21.3.3 Cytokines 21.3.4 Interleukin-26 21.3.5 Natural killer cells 21.4 Lymphoid organs 21.4.1 Thymus 21.4.2 Spleen 21.4.3 Mucosal lymphoid tissues 21.5 The major histocompatibility complex 21.5.1 The MHC class I region 21.5.2 The MHC class II region 21.5.3 The natural killer receptor complex 21.5.4 Dendritic cells 21.6 B cells and immunoglobulins 21.6.1 Immunoglobulin heavy chains 21.6.2 IGHV genes 21.6.3 Immunoglobulin Light chains 21.7 T cells and cell-mediated immunity 21.7.1 T cell receptor genes 21.7.2 Natural killer T cells and CD1 References

22. The Lagomorpha: rabbits, hares, and picas 22.1 Reproduction and lactation 22.2 Hematology 22.3 Innate immunity 22.3.1 Toll-like receptors 22.3.2 Cytokines 22.3.3 Defensins 22.3.4 Natural killer cells 22.3.5 Acute-phase proteins

xiii

323 323 324 324 324 325 325

329 330 330 331 331 331 332 332 332 333 333 333 333 333 334 334 335 335 335 336 337 337 337 337 338 339

341 342 342 342 343 343 344 344 344

xiv

Contents

22.3.6 Necroptosis 22.4 Lymphoid organs 22.4.1 Thymus 22.4.2 Spleen 22.4.3 Mucosal lymphoid tissues 22.5 Major histocompatibility complex 22.5.1 The MHC Class I region 22.5.2 The MHC class II region 22.5.3 The MHC class III region 22.5.4 Natural killer cell receptors 22.6 B cells and immunoglobulins 22.6.1 IGHM 22.6.2 Immunoglobulin Heavy chains 22.6.3 Locations 22.6.4 IGHV 22.6.5 Immunoglobulin Light Chains 22.6.6 The rabbit B cell antibody repertoire 22.6.7 Fetal liver and bone marrow 22.7 Appendix 22.8 Other mammals 22.9 T cells and cell-mediated immunity 22.9.1 TRA/D 22.9.2 TRB 22.9.3 TRG References

23. The rodents: mice, rats, and their relatives 23.1 Wild rodents versus laboratory rodents 23.2 Myomorpha (rats and mice) 23.3 Reproduction and lactation 23.4 Hematology 23.5 Innate immunity 23.5.1 Pattern recognition receptors 23.5.2 Chemokines 23.5.3 Antibacterial peptides 23.5.4 Acute-phase responses 23.6 Lymphoid organs 23.6.1 Thymus 23.6.2 Spleen 23.6.3 Mucosal tissues 23.7 Major histocompatibility complex 23.7.1 The MHC class Ia region 23.7.2 Polymorphism 23.7.3 Nonpolymorphic major histocompatibility complex class Ib molecules 23.7.4 The MHC class II region 23.7.5 Gene arrangement 23.7.6 Major histocompatibility complex class III molecules

344 344 344 345 345 345 345 346 346 346 346 347 347 348 348 349 349 349 350 351 351 351 352 352 353

355 356 357 357 357 358 358 358 358 359 359 359 359 359 359 360 361

361 362 362 362

23.7.7 The natural killer cell receptors 23.8 B cells and immunoglobulins 23.8.1 B cell subsets 23.8.2 Immunoglobulin Heavy chains 23.8.3 IGHG genes 23.8.4 IGHD genes 23.8.5 Immunoglobulin Light Chains 23.8.6 Fc receptors 23.9 T cells and cell-mediated immunity 23.9.1 TRA/D 23.9.2 TRB 23.9.3 TRG 23.9.4 Natural killer T cells 23.9.5 Thy-1 23.10 Rats (Rattus norvegicus) 23.10.1 RT1: the rat major histocompatibility complex 23.10.2 Rat natural killer cell receptor complex 23.10.3 Rat immunoglobulins 23.11 Other rodents 23.11.1 Prairie voles (Microtus ochrogaster) 23.11.2 Great gerbils (Rhombomys opimus) 23.11.3 Guinea pigs (Cavia porcellus) 23.11.4 Capybaras (Hydrochoerus hydrochaeris) 23.11.5 Hamsters (Mesocricetus ssp) 23.11.6 Mole-rats (Heterocephalus glaberi, Spalax ehrenbergi) References

363 364 364 364 364 365 365 366 366 366 367 367 367 367 368 368 368 368 369 369 369 369 370 371 371 372

24. The primates: humans and their relatives

375

24.1 Infectious disease history 24.2 Reproduction and lactation 24.2.1 Immunoglobulin transfer 24.2.2 Human colostrum and milk 24.3 Hematology 24.4 Innate immunity 24.4.1 Acute-phase proteins 24.5 Lymphoid organs 24.5.1 Thymus 24.5.2 Spleen 24.5.3 Mucosal lymphoid tissues 24.6 Major histocompatibility complex 24.6.1 Humans 24.6.2 The MHC class I region 24.6.3 The MHC class II region 24.6.4 The MHC class III region 24.6.5 Great apes

376 377 377 377 377 378 378 379 379 379 379 380 380 380 381 382 382

Contents

24.6.6 24.6.7 24.6.8 24.6.9 24.6.10

Old World monkeys New World monkeys Nonclassical MHC class I genes Natural killer cell receptors Killer cell immunoglobulin-like receptor ligands 24.6.11 Great apes 24.6.12 Old World primates 24.6.13 New World primates 24.6.14 Prosimians 24.7 B cells and immunoglobulins 24.7.1 Humans 24.7.2 Immunoglobulin heavy chains 24.7.3 IgM 24.7.4 IgD 24.7.5 IgG 24.7.6 IgA 24.7.7 CD89 24.7.8 IgE 24.7.9 IGHV 24.7.10 Immunoglobulin Light chains 24.7.11 Other primates 24.7.12 Great apes 24.7.13 Old World monkeys 24.7.14 New World monkeys 24.7.15 Prosimians 24.7.16 Immunoglobulin D 24.8 T cells and cell-mediated immunity 24.8.1 T cell antigen receptors References

25. The Afrotheria: Elephants, manatees, and their relatives 25.1 Elephants 25.1.1 Reproduction and lactation 25.1.2 Hematology 25.1.3 Innate immunity 25.1.4 Cytokines 25.1.5 Acute-phase responses 25.1.6 Adaptive immunity 25.1.7 The major histocompatibility complex 25.1.8 The natural killer cell receptor complex 25.1.9 B cells and immunoglobulins

382 383 383 384 384 385 386 386 386 386 386 387 387 387 388 388 389 389 389 389 390 390 390 390 391 391 391 392 393

397 398 398 398 400 400 401 401 401 402 402

25.2 MANATEES 25.2.1 Hematology 25.2.2 Lymphoid organs 25.2.3 The major histocompatibility complex 25.2.4 B cells and immunoglobulins 25.2.5 T cells and cell-mediated immunity 25.3 Mammalian life-spans 25.4 The r/K trade-off 25.5 Body mass and immunity References

26. Four other orders: the Xenarthra, the Scandentia, the Eulipotyphla, and the Pholidota 26.1 Xenarthra: sloths, armadillos, and anteaters 26.1.1 Reproduction and lactation 26.1.2 Hematology 26.1.3 Lymphoid organs 26.1.4 The Major histocompatibility complex 26.1.5 Natural killer cells 26.1.6 B cell responses 26.1.7 T cell responses 26.2 Scandentia. The tree-shrews 26.2.1 Innate immunity 26.2.2 Major histocompatibility complex 26.2.3 Natural killer cell receptors 26.2.4 B cell responses 26.3 Eulipotyphla. The shrews 26.3.1 Hematology 26.3.2 Lymphoid organs 26.3.3 The pancreas of Aselli 26.3.4 Vaginal tonsils 26.3.5 Hedgehogs 26.4 Pholidota. The Pangolins 26.4.1 Hematology 26.4.2 Major histocompatibility complex References Indexs

xv

404 404 404 404 405 405 407 408 408 409

411 411 412 414 414 414 414 415 415 415 415 416 416 416 417 417 417 417 418 418 418 419 419 419 423

This page intentionally left blank

Preface Over many years of teaching veterinary immunology, I often wondered about the seemingly random differences between the immune systems of domestic mammals. We have known about the immune peculiarities of individual domestic mammals for many years. Pigs have inverted lymph nodes. Cattle have unusual immunoglobulin variable regions. Sheep have 13 IgA subclasses. Horses have seven IgG subclasses. This has always seemed to be very untidy. One of my motivations for writing this text was to bring order to what seemed to be a chaotic situation. Thus, in looking at mammals as a whole, I had hoped to identify some phylogenetic pattern behind their immunologic diversity. That was not to be. The diversity of infectious and parasitic diseases as well as the enormous diversity of the mammals themselves has ensured that immune systems have to be diverse down to the species level. Likewise, the major antigen receptor systems, the major histocompatibility complex, and the antigen receptors on T and B cells have evolved as a result of an essentially random “life and death” process. The evolution of the mammalian immune systems has been empirical and opportunistic. They use whatever works. Another feature that emerged during the writing of this book is its focus on “receptors.” On reflection this is unsurprising. Immune responses require extensive “conversations” between multiple cell types as well as foreign antigens. These “conversations” are of necessity, conducted through ligand-receptor binding. Receptors thus control the immune system by controlling cell interactions. The “language” of these interactions is determined by receptors and the genes that encode them. Phylogeny thus controls the language and hence the conversation.

What are its features? This is a book about differences. We know that all mammals are warm-blooded. They suckle their young and have hair. Equally importantly, we recognize that mammals come in an enormous diversity of shapes and sizes. The same applies to their immune systems. Just as mammals diversified in response to environmental and other selective pressures, so too did their immune systems. While all conform to a similar basic pattern, it’s the differences that provide important information and interesting twists to the plot. Thus, this text summarizes the common mechanisms and pathways of the innate and adaptive immune systems. It also seeks to focus on the differences and peculiarities of each mammalian order and, if warranted, the unique features of each species. One feature of modern immunology as reflected in the literature is its use of a single flagship species to represent an entire order. This focus is not confined to humans and laboratory rodents. Other orders have their “favorite” species as well. Thus the domestic dog is the clear leader in caniforms, the bottlenose dolphin in cetaceans, and the black flying fox (Pteropus alceto) among chiropterans. However, it must be remembered that while these species may be common and convenient to study, they may not, in fact, be typical of the other members of their orders. Every living creature is obliged to invest in its defenses against microbial invasion. Every mammal that has existed over the past 300 million years lived and died in the presence of bacteria, viruses, and diverse parasites. On the other hand, microbes and mammals have also developed mutually beneficial commensal relationships. Microbes will take every opportunity to exploit animals for shelter and nutrition. Mammals are obliged to respond by investing in their defenses. Immune defense systems directed against micropredators are mandatory. Mammals have to defend not only against many different bacteria, viruses, and parasites but also against a diversity of threats of different frequencies and significance. From the skin and environmental bacteria that can infect cuts and scratches to the highly lethal pandemic agents that can eliminate whole populations. Immunity incurs costs and mammals must balance both the costs and benefits. This balance will, of necessity, be different for every species and be profoundly influenced by environmental, behavioral, and age issues. Take one specific example. Until about 150 years ago, the human primate population of planet Earth remained virtually static. Deaths and births are roughly balanced. People lived long enough to raise a few children and then died. The investment in immunity was sufficient to permit the survival of the species but rarely had the luxury of inducing xvii

xviii

Preface

prolonged and comfortable old age. On occasion, lethal epidemics drastically culled the population. The black death in Europe killed 30%
40% of the population over a few years. More drastically, it is estimated that European diseases killed about 90% of Native Americans in the century after Columbus arrived. Their immune investment, while appropriate for the healthy climates of the Americas, was inadequate—in most cases, to handle the microbial challenges presented by the European invasions. This selective process is by no means restricted to humans. Thus in 2015, 200,000 saiga antelope died in Kazakhstan as a result of bacterial pneumonia killing 88% of the population. Rinderpest, another introduced infectious disease, likely killed 80%
90% of native cattle in South and East Africa in addition to huge numbers of buffalo, giraffe, and antelope when introduced by European cattle in the 1890s. Mange has recently killed up to 90% of certain vicun˜a populations in the Andes. Infectious diseases exert strong selective pressures on populations and if their immune systems are inadequate have a potent selective effect. Consider the death toll of the COVID-19 coronavirus and what it might have been in the absence of vaccines and effective drugs. “Survival of the fittest” applies directly to the immune systems. Inadequate immunity means premature death. But challenges differ; environmental and climatic conditions influence disease severity. Infectious diseases are known to be more prevalent in the tropics. The pressures of infectious disease differ between herd animals and those that live in solitary isolation. Immune systems also differ. While immunity is critical, so too are many other body systems, and needs must be balanced. Immunity has its costs. Most mammalian species that have ever lived have gone extinct. The extant mammals are the survivors. The descendants of those whose immune systems worked. What we see today is the product of hundreds of millions of years of evolution. These are great systems but of course not infallible. Infectious agents such as coronaviruses still have the power to adapt rapidly and exploit any weakness in the defenses. They cannot be equally strong everywhere. Resistance to one agent may result in susceptibility to another. Immediate needs come first. The immune system, like ourselves, cannot tell what the future holds but must be prepared for almost every eventuality. In the words of Paul Colinvaux, “Every defense sets the evolutionary stage for a new attack, which gives the opportunity for a new defense.”

What are its weak links? As expected, the amount of data available on the immune systems of each mammalian order is highly variable. All are dwarfed by the information available on the mouse and the human so these have been taken as subjects to which other species can be compared. The immune systems of the domestic animal species, bovine, sheep, pig, horse, dog, and cat have also been described in detail. Other mammalian orders such as the monotremes, marsupials, bats, and cetaceans, have, for their specific reasons, been described in some detail but many information gaps remain. Finally, for the great majority of mammals, very little or nothing is known about their immune systems. I have described such data as is available and hope that the existence of these obvious gaps will provide a stimulus to find and examine the missing pieces. As an immunologist, I have been surprised to learn of the ongoing disputes between paleontologists and geneticists regarding the phylogeny of mammals and the timing of their major divergences. Many of these disputes appear to result from differences emerging between the fossil timescale and recently derived molecular phylogenies. As an immunologist, I have no stakes in these disputes. I have however used throughout the book, simplified phylogenies that appear to me to be logical. I hope that those who disagree with my selected divergence dates will forgive me. Other important issues that I became aware of while writing the text are the great discrepancies in the numbers of genes reported at loci such as the immunoglobulin heavy chain locus, or the number of alleles in each histocompatibility complex locus. These discrepancies reflect the state of the art. For many species, these genes have only recently been characterized and time has not permitted an in-depth investigation into these loci. As genomes are more thoroughly investigated, these numbers creep up. I have therefore simply accepted the highest reported numbers of genes in the belief that as these loci are investigated, more genes will be identified. The numbers quoted here are not likely to be the final answer.

What is next? This book does not presume to describe the immune system of every mammal, even if such information is readily available. Features of the immune system that are unexceptional are reviewed. Only if they represent a difference from the consensus system or to provide a reassurance that they are unexceptional, or for continuity’s sake, are they mentioned. This is especially the case in the mouse and human where there is much detailed information on the immune system

Preface

xix

that has yet to be investigated and confirmed in other mammals. Thus, in the case of the mouse and human and to a somewhat lesser extent in the domestic mammal species, many areas are lightly touched on. Finally, please go out and fill in the gaps. Find out about the immune systems of colugos, pangolins, anteaters, aardvarks, wombats, and tenrecs. After all, naked mole rats have provided us with important information on aging. Who knows what important findings are out there? The big picture is clear but there are large blank spots on the canvas that need to be filled in. Ian R. Tizard April 2022

This page intentionally left blank

Acknowledgments I would like to express my great appreciation to those colleagues who have reviewed chapters and provided interesting illustrations, especially Dr. Breanna Breaux, Dr. Brian Porter, and Dr. Mark Johnson, of Texas A&M University, and Dr. Joe Flanagan of the Houston Zoo. I would also like to thank Mr. Robert Tizard of the Wildlife Conservation Society for providing many fantastic images of rare mammals. As always, this book could not have been completed without the wholehearted support of my wife, Claire.

xxi

This page intentionally left blank

Acronyms Immunologists are incorrigible users of acronyms. Here are some that are widely employed in this text ADCC APC APRIL BAFF BALT BCR BoLA C CAM CD cDC CDw CDR cM CTL DAMP DAP DC dsRNA EGF Fab FACS Fc FcR FcRn FoxP3 FPT GALT GM-CSF GPI GVH HEV HLA HMGB1 HSP ICAM IDO IEL IFN Ig IK IL ILC ITAM J

antibody-dependent cell-mediated cytotoxicity antigen-presenting cell a proliferation-inducing ligand B cell
activating factor bronchus-associated lymphoid tissue B cell (antigen) receptor bovine leukocyte antigen complement cell adhesion molecule cluster of differentiation classical dendritic cell cluster of differentiation (provisional designation) complementarity determining region centimorgan, a unit of genetic distance cytotoxic T cell damage-associated molecular pattern DNAX-activating protein dendritic cell double-stranded RNA epithelial growth factor antigen-binding fragment fluorescence-activated cell sorting crystallizable fragment (of immunoglobulin) Fc receptor neonatal Fc receptor Forkhead box P3 failure of passive transfer gut-associated lymphoid tissue granulocyte-macrophage colony-stimulating factor glycosyl-phosphatidylinositol graft-versus-host (disease) high endothelial venule human leukocyte antigen high-mobility group box protein-1 heat shock protein intercellular adhesion molecule indolamine 2,3 dioxygenase intraepithelial lymphocyte interferon immunoglobulin immunoconglutinin interleukin innate lymphoid cell immunoreceptor tyrosine-based activation motif joining

xxiii

xxiv

Acronyms

JAK kb kDa KIR KLR LAIR LFA LGL LILR LPS LRC LT β2M MAb MAMP MAP kinase MDA5 MHC MIP MDSC my mya MYD88 NF-κB NK NKC NKT NLR NOS NOX NS ORF OSCAR OTU PAMP pDC pIgR PIR PKC PP PRR RIG-1 RLR ROS RNS RSS SAA SAP SLA ssRNA STAT TAP TCR TdT TGF Th cell TLR TK

Janus kinase kilobase, a measure of gene size kilodalton killer cell immunoglobulin-like receptor killer cell lectin-like receptor leukocyte-associated immunoglobulin-like receptor leukocyte function-associated antigen large granular lymphocyte leukocyte immunoglobulin-like receptor lipopolysaccharide leukocyte receptor complex lymphotoxin (or leukotriene) β2-Microglobulin monoclonal antibody microbial-associated molecular pattern mitogen-activated protein kinase melanoma differentiation-associated gene-5 major histocompatibility complex macrophage inflammatory protein myeloid-derived suppressor cells million years million years ago myeloid differentiation primary response-88 nuclear factor-κB natural killer (cell) natural killer (receptor) complex natural killer T (cell) nucleotide-binding oligomerization domain, leucine-rich repeat (receptor) nitric oxide synthase NADPH oxidase natural suppressor (cell) open reading frame osteoclast-associated receptor operational taxonomic unit pathogen-associated molecular pattern plasmacytoid dendritic cell receptor for polymeric immunoglobulin paired immunoglobulin-activating receptor protein kinase C Peyer’s patch pattern-recognition receptor retinoic acid-inducible gene 1 retinoic acid-inducible gene-like receptor reactive oxygen species reactive nitrogen species recombination signal sequence serum amyloid A (protein) serum amyloid P swine leukocyte antigen single-stranded RNA signal transducer and activator of transcription transporter for antigen processing T cell (antigen) receptor terminal deoxynucleotidyl transferase transforming growth factor helper T cell toll-like receptor thymidine kinase

Acronyms

TNF TRAIL Treg (cell) TSLP VEGF WC

tumor necrosis factor TNF-related apoptosis-inducing ligand regulatory T (cell) thymic stromal lymphopoietin vascular endothelial growth factor workshop cluster

xxv

This page intentionally left blank

Section 1

Mammalian immunology

This page intentionally left blank

Chapter 1

The evolution of the mammals and their immune systems Every multicellular organism must be able to defend itself against invasion and exploitation by microorganisms. Even the lowliest invertebrate needs a defense system. The existence, and indeed dominance of microbes have been a feature of life since the earliest times. As vertebrates evolved, they too had to maintain their integrity by excluding microbial invaders. The earliest mammals could not have survived without an effective immune system. Immunity and resistance to microbial pathogens have always been a key part of the evolutionary process. This must inevitably have involved not only constitutive defenses, the production of molecules and cells whose role was to simply kill invaders, but it also has involved induced defenses whose function is to respond to infection by rapidly counter-attacking and destroying any invaders. Every mammal ever born would have to be able to defend itself against microbial invasion if it were to survive and reproduce. Thus the evolution of immunity is as much a part of mammalian living systems as the evolution of the skeleton or the cardiovascular system. This would also have been a relatively complex process since microbial threats are also constantly evolving. No doubt innumerable arms races between invader and host have occurred in the past. The immune systems of today represent the current state of these races—but the process continues; bacteria and viruses attack, and mammals evolve to counter them [1]. Resistance and counter-resistance fuel these races in which each protagonist may gain a temporary advantage that is then countered by a move by the other. The net result is a constant pressure on the invader and defender to adapt. In the case of mammals, this also requires perpetual positive selection of key defensive pathways. This constant struggle for survival is the basis of the “Red Queen” concept -the need to run constantly to stay in the same place!

1.1

Amniotes

Amniotes are vertebrates that produce a membrane called the amnion, that surrounds and protects their developing embryos. As a result, they do not require an intermediate larval stage nor do they need to lay their eggs in water. The amnion serves to retain water while still allowing oxygen transfer and hence respiration. As a result, amniote embryos can develop either in eggs laid on land or even within the mother. This differentiates them from the preamniotes such as the fish and amphibians. The evolution of the amnion facilitated the movement of vertebrates from water to land. The first amniotes emerged around 350 million years ago (mya) during the Carboniferous period. As a result of their ability to grow and survive out of the water, the amniotes became, and remain the predominant land vertebrates. The amnion itself consists of several different membranes, including the amniotic sac that surrounds the developing fetus in the placental (Eutherian) mammals, as well as the chorion and allantois. The early amniotes also developed a leathery porous eggshell that permitted their eggs to be much larger than those of fish or amphibians. The development of a large yolk sac also permitted the developing fetus to grow to a larger size and develop fully before hatching. This in turn, also resulted in the evolution of much larger animals. As the early amniotes evolved, their fossilized skeletal remains show that they diverged into two main lines. The earliest amniotes, the anapsids, lacked an otic notch at the back margins of the skull since, not being aquatic, they no longer needed a spiracle. One branch subsequently developed a single opening (temporal fenestra) behind each eye. These animals are classified as synapsids, and they eventually gave rise to the mammals. The other branch developed two such openings and are classified as diapsids or sauropsids (Fig. 1.1). The sauropsids eventually gave rise to the reptiles, dinosaurs, and birds. Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00019-8 © 2023 Elsevier Inc. All rights reserved.

3

4

SECTION | 1 Mammalian immunology

Supratemporal fenestra DIAPSID Orbit Infratemporal fenestra

SYNAPSID Orbit

FIGURE 1.1 The structural features of the skull that define sauropsids and synapsids. The synapsids are considered the earliest precursors of what eventually became the mammals. There is no consensus on the functions of these fenestra in the skull. It is generally believed however, that they provided attachment sites for jaw muscles. The synapsids also progressively developed differentiated teeth such as the canines and molars rather than the relatively uniform teeth of amphibians.

1.2

The origins of the mammals

The synapsids are a class of amniotes that possess a single temporal opening (fenestra) on the side of the skull and a bony arch under each eye. These structures are located in the area where the jaw muscles attach. This distinctive feature of their skulls evolved around 320
305 mya during the late Carboniferous period. The synapsids are a distinctly different evolutionary branch from the reptiles (although superficially somewhat similar) and are much more closely related to living mammals [2]. The synapsids have subsequently been classified into a “primitive” subclass, the pelycosaurs, and an “advanced” subclass, the therapsids. The pelycosaurs were lizard-like animals with a sprawling gait and none have survived to the present. The therapsids, on the other hand, as proto-mammals, gradually underwent changes in their pelvis and limb bones and developed an upright gait. This would have enabled them to run faster and be more maneuverable. Modern monotremes such as the duckbilled platypus still retain the sprawling gait. The advanced synapsids, the therapsids first emerged about 310 mya during the Middle Permian period and evolved into a diverse array of shapes, sizes, and lifestyles (Fig. 1.2). They included both large and small herbivores as well as carnivores. They were the dominant land vertebrates at that time but were largely wiped out by the Permian-Triassic mass extinction event that occurred around 252 mya. The predominant surviving synapsids are known as cynodonts. The cynodonts in general became progressively smaller and more mammal-like during the Triassic period and remained relatively small during the Jurassic and Cretaceous periods. The earliest cynodont fossils are dated from the late Permian about 257 mya and are ancestral to modern mammals [3]. The synapsids, in addition, to the characteristic single temporal fenestra on the sides of their skulls progressively developed other anatomical features that we associate with mammals. For example, the earliest synapsids had a bone at the back of the skull called the quadrate that formed a joint with the lower jaw, called the articular. As the synapsids evolved, the quadrate and articular were replaced by two other bones, called the squamosal and the dentary. Instead of having multiple jawbones characteristic of reptiles, the cynodonts progressively relied on a single mandible (dentary). The remaining quadrate and articular bones progressively decreased in size and eventually either disappeared or migrated to the ear where they developed into two of the bony ossicles within the middle ear, the malleus, and incus. The stimuli for these structural changes are unclear. They may have improved hearing or even permitted an increase in brain size. Synapsids also developed differentiated teeth. Instead of a mouth full of almost identical teeth, they eventually developed a consistent tooth pattern with paired canines on the upper jaw and differentiated cheek teeth. This restructuring of the jaw and a diversified dentition would have increased their effectiveness as predators. The therapsids also gradually developed a bony palate that eventually separated the oral from the nasal cavities and permitted much

The evolution of the mammals and their immune systems Chapter | 1

FIGURE 1.2 The geologic time scales and the timing of the progressive emergence of the proto-mammals.

K-Pg event 66 Mya Periods Eras

Triassic

Permian Paleozoic 300

Jurassic

Cretaceous

200

150

Paleogene Cenozoic

Mesozoic 250

5

100

50

0

Million Years ago Prototherians Synapsids

Therapsids

Metatherians

MONOTREMES MARSUPIALS

~200 Mya Amniotes

350310 Mya

Therians

Sauropsids

Eutherians ~180 Mya

PLACENTALS REPTILES

higher respiratory rates. This would also have facilitated chewing rather than requiring their prey to be ingested whole. (They could still breathe even when their mouth was full!) The secondary palate would also have been necessary for suckling. These early synapsids also developed a relatively large brain that filled the endocranium. At some stage, some of the synapsids developed hair. It is unclear when this might have occurred since soft tissues such as skin and hair are rarely preserved as fossils. At least one beaver-like fossil from the middle Jurassic period shows unambiguous hair impressions [4]. There is also evidence of the presence of pre-mammalian therapsid hair in coprolites (fossil feces) dated to the Upper Permian (before 250 mya) [5]. It is suggested that fur may have developed at a time when dinosaurs dominated the globe and in response, mammals became small and nocturnal. As a result, they would have benefited from the additional layer of insulation [6]. Likewise, it is not known when the ability to secrete nutritional milk arose.

1.3

Mammalian phylogeny

Despite the dominance of the dinosaurs, the mammalian clade progressively developed during the Mesozoic era that followed the massive extinction event that occurred at the Permian/Triassic (P-Tr) boundary about 252 mya. This event has been described as a “greenhouse with lethal temperatures” [7]. (The P-Tr event was likely due to massive volcanic activity causing a spike in carbon dioxide levels resulting in lethal global warming). Over the 200 million years of the Mesozoic era, the earth’s climate was generally warm and there was little ice cover. However, local climatic changes may have been triggered by the break-up of the megacontinent, Pangaea into Laurasia and Gondwana around 190 mya. This was followed by the progressive fragmentation of the southern continents between 140 and 80 mya. Thus Gondwana broke up into the continents of Australasia, South America, and the Antarctic. North America separated from Laurasia at about 80 mya. North and South America eventually rejoined much more recently in the Cenozoic era about 2.7 mya. During millions of years of keeping a low profile, throughout the Mesozoic era, the early cynodonts prospered. They became more diverse and abundant. Initially, they were predominantly small insectivores and carnivores. Their trend towards differentiated teeth with canines probably permitted rapid capture and crushing of arthropods. As their size decreased, to maintain endothermy, it became necessary to develop thermal insulation such as hair in addition, to maintaining precise temperature control. Sensitive hearing and an acute sense of smell were also vital for survival. Thus middle ear bones developed. Likewise increased olfactory lobe size resulted in increased brain size. This problem and its solution have been called the “nocturnal bottleneck” [8]. The nocturnal bottleneck hypothesis suggests that the early Eutherian mammals faced severe competition from large aggressive diurnal dinosaurs. The dinosaurs, being largely ectothermic would have generally restricted their activities to the daytime hours when they were warmed by the sun. The mammals, placed under severe pressure by these aggressive predators, would only have survived by becoming nocturnal. However, this change would have stimulated the

6

SECTION | 1 Mammalian immunology

development of endothermy [9]. As a result, these early mammals became less dependent on solar heating or environmental temperature. This nocturnal lifestyle would likely also have resulted in significant changes in visual acuity. It is perhaps no coincidence that nocturnality is widespread in modern mammals, the most obvious case being the bats. It is suggested that the bats may have evolved from a nocturnal ancestor that had emerged before the K-Pg event. Like other placental mammals, they probably took advantage of the elimination of competition from the diurnal dinosaurs to diversify and expand their ranges. The first major evolutionary divergence from the main mammalian line was the emergence of the subclass Prototheria, the Monotremes. The egg-laying Monotremes that currently consist of the echidnas and the platypus survive only in Australia and New Guinea. They probably had a last common ancestor with the therian mammals about 220
200 mya during the Triassic period [10]. The echidnas probably diverged from the platypus lineage much more recently, about 19
48 mya. Like the synapsids, monotremes retain a single orifice through which they defecate, urinate, and reproduce. They lay leathery, uncalcified eggs. They feed their young with milk secreted from glands in modified skin patches on their bellies. The second major divergence from the crown mammals occurred when the placental mammals, the Eutheria, diverged from the marsupials, the Metatheria. Based on molecular genetic analysis, the divergence between the placentals and metatherians likely began around 180 mya. This is much earlier than the earliest metatherian fossil (Sinodelphis in China dated to 125 mya [11] but is consistent with the earliest Eutherian fossil Juramaia dated to 160 mya and also found in China [12]). Marsupials characteristically develop a yolk-sac placenta to feed the developing embryo in utero. They also developed a simple placenta that attaches the embryo to the uterine wall. Pregnancy is very short, and the marsupial embryo is born in a very early stage of development. As discussed in the next chapter, this short gestation period likely enables the embryo to avoid a destructive attack by the maternal immune system. The newborn marsupial climbs to a nipple within a pouch where it attaches and continues its development. Marsupials may also have a pair of “epipubic bones” that support the pouch as the young continues to grow. These bones are not unique to marsupials since they have been found in fossilized monotremes and Eutherians as well. The marsupials probably originated in Gondwana but as the continents drifted apart survived and diversified in what is now South America. Some also succeeded in reaching Australasia by way of Antarctica. (Fossil marsupials have been found in Antarctica). In the absence of competition from Eutherians they prospered. Except for bats and some rodents, the placental land mammals did not reach Australasia until the first humans arrived about 50,000 years ago. Some marsupials, the opossums remained behind and survived in South America. While marsupial diversification occurred in Gondwana in the Mesozoic, the K-Pg event largely stopped this, and subsequent radiation was confined to the isolated island continents of South America and Australia where the oldest marsupial fossil is dated to 55 mya [13].

1.3.1 Modern mammals Despite the continuing threat from the dinosaurs, the mammals survived, thrived, and continued to evolve. They remained small and insectivorous or carnivorous. Once, however, the dinosaurs had disappeared the mammals radiated explosively. There is an ongoing debate about the timing of this explosion. It should be pointed out that until recently, the appearance and stratification of fossil remains were the only methods of dating these divergences. Now, molecular genetic techniques have become available and can generate reliable molecular clocks. It is often the case however, that the results obtained by molecular methods do not agree with the fossil record. In general, molecular clocks tend to support much earlier divergence dates than does the fossil record.

1.3.2 Eutherians Following the divergence of the marsupials, the “true” placental mammals, the Eutheria continued to evolve and diversify. Modern placental mammals now dominate the planet. There are estimated to be about 6100 identified extant mammalian species plus thousands of cryptic species and they are found worldwide. The characteristic feature of the Eutherians is their mode of reproduction. Thus the Eutherian embryo attaches itself to the uterine wall using a large placenta through which the embryo receives both nutrients and oxygen. Eutherians have also succeeded in preventing immunologic rejection of the semi-allogeneic fetus and thus permit it to survive in utero until ready to take on the outside world. Eutherians have a relatively long pregnancy, and their young are often well developed at birth. This is especially obvious in the young of the large herbivores. The final stages of Eutherian development are supported by the nourishment of the newborn through milk, the defining characteristic of mammals.

The evolution of the mammals and their immune systems Chapter | 1

7

Historically, the phylogeny of mammals and other vertebrates has been determined by the examination of fossilized skeletal remains and the classification of morphological characters. As a result, there was much emphasis on skeletal details such as the number of fenestra in the skull. Unfortunately, many such classification attempts have proven to have produced incorrect results. Recently however, it has proved possible to reexamine mammalian phylogeny using molecular methods. Thus phylogeny can now be inferred by comparing the large numbers of gene sequences available for extant mammals. Consequently, there has not always been complete concordance between the paleontological phylogeny and the molecular phylogeny [14]. This is especially the case with placental mammals, especially when seeking to resolve finer-scale phylogenies. Fossil formation is a relatively infrequent event. Thus a species may be around for a very long time before one of its members just happens to be fossilized and paleontologists happen to find that fossil. As a result, a wave of large-scale molecular phylogenetic studies beginning around 2000 has forced some revisions in the timing and details of the mammalian phylogenetic tree.

1.3.3 The initial branching events There is still debate about the initial branching event, the timing, and the precise order in which these early Eutherian lineages diverged. Three models have been proposed for this process. The “long-fuse” model suggests that the initial steps in this diversification occurred early in the Mesozoic around 180 mya. The “short-fuse” model suggests that diversification was a late event, perhaps occurring B100 mya when dinosaurs still ruled. Both these models suggest that mammalian diversification significantly preceded their explosive growth after the K-Pg event. A third model, the “explosive” model suggests that the mammals started their divergence in the early Cenozoic immediately after the K-Pg event around 65 mya [15]. The long fuse model suggests that mammals diversified early but remained small and inconspicuous while the earth was dominated by large dinosaurs. The explosive model suggests that diversification and explosive growth occurred close together once the dinosaurs had been eliminated. Molecular studies using both mitochondrial and nuclear gene sequences have tended to support the short-fuse model. Thus these have confirmed the existence of four supraordinal clades of mammals that last shared a common ancestor about 110
100 mya. Because they diverged rapidly over a very short time period (less than 10 mya) there are still disagreements regarding their precise branching order and the roots of the Eutherian family tree (Fig. 1.3) [16
18].

Lower Cretaceous 110 100

90

EUARCHONTOGLIRES

Upper cretaceous 80

Paleocene 70

50

Million Years ago

Shrews Chiroptera Cetartiodactyls Perissodactyls Carnivores Pholidota Lagomorphs Rodents Scandentia Primates Dermoptera

87

97

91 105 LAURASIATHERIA

AFROTHERIA 103 XENARTHRA

Long-fuse model

60

Eocene

Short-fuse model

“Explosive” model

FIGURE 1.3 A classification of modern mammals and their orders. Mammalian paleontologists continue to differ regarding the precise timing of these divergences. Recent molecular dating studies have tended to push back divergence dates significantly when compared to estimates based on the fossil record.

8

SECTION | 1 Mammalian immunology

The first major split probably occurred between the evolving mammals in the southern hemisphere and those in the north. Thus the northern group, termed the Boreoeutheria, is estimated to have diverged from the main placental line at about 105 mya. They subsequently split again into two supraordinal groups, the Laurasiatheria, and the Euarchontoglires. This divergence likely occurred around 97 mya on the northern continent of Laurasia. The southern group is classified as the Atlantogenata. These also diverged subsequently into the Xenarthra—species classically restricted to South and Central America, and the Afrotheria, which as their name implies, consists of mammals that are almost exclusively of African origin. While the four supraordinal clades are agreed upon, their branching order is contentious. Where and when did the initial split occur? The problem remains unresolved [19]. The most favored but somewhat contentious mammalian family tree divides the Eutheria into two major groups, the Atlantogenata (born around the Atlantic) and the Boreotheria (the northern placental mammals) [16,19]. The Atlantotheria contain the Afrotheria as well as the Xenarthra. Remember that Africa and South America were joined until between 100 and 120 mya when Pangaea broke up. However, recent analysis has failed to support this split. Current opinion based on molecular genomic data suggests that the Xenarthra and the Afrotheria are sister superorders that split from the other mammals about 105 mya (95
114) and subsequently diverged [20]. Thus Afrotheria evolved and diversified in isolation on the Afro-Arabian continent. Likewise, the Xenarthra evolved on the continent of South America after it separated from Pangaea.

1.3.4 The Xenarthra The Xenarthra are of South American origin and consist of very specialized mammals classified into the Cingulata (armadillos) and the Pilosa (anteaters and sloths). While most are restricted to continental South America, the ninebanded armadillo has reached North America but only after the formation of the isthmus of Panama about 2.7 mya.

1.3.5 The Afrotheria Some of the major areas of a dispute relate to the very top level of division of the tree. In 1945 the famed paleontologist George Gaylord Simpson published “The Principles of classification and a classification of Mammals” [21]. This classic work has largely withstood the test of time. Simpson’s work had however, a fairly obvious weak link. He was unclear about the relationships of the unique African mammals such as the elephants, manatees, and aardvarks. As a result, Simpson lumped these into a group he called the Afrotheria. Molecular data now suggests that Simpson’s Afrotheria contained several groups of mammals that are only distantly related. Thus these include the African insectivores (Afroinsectiphilia), and the elephant shrews (Macroscelidea) more related to the Glires. There is however, both molecular and fossil agreement that the Paenungulata, the hyraxes, elephants, and sirenians are closely related. The Afrotheria in the broad sense contains the Tubulidentata (aardvarks); the Afrosoricida (the golden moles and tenrecs); the Macroscelidea (the elephant shrews); the Hyracoidea (hyraxes); the Sirenia (manatees and dugongs); and the Proboscidea (elephants). The last three are related and are therefore grouped into a single clade, the Paenungulata.

1.3.6 The Laurasiatheria The Laurasiatheria contain diverse Eutherian orders including the Carnivora (dogs, cats, bears as well as the marine pinnipeds), the Chiroptera (bats); the Cetartiodactyla (cows, pigs, and whales), the Pholidota (pangolins), the Perissodactyla (horses and rhinoceroses) and the Eulipotyphla (shrews and hedgehogs).

1.3.7 The Euarchontoglires The Euarchontoglires also include a diverse variety of mammals including the primates (both humans and monkeys); the Scandentia (tree shrews) and the Dermoptera (colugos). They also include the Glires (rabbits and rodents).

1.3.8 After the K-Pg event Another, much more famous, extinction event occurred at the end of the Mesozoic era called the Cretaceous-Paleogene (K-Pg) event. This occurred around 66.5 mya and resulted in the extinction of the dinosaurs (This is also called the Cretaceous-Tertiary event). The K-Pg event was triggered by a massive asteroid strike in what is now the Yucatan peninsula in Mexico [22]. The phylogenetic tree is further confused by the mass extinctions that occurred after the K-Pg

The evolution of the mammals and their immune systems Chapter | 1

110

100

90

80

70

60

Million Years ago Shrews Bats Euarchontoglires

91

Cetartiodactyls Perissodactyls Carnivores

77 97

Pangolins

Boreoeutheria

Lagomorphs

78 87

105 Eutheria

9

Rodents Scandentia Laurasiatheria Dermoptera

83 75

Primates

Atlantogenata

Armadillos

Xenarthra

Sloths

103 80

50

Aardvarks Afrosoricida Afrotheria Sirenians Elephants

FIGURE 1.4 The current consensus on the Eutherian superorders, their classification and origins. This classification will be used throughout this text.

event, the extinction of the dinosaurs, and the subsequent rapid diversification of the mammals. Thus unlike the dinosaurs, small mammals survived the K-Pg event and subsequently prospered. (Fig. 1.4). The Eutherian mammals did not expand their numbers significantly until after the K-Pg event. The recent molecular phylogenetic studies have indicated, however, that the initial diversification of the Eutherians began late in the Cretaceous period 100
70 mya. but that the extant modern families developed in the late Eocene and early Miocene epochs of the Cenozoic following the K-Pg event. The greatest diversification of the placentals occurred in the early Cenozoic epoch. This idea is supported by both the long- and short-fuse models [23]. The fossil evidence on the other hand suggests that these families first appeared later in the Cenozoic. Very few mammal fossils are found before the extinction of the dinosaurs and evidence suggests that their greatest expansion occurred after that event. The explosive model suggests that all the diversion occurred within a very short time after the K-Pg event, but this would require improbably high mutation rates and is thus considered unlikely.

1.4

The evolution of mammalian immunity

It is important not to anthropomorphize the role of the immune system. We humans put much emphasis on our need for a long healthy lifespan and freedom from lethal infections. We place much emphasis on the role of the immune system as a defense system. This is of course vitally important. In the absence of such defenses, we would simply be eaten alive. However, given that evolution is all about relative breeding success, it must be emphasized that the immune system is optimized not for long life and good health but simply for sufficient longevity and health to ensure breeding success and the raising of the young. This requires not only defeating aggressive pathogens but also reaching optimal accommodation with the body’s commensals to everyone’s benefit. Commensals and many parasites can be tolerated as long as they do not endanger the success of the species. These needs, while broadly similar between mammalian species are as diverse as the biology of the mammals themselves. A successful mouse may have very different immune requirements than a successful cow or dolphin. Animal hosts with long generation times such as the large mammals, necessarily evolve at slower rates than either short-lived mammals, or most importantly, potential pathogens. This results in asymmetric warfare. Thus even in the presence of lethal infections, several generations may have to pass before sufficiently protective genetic changes can

10

SECTION | 1 Mammalian immunology

develop in mammalian species. Temporary setbacks are common—humans call them pandemics. The signatures of these events can be detected in their genomes. Different mammalian species have different susceptibilities to specific pathogens so that the outcome of an infection in one species may not be the same in another species. In some cases, mammals, both domestic and wild, may serve as reservoir hosts for parasites or pathogens. Such reservoir species may have taken millions of years to adapt to the infection and so become carriers. In general, animals use two types of the immune system to control microbial invaders. One depends on identifying the characteristic molecules expressed by or in the invaders. These molecular patterns include the cell wall structures of bacteria and the unique nucleic acid sequences of viruses. These unique features are recognized by specialized pattern recognition receptors and so stimulate innate immunity—rapid reflexive stereotypic responses such as inflammation. Innate immune mechanisms evolved early in the evolutionary process. Thus they are critical to the survival of invertebrates. However, they play an important defensive role and as a result, have been maintained, refined, and continue to play a key role in the defense of even the most advanced vertebrates including the mammals. The second type of immune system, the adaptive immune system, depends upon the use of specialized receptors to identify specific pathogens. Generated by somatic mutation and other mechanisms, these adaptive receptors can bind and respond specifically to each of the enormous diversity of molecules, especially the proteins, that make up bacteria, viruses, and protozoa. As a result, they too are enormously diverse. Some of these specific receptors can bind to the unique structures (antigens) found on the surface of invading microorganisms. As a result, they enable the body to recognize an enormous diversity of foreign antigens and trigger defensive immune responses to an enormous diversity of invaders. This adaptive response first evolved in the jawless vertebrates but has proven so successful that all subsequent animal life forms, especially all the vertebrates, use it as their main means of defense. Just as throughout millions of years, mammalian shapes, behaviors, diets, and sizes evolved constantly to adapt to their environment, exploit food resources, reproduce successfully, and avoid being eaten by macropredators. So too they were obliged to adapt to the constant pressure placed upon them by the bacteria, viruses, fungi, and protozoa inhabiting their bodies as well as the microbial environment in which they lived. This microbial environment would have varied based on local environmental conditions such as temperature and humidity, based on food sources such as plants or meat, and based on behavior patterns such as a solitary lifestyle or living in densely populated herds or colonies. Each of these factors would have provided mammals with different pressures to evolve mechanisms that maximize survival and reproduction, successfully raise offspring, or find sufficient food in the face of constant microbial and parasite challenges. Presumably many extinct mammals failed to adapt successfully to microbial challenges and are now extinct. (See Chapter 4 for a discussion on dinosaurs and their fungal difficulties.) The extant mammals however, clearly had ancestors that adapted successfully and so survived in a microbial world. Interactions with commensals and pathogens are one of the dominant factors shaping mammalian evolution. The body must not only counter potential pathogens by the use of an effective immune system, but it must also balance the aggressive elimination of potential pathogens with the compromises necessary to maintain an incredibly useful commensal gut and body surface microbiota. Life-history patterns are shaped in many ways by the risks and benefits conferred by the immune system [24]. In the presence of aggressive parasites and pathogens, immunity will be one of the most important determinants of survival and reproductive success. Nevertheless, immunological defenses are costly to maintain and operate. Immunity requires the allocation of precious nutrient calories, and in effect competes with other critical systems as well as growth, reproduction, and temperature. It is reasonable to assume therefore that the investments made in the immune systems are the minimal commensurate with success. Success can be measured by survival and successful reproduction. As mentioned, the major components of the innate and adaptive immune systems had evolved and matured well before the emergence of mammals. Key antigen recognition molecules such as the toll-like receptors, major histocompatibility complex gene products, B cell antigen receptors, and T cell antigen receptors as well as a multitude of cytokines, both regulatory and effector, were present and functional in the earliest mammals. The earliest synapsids had T and B cells and made antibodies. During ongoing mammalian evolution, however, the system required to be “tweaked.” Numerous minor adjustments resulted in improved survival and reproductive effort in the “successful” species. Some components were not found to be beneficial and as a result, suffered a loss of function. While numerous environmental and climatic factors must have resulted in mass extinctions, so too would infectious diseases. It is possible to discern traces of these past disease events in the genomes of modern mammals. One of the most obvious methods is the evidence for positive selection in the genes of the immune system. Host defense mechanisms are under very strong selective pressure and must evolve rapidly. The negative selection caused by infectious agents is not subtle. In situations where a species population density is sufficiently high, epizootics can result in mass die-off that wipes out huge numbers of a species in a very short space of time [25]. Thus in 2015 a mass die-off killed

The evolution of the mammals and their immune systems Chapter | 1

11

.200,000 Saiga antelope (Saiga tatarica) in Kazakhstan. Caused by hemorrhagic septicemia due to Pasteurella multocida, it is estimated that 88% of the saiga in Central Kazakhstan died [25]. Presumably, however, the surviving 12% had an immunogenetic background that enabled them to survive the epizootic and provide the breeding population for future generations of antelope. This was not the first mass die-off suffered by this species. While such die-offs may not necessarily result in total species extinction, they can massively alter the genetic structure of a population and so can have long-term consequences as well. By reducing the population density of a species to a level insufficient to maintain disease transmission they will also tend to reduce their impact significantly until the organisms can no longer sustain themselves. Thus infectious agents go extinct too. A related issue relates to herd behavior. Thus the aggregation of a single species into large shoals, flocks, or herds, is an effective mechanism of defense against macropredators. On the other hand, these aggregates greatly increase the potential for infectious disease spread. Thus members become much more susceptible to micropredation by invading bacteria and viruses. Behavioral changes would also have required changes in immune system components to offset inadvertent adverse consequences. For example, the development of a carnivorous diet in predators would have introduced parasites into the intestinal tract. Social behavior such as the avoidance of sick congeners also plays a role in disease resistance [26]. The development of a large body mass would have made available a larger number of cells to attack invaders and may thus have improved survivability in the face of infectious disease challenges. This too may have contributed to the development of a K survival strategy dependent upon a long life span within a stable environment. While premature death is the ultimate negative selection pressure, diseases need not be immediately lethal to have a negative effect. Mate selection in many mammals is also based on the appearance of good health and fitness. This is not merely visual but also odor-based [26]. Appearance does however, signal resistance to parasites and perhaps other infections. These disease impacts are especially important in closed systems such as islands or specialized environments. It should also be noted that infectious agents and parasites can result in either the direct death of an animal or indirect death through predation. There are many examples, (discussed in Chapter 19), where predators selectively hunt and kill sick prey. They are easier to catch. Thus innate immune responses resulting in sickness may be effectively lethal in prey species that can neither run nor hide. This places major selective pressure on the appearance of good health. Sickness (or its appearance) is to be avoided at all costs. The flood of cytokines induced during an innate response has the potential to itself cause sickness and death [27]. The elimination of gastrointestinal parasites has the potential to cause debilitating diarrhea. Thus there is pressure on the immune system to eliminate invading pathogens without triggering sickness behaviors. This results in a tendency to use adaptive immunity whenever possible. It also means that a balance must be achieved between eliminating pathogens and the costs needed to do this. Partial elimination of parasites may be sufficient in the trade-off between immunity and health. Additionally, mammals (and birds) go to great lengths to avoid any appearance of weakness, especially in the presence of potential predators. Wild mammals may look very healthy but this may hide underlying disease or inapparent weakness. As noted later in this text, the white blood cell counts of wild mammals are consistently greater than those of their captive or domestic counterparts simply because of the relatively high background levels of infections or parasitic infestations in the wild. It must not be forgotten that pathogens evolve too so that life is a constant battle between the innovative methods adopted by parasites to avoid immune elimination and the adaptability of the innate and adaptive immune systems to counter such attacks [28]. Studies on the differences in gene content between the early mammalian lineages such as opossums, and much more recent species such as humans have identified major changes in the genes associated with innate immunity. The most significant of these influence natural killer cells and their receptors (Chapter 10) [29]. One of the major reasons why the immune system is so complex is the need for flexibility. The diversity of potential pathogens requires an equally diverse defensive system. Different pathogens invade by different routes and cause disease and damage by multiple pathways. Thus the immune system is required to be flexible. It needs to mount inflammatory responses against tissue invaders. It needs to make antibodies against extracellular bacteria. It needs to mount cellmediated immune responses against intracellular viruses and bacteria. In other words, the immune system has to be flexible and very adaptable. It achieves this by the use of multiple polymorphic gene families that can recognize the invader and select an appropriate defensive strategy. A mistake in selecting the correct response can be the difference between life and premature death. Millions of years of constant selective pressure, an eternal arms race, means that the defenses are highly effective most of the time. But the race never ends and there is also a continuous selection for the ability to effectively invade new, susceptible hosts. The pathogens never rest, but then neither does the immune system [30]. As a result of all these pressures, it is not surprising that immune system genes are among those that show the most evidence of positive selection [31]. When mammalian genomes are analyzed using sequence-based data for evidence of clusters of positively selected sites, two systems stand out—the immune system and metabolic enzymes [31].

12

SECTION | 1 Mammalian immunology

Some of these immune system components are largely predictable such as the class II molecules of the major histocompatibility complex that determine which foreign antigens can or cannot induce an immune response. Likewise, CD1a molecules are required for a mammal to respond to lipid-based antigens. Other positively selected sites include components of the innate immune system such as complement components 1a, 5, and 8α. Another such component is TLR4, the toll-like receptor that binds bacterial lipopolysaccharides [31]. These rapidly evolving gene clusters can change even in the absence of gene duplication. If we examine their place in mammalian phylogeny, we can see that most domestic animal species are relatively closely related. Even domestic pets such as dogs and cats are more closely related to farm animal species than to primates. Likewise, laboratory animals tend to cluster in a separate groups. It is unsurprising, therefore that significant differences exist among the immune systems of species of interest to veterinarians and researchers. It is also clear that if we are to understand the significance of these differences and how they evolved, we must examine the immune systems of other, unrelated mammal species as well.

References [1] Sironi M, Cagliani R, Forni D, Clerici M. Evolutionary insights into host-pathogen interactions from mammalian sequence data. Nature 2015;16:224
36. [2] Angielczyk KD. Dimetrodon is not a dinosaur: using tree thinking to understand the ancient relatives of mammals and their evolution. Evo Edu Outreach 2009;2:257
71. [3] Botha J, Abdala F, Smith R. The oldest cynodont: new clues on the origin and early diversification of the Cynodontia. Zool J Linn Soc 2007;149:477
92. [4] Ji Q, Luo Z-X, Yuan C-X, Tabrum AR, et al. A swimming mammaliaform from the middle Jurassic and ecomorphological diversification of early mammals. Science 2006;311:1123
7. [5] Bajdek P, Qvarnstrom M, Owocki K, Sulej T, et al. Microbiota and food residues including possible evidence of pre-mammalian hair in Upper Permian coprolites from Russia. Lethaia 2016;49:455
77. [6] Ruben JA, Jones TD. Selective factors associated with the origin of fur and feathers. Am Zool 2000;40(4):585
96. [7] Sun Y, Joachimski MM, Wighall PB, Yan C, Chen Y. Lethally hot temperatures during the early Triassic greenhouse. Science 2010;338:366
70. [8] Gerkema MP, Davies WIL, Foster RG, Menaker M, Hut RA. The nocturnal bottleneck and the evolution of activity patterns in mammals. Proc R Soc B 2013;. Available from: https://doi.org/10.1098/rspb.2013.0508. [9] Clark A, Portner H-O. Temperature, metabolic power and the evolution of endothermy. Biol Rev 2010;85:703
27. [10] Messer M, Weiss AS, Shaw DC, Westerman M. Evolution of the monotremes: phylogenetic relationship to marsupials and Eutherians, and estimation of divergence dates based on a-lactalbumin amino acid sequences. J Mammal Evol 1998;5(1):95
105. [11] Luo ZX, Wible JR, Yuan CX. An early cretaceous tribosphenic mammal and metatherian evolution. Science 2003;302:1934-140. [12] Luo ZX, Yuan CX, Meng QJ, Ji Q. A Jurassic Eutherian mammal and divergence of marsupials and placentals. Nature 2011;476:442
5. [13] Archer M, Flannery TF, Ritchie A, Molnar RE. First Mesozoic mammal from Australia, an early Cretaceous monotreme. Nature 1985;318:363
6. [14] Novacek MJ. Mammalian phylogeny: shaking the tree. Nature 1992;356:121
5. [15] Archibald JD, Deutchmann DH. Quantitative analysis of the timing of the origin and diversification of extant placental orders. J Mammal Evol 2001;8:107
24. [16] Foley NM, Springer MS, Teeling EC. Mammal madness: is the mammal tree of life not yet resolved? Proc R Soc B 2016;. Available from: https://doi.org/10.1098/rstb.2015.0140. [17] O’Leary MA, Bloch JI, Flynn JJ, Gaudin TJ, et al. The placental mammal ancestor and the post-K-Pg radiation of placentals. Science 2013;339:662
7. [18] Murphy WJ, Foley NM, Bredemeyer KR, Gatesy J, Springer MS. Phylogenomics and the genetic architecture of the placental mammal radiation. Ann Rev Anim Biosci 2021;. Available from: https://doi.org/10.1146/annurev-animal-061220-023149. [19] Prasad AB, Allard MW, Green (ED). Confirming the phylogeny of mammals by use of large comparative sequence data sets. Mol Biol Evol 2008;25(9):1795
808. [20] Murphy WJ, Pringle TH, Crider TA, Springer MS, Miller W. Using genomic data to unravel the root of the placental mammal phylogeny. Genome Res 2007;. Available from: https://doi.org/10.1101/gr.5918807. [21] Simpson GG. The principles of classification and a classification of the mammals. Bull Amer Mus Nast Hist 1945;85:1
350. [22] Schulte P, et al. The Chicxulub asteroid impact and mass extinction at the Cretaceous Paleogene boundary. Science 2010;327:1214-8. [23] Meredith RW, Janecka JE, Gatesy J, Ryder OA, et al. Impacts of the cretaceous terrestrial revolution and KPg extinction on mammal diversification. Science 2011;334:521
4. [24] Lochmiller RL, Deerenberg C. Trade-offs in evolutionary immunology: just what is the cost of immunity? Oikos 2000;88:87
98. [25] Fereidouni S, Freimanis GL, Orynbayev M, Ribeca P, et al. Mass die-off of Saiga antelopes, Kazakhstan, 2015. Emerg Inf Dis 2019; 25(6):1169
76.

The evolution of the mammals and their immune systems Chapter | 1

13

[26] Tizard IR, Skow L. The olfactory system. The immune sensing arm of the immune system. Anim Hlth Res Revs 2021;22(1):14
25. [27] Graham AL, Schrom EC, Metcalf JE. The evolution of powerful yet perilous immune systems. Trends Immunol 2022;. Available from: https:// doi.org/10.1016/j.it.2021.12.002. [28] Loker ES. Macroevolutionary immunology: a role for immunology in the diversification of animal life. Front Immunol 2012;. Available from: https://doi.org/10.3389/fimmu.2012.00025. [29] Dunwell TL, Paps J, Holland PWH. Novel and divergent genes in the evolution of placental mammals. Proc Roy Soc B 2017;. Available from: https://doi.org/10.1098/rspb.2017.1357. [30] Liston A, Humblet-Baron S, Duffy D, Goris A. Human immune diversity: from evolution to modernity. Nat Immunol 2021;22:1479
89. [31] Slodkowicz G, Goldman N. Integrated structural and evolutionary analysis reveals common mechanisms underlying adaptive evolution on mammals. Proc Natl Acad Sci USA 2020;117(11):5977
86.

This page intentionally left blank

Chapter 2

The evolution of viviparity The development of the embryo in the eggs of fish, amphibians, reptiles, and birds is limited since it depends entirely on the availability of nutritional reserves stored within the egg itself. The development of viviparity and lactation, therefore, opened up two other avenues through which the developing embryo and newborn could receive nutrition
through the placenta and the mammary glands. The major nutritional components of egg yolks, such as proteins, lipids, calcium, and phosphorus are transported to the yolk by a family of proteins called vitellogenins. The developing embryo in ovo can receive no other nutritional support. In contrast, Eutherian mammals no longer need vitellogenins. The vascularized chorioallantoic placenta can support the nutritional needs of the developing fetus while milk provides essential nutrients to the newly born [1]. As a result, the three ancestral vitellogenin genes were progressively pseudogenized in mammals before 30
70 mya. The only exceptions to this are the egg-laying monotremes that still retain a single functional vitellogenin gene. Conversely, caseins that have a similar nutritional function, probably appeared in the milk of the common mammalian ancestor about 200
310 mya.

2.1

The evolution of the placenta

All mammals except the monotremes rely on a placenta to support fetal growth within the uterus. In Eutherians, the placenta is the sole route by which nutrients and oxygen can reach the fetus while carbon dioxide and other waste products are removed. The placenta is thus a key feature of therian reproduction. The placenta enables mammals to both protect the developing fetus while supporting its growth and development for as long as possible. As mammals have evolved, the placenta has been subjected to selective pressure and multiple different morphological types of placenta have evolved, suggesting convergent evolution. Thus an organ with great morphological variation plays the same functional role across multiple mammalian orders. Mammalian placentas presumably evolved between 300 and 150 mya with the rise of the monotremes and the appearance of the first Metatherians and Eutherians. The development of viviparity was not however a simple single step. Too many changes are required to turn an egg-laying synapsid into a fully viviparous Eutherian. The evolution of viviparity must have involved progressive adaptations of preexisting precursor genes, causing gradual alterations in regulatory pathways and modifying many different proteins to permit their expression in the primordial placenta and uterus. A lot of things had to come together in the correct order for viviparity to succeed. Changes in uterine physiology would have had to occur together with changes in immune regulation. It would not have been a rapid conversion process. On the other hand, viviparity resulted in an enormous improvement in neonatal survival over oviparity, especially in the presence of hungry dinosaurs. Although all mammalian placentae have similar functions, they differ in both gross and microscopic structures. These differences play a significant role in both the development of maternal immune tolerance to what is essentially a semiallogeneic parasite and also in the way by which maternal immunity can be transferred to her offspring in the form of immunoglobulins. Depending upon the species, the placenta attaches to and then invades the uterine decidua. This is the specialized uterine mucosa [2]. Since the placenta is derived from the developing fetus and obtains half of its genetic material from the father it should be regarded as foreign by the maternal immune system and rejected in the same way that a foreign organ graft is rejected. The fact that it is not rejected is a remarkable feature of the mammalian immune and reproductive systems.

2.1.1 Monotremes When birds lay eggs, the fertilized ovum is encased within the eggshell within a few hours of fertilization so that almost all bird development occurs within the fully formed shell. In effect, the developing embryo in the form of a fertile egg Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00018-6 © 2023 Elsevier Inc. All rights reserved.

15

16

SECTION | 1 Mammalian immunology

Yolk Allantois Chorion Amnion Embryo

FIGURE 2.1 The embryonic membranes of a developing mammalian embryo. Both the yolk sac and the chorioallantois have the potential to absorb nutrients and so serve as placentas (red arrows). The yolk sac is important in monotremes and serves as temporary placenta early in marsupial, rodent and human pregnancies.

is expelled from the uterus almost immediately after it forms. Monotremes such as the platypus and echidnas do things differently. The egg is retained within the uterus for around 15
20 days in the echidna before being laid and incubated outside the body. About two-thirds of the embryologic development occurs within the egg which is very small by Eutherian standards. However, monotreme eggs still contain considerable quantities of yolk compared to marsupials and Eutherians. The developing embryo is also fed by endometrial glandular secretions. These nutrients are absorbed by the yolk sac through the permeable eggshell membrane. Thus the yolk sac functions, in effect, as a placenta. It consists of a vascularized trilaminar yolk membrane. This membrane is both porous and elastic and will stretch as the developing embryo grows. The process of feeding on maternal secretions is called matrotrophy. It has been argued that the monotreme yolk sac may be considered to be a form of placenta even though it is transient. Thus it comes into contact with the eggshell and nutrients can therefore pass from the endometrium, through the porous shell to the yolk. Yolk sacs also serve as early, temporary placentas in embryonic rodents and humans (Fig. 2.1).

2.1.2 Marsupials Marsupials possess noninvasive epitheliochorial placentas. Their placentas are however short-lived, and so marsupials give birth to relatively underdeveloped young [3]. The developing marsupial embryo is first enclosed by a zona pellucida, a mucoid coat, and a shell membrane. The shell membrane is secreted by the epithelial cells at the utero-tubal junction. It consists of a glycoprotein, CP4, that has significant homology to α-enolase and τ-crystallin [4]. The shell membrane persists until the blastocyst stage. Eventually, however, the shell membrane breaks down and the embryo “hatches” between 65% and 85% of the way through gestation. (Unlike monotremes where the shell membrane persists until after birth). This event occurs at about 18
22 days gestation in the tammar wallaby (Notamacropus eugenii), a species in which the entire gestation period is 27 days. The breakdown of the shell membrane triggers an inflammatory response at the fetal-maternal interface and this response may serve as the initiating stimulus for triggering birth [5]. The marsupial oocyte also contains considerably more yolk than that of Eutherians. This yolk is required to support the earliest stages of embryonic development. As the embryo develops, most of the marsupial yolk sac remains outside and is not absorbed into the embryonic mesoderm [6]. The external yolk sac forms a transient placenta during the early post-implantation period following the rupture of the shell membrane [7]. This is a non-invasive placenta derived from the fusion of the yolk sac and the chorion. An area of syncytium forms where the placenta comes into contact with maternal blood vessels. This placenta can absorb nutrients from both the uterine glands as well as from the maternal bloodstream [3]. The increasingly important role of nutrition through the placental route followed by the complex and prolonged lactation period reduced the need for yolk sac nutrition. As a result, in most mammals, the yolk sac placenta becomes vestigial after the first trimester except in rodents and rabbits [8]. In lagomorphs, rodents, some insectivores, and some Chiroptera the embryonic hemisphere of the yolk sac remains attached to the mesoderm. In most other Eutherians the entire yolk sac is invaded by embryonic mesoderm so that it is completely interiorized in the developing embryo.

The evolution of viviparity Chapter | 2

17

2.1.3 Eutherians Eutherian mammals have a placenta derived not from the yolk sac, but from the other two fetal membranes the chorion and allantois. This chorioallantoic placenta is fed by umbilical arteries that spread across the placenta to form a dense blood supply. In some species and individuals, the placenta may be divided into lobes. Once attached to the uterine wall, it consists of the combined endometrium and the trophectoderm.

2.1.3.1 Shapes and contact areas Mammalian placentas, while sharing a common basic structure, vary greatly in their gross morphology and the nature of the contact area with maternal tissue. Some species such as horses and pigs have a large area of contact between almost the entire chorioallantois and the entire uterine epithelium. This will usually develop folds and villi and is termed a diffuse type of placenta. Other species have a zonary placenta in which the placenta forms a band of tissue that encircles the fetus. There is an interdigitating contact zone around the chorionic sac. These zonary placentas are found in the Carnivora including dogs, cats, bears, and pinnipeds as well as in some herbivores such as elephants. Some placentas have multiple discrete areas of attachment with the endometrium and are classified as multicotyledonary. The interplacental areas in these cases are smooth and relatively avascular. The fetal components of these attachment points are called cotyledons. The endometrial contact sites are called caruncles while the complete complex is termed a placentome. The number of placentomes can range from 100 to 120 in sheep to four in deer. This type of placentation is a feature of ruminants. Finally, some placentas may consist of one or two relatively simple large circular attachment sites. These are a feature of primates, rabbits, and rodents and are called discoid or bidiscoid placentas. In humans, the placenta forms once the developing blastocyst is embedded into the endometrial lining of the uterus. The outer cell layer of the blastocyst eventually differentiates into the trophoblast, the outer cell layer of the placenta. The fetal blood vessels come close to, but do not connect with the maternal blood vessels. This enables the effective transfer of oxygen, and nutrients to the fetus while waste products such as urates and creatinine can diffuse into the maternal circulation.

2.1.3.2 Histologic classification Eutherian placentas differ in the number of cell layers that persist between the fetal and maternal circulations. This has significant immunological consequences. For example, both the chorioallantoic placenta and the uterine endometrium each have a surface structure that initially consists of three tissue layers. A layer of epithelial cells on the surface, underneath which are layers of connective tissue, and finally a layer of endothelial cells lining blood vessel walls. Depending upon which tissue layers are lost, three major types of placenta develop in mammals (Fig. 2.2). These differ in the precise relationship between the uterine wall and the placental tissues. They are classified as epitheliochorial, endotheliochorial, and hemochorial. It is debatable which of these three was the ancestral form of placenta

2.1.4 Epitheliochorial placentas Epitheliochorial placentas are the most superficial type of placenta since there is a minimal invasion of the maternal tissues and the uterine epithelium remains intact. In this case, all the cell layers are retained on both the fetal and maternal sides so that the epithelial cells of the fetal placenta are in close contact with the epithelial cells of the uterus. Apart from local angiogenesis resulting in increased vascularity there are minimal changes in the endometrium. Placental trophoblast cells may attach to and even fuse with maternal epithelial cells but there is no invasion of the uterus by the trophoblast. The fetal and maternal tissues interdigitate to maximize the area of contact between them. The maternal and fetal blood systems remain separated by two layers of epithelial cells and the connective tissue layers in addition to the vascular endothelia. This type of placenta is found in marsupials, horses, whales, and ruminants. It has also evolved separately in strepsirrhine primates (lemurs, bushbabies, and lorises). Many of these species tend to have singleton pregnancies although others such as pigs have large litters. Pigs have a diffuse epitheliochorial placenta. Pregnancy tends to be long, and newborns tend to be well developed (precocial). In the case of the ruminant placentomes, the trophoblast may fuse to a limited extent with the maternal epithelium, so it is classified as a synepitheliochorial placenta. The chorionic cells fuse with the epithelial cells to form hybrid multinucleated fetomaternal cells. There is no immunoglobulin transfer across this type of placenta and as a result, the newborn is completely dependent on receiving immunoglobulins through their mother’s colostrum immediately after birth.

18

SECTION | 1 Mammalian immunology

EPITHELIOCHORIAL

MOTHER

FETUS

Uterine epithelium

Chorionic epithelium

NO IMMUNOGLOBULIN TRANSFER

ENDOTHELIOCHORIAL

HEMOCHORIAL

LIMITED IMMUNOGLOBULIN TRANSFER

~ 100% IMMUNOGLOBULIN TRANSFER FIGURE 2.2 The three major structural types of placenta. Epitheliochorial placentas completely block maternal antibody transfer. Hemochorial placentas permit the effective transfer of IgG across the placenta to the fetus. Endotheliochorial placentas vary in their ability to permit IgG transfer depending on the species involved. In some species such as the elephant large quantities of immunoglobulins can pass through. In others, such as cats, very little immunoglobulin can pass.

2.1.5 Endotheliochorial placentas In an endotheliochorial placenta, the maternal uterine epithelium and its outer connective tissue layers are degraded after implantation so that the invasive trophoblast comes into direct contact with the uterine mesenchyme and endothelial cells of the maternal blood vessels. These invasive trophoblast cells include cytotrophoblasts and syncytiotrophoblasts. In effect, the fetal blood vessels in these cases are now in contact with the maternal blood vessels. The blood systems are only separated by a single layer of epithelial cells in addition to the blood vessel walls. This is a feature of placentas in the Carnivora such as cats and mink. Dogs have a zonary endotheliochorial placenta. This type of placentation is found in all four of the supraordinal mammalian clades Euarchoglires, Laurasiatherians, Xenarthra, and Afrotheria. Because of their close association with maternal tissues, this type of placenta must be significantly immunosuppressive to prevent a maternal immunological attack on the fetal cells [9]. Some carnivore placentas also possess placental hematomas or (hematophagous organs) [10]. In such cases, the fetal phagocytes digest any maternal blood cells that leak from the blood vessels. It is suggested that this is a mechanism by which the developing fetus can acquire iron. These hematomas are a feature of placentas in most carnivores, elephants, and some bats but not in hyaenas [11].

The evolution of viviparity Chapter | 2

19

2.1.6 Hemochorial placentas In a hemochorial placenta, the organ grows in such a way that by the end of the first trimester of pregnancy the invasive fetal trophoblast penetrates deep into the uterine wall and disrupts the uterine vascular endothelial cells and even the media of the uterine spiral arteries. As a result, there is fetal and maternal remodeling of the spiral arteries so that the placenta is bathed in oxygenated maternal blood. This remodeling involves dilatation of these arteries. This both reduces local blood pressure and maximizes the volume of blood bathing the intervillous space of the placenta [2]. The trophoblast cells erode all the maternal tissues including the uterine vascular epithelium so that the fetal chorionic epithelium is in direct contact with the maternal blood. As a result, only a single layer of epithelium, connective tissue, and fetal endothelium separate the blood systems. Hemochorial placentas vary according to the number of layers of trophoblast cells present. There may be one, two or three cellular layers separating the two bloodstreams. Guinea pigs, squirrels, and pacas as well as some primates including humans, all possess a haemomonochorial placenta. At the beginning of pregnancy humans and lagomorphs have a hemodichorial placenta. The two layers of cells are the syncytiotrophoblast cells underlaid by cytotrophoblasts that serve to replenish the trophoblast layer. As pregnancy progresses the cytotrophoblast layer thins and disappears [2]. The number of trophoblast cell layers determines the degree to which IgG can transfer from mother to fetus. Many species with this type of placentation tend to have short pregnancies with large litters and altricial neonates. Rats and mice have a hemotrichorial discoid placenta. Rabbits have a hemodichorial discoid placenta. Cynomolgus monkeys (Macaca fascicularis) have a haemomonochorial bidiscoid placenta. There is much debate regarding the structure of the ancestral Eutherian placenta. Some investigators believe that it was probably discoid and hemochorial [12]. Others argue that it was originally endotheliochorial in structure and subsequently evolved into hemochorial or epitheliochorial types [9].

2.2

Transfer of immunoglobulins

The placenta is of major immunological significance since it is involved, not only in regulating fetal tolerance but also controls the transfer of antibodies from the mother to the developing fetus. The placenta serves to transfer oxygen, nutrients, and hormones to the developing fetus. In addition, depending upon the type of placenta it may also transfer maternal immunoglobulins [13]. In some, but not all mammals, immunoglobulins may pass directly from the maternal bloodstream to that of her developing fetus. While ensuring that the newborn is protected against infection from the moment it is born, such a transfer depends upon the placental structure. Immunoglobulin transfer between the maternal and fetal circulations can only occur in species with endotheliochorial or hemochorial placentations. In effect, the placenta acts as a selective barrier preventing maternal leukocytes, proteins, and infectious agents from crossing from the mother to the fetus while in some species, permitting the selective passage of immunoglobulins. The route by which maternal antibodies reach the fetus is determined by the structure of the placenta. The human hemomonochorial placenta allows maternal IgG but not IgM, IgA, or IgE to transfer from mother to fetus. Thus maternal IgG can enter the fetal bloodstream, and as a result, the newborn human infant has circulating IgG levels comparable to those of its mother. However, since only IgG is transferred, not the other immunoglobulin classes, it is clear that the transfer must be an active process mediated through specific immunoglobulin (Fc) receptors. The transfer does not start until about 16
20 weeks of pregnancy but then progressively increases until birth. As a result, the newborn human infant may have a higher serum IgG concentration than its mother. Note that some mothers develop natural antibodies directed against foreign ABO blood groups on their fetal red cells, but these are usually of the IgM class and lack placental receptors, and so cannot cross to the fetus. The transfer of IgG across the hemochorial placenta is mediated by FcRn, the neonatal Fc receptor (Fig. 2.3). These receptors are expressed on the syncytiotrophoblast cells and bind IgG in a pH-dependent manner. The IgG is bound within the acidified endosomes in these cells and then transported to the interstitial space between the maternal and fetal circulations. FcRn binds to the CH3 domain of all IgG subclasses although there are significant differences in the efficiency of binding between these subclasses. In humans, for example, there is enhanced binding of IgG1. There are also differences in the transfer efficiencies between the allotypic variants of IgG3 since they bind FcRn with different affinities. In general, the efficiency of transfer of the human IgG classes is IgG1 . IgG3. IgG4 5 IgG2 [14]. FcRn binding to the antibody CH3 domain also appears to differ depending upon the immunoglobulin specificity. For example, measles-specific antibodies are almost 100% transferred, whereas antibodies to other viruses such as poliovirus are much less efficiently transferred to the fetus. It appears that this preferential antibody transfer is linked to the degree of

20

SECTION | 1 Mammalian immunology

FIGURE 2.3 The role of the neonatal Fc receptor, FcRn in the passive transfer of immunoglobulins from mother to her fetus in humans. The receptor binds IgG on the maternal side of the trophoblast, endocytoses it and transfers it to the fetal side. As a result of a change in pH the antibody-receptor complex dissociates and releases the IgG into the fetal circulation. A similar process is involved in the transfer of maternal antibodies into milk.

glycosylation of the heavy chain Fc domains [15]. This immunoglobulin glycosylation enhances not only FcRn binding but also antibody binding to the natural killer (NK) cell receptor FcγRIIIa. As a result, the transferred antibodies can selectively activate NK cells around the time of birth. Some human IgG transfer is also mediated through FcγRIII receptors expressed on the surface of the syncytiotrophoblast layer. Mice acquire their maternal antibodies partially through FcRn expression on yolk-sac-derived cells in addition to postnatal colostral transfer through suckling [16]. The polycotyledonary placentas of artiodactyls and perissodactyls are epitheliochorial and although differing in overall structure between species, still retain the six tissue layers separating the fetal and maternal blood systems. They do not permit the transfer of immunoglobulins. As a result, in horses and pigs, the newborn foal or piglet is completely dependent upon receiving immunoglobulins by way of maternal colostrum. The efficacy of such transfer depends upon adequate colostral intake and absorption from the gastrointestinal tract. Dogs have an endotheliochorial placenta in which the chorionic epithelium is in contact with the endothelium of the maternal capillaries. In this species, about 5%
10% of normal serum IgG levels are directly transferred from the mother to the puppy, but the rest must be obtained through colostrum. Interestingly cats have a similar type of placenta, but very few immunoglobulins are transferred to kittens in utero. It is also of interest to note that elephants also have endotheliochorial placentation but very large quantities of immunoglobulin are actively transferred transplacentally so that a newborn elephant has a higher level of IgG in its serum than does its mother [17].

2.3

Maternal-fetal tolerance

When mammals became viviparous and the developing fetus was retained inside its mother’s uterus, a significant immunological problem had to be overcome. The fetus risked being rejected like an allograft since it expresses paternal MHC molecules and its trophoblast lodges deep in the uterine wall. Pregnancy requires that two genetically different individuals, the mother, and her fetus, must coexist in incredibly close proximity. In a normal pregnancy, the fetus establishes and maintains itself despite this MHC incompatibility. The uterus is not a privileged site since grafts from other tissues, such as the skin, implanted in the uterine wall are readily rejected. Likewise, a mother may make antibodies against fetal blood group antigens, and these antibodies can destroy fetal red blood cells either in utero, as in primates, or following ingestion of colostrum, as occurs in other mammals. Fetal tolerance is restricted to the uterus. Embryonic allografts placed elsewhere in the body are rejected. For a successful pregnancy to occur the mother must be prevented from mounting a lethal immune response against the paternal antigens expressed by her fetus and its placenta. As placentas evolved and progressively became more invasive, the importance of this maternal-fetal tolerance became more significant. Control of the rejection process must be

The evolution of viviparity Chapter | 2

21

achieved if the benefits of the closest possible proximation of the fetal and maternal blood systems in mammals were to be obtained. Suppression of the maternal graft rejection process primarily occurs within the decidua, that part of the uterine wall that is in contact with or even actively invaded by placental trophoblast cells. The immune interactions within the decidua are not exclusively suppressive. For example, it is essential that the uterine vasculature changes accommodate the increased blood flow required to support the developing fetus. This requires remodeling of the normal uterine arteries to form flaccid large spiral arteries so allowing increased blood flow and permitting the exchange of nutrients and blood gases. The depth of decidual invasion is the subject of much cross-talk between trophoblast cells and immune cells. As a result, among the key selective pressures on the evolution of the placenta, especially an invasive hemochorial one, are those related to the selection, and expression of specific immune receptor genes. Maternal-fetal tolerance is so powerful that human surrogate mothers carrying a fully allogeneic fetus suffer a no higher rate of fetal rejection than do women with normal, semi-allogeneic fetuses. Likewise in animal agriculture, it is possible to produce interspecies hybrids by breeding a horse with a donkey or a cow with a yak. Even totally artificial pregnancies that involve, for example, placing a zebra fetus into an equine mare results in a “normal” pregnancy. Embryo transfer is an important part of modern cattle breeding.

2.4

Mechanisms of tolerance

The immunological destruction of the fetus and its trophoblast is prevented by the combined activities of diverse immunoregulatory pathways acting at the maternal-fetal interface. Fetal tolerance during pregnancy appears to take two slightly different forms. In early pregnancy, it is mediated by controlling the interactions between the extravillous trophoblast and maternal immune cells within the decidua. The second phase occurs later in pregnancy as the placenta and its circulation develop. This involves the control of immune interactions involving the villous syncytiotrophoblast (Fig. 2.4).

2.4.1 Anatomical adaptations The structure of the placenta itself may minimize the opportunities for immunologic attack. For example, in humans, the cells of the trophoblast fuse together to form an extensive syncytium that leaves no spaces between the cells for lymphocytes to penetrate. The protein syncytin-1 is a cell membrane protein produced by the placenta of humans and some primates that mediates this fusion. The gene encoding syncytin-1 is located within an endogenous retroviral element. It is therefore believed that this gene encoded the envelope protein of the retrovirus. It appears to have been integrated into the primate germline more than 26 mya. Syncytin-1-mediated trophoblast fusion and the formation of this placental barrier are required for normal human placental development [18]. The initial steps in pregnancy involve implantation of the fertilized ovum in the uterine wall and its acceptance by the maternal immune system. This implantation is regulated by hormonal signaling between the conceptus and the IMMUNOSUPPRESSIVE MOLECULES

LOCAL INHIBITION OF

Inflammation Class I and II MHC expression Complement activation Th1 cell-mediated responses

IDO Glycoproteins Blocking antibodies Uterine phospholipids

MATERNAL-FETAL TOLERANCE

REGULATORY CELLS uNK cells Treg cells MSDCs

FIGURE 2.4 The immunosuppressive mechanisms and cells involved in generating maternal-fetal tolerance. These include localized immunosuppressive functions. The presence of regulatory cells, most notably uNK cells and T reg cells, as well as locally immunosuppressive molecules such as IDO (indolamine 2,3 dioxygenase). MDSC; myeloid-derived suppressor cells.

22

SECTION | 1 Mammalian immunology

uterus. This involves hormones such as progesterone and interferons (IFNs), especially IFN-τ and IFN-γ. Blastocyst implantation is also associated with localized upregulation of inflammatory genes and the production of inflammatory cytokines. Thus implantation has a carefully regulated inflammatory component. This inflammation attracts leukocytes to the uterus, especially uterine and decidual NK cells. These cells promote local immunotolerance by secreting the suppressive cytokines, IL-10, TGF-β, and IL-1 receptor antagonist. These cause local suppression of proinflammatory Th1 and Th17 cells. They also promote the differentiation of regulatory T cells (Treg cells) that serve to establish and maintain a state of tolerance.

2.4.2 Localized immunosuppression Allografts are rejected when the recipient’s T cells recognize and attack the foreign polymorphic histocompatibility (MHC) antigens expressed on the grafted cells. This doesn’t happen in the placenta since no polymorphic (classical) class I MHC molecules (HLA-A or HLA-B in humans) are expressed on preimplantation embryos or oocytes. Likewise, there are no polymorphic MHC class Ia or class II molecules expressed on the trophoblast cells that come into contact with maternal tissues. The absence of these classical MHC molecules permits the trophoblast cells to avoid destruction by cytotoxic T cells. On the other hand, NK cells normally attack and kill cells that fail to express MHC molecules. However, human trophoblast cells express two invariant MHC class Ib molecules, HLA-E and HLA-G (Chapter 7). These inhibit NK cell cytotoxic responses. HLA-E is the ligand of an inhibitory receptor on uterine NK (uNK) cells called CD94/NKG2A. HLA-G is the ligand of the inhibitory leukocyte immunoglobulin-like receptors also found on uNK cells. HLA-G also binds to an endosomal KIR, CD158d/KIR2DL4 that activates senescence-inducing pathways on uNK cells and causes them to age prematurely [2] (Fig. 2.5). Unlike the other HLA molecules, the polymorphic class I molecule HLA-C is expressed on human extravillous trophoblast cells. It may be derived from either parent. As a result, it can differ from pregnancy to pregnancy and even between the same parents. Thus HLA-C binds to inhibitory killer cell immunoglobulin-like receptors (KIRs) on uNK cells. There are two KIR haplotypes, A and B in humans. Binding to KIR-A is inhibitory while binding to KIR-B is activating. (Mothers may be homozygous KIR-AA or BB or heterozygous AB. BB is best for a successful pregnancy). Likewise, different allotypes of HLA-C can affect NK cell activation or inhibition. Some combinations of maternal KIRs and fetal HLA-C can also activate the uNK cells that in turn may attract extravillous trophoblast cells so promoting deep placental invasion and facilitating fetal growth. Other combinations of HLA-C and KIRs may be inhibitory and impede trophoblast invasion. It is of interest to note that HLA-C is expressed in humans, chimpanzees, and gorillas. These are species in which the trophoblast characteristically invades the endometrium very deeply. On the other hand, in Old-world primates such as gibbons that do not express HLA-C, the trophoblast is much more superficial [19,20]. HLA-C tetramers bind uNK cells more effectively than they bind peripheral blood NK cells [21,22]. In cattle, classical MHC class I molecules are not expressed on the trophoblast before day 120 gestation. Equine trophoblast cells also appear to have reduced MHC class I expression. In the pig epitheliochorial placenta, activated T and NK cells have been detected [23].

Enhanced Trophoblast invasion

Increased activating signals IFNs CXCL10 IL-8

Suppressed Th17 responses

Uterine NK CELLS

Arterial remodeling

Increased inhibitory signals KIR NKG2C LILR

Suppressed cytotoxicity

FIGURE 2.5 The role of uterine NK cells is establishing and maintaining placentation. Thus while they are less cytotoxic than the circulating NK cell populations, they can contribute both to trophoblast invasion as well as spiral artery remodeling.

The evolution of viviparity Chapter | 2

23

2.4.3 Inhibition of complement activation Rodent placentas express CR1-related gene/ protein Y (Crry), a membrane-bound protein that blocks complement deposition on cell surfaces and so protects them against complement-mediated destruction [24]. In genetically engineered mice that lack Crry, their embryos die at B7.5 days as a result of placental destruction mediated by complement [25].

2.4.4 Suppression of adaptive immunity In many mammals, pregnancy is also associated with a strong skewing of the mother’s immune system in favor of Th2 (antibody-mediated) responses and a reduction in Th1 (cell-mediated) responses. This raises the interesting concept that infections that promote a strong Th1 response might reverse this skewing, compromise pregnancy, and lead to abortion. This would certainly apply to protozoan infections such as toxoplasmosis and Neospora caninum infection, as well as brucellosis. Cytokines that usually enhance MHC expression, such as IFN-γ, have no effect on trophoblast cells. The placenta can also defend itself. Syntrophoblast cells secrete exosomes that express TRAIL (TNF-related apoptosis-inducing ligand) and Fas-L on their surface. If attacked by cytotoxic T cells, TRAIL and Fas-L can bind to T cell death receptors and so trigger their apoptosis [26].

2.4.4.1 Macrophages In early human pregnancy, macrophages account for 10%
20% of the leukocyte population in the decidua. At this stage, they act as the major antigen-presenting cells. These cells have an M2 phenotype and so express high levels of IL-10, indolamine 2,3 dioxygenase (IDO), and low levels of the stimulatory receptors CD80 and CD86. As a result, they are functionally immunosuppressive. This phenotype is probably determined by cytokines such as IL-10 and M-CSF produced by nearby trophoblast cells. These macrophages also produce VEGF and the chemokine, MMP9 that promote angiogenesis and tissue remodeling. They probably also play an important role in spiral artery remodeling and trophoblast invasion by removing trophoblastic cell debris and so minimizing inflammation. (This is a role these cells also play in wound healing). In addition to all this, these macrophages also retain their primary role of destroying any invading pathogens they may encounter [27].

2.4.4.2 Th17 cells Th17 cells are CD41 T cells that promote inflammation by secreting the cytokines IL-17 and IL-22 [28]. Their differentiation is promoted by the inflammatory cytokine, TNF-α. In humans, Th17 cells appear to be largely absent from the placenta during the first trimester. However, they gradually begin to accumulate in the decidua. uNK cells can promote immune tolerance by suppressing Th17 cell activities by producing the counteracting cytokine IFN-γ.

2.4.4.3 Myeloid-derived suppressor cells Myeloid-derived suppressor cells (MDSCs) are immature bone marrow cells that normally develop into macrophages, granulocytes, and dendritic cells (DCs). MDSCs suppress cytotoxic T cell responses by expressing the inhibitory surface protein, programmed death ligand-1 (PD-L1), and by secreting immunosuppressive mediators such as arginase, IL-10, TGF-β, reactive oxygen species, nitric oxide, and peroxynitrite. Peroxynitrite causes nitrate addition to T cell antigen receptors and thus inactivates them. MDSCs also produce arginase that impairs T cell function by reducing the expression of CD3ζ, a component of the T cell antigen receptor signaling complex. Vascular endothelial cell growth factor (VEGF) promotes MDSC production by blocking DC maturation. IL-1β also promotes MDSC production. Some MDSCs promote the production of Treg cells. MDSCs may also trigger the phenotypic switch from pro-inflammatory M1 to antiinflammatory M2 macrophages. Collectively these cells effectively suppress rejection processes in the placenta. MDSCs accumulate in pregnant mice and women and are important contributors to the establishment of fetal tolerance [29].

2.5

Regulatory cells

2.5.1 Natural killer cells The most important of the regulatory processes that support fetal tolerance are those mediated by uterine natural killer (uNK) cells. In pregnant women, there are two major populations of NK cells. One population circulates in the peripheral bloodstream and plays a defensive role. The other population resides in the uterine decidua. uNK cells are distinctly different from the NK cells found in the circulation and the rest of the body. While both populations have similar

24

SECTION | 1 Mammalian immunology

properties, the circulating NK cells are primarily cytotoxic while the uterine NK cells produce both cytokines and chemokines and are poorly cytotoxic. These uNK cells mediate the vascular adaptations in the uterus needed for a successful pregnancy (Fig. 2.5). Uterine NK cells are the main maternal immune cells present in the endometrium during the establishment of the placenta. This is especially the case in species with an invasive hemochorial placenta such as humans [21]. The placental decidua contains huge populations of uNK cells and macrophages, as well as the key effector cells of the adaptive immune system, namely DCs, and T cells (especially Treg cells). In the first trimester of human pregnancy, about 70% of these immune cells are uNK cells, and about 20%
25% are macrophages [2]. T cells constitute about 3%
10% of this cell population while DCs, B cells, and mast cells are only present in low numbers. Presumably, these leukocytes are attracted to the decidua by chemokine gradients generated by stromal and trophoblast cells. This population of uNK cells undergoes a major expansion during early pregnancy when they invade the embryonic implantation site [30]. They occupy the decidua basalis and form the mesometrial lymphoid aggregate of pregnancy. These uNK cells proliferate until mid-gestation. They subsequently decline and are gradually replaced by T cells that predominate at term. uNK cells have also been identified and described in rodents, bats, pigs, and horses. uNK cells produce a diverse mixture of growth factors, angiogenic factors, and cytokines. Subpopulations of uNK cells have been identified in the decidua based on their surface antigen expression and gene expression profiles. For example, in humans, the majority of uNK cells are CD56 hi, CD16. On the other hand, peripheral blood NK cells are CD56lo and CD161. The most critical of their cell surface antigens are the NK cell MHC receptors. As described above, uNK cells possess two major inhibitory receptors, NK group 2 A (NKG2A) and leukocyte immunoglobulin-like receptor B1 (LILRB1) that bind to HLA-E and HLA-G respectively and inhibit NK cytotoxicity. However, they have many other receptors and ligands that are unique in their extreme diversity. Different combinations may have different effects on trophoblast invasion and the development of a successful pregnancy [21]. There is, in effect, a balance between MHC class I expression on the trophoblast and uNK cells in the uterine wall that together regulates trophoblast growth and invasion. The binding of the MHC ligands to their receptors on uNK cells suppresses their cytotoxicity and also triggers epigenetic changes in their DNA, especially during a first pregnancy. This subgroup of uNK cells, called pregnancy-trained uterine NK cells are present in the decidua after repeated pregnancies. They have a unique transcriptome and epigenetic signature [31]. Their activation leads to increased production of IFN-γ and vascular endothelial growth factor (VEGF)-α. In effect, this results in a form of “trained” immunity that plays a role in facilitating subsequent pregnancies [22]. As the fetus grows within the uterus, the placenta and the uterine vasculature must grow with it. uNK cells secrete chemokines and cytokines that can activate fetal trophoblast cells. As a result, the NK cells and trophoblast cells act together to remodel the uterine spiral arteries. This ensures that the flow of oxygenated blood reaching the placenta increases sufficiently to permit uninterrupted fetal growth. If this spiral artery remodeling is insufficient and the fetus fails to get enough blood, then this may result in fetal growth restriction resulting in pre-eclampsia or miscarriage [25]. In mice, the dilation of the spiral arteries appears to be mediated by uNK-derived IL-8, type 2 IFNs, and by the chemokine CXCL10 [32].

2.5.2 Regulatory T cells In the first trimester of pregnancy in humans, there is an increase in the number of decidual T cells. These T cells in the human decidua primarily use α/β 1 antigen receptors with relatively few γ/δ1 T cells. CD81 cells account for 45%
75% of the decidual T cells while CD41 are about 30%
45%. There is also an increase in placental CD41 cells in the late term associated with parturition. Th1 cells constitute about 30%, Th2 5% and Th17 cells 2%
5% of human decidual T cells. In humans and mice CD41, CD25hi, and FoxP31 Tregs account for about 5%
20% of these decidual T cells. Placental T cells secrete antiviral molecules including IFN-γ and some primate-specific miRNAs that restrict viral infections in a paracrine and autocrine manner. CD41 FoxP31 Treg cells play a critical role in preventing fetal rejection. Estrogen treatment and pregnancy both induce FoxP3 expression, as does the presence of a seminal plasma [33]. In the first trimester of pregnancy, there is a progressive increase in the number of decidual Treg cells. In humans, substantial numbers of maternal T cells cross the placenta to reside in fetal lymph nodes. These induced Treg cells suppress maternal responses to paternal antigens probably through IL-10 and TGF-β production while at the same time suppressing the production of Th1, Th2, and Th17 cells. Treg numbers steadily increase during the second and third trimesters in humans before declining late in gestation. This decline in Treg numbers may assist in promoting parturition. These fetal-specific Treg cells can persist beyond parturition. In subsequent pregnancies, their numbers increase rapidly [2].

The evolution of viviparity Chapter | 2

25

In mice, Tregs are induced around the time of conception by cytokines found in semen [33]. Their numbers expand as pregnancy proceeds. In humans, these decidual Tregs may serve to suppress cytotoxic T cell responses against paternal alloantigens. They probably also regulate the functions of the other immune cells, uNKs, macrophages, and DCs. They may also participate in tissue remodeling. Although all these cells and mechanisms minimize maternal sensitization by allogeneic fetal cells, cytotoxic T cells or antibodies do develop during pregnancy. We know that pregnant females may mount an immune response to their fetuses. For example, in pregnant mares, placental cells invade the uterine wall to form structures called endometrial cups [28]. This is mediated by the T cell cytokine, IL-22. These placental cells in turn stimulate a strong immune response against paternal MHC antigens around day 60 of gestation [34]. As a result, the cups become surrounded by large numbers of CD41 and CD81 T cells, macrophages, and plasma cells. Presumably, these cells attack, and this eventually leads to degeneration of the endometrial cups around 120 days of pregnancy. Despite this immune attack, however, the pregnancy is unaffected. These attacking cells may mount Th2 and Treg responses dominated by IL-10 production that does not threaten the pregnancy rather than cytotoxic Th1 and inflammatory Th17 responses that may lead to fetal rejection.

2.6

Other immunosuppressive mechanisms

Indoleamine dioxygenase (IDO) is a heme-containing enzyme found only in mammals. It is expressed in large amounts in the placenta. It catalyzes the oxidation of the essential amino acid tryptophan into L-kynurenine. IDO expression is under the control of IFNγ, TNF-α and the chorionic gonadotrophins [35]. The IDO substrate tryptophan is an essential amino acid. As a result of its degradation, IDO can eliminate pathogens by nutrient depletion. However, tryptophan is also essential for normal immune function so IDO also plays an immunosuppressive role and can induce fetal tolerance. T cells are especially susceptible to tryptophan depletion. They become locked in their G1 phase and cannot proliferate. IDO blocks Th1 and Th17 responses and promotes apoptosis of cytotoxic T cells. Thus it causes a marked skewing away from a type 1 towards a type 2 immune response. Inhibitors of IDO permit maternal rejection of allogeneic fetuses in mice. Treg cells upregulate IDO expression in DCs. In addition, IDO induces trophoblast HLA-G expression, suggesting that these molecules interact to maintain pregnancy. IDO homologs are found in diverse mammalian species and a paralog of IDO termed IDO2 has been identified in humans and mice. IDO and IDO2 are encoded by similar genes, and these are located adjacent to each other on chromosome 8 in both species. IDO probably developed as a result of the duplication of the IDO2 gene. The IDO gene first appeared in pre-vertebrates at least 500 mya. This proto-IDO appears to have acted as an oxygen carrier while the ability to degrade tryptophan is associated with its appearance in mammals. IDO is expressed in mouse syncytiotrophoblast cells. It blocks T cell proliferation and interferes with the recognition of paternal MHC molecules by maternal T cells. In pregnant humans, IDO is expressed in the vascular endothelium of the decidua and the villous chorion but not in the trophoblast. Its highest concentration is found at the fetomaternal interface. It is likely therefore that it only exerts its influence within this very restricted zone. IDO may also be involved in the initial invasion of the endometrium by trophoblastic cells [36].

2.6.1 Glycoproteins The Eutherian Fetoembryonic Defense System (Eu-FEDS) hypothesis suggests that soluble and cell-surface associated glycoproteins expressed on gametes also suppress immune responses and so inhibit fetal rejection. The theory suggests that some specific oligosaccharides are linked to these glycoproteins. The major glycoproteins implicated include alpha-fetoprotein, cancer antigen 125 (CA125, a large mucinous glycoprotein), and glycodelin-A [37]. Galectins are known to regulate immune responses by binding to cell surface carbohydrates (glycans). One subfamily of these galectins is specifically expressed in the placenta of primates, especially in the syncytiotrophoblast. These specific galectins induce apoptosis in T cells. Thus they may also serve as local immunosuppressive agents [38].

2.6.2 Cytokines The fetus does not depend entirely on maternal mechanisms for its protection. The placenta is a source of many immunosuppressive molecules, including estradiol and progesterone, and possibly also chorionic gonadotropin. In addition, some pregnancy-associated glycoproteins, including α2-macroglobulin, α-fetoprotein, the major protein in fetal serum, and placental IFNs, have immunosuppressive properties. In mammals, unique IFNs (IFN-ω in humans, horses, and

26

SECTION | 1 Mammalian immunology

dogs; IFN-τ in ruminants; and IFN-γ and IFN-δ in pigs) produced by the embryonic trophoblast act as signaling proteins between the embryo and mother during early development. These IFNs also inhibit lymphocyte activities. Amniotic fluid is rich in immunosuppressive phospholipids.

2.6.3 Blocking antibodies Up to 90% of pregnant mares make antibodies to foal MHC class I molecules [34]. Similar antibodies develop in multiparous sheep and cattle. In some mouse strains, up to 95% of pregnant animals make antibodies against fetal MHC molecules. Up to 40% of women make antibodies to fetal MHC molecules after giving birth. The presence of these antibodies has no adverse effect on the course of the pregnancy. On the contrary, the maternal immune response may stimulate placental function. In mice, hybrid placentas are larger than the placentas of inbred animals, and females tolerant to paternal antigens have smaller placentas than intolerant females. Other studies show that mothers sensitized to paternal MHC molecules have better fetal survival. This may be due to the stimulatory effects of IL-3 and granulocytemacrophage colony-stimulating factor from maternal T cells on trophoblast growth. It is also relevant to note that in cattle there is a clear association between retention of the placenta after calving and its MHC class I haplotype. MHC class I compatibility between a mother and her calf increases the risk of a retained placenta, whereas MHC class II compatibility has no effect. It has been suggested that the expulsion of the placenta after birth may be due, at least in part, to an allograft response [39]. Some antibodies made by the mother against fetal antigens may coat placental cells, preventing their destruction by maternal T cells. These blocking antibodies can be eluted from the placenta and shown to suppress other cell-mediated immune reactions against paternal antigens, such as graft rejection. The absence of these blocking antibodies accounts for some cases of recurrent abortion in women [40]. Nevertheless, it has also been shown that immunodeficient mice can have successful pregnancies. Despite the previous discussion, if the antigenic differences between the mother and her fetus are too great, then pregnancy may not go to completion. Studies on xenogeneic hybridization of two different mouse species show that their embryos develop until mid-gestation and are then attacked and destroyed by maternal lymphocytes. Similarly, donkey embryos transferred to horse mares are destroyed by large numbers of maternal lymphocytes. Mild immunosuppression is a consistent feature of late pregnancy and the early postpartum period. Pregnant animals may have minor deficiencies in cell-mediated immune reactivity to nonfetal antigens. Treg cells can cause systemic immunosuppression in pregnant females. Thus in pregnant women, autoimmune diseases may undergo remission. Similarly, their antibody responses to some vaccines such as influenza may be reduced. Dairy cows experience a periparturient depression in neutrophil function and reduced T cell cytotoxicity and cytokine production [41]. This suppression appears to be due to multiple causes including the stress of parturition, the production of glucocorticoids, the loss of immunoglobulins into colostrum, and a negative energy balance. In mares, lymphocyte responses to mitogens drop from four weeks before, to five weeks after parturition. NK cell activity in pigs drops at the end of gestation to reach a low point two to three weeks after parturition. Ewes in late pregnancy may show a reduction in the levels of some immunoglobulins such as IgG1. This may be due to alterations in helper T cell function or, more plausibly, simply to diversion of the IgG1 into the mammary gland to produce colostrum. This suppression may be clinically significant in parasitized animals, in which the immune response barely controls the parasite. Parturition can be considered to be mediated in part by a sterile inflammatory response following the invasion of the uterus by large numbers of leukocytes. Granulocytes and macrophages invade the fetal-maternal interface in late pregnancy. These inflammatory responses result in degradation of the fetal membranes and so may contribute to the induction of labor [42]. These cells secrete a mixture of cytokines, proteases (especially collagenases), prostanoids, and chemokines. The cytokines stimulate uterine stromal cells to amplify the process; and the collagenases remodel collagen, weakening fetal membranes, softening the cervix, and increasing the contractility of the myometrium. All this leads to fetal expulsion. After delivery, the uterus involutes as most of the leukocytes leave or are destroyed.

2.6.4 Microchimerism From early in pregnancy, fetal cells may be detected in maternal blood. Their numbers increase progressively during gestation. Conversely, maternal cells can be detected in human fetal tissues beginning in the second trimester. These fetal cells may persist in mothers for many years while the maternal cells may persist in their children until adulthood [43]. It is unclear what the functions of these microchimeric cells are, but it appears that those present in the mother may promote the success of future pregnancies.

The evolution of viviparity Chapter | 2

27

The transfer of fetal cells to the mother appears to be a common feature of mammalian pregnancy. Thus in rhesus macaques (Macacca mullata), fetal cells can be found among their blood mononuclear cells as well as in organs such as the heart, liver, and spleen. Likewise, fetal cells are found in many tissues in pregnant mice [43]. The progressive accumulation of these fetal cells correlates with a rise in the number of CD41 Treg cells. The numbers of Treg cells double during pregnancy in mice. Allogeneic pregnancies promote a greater expansion of Treg cells than do syngeneic pregnancies in mice. These Treg cells appear to be largely specific for fetal antigens. Maternal cells also accumulate within the developing fetus during pregnancy. As in the mothers, these foreign cells are not attacked by cytotoxic T cells. They appear to stimulate the fetus to produce maternal antigen-specific Treg cells. This tolerance to the microchimeric cells persists after pregnancy. The fetal cells may persist in mothers for decades after birth and in their offspring throughout postnatal development. Breastfeeding appears to be important in maintaining this tolerance. This is probably mediated by further acquisition of maternal cells by the newborn during suckling (Chapter 3). One apparent benefit is improved outcomes of subsequent pregnancies resulting from breeding with the same sire. Thus the incidence of pre-eclampsia is reduced. Microchimerism may promote tolerance in mothers to promote the survival of genetically similar siblings [43].

2.6.5 Adaptive immunity Newborn mammals, mount adaptive responses skewed toward antibody-mediated type 2 rather than cell-mediated type 1 responses. Thus they favor antibody responses over cell-mediated immunity. This imbalance may result from residual immunosuppression persisting after pregnancy. Mononuclear cells from newborn foals are unable to express IFN-γ. Their Th2 cells rapidly differentiate, whereas neonatal Th1 cells are slow to develop. Excess IFN-γ may cause placental damage, so this skewing is not accidental. IFN-γ production gradually increases through the first six months of a foal’s life to reach adult levels within a year when the acquired responses revert to the balanced adult pattern. Unless additional immunological assistance is provided, however, organisms that present little threat to an adult may kill newborn mammals. This immunological assistance is provided by antibodies transferred from the mother to her offspring across the placenta or through the colostrum. Maternal lymphocytes may also be transferred to the fetus through the placenta or to newborn mammals through the colostrum.

2.6.6 Sperm Allogeneic sperm can successfully and repeatedly penetrate the female reproductive tract without provoking graft rejection. One reason for this is that seminal plasma is immunosuppressive [33]. Sperm exposed to this fluid are nonimmunogenic, even after washing. Prostatic fluid, one of the immunosuppressive components of seminal plasma, also inhibits complement-mediated hemolysis. Seminal plasma promotes the growth of Treg cells that subsequently migrate to the endometrium and promotes tolerance to paternal alloantigens. Not only does seminal plasma cause the recruitment of monocytes to the vaginal wall, but it also modulates the development of monocyte-derived DCs and directs them to differentiate towards a regulatory subset. They may promote fertility by inducing tolerance to paternal alloantigens, but they may also increase susceptibility to infectious agents. Nevertheless, occasional cases of infertility resulting from the production of anti-sperm antibodies in the uterus and vagina do occur.

References [1] [2] [3] [4] [5] [6] [7] [8] [9]

Brawand D, Walhi W, Kaessmann H. Loss of egg yolk genes in mammals and the origin of lactation and placentation. PLoS Biol 2008;6(3):e63. Ander SE, Diamond MS, Coyne CB. Immune responses at the maternal-fetal interface. Sci Immunol 2019;4:eaat6114. Roberts RM, Green JA, Schultz LC. The evolution of the placenta. Reproduction 2016;152(5):R179
89. Menkhorst E, Nation A, Cui S, Selwood L. Evolution of the shell coat and yolk in Amniotes: a marsupial perspective. J Exp Zool (Mol, Dev Evol,) 2009;312B:625
38. Griffith OW, Chavan AR, Protopapas S, Maziarz J, et al. Embryo implantation evolved from an ancestral inflammatory attachment reaction. Proc Natl Acad Sci USA 2017;. Available from: https://doi.org/10.1073/pnas.1701129114. Freyer C, Renfree MB. The mammalian yolk sac placenta. J Exp Zool 2009;312B:545
54. Guernsey MW, Chuong EB, Cornelis GC, Renfree MB, Baker JC. Molecular conservation of marsupial and Eutherian placentation and lactation. eLife 2017;. Available from: https://doi.org/10.7554/eLife.27450.001. Furukawa S, Kuroda Y, Sugiyama A. A comparison of the histologic structure of the placenta in experimental animals. J Toxicol Pathol 2014;27:11
18. Carter AM, Enders AC. The evolution of epitheliochorial placentation. Annu Rev Anim Biosci 2013;l(1):443
67.

28

SECTION | 1 Mammalian immunology

[10] Sarli G, Castagnetti C, Blanco C, Ballotta G, et al. Canine placenta histological findings and microvascular density: the histological basis of a negative neonatal outcome. Animals 2021. Available from https://doi.org/10.3390/ani11051418. [11] Enders AC, Carter AM. What can comparative studies of the placenta tell us?
A review. Placenta 2004;25:1
7. [12] Wildman DE, Chen C, Erez O, Grossman LI, et al. Evolution of the mammalian placenta revealed by phylogenetic analysis. Proc Natl Acad Sci USA 2006;103(9):3203
8. [13] Chucri TM, Monteiro JM, Lima AR, Salvadori MLB, et al. A review of immune transfer by the placenta. J Repro Immunol 2010;87:14
20. [14] Clements T, Rice TF, Vamvakas G, Barnett S, et al. Update on transplacental transfer of IgG subclasses: impact of maternal and fetal factors. Front Immunol 2020. Available from: https://doi.org/10.3389/fimmu.2020.01920. [15] Jennewine MF, Goldfarb I, Dolatshahi S, Cosgrove C, et al. Fc glycan-mediated regulation of placental antibody transfer. Cell 2019;178:202
15. [16] Kim J, Mohanty S, Ganeson LP, Hua K, et al. FcRn in the yolk sac endoderm of mouse is required for IgG transport to the fetus. J Immunol 2009;182(5):2583
9. [17] Nofs SA, Atmar RL, Keitel WA, Hanlon C, et al. Prenatal passive transfer of maternal immunity in Asian elephants (Elephas maximus). Vet Immunol Immunopathol 2013;153:308
11. [18] Mi S, Lee X, Li X-P, Veldman GM, et al. Syncytin is a captive retroviral envelope protein involved in human placental morphogenesis. Nature 2000;403:785
9. [19] Carter AM. Comparative studies of placentation and immunology of non-human primates suggest a scenario for the evolution of deep trophoblast invasion and an explanation for human pregnancy disorders. Reproduction 2011;141:391
6. [20] Carter AM, Enders AC. Comparative aspects of trophoblast development and placentation. Repro Biol Endocrinol 2004;. Available from: https://doi.org/10.1186/1477-7827-2-46. [21] Colucci F, Boulenouar S, Kleckbusch J, Moffett A. How does variability of immune system genes affect placentation? Placenta 2011;32 (8):539
45. [22] Colucci F. The immunological code of pregnancy. Science 2019;365:862
3. [23] Stas MR, Koch M, Stadler M, Sawyer S, et al. NK and T cell differentiation at the maternal-fetal interface in sows during late gestation. Front Immunol 2020;11. Available from: https://doi.org/10.3389/fimmu.2020.582065. [24] Wu X, Spitzer D, Mao D, Peng SL, et al. Membrane protein Crry maintains homeostasis of the complement system. J Immunol 2008;181 (4):2732
40. [25] Xu C, Mao D, Holers VM, et al. A critical role for murine complement regulator Crry in fetomaternal tolerance. Science 2000;287:498
502. [26] Skopets B, Li J, Thatcher WW, et al. Inhibition of lymphocyte proliferation by bovine trophoblast protein-1 (type I trophoblast interferon) and bovine interferon-αI1. Vet Immunol Immunopathol 1992;34:81
96. [27] Hsu P, Nanan RK. Innate and adaptive immune interactions at the fetal-maternal interface in healthy human pregnancy and pre-eclampsia. Front Immunol 2014. Available from: https://doi.org/10.3389/fimmu.2014.00125. [28] Brosnahan MM, Miller DC, Adams M, Antczak DF. IL-22 is expressed by the invasive trophoblast of the equine (Equus caballus) chorionic girdle. J Immunol 2012;188:4181
7. [29] Ostrand-Rosenberg S, Sinha P, Figley C, Long R, et al. Myeloid-derived suppressor cells (MSDCs) facilitate maternal-fetal tolerance in mice. J Leuko Biol 2017;101:1091-101. [30] Matson BC, Caron KM. Uterine natural killer cells as modulators of the maternal-fetal vasculature. Int J Dev Biol 2014;58:199
204. [31] Valencia-Ortega J, Saucedo R, Pena-Cano MI, Hernandez-Valencia M, Cruz-Duran G. Immune tolerance at the maternal-placental interface in healthy pregnancy and pre-eclampsia. J Obstet Gynaecol Res 2020;46:1067
76. [32] Wallace AE, Whitley GS, Thilaganathan B, Cartwright JE. Decidual natural killer cell receptor expression is altered in pregnancies with impaired vascular remodeling and a higher risk of pre-eclampsia. J Leukoc Biol 2015;97:79
86. [33] Lenicov FR, Rodrigues CR, Sabatte J, Cabrini M, et al. Semen promotes the differentiation of tolerogenic dendritic cells. J Immunol 2012;189:4777
86. [34] Lunn P, Vagnoni KE, Ginther OJ. The equine immune response to endometrial cups. J Reprod Immunol 1997;34:203
16. [35] Munn DH, Zhou M, Attwood JT, et al. Prevention of allogeneic fetal rejection by tryptophan catabolism. Science 1998;281:1191
3. [36] Durr S, Kindler V. Implication of indolamine 2,3 dioxygenase in the tolerance towards fetuses, tumors, and allografts. J Leuko Biol 2013;93:681
7. [37] Clark GF. The role of glycans in immune evasion: the human fetoembryonic degense system hypothesis revisited. Mol Hum Repro 2014;20:185
99. [38] Than NG, Romero R, Goodman M, et al. A primate subfamily of galectins expressed at the maternal-fetal interface that promote immune cell death. Proc Natl Acad Sci U S A 2009;106:9731
6. [39] Joosten I, Sanders MF, Hensen EJ. Involvement of major histocompatibility complex class I compatibility between dam and calf in the aetiology of bovine retained placenta. Anim Gen 1991;22:455
63. [40] Hwang J-L, Ho H-N, Yang Y-S, Hsieh C-Y, et al. The role of blocking factors and antipaternal lymphocytotoxic antibodies in the success of pregnancy in patients with recurrent spontaneous abortion. Gyne Endocrinol 1992;58(4):691
6. [41] Aleri JW, Hine BC, Pyman MF, Mansell PD, et al. Periparturient immunosuppression and strategies to improve dairy cow health during the periparturient period. Res Vet Sci 2016;108:8
17. [42] Gomez-Lopez N, Guilbert LJ, Olson DM. Invasion of the leukocytes into the fetal-maternal interface during pregnancy. J Leuko Biol 2010;. Available from: https://doi.org/10.1089/jlb.12098796. [43] Kinder JM, Stelzer IA, Arck PC, Way SS. Immunological implications of pregnancy-induced microchimerism. Nat Rev Immunol 2017;17(8):483
94.

Chapter 3

The evolution and role of lactation When he first devised his system of taxonomy in 1758 Carolus Linnaeus chose to group those animals whose females produced milk into a single class that he called the Mammalia. As a result, he paired the land mammals with the whales that, until then, had been considered to be fish. Linnaeus recognized that lactation is one of the defining features of mammals,—animals whose females feed their young milk through mammary glands,—mammae. In addition to providing nourishment for the newborn, milk also serves to protect against infections. When the newborn mammal leaves the sterile uterus and enters the microbiologically rich external environment, it requires temporary protection to ensure that it can resist this initial microbial invasion until a stable commensal population is established. This protection complements that which may be transferred (or not transferred) across the placenta. Lactation provides mammals with the ability to feed their newborns in any environment where the adults can survive and thrive. In effect, viviparity and lactation are inextricably linked. Both provide for the nutritional needs of developing offspring. Placentation for the first part of the growth period and lactation for the second (Fig. 3.1). Lactation may also persist for years in very large mammals such as humans, great apes, elephants, or toothed whales. On the other hand, it may be as short as 3
4 days in arctic seals [1]. The composition of milk varies among mammalian species as a result of differences in gene function and physiology, especially the structure of the placenta. Presumably, these variations in composition reflect adaptations to different environments including exposure to specific microbes in addition, to different reproductive strategies and growth requirements. These differences are broadly based on the developmental state of the newborn. Species such as cattle and antelopes with precocial young born in full view of potential predators have different nutritional needs than do altricial species where the young are underdeveloped but sheltered or otherwise hidden so that there is no need to move rapidly. The most extreme example of this is the situation in marsupials where their extremely altricial young move to a pouch or skin fold where they require very much longer lactation periods than do Eutherian mammals.

3.1

The origins of lactation

Lactation evolved long before the development of viviparity. Milk production probably originated more than 310 mya as a glandular skin secretion in the early synapsids [2]. This corresponded to a time when relatively soft skin evolved rather than the previously standard hard, impermeable scales. The early synapsids were egg-layers. Their eggs had parchment-like shells that can lose water rapidly when exposed to dry conditions. These could not have been incubated in an open nest. Egg laid Monotremes

Pouch/ Lactation Hatch Placenta

Marsupials

Pouch/ Lactation

FIGURE 3.1 Lactation can be considered an extension of pregnancy. Thus in monotremes and marsupials, the highly immature neonate forgoes nutrition through the placenta in order to receive it in the form of milk. Instead of residing within the uterus, the neonate develops in another location, the pouch. One possible reason for this procedure is a need to get out of the uterus before the allograft rejection process destroys the developing embryo. In effect therefore the developmental stages in marsupials and monotremes represent an intermediate step in the transition from oviparity to viviparity.

Implantation Birth Placenta Eutherians Implantation

Uterus

Lactation Birth

Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00009-5 © 2023 Elsevier Inc. All rights reserved.

29

30

SECTION | 1 Mammalian immunology

They were likely held close to the body, perhaps in a pouch. It is proposed therefore that the earliest skin secretions in the synapsids may have served to keep their newly laid eggs moist. These eggs can take up water across the eggshell. This would have been especially important as these early mammals developed endothermy and so incubated their eggs above ambient temperature. Nutritional roles likely came later [3]. Anatomically speaking, these skin secretory glands probably arose from hair follicle-associated apocrine glands. Both apocrine glands and mammary glands secrete fluids by exocytosis of secretory vesicles. They were probably fully developed well before the emergence of the monotremes having arisen from ancestral apocrine glands. The first phase of their evolution likely resulted in the development of an elongated ectodermal field committed to mammary gland development, the mammary ridge. Along this ridge, between the armpit and the groin, a local thickening of the epidermis progressively developed mammary hair follicles with associated glands. One or more mammary placodes then developed further to form a mammary bulb. From the bulb there emerged primary sprouts that underwent canalization and led through the epidermis field. This suggestion is supported by the situation in monotremes. Echidnas employ a “mammary patch” without a nipple while the platypus has two, fur-covered nipples. This relationship is also seen in marsupials where there is a developmental association between mammary glands and hair follicles. The parchment-like eggshells of the monotremes are porous. As a result, their eggs can desiccate readily. Consequently, monotremes incubate their eggs within a moist skin fold. This not only prevents desiccation but over time, its secretions have evolved to become nutritionally rich. In marsupials such as opossums, the mammary glands first develop as hair follicles, but the hairs are subsequently shed leaving a duct (the galactophore) through which the milk secretions can reach the surface of the skin. These early developments in the protomammary glandular tissue were accompanied by a progressive loss of function of the vitellogenins, the proteins that transport nutrients to the egg yolk [4]. As the need for a functional yolk sac declined, its role was taken over, first by lactation as in the monotremes, and subsequently by the placenta as in the marsupials. Vitellogenins were thus superfluous and as a result, their genes were eventually pseudogenized and lost from the Eutherian genome. It is probable that the evolution of the mammary gland, like viviparity, involved the progressive adaptation of preexisting precursor genes, involving alterations in regulatory sequences to permit their expression in the primordial mammary tissue. Genetic alterations permitted preexisting proteins to assume new functions. As discussed below, the likely first functions of these new skin structures, in addition to providing moisture, were to provide passive immune protection to the neonate by passing on antimicrobial proteins. In addition to protecting the neonate, these early secretions also provided essential nutrition. This nutrition benefited not only the newborn but also its necessary adjunct, the intestinal microbiota. Indeed, suckling and maternal licking would have served as effective methods of microbial transfer resulting in the early colonization of the gastrointestinal tract with the “right” microbes. Mammary gland development was slow and progressive across different taxa. Thus in monotremes, the mammary placode spread out to form the mammary bulb from which more than 100 primary sprouts descend into the mesenchyme. At their distal ends, they empty into the infundibula of hair follicles. The platypus has a relatively large mammary gland extending almost a third of her body length and possessing two fur-covered nipples [5]. In marsupials, a flask-like mammary bulb elongates as a sprout but then hollows out to form teats. In Eutherians such as cats, horses, cows, and humans the distal ends of these pilosebaceous structures develop into mature milk ducts. The first proto-mammary glands probably secreted a complex mixture of proteins that eventually evolved into nutrient-rich milk. Calcium-binding phosphoproteins may have initially played a role in delivering calcium to the eggshell but eventually evolved into the caseins that provided amino acids, as well as calcium and phosphorous. Importantly, antimicrobial proteins were also incorporated into the mixture, including an ancestral form of butyrophilin as well as xanthine oxidoreductase (XOR). These could have been incorporated into fat droplets. One such protein eventually mutated into alpha-lactalbumin and acquired the ability to carry novel milk sugars and oligosaccharides. XOR in milk catalyzes the formation of uric acid. It may also have antimicrobial properties. The butyrophilins are members of the immunoglobulin superfamily expressed in multiple tissues. Butyrophilin 1A1 is found in milk fat droplets and the thymus. It appears to play a role in the regulation of T cells. The ancestral butyrophilin was likely a defensive transmembrane protein in secretory cells. By 200 mya, the early mammals had begun to develop endothermy so that their warm eggs were even more susceptible to fluid loss. They were initially small in size thus requiring a limited incubation period. This, together with the need for rapid early growth made post-hatching nourishment essential. As a result, there was a progressively increased dependence on milk for nourishment. By about 170 mya, milk had superseded the fetal egg yolk as the major source of early nourishment [6]. It is a much more efficient way to provide nutrients to the developing young and in marsupials and monotremes, lactation rather than gestation remains the prime contributor to neonatal growth [7].

The evolution and role of lactation Chapter | 3

31

TNF-D Upregulates

NF-NB Upregulates

IFN-J

Prolactin Xanthine oxoreductase

NITRIC OXIDE Reactive oxygen species Antibacterial activity

Milk fat droplet production Milk fat secretion

FIGURE 3.2 The xanthine oxoreductase (XOR) production pathway. XOR production is stimulated by tumor necrosis factor acting through the NFκB pathway. This enzyme has both significant antimicrobial properties as well as being required for fat droplet formation. This is an example of the overlap between the shared defensive and nutritional properties of milk components.

3.2

The functions of milk

3.2.1 Nutritional functions Nourishment of the neonate is a very obvious role of milk. It is the initial food source of young mammals, and their success testifies to its benefits. It essentially provides the neonate with a balanced mixture of essential nutrients including energy-rich lipids, minerals such as iron, vitamins, and proteins in addition to water. It is delivered in a highly absorbable and digestible form. It does this at a relatively minimal energy cost to the mother. Thus lactation mobilizes the essential biomolecules and combines them in a form that optimizes the well-being of neonates. Every mammal depends on milk for its early survival. It provides the essential nutrients for small altricial neonates from monotreme hatchlings and marsupial neonates to highly precocious offspring such as hooded seals that suckle for only four days but double their mass in that time [1]. Secreted milk is highly variable in composition depending upon the species. For example, there are only traces of fat (less than 1%) in rhinoceros’ milk whereas there may be as much as 60% fat in the milk of seals that breed on ice floes [8]. Milk also contains many unique proteins including different caseins, beta-lactoglobulin, and alphalactalbumin, as well as membrane-enclosed lipid droplets, and a complex mixture of sugars and oligosaccharides.

3.2.2 Intestinal development Milk plays a role in the early development of the gastrointestinal tract by influencing both its microbial colonization and mucosal immune development. MicroRNAs (miRNAs) are small single-stranded RNA molecules about 22 nucleotides in length, that regulate gene expression. They bind to the 3’ untranslated region of the coding regions of messenger RNAs and promote their degradation. miRNAs regulate many different biological processes including embryonic development, cell differentiation, and apoptosis. Their expression patterns and regulatory functions may be tissuespecific. Some miRNAs regulate gut development and mucosal immunity in newborns. miRNAs are present in many body fluids including both milk and colostrum [9]. They are shed into the colostrum from lactocytes within microvesicles and exosomes. These colostral miRNAs are absorbed from the intestine and enter the circulation of newborn mammals. Once in the circulation, miRNAs influence the development of the mucosal immune system, especially during the first week of life. The miRNAs likely regulate IL-6 and IL-17 signaling as well as T cell differentiation in the jejunum and ileum. These exosomes are more plentiful in colostrum than in milk. The commensal microbiota also influences miRNA expression and coordinates intestinal colonization and immune system development. Thus the expression of miRNAs in the small intestine of neonatal dairy calves is correlated with their total bacterial load and the presence of specific bacterial groups such as Lactobacilli and Bifidobacteria. This is clearly selective nourishment [10].

32

SECTION | 1 Mammalian immunology

3.2.3 Protective functions The innate immune system evolved millions of years before the appearance of mammals. The innate system has both constitutive and induced components. The constitutive component primarily relies on the continual presence of specialized antimicrobial proteins and peptides to destroy invading microbes. Many of these antimicrobial proteins are also found in milk. As a result, many milk components are protective without inducing significant innate responses in the form of inflammation. Many are not unique to the mammary gland but are shared with the innate immune system elsewhere in the body. As a result, there is a persuasive argument to be made that the mammary gland and lactation first evolved as a method of delivering passive innate protection to the neonate before assuming a nutritional role [11]. In addition, to maternal antibodies, milk contains many potent antimicrobial components. These include the ironbinding glycoprotein lactoferrin, the enzymes lysozyme, lactoperoxidase, and XOR, in addition, to oligosaccharides, glycoproteins, defensins, soluble CD14, and glycolipids. Colostrum also contains many viable leukocytes,—neutrophils, macrophages, and lymphocytes. Even when dead these cells may serve a protective function. For example, lactoferrin released from dying milk neutrophils inhibits microbial growth by chelating iron. Lactoperoxidase is also released from milk neutrophils. Neutrophil defensins have significant antimicrobial activity. Thus milk has a dual function being both a source of nutrition and protective fluid. Vorbach et al. have developed evidence to support the theory that the mammary gland is derived from modified skin glands whose primary function was to mount a defensive response to protect damaged skin and perhaps also the surface of eggs. Thus they suggest that lactation evolved in part from a local innate immune response. Its first role was to provide local antimicrobial protection, and this subsequently evolved into the provision of nutritional support for newborn hatchlings in monotremes or other suckling newborns [12]. They have pointed out that two enzymes found in milk, XOR and lysozyme play a dual role in that both have potent antimicrobial activity as well as being important for the utilization of milk nutrients. These molecules influence the nutritional properties of milk since they are involved in purine metabolism and the generation of milk fat globules (Fig. 3.2). XOR is involved in bactericidal activities, the production of reactive oxygen species, and the recruitment and activation of neutrophils and macrophages. It has been suggested that XOR is a protein that may have been originally produced in the primordial skin secretions. In addition, alpha-lactalbumin, a subunit of the lactose synthase complex also appears to have evolved as a result of duplication of the lysozyme gene. Lysozyme is an enzyme that kills Gram-positive bacteria by digesting their cell walls, while lactose is a key nutritional component of milk. Lactose is a disaccharide only metabolized by certain microorganisms. As described below, the lactose can be attached to other oligosaccharides that may selectively influence commensal bacterial growth. It has also been pointed out that both the immune and nutritional components of milk share the NF-κB pathway and the Janus kinase (JAK)/ signal transducer and activator of transcription (STAT) pathways. NF-κB regulates the production of lactogenic hormones, prolactin, and oxytocin. Acting through the JAK/STAT pathway, prolactin stimulates the expression of XOR.

3.3

Lactation and the microbiota

Milk contains large quantities of complex oligosaccharides. For example, in human milk, there are about 200 unique neutral and anionic oligosaccharides that may contain up to 22 sugar monomers. These monosaccharides include Dglucose, D-galactose, N-acetylglucosamine, L-fucose, and N-acetylneuraminic acid with a lactose core. There is however much diversity among individual mothers [13]. While milk is an important source of nutrients it is also clear that human infants cannot utilize these monosaccharides. They are indigestible. This raises the question, what is their function? It appears that they serve as an important carbon source for many intestinal bacteria. In effect, they act as selective prebiotics that promotes the growth of key members of the microbiota. Thus Bifidobacterium longum biovar infantis can readily utilize human milk oligosaccharides. This bacterium appears to have coevolved with the production of milk oligosaccharides. It is also recognized that certain gut pathogens bind to intestinal epithelium through cell surface oligosaccharide receptors. Thus the milk oligosaccharides may also serve a protective function by out-competing potential pathogens for these receptors [14]. A healthy gastrointestinal tract also requires that the intestinal commensal microbiota have a stable relationship with the mammalian immune system. Maternal colostrum and milk initiate gastrointestinal protection against potential pathogens in the form of large quantities of secretory IgA (sIgA). This IgA reaches the milk and colostrum by active transport across the mammary gland epithelial cells. When suckled and unsuckled mice are compared, some important differences become readily apparent [15]. Thus early exposure to sIgA prevents the translocation of some bacteria across the mouse gut wall into draining lymph nodes. By weaning time, suckled mice have a significantly different gut

The evolution and role of lactation Chapter | 3

33

microbiota than mice that did not receive milk IgA and these differences persist into adulthood. Early exposure to sIgA also results in an altered pattern of gene expression in intestinal epithelial cells, including the expression of proinflammatory genes and those involved in intestinal tissue repair [16].

3.3.1 The gut-mammary axis Once established, the presence of the gut microbiota drives B cell IgA production by plasma cells within the intestinal wall. The gut microbiota is constantly generating antigens that are sampled and processed by M cells and dendritic cells overlying inductive lymphoid sites such as the Peyer’s patches. These processed antigens trigger B cell responses and their differentiation into IgA-secreting plasma cells. Thus the IgA system is constantly responding to the microbiota. During pregnancy, many of these responding B cells leave the mother’s Peyer’s patches, enter her circulation and eventually colonize the developing mammary gland under the influence of the chemokine CCL28. It is these IgA1 B cells and plasma cells that subsequently secrete the IgA that is found in milk [17]. As a result, the suckling newborn will ingest large quantities of IgA. Some of this IgA is directed specifically against enteric pathogens while some lowaffinity antibodies may also promote the growth of beneficial commensals (Fig. 3.3). Different commensals differ in their ability to induce SIgA production. In mice, organisms such as Bacteroides acidfaciens and Prevotella buccalis appear to be indispensable for programming this maternal IgA synthesis. Thus the intestinal microbiota is, in part, responsible for the production of much of the IgA in milk [18].

3.4

Adaptive immunity

The evolution of the placenta in Eutherians removed much of the need for fetal nutrition in the early stages of development. However, marsupials with their very simple placentas and short gestation periods make an early switch and rely on lactation for nutrition to a much greater degree than do Eutherians. Marsupial milk also undergoes greater changes in its composition during lactation than does Eutherian milk (Chapter 13). The amount and composition of the first milk, the colostrum, very much depends on the nature of the maternal placentation and the ability of the mother to passively transfer immunity to her young. Species that have significant transplacental immunoglobulin transfer such as primates and lagomorphs do not require much additional transfer via the colostrum. Conversely in ungulates where placental transfer is minimal, then colostrum must be rich in immunoglobulins. Their young are often agammaglobulinemic when born. In these species, the intake and absorption of sufficient colostrum are mandatory if the newborn is to be adequately protected against systemic infections. Their B cells have never had to encounter or respond to foreign antigens. They are unexercised in this role and as a result, require time to respond and generate protective levels of immunoglobulin. They need immediate immunological assistance in the form of IgG. As a result, there are great differences in the composition of colostrum between different mammalian orders [19].

Mammary gland Y

Y

Y

Y

Y

Y

IgA

Y Y

Peyer’s Patch

Y

Antimicrobial IgA response

Y

Y Y

Antimicrobial IgA antibodies

Y

Y

Y

Y

Y

Microbial antigens

Milk

Y

Y

Y

Y

IgA+ B cell

Maternal intestine Neonatal Intestine

FIGURE 3.3 During pregnancy, bacteria in the mother’s intestinal tract stimulate a local IgA response in her intestinal lymphoid tissues, especially her Peyer’s patches. B cells from these patches enter the circulation and colonize the developing mammary gland. The IgA produced by these B cells is secreted into the milk and eventually ingested by the sucking newborn. Thus the newborn ingests milk rich in antibodies directed against intestinal bacteria.

34

3.5

SECTION | 1 Mammalian immunology

Colostrum

3.5.1 Production Colostrum contains the accumulated secretions of the mammary gland over the last days of pregnancy and the first few days after the young have been born. Colostrogenesis is driven by pregnancy hormones. In ruminants, colostrum production begins 3
4 weeks prior to birth and ends a few hours after birth. Most of the protein in colostrum and milk is produced by lactocytes within the mammary gland and secreted by exocytosis. However, the rest of the proteins such as albumin, sIgA, IgG, and insulin are extracted from the bloodstream by endocytosis on the basolateral surface of the lactocyte and transported across it to be exocytosed at the apical membrane. Mammals whose placentas permit significant transplacental passage of IgG generally produce colostrum that is not very different from milk in its immunoglobulin content. These immunoglobulin-low species also include primates, lagomorphs, and rodents.

3.5.1.1 The neonatal Fc receptor Most IgG in milk and colostrum comes from the maternal bloodstream although there are also plasma cells found within the mammary tissue. Neonatal immunoglobulin receptors (FcRn) are expressed on mammary gland ductal and acinar cells [20]. Serum immunoglobulins and albumin are taken up by FcRn in mammary epithelial cells, incorporated into endosomes, and transferred from serum into colostrum. Colostrum is therefore rich in IgG. The Fc receptor of the neonate, FcRn, is a heterodimer that is classified as a nonclassical class I MHC protein. Its α-chain forms heterodimers with β2-microglobulin (β2 M). Its ligands are IgG and serum albumin. FcRn is expressed on intestinal, pulmonary, and mammary epithelial cells, as well as on kidney podocytes and hepatocytes [21]. It is also present on many leukocytes, especially monocytes and dendritic cells. It binds IgG at pH , 6.5 but rapidly releases it at pH 7. Thus it binds IgG within acidified epithelial cell endosomes but releases it on exposure to the higher pH in extracellular fluids. The Fc region of the IgG is believed to bind to the α1 domain of the FcRn. Serum albumin binds to the α2 domain [22]. IgG subclass affinity for FcRn is also very variable. Different haplotypes of the FCGRT gene that encodes the FcRn α chain are associated with different milk IgG levels in both colostrum and neonatal calves. Haplotypes of the β2 M gene also influence IgG concentrations in calves [23].

3.5.2 Immunoglobulin transfer Young mammals that suckle soon after birth ingest colostrum. Naturally suckled calves ingest an average of two liters of colostrum, although some individuals may ingest as much as six liters. In these neonatal mammals, protease activity in the digestive tract is low for the first 24 hours after birth and is further reduced by trypsin inhibitors in colostrum. As a result, colostral proteins, especially IgG are not degraded but reach the ileum intact. These colostral immunoglobulins are endocytosed by enterocytes and bind to their FcRns. Once bound to endosomal FcRn, the immunoglobulin molecules are transported across the enterocytes and transferred to the lacteals and possibly the intestinal capillaries. Eventually, the absorbed immunoglobulins reach the bloodstream and newborn mammals obtain a massive transfusion of maternal immunoglobulins (Fig. 3.3). The duration of colostrum production appears to be directly related to the development state of the newborn mammals. Thus species with precocious young that need to move rapidly if necessary, such as calves and lambs, generally drink sufficient colostrum and acquire the IgG that they need within a few hours. Conversely, species with very altricial young such as dogs and bears, that require prolonged maternal care in a den, will require a much longer period of colostrum production than those with precocious young. Thus in the giant panda, the transition from IgG-rich colostrum to mature milk takes 30 days. During this time presumably, the growing cubs will continue to consume colostrum and absorb IgG through their gastrointestinal mucosa [24]. Likewise, rodents gradually transition from colostrum to milk over their entire lactation [21].

3.5.3 Composition Colostrum contains locally synthesized proteins as well as proteins actively transferred from the bloodstream under the influence of estrogens, prolactin, and progesterone. These include proteins that serve both a nutritional function such as the caseins and as well as the major immunoglobulin classes, IgM, IgG, and IgA. Colostrum also contains many cytokines and regulatory proteins. For example, bovine colostrum contains significant amounts of IL-1β, IL-6, TNF-α, and

The evolution and role of lactation Chapter | 3

35

IFN-γ. It contains members of the insulin-like growth factor family (IGF) as well as IGF-binding proteins and lactadherin. These cytokines and regulatory proteins are functional and available to promote the development of the immune and digestive systems in young animals. Compared to milk, colostrum also contains more sodium and chloride and less potassium and lactose. As described above, colostrum from species in which immunoglobulins do not cross the placenta contains very high levels of IgG that drop as lactation proceeds until IgA becomes the predominant class in milk. In bovine colostrum, for example, immunoglobulins can constitute up to 70%
80% of the colostral proteins, but this however varies greatly. The concentration of these may reach more than 150 mg/mL [21]. Of these, IgG constitutes 65%
80%. All of the IgG, most of the IgM, and about half of the IgA in bovine colostrum are actively transferred from the bloodstream to the mammary gland [25]. In milk, in contrast, only 30% of the IgG and 10% of the IgA are so derived; the rest are produced locally by B cells and plasma cells residing within the mammary gland. These immunoglobulin-high species include the cetartiodactyls (whales, camels, pigs, and ruminants).

3.5.4 Colostral lymphocytes Colostrum is full of lymphocytes, but milk is not. For example, sow colostrum contains between 1 3 105 and 1 3 106 lymphocytes/mL. Of these, 70% to 80% are T cells. Bovine colostrum also contains between 3 3 104 and 1 3 105 lymphocytes/mL, about half of which are T cells. Colostral lymphocytes may survive up to 36 hours in the intestine of newborn calves, and some may penetrate the epithelium of Peyer’s patches and reach the lacteal ducts or the mesenteric lymph nodes. Within two hours after receiving colostrum that contained labeled cells, these maternal lymphocytes appeared in the bloodstream of newborn piglets. As a result, cell-mediated immunity can also be transferred to newborn mammals. Piglets that had received these colostral cells showed enhanced T cell responses to mitogens compared with control mammals [26]. Cell-containing and cell-free colostrum have been compared for their ability to protect calves against enteropathic E. coli. The calves receiving colostral cells excreted significantly fewer bacteria than the calves receiving cell-free colostrum [27]. The concentration of IgA- and IgM-specific antibodies against E. coli in the serum of neonatal calves was higher in those that received colostral cells than in those that did not. The calves that received colostral cells also had better responses to the mitogen concanavalin A and foreign antigens such as sheep erythrocytes. The mechanisms of this protective effect are unclear but the monocytes of calves that received colostral cells are more capable of processing and presenting antigens. Transcriptome analysis of colostral T cells in sows has also indicated that they are more activated than peripheral blood T cells. The CD81 T cells in bovine colostrum can produce large quantities of IFN-γ, that may influence the early development of Th1 responses in neonatal calves [28]. Transfer of cell-mediated immunity by bovine milk lymphocytes has also been demonstrated. Pregnant cows were vaccinated against the bovine virus diarrhea virus (BVDV). Blood lymphocytes from calves that received cell-free colostrum from these cows were unresponsive to the BVDV antigen. In contrast, lymphocytes from calves that received colostrum containing live cells showed enhanced responses to BVDV antigen one and two days after colostral ingestion. The lymphocytes of calves that received whole colostrum showed enhanced mitogenic responses to maternal and unrelated leukocytes after 24 hours [29]. They also responded to the nonspecific stimulant staphylococcal enterotoxin B. In contrast, the lymphocytes of calves that received acellular colostrum did not. Ingestion of maternal colostral leukocytes immediately after birth stimulates the development of the neonatal immune system [30]. Similar phenomena also occur in the horse. Mononuclear cells in mare’s colostrum are mainly CD41 and CD81 T cells [31]. These CD81 cells are enriched when compared to blood. When stimulated with phorbol myristate acetate and ionomycin they produce more IL-17, similar IFN-γ, and less IL-4 and IL-10 when compared to blood cells [32]. Thus these cells are polarized towards IFN-γ and IL-17 production. This phenotype is generally considered to promote inflammation. As discussed in the previous chapter, pregnancy requires the development of an immune regulatory phenotype mediated by dendritic cells. In humans, this regulatory state does not end immediately at birth but extends into the neonatal period. This, together with the vertical transmission and establishment of the microbiota play a key role in regulating the developing immune system of the newborn. Thus Treg cell numbers are increased in both the fetus and the mother. Breast milk contains both bacteria and maternal cells and this combination can play a role in regulating neonatal immune development. In humans, it is clear that exposure of the neonate to maternal cells while breastfeeding serves to drive the maturation of Treg cells and “tolerized” the infant towards certain maternal antigens [33]. Thus Treg populations expand greatly during the first three weeks of life, and this effect is much greater in breastfed infants than in formula-fed babies. These T regs appear to be activated since they show increased expression of HLA-DR. However, they are relatively unresponsive to the presence of maternal cells

36

SECTION | 1 Mammalian immunology

Transplacental transfer of cells during pregnancy can also promote specific tolerance to maternal antigens and promotes reciprocal microchimerism that can last for many years. This promotes fertility in female offspring in nextgeneration pregnancies [34,35]. Evidence from both humans and other mammals suggests that maternal cells in colostrum can also colonize infant mucosal tissues thus resulting in neonatal microchimerism. The cells involved are most likely maternal stem cells that can account for up to 6% of maternal breastmilk cells. Many are also likely to be immune cells,—most likely regulatory T cells (Tregs). These may induce tolerance to some, noninherited maternal antigens. Thus in combination with the microchimerism that occurs during pregnancy, these transferred cells may contribute to the immune system and intestinal development.

3.6

Milk

In general, once mammary production has switched from colostrum to milk, the differences between species shrink significantly as immunoglobulin levels drop (Fig. 3.4) [36].

3.6.1 Milk immunoglobulins The secretions of the mammary gland gradually change from colostrum to milk. Milk composition can also vary greatly during lactation. Likewise, the domestication of cattle specifically for milk production has resulted in alterations in the composition of both colostrum and milk, especially in milk fat levels. In humans (primates) for example, IgG accounts for about 10% of total immunoglobulin content in milk (Fig. 3.4). As lactation progresses and colostrum changes to milk, differences among species emerge. In primates, IgA predominates in both colostrum and milk. In human milk, the level of IgG averages about 14.71 g/L during the first year of lactation but can climb to about 19 g/L in the second year [37]. In pigs and horses, IgG also predominates in colostrum, but its concentration drops rapidly as lactation proceeds so that IgA predominates in milk.

3.6.1.1 IgG The IgG transferred through the colostrum to the neonate by the FcRn represents the results of the mother’s history of antigen exposure, B cell responses, and somatic hypermutation. This maternal IgG in effect represents the

1-3

MILK

110-620

COLOSTRUM 1.4 2.8

Dog 37

90 140

22 SERUM

250

250 Human

300-1300

100-700

FIGURE 3.4 The level of immunoglobulins in colostrum and milk of selected domestic mammals and humans. While there is great variation in these levels between individuals, it is clear that the situation in bovine milk and colostrum is very different from that in humans and other single stomached animals. The immunoglobulin levels in milk are expressed in mg/dL.

10-50

300-500 Pig

Human 10-20

5-10 10-50 20-50

2400-8000 IgM

IgG

50-750

50-100

IgA Cow, sheep, horse, pig

Cow

Horse

The evolution and role of lactation Chapter | 3

37

immunological experiences of the mother. Maternal antibodies act on the immune system of the newborn during a critical imprinting period and exert a lifelong influence on the newborn’s immune development [38]. This influence may be stronger than some genetic predispositions! Thus maternal antibodies can enhance the newborn immune responses to some antigens and suppress their responses to others. They may also influence Th1/Th2 polarization and the subsequent development of allergies.

3.6.1.2 IgM IgM is present in colostrum and milk in small quantities. In human mothers, it begins at a concentration of 2.81 6 2.74 mg/L. However, it declines progressively as lactation proceeds and is not detectable in all breast milk samples [37].

3.6.1.3 IgA and IgE As described above, B cells producing IgA in the mother’s intestine migrate to her mammary gland during pregnancy—the common mucosal immune system. Thus this milk IgA is directed against potential maternal pathogens as well as selected members of the gut microbiota. Transepithelial transport of IgA occurs via the polymeric Ig receptor that normally binds secretory component (SC) and IgM. Ruminant milk is rich in both IgG1 and IgA. Nonruminant milk is rich in IgA. For the first few weeks in life, while protease activity is low, these immunoglobulins can be found throughout the intestine and in the feces of young mammals. As the digestive ability of the intestine increases, eventually only SIgA molecules remain intact. The amount of IgA provided by milk can be large; for instance, a 3-week-old piglet may receive 1.6 g daily from sow’s milk. In the case of humans, mature milk contains up to 7.55 g/L of IgA [37]. (Colostrum contains up to 12 g/L.) A breast-fed infant about four months of age may consume about 75 mg/kg of IgA per day (B50 g). IgE production in milk has been investigated in horses [39]. Although IgE is present in mare’s milk and transmitted to the suckling foal, its level in foal serum drops to a very low level by 6 weeks of age. IgE synthesis by foals begins at about 9
11 months of age, and at that time a pattern of relatively high or low IgE levels is established. These levels are not correlated with the levels resulting from suckling. Total levels of IgE in young horses (and their susceptibility to allergies) are mainly determined by genetic factors and their microbiota. IgE is also transferred in sheep colostrum. Colostral IgE levels are significantly higher than in ewe’s serum. IgE is absent from pre-suckling lamb serum but can rise to adult levels by two days after birth. It then declines steadily over several weeks. The milk of ungulates is nutritionally very rich since it is required for the nutrition of active, precocial young. In addition, newborns can supplement maternal milk with solid grazing material. In most ungulates, milk is richer in carbohydrates and fats than is colostrum, reflecting its importance as an energy source. In the related cetaceans such as the baleen whales, their very high-fat concentration in milk relative to colostrum is probably also due to the need for young whales to develop layers of insulating blubber before migrating to colder environments. Colostrum also contains SC both in the free form and bound to IgA. Young pigs and probably other young mammals have large amounts of the free SC in their intestine. Colostral IgA and, to a lesser extent, IgM can bind this SC, which may then prevent their absorption through the intestinal wall. The duration of intestinal permeability also varies among species and immunoglobulin classes. In artiodactyls and carnivores, permeability is highest immediately after birth and declines after about six hours because of the replacement of FcRn-bearing enterocytes by more mature cells that do not express this receptor. As a rule, absorption of all immunoglobulin classes drops to a very low level after about 24 hours of age. Multiple factors affect the duration of intestinal permeability. Feeding colostrum tends to hasten closure, whereas a delay in feeding results in a slight delay in closure (up to 33 hours). In piglets, the ability to absorb immunoglobulins may be retained for up to four days if milk products are withheld. The presence of the mother may be associated with increased immunoglobulin absorption. Thus calves fed measured amounts of colostrum in the presence of their mother absorb more immunoglobulins than calves fed the same amount in her absence. In laboratory studies in which measured amounts of colostrum are fed to calves, there is a great variation (25%
35%) in the quantity of immunoglobulins absorbed. Rabbits exclusively transfer their immunoglobulins via the placenta and their young are born in a very altricial state. As a result, there are minimal differences in protein, fat, and carbohydrate between their first milk and later milk. In effect, rabbits do not require specialized colostrum. A similar situation occurs in primates. In cats, colostral protein concentrations are lower than in mature milk [40].

38

SECTION | 1 Mammalian immunology

Unsuckled mammals normally have very low levels of immunoglobulins in their serum. The successful absorption of colostral immunoglobulins immediately supplies them with serum IgG at a level approaching that found in adults. Peak serum immunoglobulin levels are normally reached between 12 and 24 hours after birth. After absorption ceases, these passively acquired antibodies decline through normal metabolic processes. The rate of decline differs among immunoglobulin classes, and the time taken to decline to nonprotective levels depends on their initial concentration. In calves, the serum half-life of colostral-derived IgG is about 28 days. As intestinal absorption is taking place, simultaneous proteinuria may occur [41]. This is due to intestinal absorption of very small proteins such as β-lactoglobulin that can be excreted in the urine. In addition, the glomeruli of newborn mammals are permeable so that the urine of neonatal ruminants contains intact immunoglobulin molecules. This proteinuria ceases with the termination of intestinal absorption. Urine from puppies collected 24 hours after birth contains relatively large amounts of IgG, IgM, and IgA. The amount declines over time so that IgM is undetectable by 14 days, although there may still be significant amounts of IgG and IgA present. Over the first two weeks of life, the puppy’s glomeruli mature and acquire the ability to retain macromolecules.

References [1] Oftedal OT, Bowen WD, Boness DJ. Energy transfer by lactating hooded seals and nutrient deposition in their pups during the four days from birth to weaning. Physiol Zool 1993;66(3):412
36. [2] Oftedal OT. The mammary gland and its origin during synapsid evolution. J Mammary Gland Biol Neoplasia 2002;7(3):225
52. [3] Oftedal OT. The origin of lactation as a water source for parchment-shelled eggs. J Mammary Gland Biol Neoplasia 2002;7930:253
66. [4] Brawand D, Wahli W, Kaessermann H. Loss of egg yolk genes in mammals and the origin of lactation and placentation. PLoS Biol 2008;6:e63. [5] Grant TR, Griffiths M, Leckie RMC. Aspects of lactation in the platypus(Ornithorhynchus anatinus) in waters of New South Wales. Aust J Zool 1983;31:881
9. [6] Oftedal OT. The evolution of milk secretion and its ancient origins. Animal 2012;6(3):355
68. [7] Jenness R. Lactational performance in various mammalian species. J Dairy Sci 1986;69:869
85. [8] Oftedal OT, Iverson SJ. Phylogenetic variation in the gross composition of milks. In: Jensen R, editor. Handbook of milk composition. ,New York, NY: Academic Press; 1995, p. 749
89. [9] Liang G, Malmuthuge N, Guan Le L, Griebel P. Model systems to analyze the role of miRNAs and commensal microflora in bovine mucosal immune system development. Mol Immunol 2015;66:57
67. [10] Zivkovik AM, German JB, Lebrilla CB, Mills DA. Human milk glycobiome and its impact on the infant gastrointestinal microbiota. PNAS 2011;108:4653
8. [11] McClellan HL, Miller SJ, Hartmann PE. Evolution of lactation: nutrition v protection with special reference to five mammalian species. Nutr Res Rev 2008;21:97
116. [12] Vorbach C, Capecchi MR, Penninger JM. Evolution of the mammary gland from the innate immune system? Bioessays 2006;28:606
16. [13] German JB, Freenan SL, Lebrilla CB, Mills DA. Human milk oligosaccharides: evolution, structures and bioselectivity as substrates for intestinal bacteria. Nestle Nutr Workshop Pediatr Program 2008;62:205
22. [14] Newburg DS, Ruiz-Palacios GM, Morrow AL. Human milk glycans protect infants against enteric pathogens. Annu Rev Nutr 2005;361:15
23. [15] Rogier EW, Frantz AL, Bruno MEC, Wedlund L, et al. Secretory antibodies in breast milk promote long-term intestinal homeostasis by regulating the gut microbiota and host gene expression. Proc Natl Acad Sci USA 2014;111(8):3074
9. [16] Macpherson AJ, Koller Y, Mccoy KD. The bilateral responsiveness between intestinal microbes and IgA. Trends Immunol 2015;36:460
70. [17] Wilson HL, Obradovic MR. Evidence for a common mucosal immune system in the pig. Mol Immunol 2015;66:22
34. [18] Usami K, Niimi K, Matsuo A, Suyama Y, et al. The gut microbiota induces Peyer’s patch-dependent secretion of maternal IgA into milk. Cell Rep 2021. Available from: https://doi.org/10.1016/j.celrep.2021.109655. [19] Langer P. Differences in the composition of colostrum and milk in Eutherians reflect differences in immunoglobulin transfer. J Mammalology 2009;90(2):332
9. [20] Schnulle PM, Hurley WL. Sequence and expression of the FcRn in the porcine mammary gland. Vet Immunol Immunopathol 2003;91:227
31. [21] Baumrucker CR, Macrina AL, Bruckmaier RM. Clostrogenesis: role and mechanism of the bovine Fc receptor of the neonate (FcRn). J Mamm Gland Biol Neoplasia 2022. Available from: https://doi.org/10.1007/s0911-021-09506-2. [22] Pyzik M, Rath T, Lencer WI, Baker K, Blumberg RS. FcRn: the architect behind the immune and nonimmune functions of IgG and albumin. J Immunol 2015;194:4595
603. [23] Clawson ML, Heaton MP, Chitko-McKown CG, Fox JM, et al. Beta-2-microglobulin haplotypes in US beef cattle and association with failure of passive transfer in newborn calves. Mamm Genome 2004. Available from: https://doi.org/10.1007/s00335-003-2320-x. [24] Griffiths K, Hou R, Wang H, Zgang Z, et al. Prolonged transition time between colostrum and mature milk in a bear, the giant panda, Ailuropoda melanoleuca. Roy Soc Open Sci 2015. Available from: https://doi.org/10.1098/rsos.150395. [25] Hurley WL, Theil PK. Perspectives on immunoglobulins in colostrum and milk. Nutrients 2011;3:442
74. Available from: https://doi.org/ 10.3390/nu3040442.

The evolution and role of lactation Chapter | 3

39

[26] Hlavova K, Stepanova H, Faldyna M. The phenotype and activation status of T and NK cells in porcine colostrum suggest these are central/ effector memory cells. Vet J 2014;202:477
82. [27] Donovan DC, Reber AJ, Gabbard JD, et al. Effect of maternal cells transferred with colostrum on cellular responses to pathogen antigens in neonatal calves. Am J Vet Res 2007;68:778
82. [28] Ghosh MK, Nguyen V, Muller HK, Walker AM. Maternal milk T cells drive development of transgenerational Th1 immunity in offspring thymus. J Immunol 2016;197:2290
6. [29] Reber AJ, Hippen AR, Hurley DJ. Effects of the ingestion of whole colostrum or cell-free colostrum on the capacity of leukocytes in newborn calves to stimulate or respond in one-way mixed leukocyte cultures. Am J Vet Res 2005;66:1854
60. [30] Reber AJ, Donovan DC, Gabbard J, et al. Transfer of maternal colostral lymphocytes promotes development of the neonatal immune system II. Effects on neonatal lymphocytes. Vet Immunol Immunopathol 2008;123:305
13. [31] Perkins GA, Goodman LB, Wimer C, Freer H, et al. Maternal T-lymphocytes in equine colostrum express a primarily inflammatory phenotype. Vet Immunol Immunopathol 2014;161:141
50. [32] Hagiwara K, Domi M, Ando J. Bovine colostral CD8-positive cells are potent IFN-γ-producing cells. Vet Immunol Immunopathol 2008;124:93
8. [33] Wood H, Acharjee A, Pearce H, Quraishi MN, et al. Breastfeeding promotes early neonatal regulatory T cell expansion and immune tolerance of non-inherited maternal antigens. Allergy 2012;76:2447
60. [34] Moles J-P, Tuaillon E, Kankasa C, Bedin A-S, et al. Breastmilk cell trafficking induces microchimerism-mediated immune system in the infant. Pediatr Allergy Immunol 2018;29:133
43. [35] Kinder JM, Stelzer IA, Arck PC, Way SS. Immunological implications of pregnancy-induced microchimerism. Nat Rev Immunol 2017;17 (8):483
94. [36] Goldsmith SJ, Dickson JS, Barnhart NM, Toledo RT, et al. IgA, IgG, IgM and lactoferrin contents of human milk during early lactation and the effect of processing and storage. J Food Prot 1983;46(1):4
7. [37] Czosnykowska-Likacka M, Los-Kuberka J, Krolak-Olejnik B, Orczyk-Pawilpwicz M. Changes in human milk immunoglobulins during prolonged lactation. Front Pediatr 2020. Available from: https://doi.org/10.3389/fped.2020.00428. [38] Fink K, Zellweger R, Weber J, et al. Long-term maternal imprinting of the specific B cell repertoire by maternal antibodies. Eur J Immunol 2008;38:90
101. [39] Marti E, Ehrensperger F, Burger D, et al. Maternal transfer of IgE and subsequent development of IgE responses in the horse (Equus caballus). Vet Immunol Immunopathol 2009;127:203
11. [40] Day MJ. Immune system development in the dog and cat. J Comp Pathol 2007;137:810
15. [41] Schaefer-Somi S, Baer-Schadler S, Aurich JE. Proteinuria and immunoglobulinuria in neonatal dogs. Vet Rec 2005;157:378
82.

This page intentionally left blank

Chapter 4

Endothermy and immunity Endothermy, the ability to maintain a raised body temperature, is an expensive property since it requires a massive amount of caloric intake in order to fuel this body heat. Its costs are significant, and its benefits must therefore be proportional. The nature of the benefits conferred by endothermy remains a matter of debate. The reason for this is that endothermy involves three distinct processes. First, it requires systems for generating body heat. This requires a high metabolic rate to release the energy from food sources. In practice, an endothermic mammal requires between six and ten times greater caloric intake than does an ectothermic reptile of the same body mass. Second, it requires regulatory mechanisms that can maintain this body temperature within fairly narrow limits. This state, called homeothermy, requires maintaining a constant temperature with a very high level of stability, generally varying no more than two degrees from a constant baseline temperature. Thus mammals must have the ability to rapidly lose, or gain heat as needed to maintain temperature stability. Third, it requires a high aerobic scope. An ability to increase a mammal’s aerobic metabolic rate above maintenance levels. This may require as much as 10
15 times its basal metabolic rate. Heat is generated as a byproduct of the biochemical reactions that convert food energy into adenosine triphosphate (ATP). In the absence of external work activity, heat is generated simply by the process of maintaining the structural and metabolic integrity of body organs. Thus about 70% of the heat production in a resting mammal is generated within the abdominal organs such as the liver, kidney, and intestine, as well as the brain. Yet, these organs only account for 8% of body weight. Only during aerobic exercise do the skeletal muscles generate more heat although they account for about 45% of body mass [1]. (Heat generation also requires a greater number of mitochondria in muscle cells and the ability to deliver oxygen to them.) Another major source of heat production in some mammals is brown adipose tissue. This is a highly specialized form of adipose tissue located throughout the body at sites near the scapula, neck, adrenal glands, and deep blood vessels [1,2]. Clearly, an elevated body temperature will affect many different biochemical reactions including rates of enzymic reactions, diffusion of metabolites, and many different physiological processes including both innate and adaptive immune responses. The warmth not only affects the animal but also its developing fetuses, its commensals, as well as any invading microorganisms. As a result, its specific benefits may be difficult to identify. In addition, the effects of heat on the immune system and on any invading pathogen will vary depending upon the host and the invasive bacterial or viral species [3]. On the other hand, endothermy is potentially highly beneficial. It makes possible a high rate of sustainable aerobic activity that is vastly greater than any that can be generated by any possible ectotherm. (Ectothermic animals can do so for short periods using anaerobic processes but these are not sustainable.) This results in obvious advantages such as increased endurance that assists in successful predation or predator avoidance, the ability to move freely (agility) and migrate as well as the size of the territory that can be occupied and exploited. Theory suggests that the benefits obtained from increased aerobic capacity such as increased sustainable activity are sufficient to explain the evolution of endothermy. A slight increase in the number of functional mitochondria in muscle cells may have conferred a significant advantage in chasing prey. The higher metabolic activity may have raised body temperature somewhat, thus effectively preventing the onset of torpor at low temperatures, a major survival advantage [4].

4.1

The evolution of endothermy and homeothermy

It has long been assumed that endothermy, like viviparity or lactation, is too complex a process to have evolved in a single step. Thus one function such as high metabolic rate may have emerged first, while homeothermy evolved on separate occasions. It is unclear which came first. From an immunological viewpoint, however, given the protection afforded by a fever, it is possible that the ability to mount a febrile response emerged first and this advantage in immune function could then have been maintained by the evolution of effective thermoregulatory processes. Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00021-6 © 2023 Elsevier Inc. All rights reserved.

41

42

SECTION | 1 Mammalian immunology

A key feature of mammals is their ability to maintain a constant body temperature irrespective of environmental conditions [4]. The very first mammals, the ancestral synapsids, are believed to have been relatively small. Their size increased gradually over time to reach a peak with the primitive therapsids. They then gradually decreased in size again through the cynodonts. Thus the first true mammals were small, rodent-sized creatures. It has been suggested that the evolution of endothermy tracked these size changes and it is likely therefore to have evolved in therapsids during the Permian period [5]. Endothermy probably evolved in a line of small carnivorous or insectivorous synapsids. Given their need for significant agility when climbing trees to avoid predation, these early endothermic mammals may also have been arboreal. The high surface area to body mass ratio in these small animals would have required an increased metabolic rate to compensate for heat loss in a low-temperature environment. A variation in this hypothesis suggests that endothermy was associated with adaptation to a nocturnal, insectivorous lifestyle. An erect gait may also have permitted the increase in ventilation rates needed to sustain this enhanced aerobic activity. In order to maintain activity, especially in small nocturnal mammals, the basal metabolic rate had to increase. Any temperature increase may have been minimal at first, but as mammals became more diurnal then, the advantages acquired from elevated body temperatures may have continued and promoted the trend. It does appear that cynodonts were probably the first protomammals to increase in their basal metabolic rate, but there is no agreement as to how and when endothermy first developed. Homeothermy, the ability to maintain a constant body temperature, probably proceeded in step with endothermy, at least initially. However, as mammals evolved, it is probable that baseline body temperature progressively increased over time to eventually reach optimal values in extant Eutherian mammals. This is exemplified by the fact that the most basal of the living mammals, the monotremes, while endothermic, have a relatively low body temperature. For example, the platypus has a body temperature of 32 C while the echidnas range from 30 C to 32 C [6] (Fig. 4.1). In both species, body temperatures also tend to be unstable, varying diurnally by as much as 2 C
4 C [7]. Even some marsupials have a relatively low metabolic rate and body temperature when compared to Eutherians. They generally have a level of metabolism about 30% less, and a mean body temperature that is 2 C
3 C lower than is usual for Eutherians [8]. For example, perameloid marsupials (the bandicoots) have a body temperature of 35 C [9]. Even the Eutherian Xenarthra may have a relatively low body temperature. The nine-banded armadillo (Dasypus novemcinctus) has a mean body temperature of 35 C [10]. Likewise, Madagascar hedgehog tenrecs (Echinops telfairi), members of the Afrotheria, and thus relatively primitive placentals appear to maintain homeothermy in the 27 C
30 C range. Perhaps this was the normal body temperature range of Triassic mammals [11] (Box 4.1).

Degrees celsius

Fungal optimum 36.7 35 Armadillo 34-35 Platypus 32 Echidna 31 30

I I I I I I I I I I I I I I I I

40 Bovine Mouse Human Elephant Gorilla

38-39 36-38 37 36 35.5

Wallaby 36 Bandicoots 35

Tenrec 27-30

Two toed sloth 24-33

25

FIGURE 4.1 The body temperatures of some of the basal mammalian clades such as the monotremes, marsupials and Xenarthra tend to be significantly lower that most Eutherians. This supports the concept of a progressive increase in body temperature as mammals evolved. Currently most Eutherian mammals have a core body temperature in a narrow range just slightly higher than the maximum tolerated by fungal pathogens.

Endothermy and immunity Chapter | 4

43

BOX 4.1 Dropping Body Temperatures In order to determine normal body temperature, it has been usual to simply measure the temperatures of several thousand healthy individuals and determine the average. However, there are degrees of “healthy” and it is perhaps inevitable that some apparently normal individuals may in fact be suffering from a minor infection with a mild fever. Recent analysis of “normal” body temperatures suggests that men born in the early 19th century had temperatures 0.59 C higher than men today. In women, average body temperature has dropped by 0.32 C since the 1890s. It has been suggested that the late 1800s, many individuals would have been suffering from inapparent infections such as syphilis, tuberculosis, and periodontal disease and may have been mildly febrile. Over the past century and a half, the prevalence of these infections have greatly diminished in developed societies resulting in decreased inflammation, improved health, and as a result, an average lower body temperature. Protsiv M, Ley C, Lankester J et al. Decreasing human body temperature in the United States since the industrial revolution. eLife 2020; https://doi.org/10.7554/eLife.49555

4.2

The benefits of endothermy

Four major theories regarding the benefits of endothermy have been proposed: First, a high stable body temperature enables an animal to function in low environmental temperatures. As pointed out above, mammals have few problems with the sluggishness and vulnerability associated with low temperatures including cold nights. Consider, for example, the degree to which environmental temperatures drop at night in low humidity regions such as deserts. Endothermy thus expands the daily, geographic, and altitudinal ranges that mammals can operate at. Endotherms can function during the cool parts of the day or in cooler climates. In addition, endothermy enables them to avoid the need for exposure to solar radiation to induce a behavioral fever and the vulnerabilities associated with being exposed in such a manner. Ectotherms tend to be restricted to hot climates and as a result, are not free to expand their ranges indefinitely. The advent of endothermy opened up hitherto unavailable habitats and their resources. Mammals occupy much larger northsouth geographic ranges than reptiles or amphibians. Second, the increased aerobic capacity of an endotherm permits longer sustained muscular effort and results in increased stamina. This would be of considerable benefit both when chasing prey or even when seeking to evade predators. In effect, endothermy permits rapid movement for a longer period of time. Third, an elevated body temperature probably hastens the development of the young in utero. This would effectively shorten gestation periods while enhancing the benefits of viviparity. This may be an important factor. For example, the Lesser hedgehog tenrec (E. telfairi) of Madagascar as a “protoendotherm” has a normal body temperature in the range of 27 C
30 C. However, dissection shows that females have a mass of brown adipose tissue surrounding their uterus [12]. This brown adipose tissue is functional and actively thermogenic. Thus it may serve as a biological "blanket" to warm the developing fetus in utero and as a result, accelerate its development. In humans, brown adipose tissue arises late in the second trimester of pregnancy. By birth, an infant has both visceral and subcutaneous deposits. This brown fat protects the newborn infant from the cold until it develops the ability to shiver [13]. Finally, endothermy may make the body inhospitable to many potential pathogens. The body temperature of mammalian endotherms may exceed the maximum temperature tolerance of some bacteria and fungi. It may therefore limit the growth and invasiveness of pathogenic microbes. In addition, it significantly enhances the effectiveness of both innate and adaptive immune responses. Endothermy is also associated with the ability to develop a fever and a sickness response that improves not only disease resistance, but also results in other important behaviors such as malaise and loss of stamina.

4.3

The role of brown adipose tissue

One marked feature of endothermy is the occurrence of thermogenesis in the absence of shivering. In other words, heat is produced in the absence of muscular activity [5]. A feature of Eutherian mammals is the presence of heat-generating brown adipose tissue [2]. This tissue facilitates heat production by mitochondria by uncoupling the oxidation of foodderived nutrients from chemical energy ATP production. Brown adipose tissue contains large numbers of mitochondria, extensive vascularization, and importantly, the presence of uncoupling protein 1 (UCP1) (originally called thermogenin). In response to reduced body temperatures, brown and beige adipocytes can burn triglycerides or glucose to release

44

SECTION | 1 Mammalian immunology

FIGURE 4.2 The mechanism of heat production in brown adipose tissue. Uncoupling protein 1, diverts the protons generated by the respiratory chain into the mitochondrial matrix where their energy is released as heat rather than being stored in the form of adenosine triphosphate.

Fatty acids O2

ADP

ATP synthase

Oxidation

Electron transport chain + + H H H+ + Protons H+ H + H+ H UCP 1 MITOCHONDRIAL MATRIX

ATP synthesis

Heat generation

energy in the form of heat [13]. For this, they employ their stored fat droplets, their high numbers of mitochondria as well as UCP1. The major mechanism of generating heat in brown adipocytes is through the mitochondrial proton leak. The proton leak is enhanced in brown adipose tissue as a result of the activities of UCP1. UCP1 is an inner mitochondrial enzyme protein that dissipates the flow of protons generated by electron transport in the respiratory chain. Respiration is therefore uncoupled from ATP synthesis. As a result, it doesn’t form the phosphate bonds needed for ATP synthesis so the energy generated is released in the form of heat (Fig. 4.2). UCP1 plays a key thermoregulatory role in neonatal and small-bodied placental mammals, especially those living in cold environments [14]. While the gene encoding UCP1 has been detected in all Eutherians tested, inactivating mutations are present in members of at least eight mammalian orders. Analysis has shown that UP1 inactivation is associated with reduced metabolic activity in Xenarthra and the pangolins. In other orders, the appearance of the inactivating mutations appears to have coincided with the 30 my period of global cooling during the Paleogene that resulted in a rapid increase in mammalian body masses. Some mammals such as wooly mammoths, and Steller’s sea cows (Hydrodamalis gigas) increased greatly in size and so developed extreme cold hardiness in the absence of UCP1 [14]. Likewise, Afrotheria such as elephants, and sirenians with a large body mass and extensive fat deposits also lack this enzyme. Cetaceans also lack UCP1. As a result, they must rely on extensive deposits of insulating blubber to maintain their body temperature. Unsurprisingly, small Xenarthra such as armadillos, with a normally low body temperature or those with a very low metabolic rate such as sloths, lack UCP1 as well. White adipocytes, the major fat storage sites in obese individuals can undergo “browning.” Brown adipocytes can be generated from white adipocytes by exercise, diet, and some immune pathways. Partially browned white adipocytes are termed “beige” adipocytes. They can therefore increase heat production even in the absence of brown adipocytes. This process involves the remodeling of lipid droplets and an increase in both mitochondrial numbers and vascularity. Lipid droplet remodeling promotes rapid lipid mobilization and oxidation. The heat generated by the browning of white adipose tissue is regulated by the immune system. Both white and brown adipose tissues contain large numbers of lymphoid cells including M2-macrophages, eosinophils, innate type 2 lymphoid (ILC2) cells, and Treg cells. Many T cells are also normal residents of adipose tissues and are important regulators of energy release and thermogenesis. Several cytokines play a role in this process [15]. Thus the activated T cells release IL-4 and IL-13 to activate the M2-macrophages. These macrophages are cold-sensitive and in turn release IL-1, IL-18, and IL-33. IL-33 is a major stimulator of thermogenesis [16]. The IL-33 acts directly on adipocytes to stimulate fat mobilization and thermogenesis. IL-33 also stimulates ILC2 cells that can activate the thermogenic pathway and promote browning. γ/δTh17 cells act on adipose tissue by stimulating stromal cell production of IL-33. Other cytokines that play a role in fat metabolism include IL-6, a mediator of the acute phase response that triggers thermogenesis as do IL-15 and IL-18. It has recently been demonstrated that when IL-33 acts on T cells it activates mTORC1 (mitochondrial target of rapamycin 1). This in turn promotes the expression of glucose transporter 1 and glycolytic enzymes. As a result, these T cells increase both glucose uptake and lactate production. Thus these T cells undergo accelerated aerobic glucose metabolism that will also release energy in the form of heat [17]. Interleukin-33 also acts on macrophages to cause mitochondrial “rewiring.” It activates UCP2 which in turn uncouples the macrophage respiratory chain and as a result

Endothermy and immunity Chapter | 4

45

promotes their differentiation into alternatively activated phenotypes. This effectively promotes wound healing and the resolution of infection [18]. It should be pointed out that loss of adipose tissue is a feature of chronic inflammatory diseases of mammals such as tuberculosis and results in significant loss of body weight—consumption.

4.4

Fevers

The most obvious systemic response to infection is the development of a fever. In the case of endothermic and homeothermic mammals, fevers result from an overall increase in metabolic rate (Fig. 4.3). Thus the development of a fever, 2 C above baseline body temperature, requires a 20% increase in basal metabolism. Many cytokines can act as endogenous pyrogens. These include IL-1α, IL-1β, IL-6, IFN-α, TNF-α and TNF-β all of which can trigger a rise in body temperature. These cytokines induce cyclooxygenase-2 (COX-2) production in the preoptic nucleus of the hypothalamus as well as the limbic system. This results in prostaglandin-E2 (PGE2) production. The PGE2 acts on warm-sensitive neurons to slow their firing rate and so alter the body’s thermostatic set-point. Ceramide, a bioactive lipid may act as a second messenger independent of PGE2 and is likely important in the early stages of fever generation [19]. Transient receptor potential cation channels of the vanilloid subtype (TRPV) especially TRPVs 1
4 on neurons also act as cellular thermosensors. TRPV1 is a major regulator of body temperature outside the nervous system. These changes in the set-point result in significant behavioral changes. For example, affected animals conserve heat by vasoconstriction and increase their heat production by shivering, thus raising their body temperature until it reaches the new set-point. High-mobility group band protein-1 (HMGB1) is a potent sickness-inducing cytokine. Although IL-1, IL-6, and TNF-α have long been known to cause sickness behavior, it is now clear that these three molecules induce the slow release of HMGB1 from macrophages. HMGB1 has been implicated in food aversion and weight loss by its actions on the hypothalamic-pituitary axis. It mediates endotoxin lethality, arthritis, and macrophage activation. The inflammation induced by necrotic cells is caused in part by the escape of HMGB1 from disrupted nuclei and damaged mitochondria. As discussed in Chapter 8, T cells that employ γ and δ chains in their antigen-binding receptor, play a key role in innate immune responses. Some of them produce IL-17 in response to the cytokines IL-1β and IL-23 produced during inflammation. These long-lived γ/δ T cells accumulate within non-lymphoid organs such as the lungs, dermis, and uterus. In addition, many of these γ/δ T cells accumulate within adipose tissue. Here they stimulate stromal cells to produce IL-33. Mice that lack either γ/δ T cells or IL-17 have lower body temperatures than wild-type mice. Thus this regulatory cell subset appears to control adaptive thermogenesis. It has been claimed that the maintenance of the core body temperature relies upon the activities of these tissue-resident γ/δ17 T cells [20]. INFECTION Exogenous pyrogens

Endogenous pyrogens IL-1 TNF IFN, IL-6 HMGB-1

Inflammasomes

Prostaglandin E2 production

HYPOTHALAMUS Thermosensitive neurons Triggers cytokines IL-33 IL-17 Increased heat production

Increased heat retention Shivering (chills), piloerection

Body temperature rises

FIGURE 4.3 Mechanisms of a febrile response. Cytokines acting through the thermoregulatory centers of the hypothalamus, change the body’s set-point. The body subsequently increases heat production and minimizes heat loss. This results from increased thermogenesis caused by IL-17 and IL-33 as well as increased heat retention as a result of thermoregulatory reflexes such as shivering.

46

SECTION | 1 Mammalian immunology

4.4.1 Fevers and endothermy The immune system is thermally regulated. As a result, the ability to raise body temperature in response to infection significantly enhances both innate and adaptive immune responses. “Fever is good for the immune system” [21]. It appears that the development of endothermy results in enhanced resistance to infections. This is accomplished without suffering the disadvantages of maintaining a behavioral fever. A behavioral fever is generated when an ectotherm deliberately exposes itself to a heat source such as direct sunlight in order to allow its body temperature to rise and hence promote its resistance to infection. This works best in tropical climates and in effect, restricts the geographic range of ectotherms and their selection of microclimates. Endotherms have no such limitation, either seasonal or geographic. However, fevers push body temperature to an optimal level for immune system effectiveness. In effect, they can prime the immune system so that a modest increase in body temperature will result in enhanced immune function and hence protection [21]. In the case of ectotherms, the adaptative response of invasive pathogens likely results in a “race” between the animal’s ability to raise its temperature and the adaptability of the infectious agent. At some point, the animal would benefit by simply maintaining its body temperature at a higher constant level. In other words, mammalian homeostatic endothermy. Vertebrates generally mount enhanced innate and adaptive immune responses at higher body temperatures. Conversely, low temperatures in ectotherms are immunosuppressive. For example, in chilled fish, the lag period following vaccination may be long or may result in a complete absence of detectable antibodies. Only certain phases of the antibody response are temperature dependent. For example, secondary immune responses can be elicited at low environmental temperatures provided primary immunization is carried out at a warm temperature. The cells that are sensitive to low temperatures in fish are helper T cells. The failure to respond is due to a loss of T cell membrane fluidity and responsiveness to cytokines. Acclimatization to low temperatures can also occur. For example, goldfish that are acclimated to a low water temperature may be able to produce a similar number of antibody-forming cells to those that have been kept in warmer water. Certain T cell-dependent mitogens are ineffective in fish held at low temperatures, again implying that the target cell is a helper T cell. Although endotherms such as mammals develop a fever when infected, ectotherms are unable to change their body temperature by physiological mechanisms. As a result, they cannot develop a fever if maintained in a constant temperature environment. If, however, they are maintained in an environment with both cool and warm areas, they can cycle between these areas and so maintain their body temperature within defined limits. For example, it has been observed that healthy iguanas (Dipsosaurus dorsalis) normally cycle to maintain their temperature between 37 C and 41 C. However, iguanas infected with the bacterium Aeromonas hydrophila modify their behavior so that they spend more time in a warm environment. As a result, their temperatures cycle between 40 C and 43 C. Once the bacterial infection is cured, the iguanas resume their normal behavior. Thus the iguanas effectively induce fever through their behavior. A similar behavioral fever is seen in goldfish and zebrafish kept in two interconnected tanks maintained at different temperatures. In response to microbial infection, the fish will choose to spend more time in the warmer water, effectively raising their body temperature. The benefits of this are obvious because, as pointed out earlier, their immune systems function more efficiently at higher temperatures.

4.4.2 Hibernation Some mammals hibernate when food is seasonally unavailable. These include most notably bears, bats, and some rodents. In the prior summer, their gut microbiota is highly efficient, food intake is high, and these promote adiposity. As a result, bears enter hibernation in an obese state [22]. During hibernation, periods of metabolic depression known as torpor are interspersed by a transient activity called arousal. During torpor, the animal’s temperature may fall to less than 10 C. This affects both their innate and adaptive immune systems. Torpor drastically reduces blood leukocyte numbers, complement levels, phagocytosis, TNF-α and IFN-γ production, T cell proliferation, and antibody synthesis [23]. For example, if bats are cooled to about 8 C, they cease antibody production, but rewarming permits rapid resumption of antibody synthesis. This cessation of the antibody response in hibernating bats may allow them to act as persistent carriers of viruses such as rabies but increases their susceptibility to some infections, as shown by the development of white-nose syndrome, a fungal infection in hibernating bats (Chapter 18).

4.4.3 The costs While on balance, the ability to develop a fever is a positive adaptive response to infections, it does result in a temporary inability to function normally in a social context. Humans take to their beds, other mammals do not have a choice.

Endothermy and immunity Chapter | 4

47

Thus the development of a fever, while of benefit in confronting a microbial pathogen, results in an overall reduction in activity. While conserving energy to fight infections, sickness behavior may result in increased vulnerability to macropredators. In addition, there may be social costs within animal groups. For example, when vervet monkeys (Chlorocebus pygerythrus) were monitored with temperature sensors, it was found that febrile monkeys spent more time resting and less time feeding, just like sick humans! Fevers did not, however, influence the time spent interacting with other conspecifics. Nevertheless, the febrile monkeys were targeted with twice as much aggression by their troop mates and were six times more likely to become injured. Thus sickness behavior has significant costs for gregarious mammals and important ecological consequences [24]. Given these disadvantages of suffering sickness, many species attempt to suppress their sickness behavior. This however, also has costs. In one notable example, lizards that were actively prevented from developing a behavioral fever suffered from higher mortality than those that were permitted to do so [25].

4.4.4 Fevers and innate immunity Combating infection requires a major shift in metabolic priorities [26]. Given that sickness consistently suppresses food intake, the body must mobilize reserves of protein and energy to support the initial acute phase response as well as increased thermogenesis. Fevers are protective in that they enhance many innate immune responses (Fig. 4.4). For example, a high body temperature enhances many neutrophil functions. It promotes their release from the bone marrow, their trans-endothelial migration, and their chemotactic responses so increasing their accumulation within tissues. Fever range temperatures stimulate their respiratory burst so enhancing neutrophil functions and bacteriolytic activity. (This reaction also generates significant heat.) [27] Fever also accelerates caspase-dependent neutrophil apoptosis. Warmth increases neutrophil recruitment to inflammatory sites such as the lungs. Fever modulates the outcomes of infection. Thus it reduces virus replication within cells and increases the sensitivity of bacteria to complement-mediated lysis. Fever increases blood flow to infected tissues thus speeding the mobilization of effector cells. Raised body temperatures cause dendritic cells to mature; enhance multiple macrophage functions such as phagocytosis, the release of NO and cytokines, and the expression of TLR2, TLR4, and MHC molecules. They decrease the response times of many immune systems [21]. Fever enhances the phagocytic abilities of dendritic cells (DCs) as well as NK cell cytotoxicity. It acts on target cells to upregulate the NKG2D ligand, and MICA as well as promote receptor clustering [28]. Fever upregulates phagocytosis by DCs and their expression of TLR2 and TLR4. It enhances DC migration to lymphoid organs. The activated DCs enhance CD4 T cell proliferation and directs them towards Th1 responses. Heated macrophages upregulate their production of TNF-α through the actions of HSP70. (HSP70 speeds up ATP hydrolysis). Elevated body temperature also upregulates their production of nitric oxide synthase (iNOS) and hence NO production.

NEUTROPHILS Enhanced responses to LPS Enhanced release of ROS and NO Increased neutrophil infiltration Release of neutrophils from the BM

FIGURE 4.4 The effects of febrile temperatures on innate immune responses.

NK CELLS Enhanced NK activities Upregulated MICA on target cells NKG2D clustering on NK cells

FEVER RANGE TEMPERATURES

MACROPHAGES IL-33 induced rewiring uncouples respiratory chain Activation of GATA 3 Activated aggressive cells

DENDRITIC CELLS Activation and enhanced migration

48

SECTION | 1 Mammalian immunology

FIGURE 4.5 The effects of febrile temperatures on T cell functions.

Upregulates GATA3 Downregulates Tbet Promotes Th2 responses IL-4, IL-5 and IL-13 pathways Suppresses IFN-J pathways

Increased binding to some integrins Promotes T cell transmigration through D4 integrins Promotes T cell trafficking

FEVER RANGE TEMPERATURES

Enhances CD8-mediated cytotoxicity

Promotes Th17 differentiation and functions

Inhibits T cell apoptosis

4.4.5 Fevers and T cell functions Higher body temperatures increase the binding affinity of many immune system receptor-ligand pairs. As a result, they enhance lymphocyte, specifically T cell, behavior and effectiveness (Fig. 4.5). Fevers enhance the passage of T cells across high endothelial venule walls, promote immunological synapse formation, and inhibit T cell apoptosis. Activated CD81 T cells incubated at 39 C show augmented metabolic activity and T cell effector functions even without affecting their proliferation. Transcriptome profiling has shown upregulation of mitochondrial pathways, and this is responsible for the enhancement of metabolic activity and T cell functions. In effect, a fever can optimize the functional responses of CD81 T cells [29]. Fevers promote T cell trafficking and enhance immune surveillance during infection through a sensory pathway that involves heat shock protein 90 (HSP 90) and α4 integrins. Thus when mouse T cells are held at 40 C for 12 hours, they develop an increased ability to bind to ligands of the α4β1 integrin (VCAM1) and α4β7 integrins (MAdCAM1). As a result, these heat-treated T cells develop an enhanced ability to migrate across membranes coated with these integrins [30]. The febrile T cells also upregulate various HSPs. It has been shown however, that only HSP90 is selectively bound to the α4 integrins. and this is enhanced at febrile temperatures. When wild-type mice are exposed to elevated temperatures for six hours, their T cells show increased adherence to high endothelial venules and trafficking into lymph nodes and Peyer’s patches. Fever-induced enhancement of the HSP90-α4 axis is the key to this increased trafficking [30]. Fever range temperatures enhance CD8 effector cell differentiation . Fever range temperatures also modulate cytokine production by T cells [31]. Thus when primed CD41 T cells are maintained at a moderate fever temperature (39 C) they tend to function as Th2 cells producing IL-4, IL-5, and IL-13 while suppressing the production of the Th1 cytokine IFNγ. This is also associated with upregulation of the transcription factor GATA3 and downregulation of Tbet. It is also clear that febrile temperatures control the differentiation and plasticity of Th17 cells. As a result, they produce greater quantities of IL-17F and IL-22. These cells and the interleukins they produce play a critical role in stimulating acute inflammation and neutrophil activation. Both activities are promoted at higher temperatures in vivo. Treatment with antipyretic drugs reduces Th17 cell differentiation in vivo. This pathway appears to function by enhancing SUMOylation of the SMAD4 transcription factor by facilitating its nuclear localization [32].

4.4.6 Fevers and B cell functions Changes in body temperature also influence B cell responses but this is largely a secondary consequence of alterations in Th cell functions [33].

4.4.7 Fevers and bacterial diseases Fevers not only influence immune functions, but they also have direct effects on invading microorganisms. Fever-range temperatures may damage microbial proteins, membrane lipids, or RNA and disrupt DNA synthesis, especially in rapidly dividing bacteria [27]. All microbes have an upper-temperature limit above which their viability and growth are impaired. Thus impairment of either of these features is likely to benefit the infected animal by limiting the number of

Endothermy and immunity Chapter | 4

49

invaders that the immune system is required to destroy [3]. The significance of this growth inhibition varies between different microbes although as discussed below, it appears to be especially important in the control of fungal pathogens. Many bacteria also have an optimal replication rate at temperatures below 37 C. Bacteria readily cause fevers by releasing what has been characterized as exogenous pyrogens. Signaling through toll-like receptors such as TLR4, these bacterial components trigger the release of the cytokines that act on the thermoregulatory centers of the brain. In addition, some pathogenic bacteria are highly temperature sensitive and cannot survive at fever-range temperatures. One such example is Treponema pallidum, the cause of syphilis. In 1927, an Austrian doctor, Julius WagnerJauregg, won the Nobel Prize for Medicine for curing insanity! Patients with late-stage syphilis develop “general paralysis of the insane” as a result of severe brain damage. Wagner-Jauregg observed that syphilis patients who also had malaria developed high fevers and improved clinically. Therefore he infected his neurosyphilitic patients with live Plasmodia. This malaria therapy worked! The malaria parasites triggered an innate immune response, and the patients developed a very high fever, and the high body temperature killed T. pallidum, the cause of neurosyphilis. It resulted in full remission in about 30% of cases. It can be argued that simply increasing body temperature beyond the thermotolerance limits of a microbe will not always work. Microbes adapt much faster than vertebrates and as a result, will be rapidly selected for survival at higher temperatures. They will eventually become tolerant of temperatures achieved during fevers.

4.4.8 Fevers and viral diseases Viruses cause fevers, such as yellow fever, and viral hemorrhagic fevers. They too act by stimulating the TLRS and other intracellular pattern recognition receptors. Febrile temperatures enhance type I interferon responses. These type 1 interferons can cause a fever in addition to muscle pain as well as “flu-like symptoms.” It is also apparent that the antiviral effects of type 1 interferons are enhanced at higher temperatures. Interferons appear to be less active at lower temperatures, especially in chilled limb extremities [34].

4.5

Fevers, fungi, and the rise of the mammals

4.5.1 Fungi and endothermy In general, mammals, unlike ectotherms, are very resistant to systemic fungal infections. While there are about 1.5 million known species of fungi, only about 150 can cause disease in mammals. Very few of these cause systemic disease. It has been suggested that this antifungal resistance is linked to mammalian endothermy and homeothermy [35]. Fungal pathogens generally have low thermal tolerance, and many will not grow at body temperatures. Most human systemic fungal infections occur not in healthy individuals but immunosuppressed individuals such as AIDs patients or those undergoing immunosuppressive cancer therapy. In such cases, fungal commensals are considered opportunistic pathogens. In addition, unlike bacterial or viral infections, systemic fungal diseases are not usually contagious. Most fungal infections tend to be superficial such as Malassezia yeast infections in cases of atopic dermatitis. Skin is also cooler than the deep body. This resistance to systemic mycoses has been attributed not only to an effective immune system but also to a raised body temperature. This theory suggests that mammalian body temperature effectively excludes potential fungal pathogens that cannot survive at such a high temperature. Mammalian resistance to such fungal infections is much greater than for example, fish or amphibians, or even invertebrates. Frogs appear to be especially vulnerable to Chytrids (Batrachochytrium dendrobatidis). It is suggested that their lower body temperature makes them much more vulnerable to fungal invasion. Frogs can be cured of chytridiomycosis by raising their body temperature to 37 C [36]. While most mammals have an effective body temperature in the region of 36 C
38 C, the duck-billed platypus (O. anatinum) has a lower-body temperature
around 32 C. As a result, the platypus is susceptible to infection with a frog fungal pathogen, Mucor amphiborium that cannot survive at temperatures higher than 36 C and thus cannot infect hotter mammals [37]. The fungus causes severe ulcerative skin disease in the platypus. Another mammalian example occurs in bats. Over the summer months, bats are active and maintain their body temperature within the normal mammalian range. However, they hibernate over winter and their core temperature drops to 10 C
20 C. As a result, they become vulnerable to infection with the fungus that causes White-nose syndrome, Geomyces destructans [38]. In this disease, white fungal growth develops on their muzzles, ears, and wing membranes. It kills the bats by blocking nasal passages and causing extensive skin ulceration. This syndrome has killed over 75% of bats in some North American colonies. If appropriately treated, and the temperature of the bats allowed to rise by housing them at increased

50

SECTION | 1 Mammalian immunology

temperatures, fungal growth is inhibited, and affected bats may recover. White-nose syndrome also confirms that fungal diseases have the potential to cause the extinction of vulnerable species. It is of relevance to note that since fungi lose their ability to grow as body temperature increases, and that such endothermy has significant energy costs, the optimal body temperature to prevent fungal growth but not incur excessive energy costs can be determined. The calculated optimum is 36.7 C, a figure that is remarkably close to average mammalian body temperatures [39]. This temperature can prevent the growth of most fungal species while at the same time minimizing excessive metabolic expenditure. Given the importance of endothermy and homeothermy in resisting fungal invasion, it has been speculated that this may have played a key role in the survival of mammals and the extinction of the dinosaurs after the K-Pg event. It has been suggested that as a result of their elevated body temperature, mammals were protected against, and so survived a wave of fungal infections following the K-Pg event at the end of the Cretaceous period. It is currently believed that a massive body crashed into the earth near Chicxulub in the Yucatan peninsula in Mexico about 66 mya. The resulting tidal waves, clouds of dust, and sulfur aerosols resulted in darkened skies and global cooling that lasted for at least six months. This would have shut down photosynthesis while the cool humid climate may have persisted for several years. It would have resulted in massive global deforestation. This massive destruction of plants would have generated huge quantities of decomposing material that would have resulted in extreme fungal proliferation. Analysis of plant material in coal deposits associated with the K-Pg event has shown that there was an enormous increase in fungal spore counts at that time indicating that this “composting” was accompanied by a global fungal bloom. The fungal bloom would likely have lasted for only a few years before photosynthesis returned to normal [40]. Casadevall has suggested that this extraordinary fungal bloom could have directly affected ectotherms such as some of the dinosaurs [35]. Stressed, starving, and cold animals would have been unable to raise their body temperature in response to infection (behavioral fever) and as a result, would have been selectively infected and eliminated as a result of fungal infections. Mammals would have been much more resistant to these fungi. Cold dinosaur eggs would have been vulnerable to fungal invasion while viviparous mammals would have been able to survive. It is perhaps no coincidence that those other endotherms, the birds, also survived the K-Pg mass extinction event.

References [1] Mackowiak PA. Concepts of fever. Arch Intern Med 1998;158:1870
81. [2] Cannon B, Nedergaard J. Brown adipose tissue: function and physiological significance. Physiol Rev 2004;84:277
359. [3] Casadevall A. Thermal restriction as an antimicrobial function of fever. PLoS Pathog 2016;12(5):e1005577. Available from: https://doi.org/ 10.1371/journal.ppat.1005577. [4] Kemp TS. The origin of mammalian endothermy: a paradigm for the evolution of complex biological structure. Zoo J Linn Soc 2006; 147(4):473
88. [5] Jastroch M, Seebacher F. Importance of adipocyte browning in the evolution of endothermy. Phil Trans Roy Soc B 2020;. Available from: https://doi.org/10.1098/rspb.2019.0134. [6] Grigg GC, Beard LA, Barnes JA, Perry LI, et al. Body temperature in captive long-beaked echidnas (Zaglossus bartoni). Comp Biochem Physiol Pt A 2003;136:911
16. [7] Grant TR. Body temperatures of free-ranging platypuses, Ornithorhynchus anatinus (Monotremata) with observations on their use of burrows. Aust J Zool 1983;31(2):117
22. [8] Dawson TJ, Hulbert AJ. Standard metabolism, body temperature, and surface areas of Australian marsupials. Am J Physiol 1970;218 (4):1233
8. [9] Hulbert AJ, Dawson TJ. Standard metabolism and body temperature of perameloid marsupials from different environments. Comp Biochem Physiol 1974;47A:583
90. [10] Johansen K. Temperature regulation in the nine-banded armadillo (Dasypus novemcinctus mexicanus). Physiol Zool 1961;34(2):126
44. [11] Bakker RT. Dinosaur physiology and the origins of the mammals. Evolution 1971;25(4):636
58. [12] Oelkrug R, Goetze N, Exner C, Lee Y, et al. Brown fat in a protoendothermic mammal fuels Eutherian evolution. Nat Comm 2013;. Available from: https://doi.org/10.1038/ncomms3140. [13] Cypess AM. Reassessing human adipose tissue. N Engl J Med 2022;386(8):768
79. [14] Gaudry MJ, Jasroch M, Treberg JR, Hofreiter M, et al. Inactivation of thermogenic UCP1 as a historical contingency in multiple placental mammal clades. Sci Adv 2017;3:e1602878. [15] Garcia MC, Pazos P, Lima L, Dieguez C. Regulation of energy expenditure and brown/beige thermogenic activity by interleukins: new roles for old actors. Int J Biomed Sci 2018;. Available from: https://doi.org/10.3390/ijms19092569. [16] Kohlgruber AC, Gal-Oz ST, LaMarche NM, Shimakazi M, et al. gd T cells producing interleukin-17A regulate adipose regulatory T cell homeostasis and thermogenesis. Nat Immunol 2018;19(5):464
74.

Endothermy and immunity Chapter | 4

51

[17] Liang Y, Wang X, Wang H, Yang W, et al. IL-33 activates mTORC1 and modulates glycolytic metabolism in CD81 T cells. Immunology 2022;165:61
73. [18] Faas M, Ipseiz N, Ackermann J, Culemann S, et al. IL-33-induced metabolic reprogramming controls the differentiation of alternatively activated macrophages and the resolution of inflammation. Immunity 2021;54:2351
546. [19] Walter EJ, Hanna-Jumma S, Carreretto M, Forni L. The pathophysiological basis and consequences of fever. Crit Care 2016;. Available from: https://doi.org/10.1186/s13054-016-1375-5. [20] Papotto PH, Siva-Santos B. Got my γδ17 T cells to keep me warm. Nat Immunol 2018;19:427
9. [21] Logan ML. Did pathogens facilitate the rise of endothermy? Ideas Ecol Evol 2019;12:1
8. [22] Sommer F, Stahlman M, Ilkayeva O, Arnemo JM, et al. The gut microbiota modulates energy metabolism in the hibernating brown bear ursus arctos. Cell Rep 2016;14:1655
61. [23] Bohr M, Brooks AR, Kurtz CC. Hibernation induces immune changes in the lung of 13-lined ground squirrels (Ictidomys tridecemlineatus). Dev Comp Immunol 2014;47:178
84. [24] McFarland R, Henzi SP, Parrett L, Bonnell T, et al. Fevers and the social costs of acute infection in wild vervet monkeys. Proc Natl Acad Sci USA 2021;118(44):e207881118. [25] Lopes PC. When is it socially acceptable to feel sick? Proc R Soc B 2014;. Available from: https://doi.org/10.1098/rspb.2014.o218. [26] Lochmiller RL, Deerenberg C. Trade-offs in evolutionary immunology: just what is the cost of immunity? Oikos 2000;88:87
98. [27] Wrotek S, LeGrand EK, Dzialuk ZA, Alcock J. The meaning of fever in the pandemic era. Evolution Med Public Health 2021;. Available from: https://doi.org/10.1093/emph/eoaa044. [28] Evans SS, Repasky EA, Fisher DT. Fever and thermalregulation of immunity: the immune system feels the heat. Nat Rev Immunol 2015;15(6): 335
49. Available from: https://doi.org/10.1038/nri3843. [29] O’Sullivan D, Stanczak MA, Villa M, Uhl FM, et al. Fever supports CD8 1 effector T cell responses by promoting mitochondrial translation. Proc Natl Acad Sci USA 2021; 118(25): e2023752118. [30] Lin C, et al. Fever promotes T lymphocyte trafficking via a thermal sensory pathway involving heat shock protein 90 and α4 -integrins. Immunity 2019;50:137
51. [31] Umar D, Das A, Gupta S, Chattopadhyay S, et al. Febrile temperature change modulates CD4 T cell differentiation via a TRPV channelregulated Notch-dependent pathway. Proc Natl Acad Sci USA 2020;117(36):22357
66. [32] Wang X, Ni L, Wan S, Zhao X, et al. Febrile temperature critically controls the differentiation and pathogenicity of T helper 17 cells. Immunity 2020;52:328
41. [33] Hanson DF. Fever, temperature and the immune response. Ann NY Acad Sci 1997;813:453
64. [34] Prow NA, Tang B, Gardner J, Le TT, et al. Lower temperatures reduce type 1 interferon activity and promote alphaviral arthritis. PLoS Pathog 2017;. Available from: https://doi.org/10.1371/journal.ppat.1006788. [35] Casadevall A. Fungi and the rise of the mammals. PLoS Pathog 2012;8(8). Available from: https://doi.org/10.1371/journalppat.1002808. [36] Fites JS, Ramsey JP, Holden WM, et al. The invasive Chytrid fungus of amphibians paralyzes lymphocyte responses. Science 2013;342:366
70. [37] Obendorf DL, Peel BF, Munday BL. Mucor amphibiorum infection in platypus (Ornithorhynchus anatinus) from Tasmania. J Wildl Dis 1993;29(3):485
7. [38] Blehert DS, Hicks AC, Behr M, Meteyer CU, et al. Bat white nose syndrome: an emerging fungal pathogen? Science 2009;323:227. [39] Bergman A, Casadevall A. Mammalian endothermy optimally restricts fungi and metabolic costs. mBioi 2010;1(5):e00212-10. Available from: https://doi.org/10.1128/mBio.00212-10. [40] Vajda V, McLoughlin S. Fungal proliferation at the cretaceous-tertiary boundary. Science 2004. Available from: https://doi.org/10.1126/ science.1093807.

This page intentionally left blank

Chapter 5

The microbiota
immune system relationship Mammals are metagenomic. This means that they contain not only their species-specific mammalian genes but also the genes of all the microbes that live in and on their bodies. The evolution and development of the immune system have been profoundly influenced by the presence of billions of bacteria that constitute the body’s microbiota, especially within the gastrointestinal tract. Mammals possess about 20,000 protein-encoding genes while their microbiota may collectively possess at least another 10 million. As a result, the adaptive abilities of mammals are determined not only by their genomes but also by the genetic repertoire of their associated commensal population. Every mammal that has ever lived has been obliged to accommodate a large and diverse commensal microbiota. Mammals have adapted to the presence of these organisms and indeed, have evolved specific organs such as the rumen or cecum to house them and exploit their beneficial symbiotic activities. The composition of the mammalian microbiota is affected by multiple factors including both phylogeny and diet. In effect, the microbiota has co-diversified with their hosts. This is important since it is also abundantly clear that the microbiota also plays a key role in the development and proper functioning of the immune system. Some commensal bacterial clades are found across many different mammalian genera while others are conserved within specific lineages. These conserved clades are linked to mammalian phylogeny. They co-diversify with their hosts and are often vertically transmitted, or their abundance may be determined by the host genotype. Their evolutionary patterns are consistent with natural selection [1]. The great majority of the body’s microbiota live within the lower intestinal tract. Here they provide metabolic functions such as the digestion of cellulose that releases otherwise unavailable energy. The development of new dietary habits is one of the major drivers of the evolution of new species. This has impacted the evolution of mammals. Thus much of the modern diversification of mammals occurred in the quaternary period [1.8 million years ago (mya)] when, as a result of a drop in CO2 levels and climate change, there was a massive expansion of C4-grasslands. This new food source favored the expansion of herbivores with high crowned teeth and much longer gut retention times enabling the less digestible grasses to be effectively degraded by their microbiota. Thus these changes in diet and behavior also resulted in significant changes in the intestinal microbiota. In general, the earliest mammals are believed to have been primarily carnivores and insectivores. However, the modern gut microbiota did not evolve directly from carnivores. Much of the herbivore microbiota was probably acquired independently from their forage [2]. About 80% of modern mammals are herbivores and herbivores are present in most mammalian lineages. It is clear therefore that diet is the major factor that influences the intestinal microbial community. (This is debated by some, and it depends on scale. Phylogeny plays an important role as well, perhaps at the species level). Studies by Ley et al. have indicated that while herbivore microbiotas consist of about fourteen different bacterial phyla, omnivores contain about twelve, and carnivores have only six phyla. Diet is thus the dominant predictor of microbiome diversity [2]. Once mammals adapted to a plant-based diet it vastly increased the calories available to them and permitted population growth and subsequent radiations. By harnessing the diverse genomes present in the microbiota, herbivorous mammals enhanced their metabolic potential and generated new pathways to obtain food calories. Thus they increased their ability to extract energy from plant structural carbohydrates and to obtain essential vitamins and minerals. To obtain the maximal nutrition from a plant-based diet, especially from complex structural oligosaccharides such as cellulose, gut retention times had to be lengthened to permit bacterial fermentation. This retention occurred either in the foregut as in ruminants or in the hindgut as in most other herbivores. The microbiota of herbivores generally clusters into two distinct groups corresponding to foregut and hindgut fermenters. Thus foregut fermenters such as sheep, Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00001-0 © 2023 Elsevier Inc. All rights reserved.

53

54

SECTION | 1 Mammalian immunology

kangaroo, okapi, giraffe, and cattle form one group. The hindgut fermenters such as elephant, horse, rhinos, capybara, mole-rat, and gorilla cluster in another. As a result of these changes in their diets, herbivores were also obliged to develop a collaborative relationship with the microbes in their environment [3]. As a result, they evolved systems dependent upon the presence of the microbiota for their optimal efficiency while they also regulate the composition and behavior of the microbial populations. The surfaces of the animal body consist of many stable, nutrient-rich ecosystems where microbes can thrive. Each surface that connects to the exterior is populated by enormous numbers of bacteria, archaea, fungi, and viruses. The bacteria are the best studied of these. Thus bacteria live on the skin, in the respiratory tract, in parts of the genitourinary tract, and sometimes within the body but mainly within the gastrointestinal tract. It has been estimated that in an animal body, at least half of all the cells are microbial [4]. As a result of their life-long, intimate association with body surfaces, the microbiota can be considered to be an integral part of the body—“a virtual organ.” As such, they influence both innate and adaptive immunity and conversely, they are influenced by signals generated by the host immune system. This has given rise to the concept that mammals and their microbiota together form “superorganisms” that share nutrition and exchange energy and metabolites and whose complex interactions are regulated in large part by immune mechanisms.

5.1

Herbivores

The large herbivorous mammals have some additional complexities associated with their diet and lifestyle. For example, domestic herbivores contain massive amounts of microbial material in their rumen and large intestine. These reflect the major role of the microbiota—providing nutrition by extracting energy from complex, plant-derived polysaccharides such as celluloses. Carbohydrate digestion is a primary function of the gastrointestinal microbiota in most herbivores. Many members of the microbiota are optimized to ferment oligosaccharides and thus generate energy-rich short-chain fatty acids (SCFAs). The fatty acids in turn can be used as energy sources by other, more specialized bacteria. It is estimated that in an omnivore such as a human, as much as 10% of daily energy needs are provided by colonic fermentation. This figure is much higher for obligate herbivores such as ruminants, camels or rabbits. Microbial metabolism permitted mammals to adapt to otherwise noncompetitive lifestyles. For example, mice with a conventional microbiota need to consume 30% fewer calories than do germ-free mice to maintain their body weight. The microbiota can extract much more energy from food and it is no coincidence therefore that most extant mammals are herbivores.

5.2

Carnivores

The gastrointestinal tract of carnivores is less complex than that of herbivores given that the energy from ingested muscle or adipose tissue is more readily available and thus they require a less diverse microbiota. It should however be pointed out that carnivores will inevitably consume parts of the gastrointestinal tracts of their prey. Thus they will also ingest many of the organisms from their prey microbiota. This is not necessarily a bad thing since few of the ingested organisms are likely to be significant pathogens. That said, many carnivores are not fussy about who they eat, and they generally prefer to catch and eat the easiest prey available. Thus sick and injured animals are disproportionately more likely to be caught and eaten. An effective vomiting reflex will confer some protection from “food poisoning.” However, carnivores and their microbiota have evolved in such a way that they are relatively resistant to ingested pathogens. Thus defensive mechanisms such as highly acidic stomach contents and the production of multiple defensins by intestinal Paneth cells play important roles in maintaining gut homeostasis in these species.

5.3

The microbiota
immune relationship

It is now well accepted that the intestinal microbiota also contributes to local host defenses and has effects on the immune system that extends throughout the body. Nutrients, antigens, and microbial metabolites are continually released into the body where they influence immune cell and inflammatory pathways. The immunological defenses on body surfaces are therefore faced with the task of co-existing with the microbiota while at the same time controlling any inadvertent or opportunistic invasion through breaks in the epithelial barriers [5]. As a result, the gut microbiota has greatly influenced the evolution of key traits in both the innate and adaptive immune systems. Conversely, the adaptive immune system evolved in part, to maintain control of the dense and diverse microbiota. The intestinal environment also plays a role in regulating the microbiota. Thus the ability of a bacterium

The microbiota
immune system relationship Chapter | 5

55

to colonize the gut will depend on pH, oxygen tension, peristalsis rate, metabolite concentrations, and of course the presence of immunoglobulin A (IgA) and possibly IgD. The immune system and the microbiota interact in three important ways. First, the immune defenses are important in keeping the microbiota within limits and constraining any tendency to overgrowth or tissue invasion. Second, the symbionts can outcompete potential pathogens and thus serve a protective function. Third, signals from the symbionts influence the proper development of the immune system. Selection for host control of the microbiota has been required for the optimization of diets and the emergence of foregut fermenters. Likewise, the evolution of the immune systems has been driven in part by the need to edit the composition of the gut microbiota in order to maximize immune effectiveness. Thus there has been a need for the immune systems to eliminate or exclude organisms that have adverse effects on survival while at the same time tolerating or even stimulating the growth of those organisms that are beneficial. Certainly, studies on germ-free animals reveal numerous deficiencies in their adaptive immune responses (Fig. 5.1). Likewise inappropriate immunological or inflammatory attacks on the microbiota can result in digestive diseases such as celiac disease or ulcerative colitis. Loss of toll-like receptors (TLRs) can also result in adverse changes in the gut microbiota [6]. The heritability of the microbiota depends largely on how the bacteria are transmitted, - vertically or horizontally. Vertically transmitted organisms from the mother usually arrive first and occupy their preferred niches. Horizontally acquired organisms from the environment arrive later and may have a harder time establishing themselves [7]. Horizontal transmission is the predominant transmission mode of bacteria to mammals. Their establishment and survival will depend upon the host’s immune responses. Generally, mammals use innate mechanisms to distinguish between symbionts and pathogens through the detection of pathogen-associated molecular patterns (PAMPs). Once established, innate immune mechanisms then regulate these organisms and if required, segregate them to specific host sites. This is especially the case in the gut lumen where the microbiota is separated from the epithelial cells by a layer of mucus containing antimicrobial defensins and antibodies of the IgA class. In this case, a combination of adaptive and innate immunity is required. Immune system genes are among the most rapidly evolving in mammalian genomes. While much of this is attributed to constant threats from potential pathogens—the “red-queen” process; it is also due to alterations in the gut microbiota in response to environmental and dietary changes. Regulation of the composition of the millions of organisms in the gut microbiota is a very complex and continuous task. The microbiota will amplify any environmental signals affecting host fitness. As a result, some mammals may become “evolutionarily addicted” to certain specific gut microorganisms. Except in rare cases, the same bacterial phyla tend to predominate in mammalian intestines across each class depending upon their diet—herbivore, carnivore, or omnivore. Mammalian evolutionary history also affects the composition of the mammalian gut microbiome. One major exception is the bats (Chiropterans) whose microbiota more closely resembles that of small birds than small mammals. The bat microbiota consists predominantly of Proteobacteria, and they tend to be similar to other bats regardless of diet. The bats have a great diversity of diets including nectar eaters, carnivores, blood drinkers, fruit eaters, and insectivores. These do not appear to correlate well with the composition of their microbiota. Thus in bats, there is a very weak concordance with phylogenetic history [8]. In other mammal lineages, the microbiota tends to be associated with a specific lineage as determined by diet. Thus there are strong phylogenetic influences on the composition of the microbiota. Despite very different diets and physiology, the mammalian microbiota appears to have diverged at roughly the same rate over the past 75 million years. As taxa diverge so too do Decreased frequency of CD4 and CD8 T cells Altered cytokine polarization - Th2 bias Decreased expression of IL-22 Thymic abnormalities Small Peyer’s patches. Lymph nodes and spleen lack germinal centers

Reduction in DE T cell diversity Defective production of IgA and IgG1 Hypogammaglobulinemia

Reduced intraepithelial lymphocytes in the intestine

FIGURE 5.1 In the absence of a commensal microbiota, the adaptive immune system fails to develop properly. Animals such as mice and pigs can be raised for extended periods of time in “germ-free” isolators. As a result, significant differences emerge between the structure and function of the immune systems of conventional and germ-free animals. These differences point to the essential contribution that the microbiota makes to normal immune development.

56

SECTION | 1 Mammalian immunology

their microbiota. This rate is approximately 2% of the shared 97% of operational taxonomic units lost every 10 million years. However, the gut microbiota of the Artiodactyla has evolved at a faster rate while conversely, that of the Chiropterans has evolved much more slowly. Nevertheless, even within individuals such as mice and humans, changes in diet can bring about rapid alterations in microbial community structure [9]. The association between diet and phylogeny occurs across all the mammalian gut communities. Dietary adaptation is the major driving force for their composition and convergence. Transitions to herbivory appear to have had especially large effects (with the notable exception of the giant panda that although an herbivore, possesses a gut microbiota that is closely related to that of other carnivorous and omnivorous bears) [10]. Thus mammals with herbivorous diets tend to have similar microbiota. Insectivory and myrmecophily are also associated with microbiota convergence across distantly related species such as anteaters, aardvarks, and aardwolves [11]. Within the cetaceans, the baleen whales contain bacterial populations that can degrade chitin, the major constituent of krill exoskeletons. These changes in the cetacean microbiota may be facilitated by the presence of the artiodactyl chambered stomach. Despite their specialized diet, their microbiota shares many similarities with the terrestrial foregutfermenting herbivores. The human microbiome has rapidly diverged from that of the great apes over a relatively short time period. Recent changes in human lifestyles such as the switch to cooked foods, the advent of agriculture, urban living, a higher meat intake, and high-fat diets have also resulted in significant changes. These changes have had secondary immunological effects including an increased prevalence of allergic and autoimmune diseases. Despite the above, certain gene losses are associated with an obligate carnivorous diet. Cetaceans such as the dolphins and baleen whales appear to have lost the gastrointestinal host defense gene NOX1. Thus NOX1 is a transmembrane NADP oxidase that generates reactive oxygen species in the colonic epithelium. It is activated in response to microbial invasion thus killing pathogens and triggering the wound healing process [12]. Its loss is possibly related to decreased gut microbiome diversity in these marine mammals. N-glycolylneuraminic acid (Neu5Gc) is a form of sialic acid found on the surface of many cells in most mammals but not in humans. The human-specific loss is due to a mutation in the Cytidine monophospho-N-acetylneuraminic hydroxylase (CMAH) gene estimated to have occurred around 2
3 mya. As a consequence, there is an increased abundance of its precursor N-acetylneuraminic acid expressed in human cells. In addition, most humans have autoantibodies directed against Neu5Gc [13]. Other mammals that lack Neu5Gc include the mustelids such as ferrets, mink, and their relatives. The origin of these antibodies in humans is unclear since most bacteria cannot make this glycan. However, the mustelids consume a diet of small mammals that contain large amounts. Other mammals that lack Neu5Gc include New-world primates, pinnipeds, raccoons, hedgehogs, some bats, toothed whales, white-tailed deer, and monotremes. While the CMAH gene has clearly been lost on multiple occasions, the benefits of such a loss are unclear.

5.4

The location of the microbiota

5.4.1 Skin Normal mammalian skin harbors trillions of microorganisms. It has been estimated that up to a billion bacteria may live on a square centimeter of human skin. These are found on the keratinocyte surface and extend deep into sebaceous glands and hair follicles. The skin is not, however, a hospitable environment. The outer keratinocytes are constantly shed and replaced by new cells from below. It is cool in some areas and hot in others. Haired skin is a very different environment from that at mucocutaneous junctions. Some areas of skin may be very dry, have a high salt content, and are nutrient-poor. Other areas may be moist but bathed in a complex mixture of proteases, lysozyme, and antimicrobial peptides such as β-defensins and cathelicidins. The skin microbiota of dogs varies greatly between different individuals and different skin sites [14]. The highest microbial diversity was found in the axilla and the dorsum of the nose. On average about 300 different bacterial species have been identified on the dorsal canine nose [15] (Fig. 5.2). One significant factor that influences the diversity of the skin microbiota is the presence or absence of fur. As a result, human skin has a less diverse microbiota than other mammals [15]. The composition of the skin microbiota shows similarities between phylogenetically related mammals again suggesting the occurrence of coevolution. For example, there is significant congruence between the skin microbiota of cattle and horses [15]. The skin microbiota can be divided into a resident population that is relatively stable and consistent—a true commensal population, and a population of transient bacteria that may only persist for hours or days. Both populations can contain a mixture of commensals and potential pathogens, yet invasion and disease are relatively uncommon. Large populations of Proteobacteria and Oxalobacteriaceae predominate. The precise composition of this microbiota also

The microbiota
immune system relationship Chapter | 5

57

FIGURE 5.2 The diverse composition of the microbiota of the skin of the dog. From Rodrigues Hoffmann A, Patterson AP, Diesel A, Lawhon SD, et al. The skin microbiome in healthy and allergic dogs. PLoS One 2014 8;9(1):e83197, doi: 10.1371/journal.pone.0083197.

depends on the location (hairy, wooly, or bald skin; back vs skin in the axilla or groin or ear) and the presence of ectoparasites. There is also great variation between individual animals. In mice, it has been shown that the skin microbiota influences local inflammatory and T cell responses. Epidermal Th17 and CD81 T cells are especially affected. The microbiota controls the balance between effector and regulatory T cells within skin tissue. They influence keratinocyte production of IL-1 and its effects on epidermal dendritic cells and thus control local T cell responses. Skin bacteria can also activate antigen-specific T cells across the intact epithelium. However, the presence of Treg cells in neonatal skin mediates tolerance to commensal bacteria at a time when the microbiota are becoming established [16].

5.4.2 The respiratory tract Like all external surfaces, the upper respiratory tract of mammals houses a dense and complex microbiota. It has been calculated that a human inhales 105 organisms/day just breathing normally. Many nasal bacteria are also found on the skin while others are common environmental bacteria. Deeper in the airways, Neisseria and Gram-negative cocci are common [17]. Healthy lungs harbor a complex microbiota, closely related to, but much less dense than that found in the upper respiratory tract. The bronchi contain about 2000 bacterial genomes per cm2. Lung tissues contain between 10 and 100 bacterial cells per 1000 lung cells. These include both aerobes and anaerobes and like other surfaces, the populations differ greatly between individuals. These organisms generally live within the mucus layer and include bacteria, yeasts, and viruses including bacteriophages. Pathobionts are also present and may induce disease in immunodeficient individuals. The best example of these is the fungi of the genus Pneumocystis that can cause lethal pneumonia in immunosuppressed humans. In the absence of the microbiota, the airways are prone to mount exaggerated Th2 responses. The presence of a microbiota induces Treg activity that suppresses this. This probably explains the protective effects of inhaled microbial antigens (as in a farming environment) on the development of allergies. Increased dietary fiber also has a protective effect on allergic airway inflammation in mice as a result of increased levels of circulating SCFAs.

58

SECTION | 1 Mammalian immunology

The intestinal microbiota also regulates pulmonary adaptive responses. Thus segmented filamentous bacteria (SFB) in the intestine can influence pulmonary immunity to bacteria and fungi, probably through their generation of retinoic acid [18]. Conversely, influenza infection in the lungs generates type I interferons. These in turn induce changes in the gut microbiota such as depletion of obligate anaerobic bacteria and an increase in Proteobacteria leading to intestinal dysbiosis.

5.4.3 The genitourinary system In adult female mammals, the healthy cervicovaginal microbiota is usually dominated by lactobacilli and other lactic acidproducing bacteria. The vagina is also lined by a squamous epithelium composed of cells rich in glycogen. When these cells desquamate, the glycogen provides a substrate for the lactobacilli that, in turn, generate large quantities of lactic acid. This reduces the pH to a level that protects the vagina against invasion by many pathogenic bacteria and yeasts. Glycogen storage in the vaginal epithelial cells is stimulated by estrogens and thus occurs only in sexually mature animals.

5.4.4 The gastrointestinal tract The gut microbiota is a complex community containing bacteria, archaea, fungi, and viruses. The most obvious of these are trillions of bacteria belonging to hundreds of different species. It has been estimated that the canine small intestine harbors more than 200 different bacterial species while the canine and human colons may house over a thousand species each [19]. Each mammal’s microbiota is unique, and its composition is determined by diet, structural, genetic, and environmental factors. The composition of the microbiota also changes along the gastrointestinal tract under the influence of nutrient availability and the local microenvironment. In most mammals, they are dominated by two phyla, the Firmicutes and the Bacteroidetes, with lesser numbers of Actinobacteria and Proteobacteria and many minor phyla such as the Fusobacteria and the Verrucomicrobia. The Firmicutes consist of mainly Gram-positive bacteria. Many are sporeforming. Important members include the Clostridia which may be beneficial or pathogenic. They also include potentially pathogenic Streptococci and Staphylococci. The Actinobacteria are also Gram-positive bacteria with different G 1 C content than the Firmicutes. The Bacteroidetes are Gram-negative bacteria that ferment indigestible plant carbohydrates to produce SCFAs. The Proteobacteria contain Gram-negative enterobacteria such as Escherichia coli and Klebsiella. Bacterial counts in the canine and feline duodenum are in a range from 102 to 109 per gram of content. In the colon, the count ranges from 109 to 1011 colony-forming units/g. The human colon is estimated to contain 3.8 3 1013 bacteria [4].

5.4.4.1 Foregut fermenters The surface of the rumen is covered by stratified squamous epithelium. Thus its defenses have more in common with skin rather than with the rest of the gastrointestinal tract. Although ruminal epithelium is largely leak-proof, the presence of such a large source of microbial antigens suggests that some provision must be made for defense against leakage. Disturbances in rumen metabolism, often caused by feeding very high-energy diets result in changes in the ruminal microbiota leading to an increase in fatty acid and ethanol production, a drop in rumen pH, and the development of rumen acidosis. This, in turn, results in local inflammation, the opening of intercellular junctions, and a disruption in the barrier function of the ruminal squamous epithelium. This can permit bacterial PAMPs such as endotoxins, flagellae, and other microbial products to cross the ruminal wall and enter the bloodstream. This may lead to endotoxemia, a systemic innate immune response, an acute-phase response, and prolonged systemic inflammation. While much is known about the relationships between the intestinal immune system and lymphoid tissues in simplestomached animals, much less is known about the interactions between the rumen and the immune system. The ruminal lymphatics drain to many different lymph nodes. TLRs such as TLR4, and cytokines such as IL-1β, IL-10, and caspase-1 can be found in the ruminal walls suggesting that inflammation can readily occur here. Interferon-γ is found in ruminal contents as are T and B cells. It appears, that the ruminal microbiota may communicate with its associated lymphoid tissues and so promote regulatory responses. There is little evidence that the ruminal microbiota directly shapes the development of the immune system, but it has been suggested that the complex innate immune systems of ruminants, especially their many diverse antimicrobial peptides, may have evolved in response to this potential source of microbial invasion.

5.4.4.2 Hindgut fermenters Hindgut fermenters such as the horse are monogastric herbivores that digest fibrous plant material by anaerobic fermentation in the cecum and colon [20]. The SCFAs produced are absorbed through the mucosa. Firmicutes are the main

The microbiota
immune system relationship Chapter | 5

59

bacterial phylum (especially Clostridia), but the predominant organism is a member of the Verrucomicrobia. Other phyla found in the equine large intestine include Spirochetes, Fibrobacteres, Ruminococcus, and Bacteroidetes. Under certain conditions such as carbohydrate overload, the microbiota of the large intestine change drastically as a result of which their pH drops, and bacterial PAMPs escape into the bloodstream this may result in the development of systemic inflammation.

5.5

The functions of the microbiota

5.5.1 Nutritional efficiency The composition of the intestinal microbiota is critically dependent on diet. Thus the microbiota in mammals fed a low fat, high plant polysaccharide diet is very different from that in mammals on high fat, high sugar, and low plant polysaccharide diet. There are great differences in the microbiota of African children compared to European children or in Indian cattle compared to American cattle. Environmentally induced fluctuations in the intestinal microbiota can result in the host adjusting its metabolic and immunologic performance in response to nutritional and environmental changes. The intestinal microbiota also changes during pregnancy. In mice, during the third trimester, there is a reduction in species richness. When transferred to germ-free mice, late pregnancy microbiota-induced greater adiposity and reduced insulin sensitivity. This effect is beneficial in a normal pregnancy since it supports fetal growth and the onset of lactation. The maternal microbiota also drives early postnatal immune development including the potential for young animals to develop type 2 immune responses and hence allergies.

5.5.2 Intestinal protection The microbiota plays a role in protecting animals against invasion by pathogens and prevents the overgrowth of pathobionts. They do this by competing for essential metabolites and nutrients, and by inducing intestinal immune responses [21]. By fully occupying and exploiting the intestinal environment, commensal bacteria outcompete and so block subsequent colonization by less adapted pathogenic bacteria. The microbiota also modifies local environmental conditions by keeping both pH and oxygen tension low. This is also influenced by the diet; for example, the intestine of milk-fed animals contains many lactobacilli that produce bacteriostatic lactic and butyric acids. These acids inhibit colonization by E. coli, so young animals suckled naturally tend to have fewer digestive disturbances than animals weaned early in life.

5.5.3 Development of the immune system The development of the gastrointestinal lymphoid tissues begins well before birth. However, their complete maturation and the generation of IgA-secreting B cells and an intestinal T cell population only occur after birth. The microbiota effectively recruits immune cells to mucosal surfaces and drives the development and organization of many of the major lymphoid organs [22]. It has long been possible to derive animals such as mice or piglets by cesarean surgery and raise them on sterile food within sealed isolators in such a way that they are free of microbes (Fig. 5.1). Compared to conventionally raised animals, these isolator-raised, “germ-free” animals have fewer and smaller Peyer’s patches, smaller mesenteric lymph nodes, and fewer CD41 T cells in the lamina propria of the gut wall. They have fewer intraepithelial T lymphocytes (IELs) within their intestinal epithelium. These IELs have reduced expression of TLR and MHC class II molecules, as well as reduced cytotoxicity. Systemic immune defects are also apparent. Germ-free mice have fewer CD41 T cells in the spleen and fewer and smaller germinal centers as a result of reduced B cell numbers. Their production of macrophages and neutrophils by bone marrow stem cells is impaired. Their immunoglobulin levels are only about 2% of normal. If exposed abruptly to the external microbial environment, they are vulnerable to bacterial infections. The presence of the microbiota is also necessary for the production of tertiary lymphoid structures such as cryptopatches and isolated lymphoid follicles [23]. Mammals have evolved two strategies to generate diverse B cell populations. Thus mice and humans rely mainly on random rearrangements of the immunoglobulin heavy and light chain V, D, and J genes during B cell development within the bone marrow. Other mammals such as cattle, sheep, pigs, and rabbits use an alternative strategy. These species undertake an initial burst of B cell proliferation with limited receptor diversification in utero. These newly produced B cells then migrate to the gut-associated lymphoid tissues after birth where in the second phase of development, they expand both their numbers and their receptor diversity. As described in Chapter 17, in sheep, the ileal Peyer’s

60

SECTION | 1 Mammalian immunology

patch is the site of B cell repertoire expansion while the jejunal Peyer’s patch is the source of antigen-specific IgA responses. In rabbits (Chapter 22) the appendix appears to be the major site of this B cell diversification. The process of microbial-driven B cell diversification and IgA production appears to depend upon the presence of a select subgroup of bacteria within the microbiota. For example, a combination of Bacteroides fragilis and Bacillus subtilis can induce B cell development and VDJ diversification in germ-free rabbits. Neither species alone has this effect suggesting that two signals are needed. Once B cell proliferation and diversification are triggered, the microbiota continues to regulate any additional receptor diversification. It is believed that microbial products trigger these B cell responses by binding to their TLRs and activating NF-κB pathways. Bacterial superantigens might also trigger a polyclonal B cell response and drive the process by preferentially stimulating the production of B cells expressing certain V domains on their antigen receptors.

5.5.4 Regulation of immunity Bacteria, be they on the skin, respiratory tract, genital tract, or intestine, communicate directly and effectively with their host’s immune system. Indeed, this interaction is essential to the proper functioning of the innate and adaptive immune responses. Alterations or imbalances in the microbiota may have profound effects on the functions of the immune system [24]. Dietary plant fibers consist of complex carbohydrates. When digested by Clostridia in the cecum and colon, these complex carbohydrates generate large amounts of SCFAs such as butyrate, propionate, and acetate that suppress macrophage functions and promote the production of intestinal Foxp31 Treg cells (Fig. 5.3). Butyric acid also has antiinflammatory effects since it prevents epigenetic changes by inhibiting histone deacetylases. Butyrate also increases barrier functions by stimulating enterocytes and increasing their transcription of mucin genes, goblet cell differentiation, and mucus production [25]. It can also enhance some bovine neutrophil functions. As a result, high fiber diets play a key role in regulating intestinal inflammation in herbivores and omnivores [26]. COMPLEX PLANT POLYSACCHARIDES (FIBER)

Gut microbiota SCFAs BUTYRATE ACETATE PROPIONATE Formate Lactate Succinate

INCREASED Epithelial integrity Mucus production Retinoic acid Tissue repair Wound healing

DECREASED Inflammation Allergic airway responses

Immune responses IgA production Treg production Respiratory burst Phagocytosis

FIGURE 5.3 The role of short-chain fatty acids in immune regulation. These fatty acids are generated by the digestion of complex carbohydrates and are also a major source of energy in herbivores. They are produced in high quantities by the microbial digestion of high fiber diets. In the absence of adequate fiber intake by omnivores such as humans, immune imbalances resulting in allergic diseases may develop.

The microbiota
immune system relationship Chapter | 5

61

If the intestinal microbiota increases its production of acetate, then this will activate the parasympathetic nervous system and promote glucose-stimulated insulin production [27]. This also stimulates the production of a hunger hormone called ghrelin that enhances appetite leading to increased food intake and obesity. Among the intestinal microbiota, some species play a key role in regulating immune responses. Among the most important are the Clostridial clusters. Several of these bacteria (Clostridia clusters IV, XIVa, and XVIII) specifically induce Treg cells and IL-10 production in the gut. These Clostridia may form a biofilm over the epithelium and enhance the release of transforming growth factor-β (TGF-β) and IDO from enterocytes. They also promote mucus production by goblet cells. In addition to the Clostridia, the capsular polysaccharide-A of B. fragilis triggers IL-10 production through the TLR2-MyD88 pathway. Numerous Clostridial species can trigger IL-10 production by a non-MyD88 pathway involving TGF-β production by intestinal epithelial cells. Other mechanisms such as desensitization of TLRs to bacterial PAMPs and bacterial stimulation of IL-10 and IL-2 production by Treg cells also help minimize inflammation. Additionally, some commensal bacteria actively suppress intestinal inflammation. For example, Lactobacilli and Bacteroides inhibit the innate signaling pathways triggered by TLRs and NLRs. A common commensal, Bacteroides thetaiotaomicron inhibits NF-κB signaling, and intestinal lactobacilli prevent degradation of the NF-κB inhibitor IκB. The presence of Clostridial clusters in the colon also increases the numbers of IL-10-producing Treg cells in distant tissues such as the spleen and lung and they play a role in inhibiting allergic responses. Thus T cells educated by commensal bacteria may emigrate from the gut to remote tissues and determine the body’s T cell balance.

5.5.4.1 Immunity to the microbiota For many years, it was believed that the role of the immune system was simply to ensure the complete exclusion of all invading microbes by distinguishing between self and not-self and eliminating foreign antigens. We now know however that that decision alone is insufficient to ensure optimal health. The immune system must also determine the degree of threat posed by the microbes it encounters and adjust its response accordingly. It must be relatively tolerant of the microbiota or food antigens while, at the same time, being highly responsive to invading pathogens. This discrimination is determined in part by the way in which enteric antigens are processed and the behavior of enterocytes. It is also determined by intestinal T and B cells as well as by the gut microenvironment [28]. (Fig. 5.4) The presence of the intestinal microbiota must either be tolerated or ignored if an animal is to remain healthy. An animal cannot afford to act aggressively towards its own microbiota. The presence of all these bacteria and their products has the potential to trigger massive acute inflammation. But, this inflammation must not happen unless necessary for the defense of the body.

5.5.4.2 Enterocytes The intestinal epithelial cell layer is not simply a barrier. It is a highly responsive tissue that employs innate and adaptive immune cells to restrain the microbiota without triggering unnecessary inflammation but is always ready to activate more potent defensive responses should the need arise. Enterocytes interact with the intestinal microbiota. They can produce peptides that kill or inactivate bacteria and as a result, shape their composition. Enterocytes block access of intact antigens to the lamina propria. They ensure that a balance exists between inflammation and tolerance. They secrete and respond to regulatory cytokines. They display processed antigens to dendritic cells. Within the epithelium and the underlying lamina propria are IELs that upon appropriate stimulation by microbiota-induced IL-1 or IL-23 can regulate their differentiation into effector or regulatory cells [29]. Enterocytes express receptors for many microbial-associated molecular patterns (MAMPs) including TLRs-1, -2, -3, -5, and -9 as well as NOD-2 [30]. When exposed to MAMPs, recruitment of MyD88 and TRIF result in NF-κB and

IgA Glycocalyx

Tight junctions

Antimicrobial LUMEN peptides Outer loose mucus layer Inner dense mucus layer

Paneth cells Plasma cells

Enterocytes

FIGURE 5.4 Some of the many defensive mechanisms protecting the intestinal wall from microbial invasion.

62

SECTION | 1 Mammalian immunology

MAP kinase activation and cytokine synthesis. In practice, bacterial signals trigger the production of some antimicrobial peptides and cytokines but not inflammation. This is because the pattern recognition receptors are not expressed on the luminal side of enterocytes, where they would normally come into contact with the commensals. Instead, they are located at the base of the cells and in intracellular locations. Thus they are only activated if bacteria penetrate the epithelial barrier. By blocking microbial invasion, enterocytes also prevent the development of inflammation within the intestinal wall. Some of the cytokines produced by enterocytes influence the regulatory activities of antigen-processing macrophages and dendritic cells. IL-10 inhibits the TLR-MyD88 pathway, whereas IL-2 inhibits TLR-independent pathways [31]. The antimicrobial peptides found within the inner mucus layer of the gut keep most of the microbiota from contacting the enterocytes and thus ensure that the microbiota remains within the intestinal lumen. Thus they not only protect the host from microbial invasion but also from the potentially harmful inflammatory responses that may result if MAMPs are absorbed into the body.

5.5.4.3 Group 3 innate lymphoid cells Group 3 innate lymphoid cells (ILC3) play an important role in regulating the interactions between the microbiota and its host [32]. They respond to IL-23, IL-1β, and thymic stromal lymphopoietin (TSLP) from dendritic cells by producing IL-17 and IL-22. These attract neutrophils and promote the production of antimicrobial peptides, especially C-type lectins of the Reg3 family [33]. Reg3 lectins interact with the mucus layer to maintain a relatively bacteria-free zone adjacent to the mucosal surface. ILC3 cells can also activate B cells and induce IgA production [34]. They can promote tolerance to food antigens by producing GM-CSF that in turn promotes Treg production. Their production is regulated by the aryl hydrocarbon receptor. The Reg3 family of proteins contain multiple subtypes. Mice have four of these subtypes while humans have only three. In addition, microbiota-generated butyrate suppresses ILC3 cell production in terminal ileal Peyer’s patches [35].

5.5.4.4 B cell functions There is more immune activity in the intestine than in all the other lymphoid tissues combined. It has been estimated that more than 80% of the body’s activated B cells are found in the intestine. Their function is to produce antibodies, specifically IgA to exclude invaders and defend against possible invasion by the microbiota. Although the microbiota are separated by the inner mucus layer and glycocalyx from direct contact with enterocytes, intestinal dendritic cells can extend their processes into the intestinal lumen to sample the microbiota. These bacteria may then persist within the dendritic cells for several days while they are carried to the lamina propria and mesenteric lymph nodes where they are presented to B cells. In addition, some bacteria are taken up by specialized antigencapturing M cells, penetrate the Peyer’s patches, and persist within the tissues. The B cells respond by producing IgA, which modifies the composition of the microbiota and blocks further mucosal penetration. Bacteria are thus prevented from breaching the mucosal barrier by the ongoing IgA response. The mesenteric lymph nodes form an additional barrier that also prevents the commensals from reaching the systemic immune system.

5.5.4.5 Immunoglobulin A and the microbiota The commensal bacteria are continually producing a diverse mixture of metabolites that effectively “educate” the immune system. Many such metabolites are antigenic and hence stimulate mucosal IgA production and secretion. This IgA, in turn, affects the composition of the microbiota while at the same time preventing invasion by pathobionts [36]. Intestinal mucosal plasma cells secrete IgA that is bound to the polymeric immunoglobulin receptor (secretory component) on enterocytes and then exported through them to the gut lumen. Much of this IgA consists of low-affinity antibodies directed against antigenic glycans found on commensals. Organisms targeted by this IgA alter their functions and metabolism as a result of being coated with the IgA. Thus IgA affects their susceptibility to bacteriophages and reduces their motility but also protects them against bile acid toxicity [37]. IgA also acts to both prevent and promote bacterial colonization and in effect, controls the diversity of the microbiota. However, these IgA-microbial interactions may be disrupted by abrupt dietary changes as well as by under- or overnutrition [38]. One factor that appears to directly affect the IgA response is the presence of acetate, one of the fatty acids generated by digestion of a high fiber diet. Acetate promotes the production of IgA in the colon and affects the ability of IgA to bind to certain enterobacteria. The acetate appears to act by activating a subset of Th2 cells that promote T-cell-dependent IgA production [26]. While the production of large quantities of low-affinity IgA has a direct effect on the diversity of the microbiota, the reverse is also the case. Thus newborn mice are born without a significant microbiota and rely on their mother as

The microbiota
immune system relationship Chapter | 5

63

the primary source of organisms for intestinal colonization. This early colonization of the newborn gut appears to trigger intestinal B cells to produce large quantities of low-affinity IgA. This stimulus requires help from either T cells or ILC3 cells to induce the B cell response [39]. Some of this “innate” IgA enables selected organisms to enter Peyer’s patches, be processed, and trigger a positive feedback loop of IgA production.

5.5.4.6 The gut-mammary axis The gut microbiota act on the intestinal lymphoid tissues such as Peyer’s patches to drive IgA production by B cells. Thus the IgA1 B cell system is constantly responding to the microbiota. During pregnancy, many of these responding B cells emigrate from the Peyer’s patches, circulate, and eventually colonize the developing mammary gland. They do this by expressing the chemokine receptor CCR10 that binds its ligand, CCL28 produced by the mammary epithelial cells. These IgA1 B cells colonize the gland and subsequently produce the IgA that is found in milk. As a result, suckling newborns ingest large amounts of IgA. As in the maternal intestine, this IgA has a selective role. Some of the IgA is directed specifically against enteric pathogens and is protective, while IgA molecules of other specificities can also promote the growth of beneficial commensals. In mice, organisms such as Bacteroides acidfaciens and Prevotella buccalis appear to be indispensable for programming this maternal IgA synthesis within the Peyer’s patches. Thus the intestinal microbiota is directly responsible, for the production of specific IgA in milk [40] (Fig. 3.3).

5.5.4.7 IgD and the microbiota Recent studies on the role of IgD have indicated that B cell class switching from IgM to IgD in mice is initiated by the intestinal microbiota! In this species, much IgD is produced by the B cells of the intestinal mucosa-associated lymphoid tissues. However, it is not produced in germ-free mice. Nor is it produced in MyD88-deficient mice, indicating that its production is triggered through TLRs. The IgD produced in the intestine is active against the intestinal bacteria. Microbiota-associated IgD class switching can also be detected in the nasal mucosa. Thus IgD may play a role in regulating the normal microbiota [41].

5.5.4.8 T cell functions The key to successful accommodation with the intestinal microbiota depends on the body’s ability to control inflammation in the gut wall. This is achieved by maintaining a balance between proinflammatory Th17 cells and antiinflammatory Treg cells (Fig. 5.5). PATHOBIONTS

IL-1 IL-6 IL-23

Th17

COMMENSALS

IL-10 Retinoic acid

Treg

PRO-INFLAMMATION ANTI-INFLAMMATION FIGURE 5.5 The balancing role of Treg cells and the gut microbiota. T cells must be prepared to attack and destroy invaders, but they must also distinguish between pathogens and commensals. This is achieved by balancing Treg and Th17 responses.

64

SECTION | 1 Mammalian immunology

Intestinal helper T cell phenotypes are “plastic” and precursor cells can differentiate into either Treg or Th17 cells while Th17 cells may further differentiate into Th1 cells. This differentiation is regulated by signals from the microbiota [42]. In effect the microbiota program the T cell system to optimize its function. Additionally, helper T cell differentiation is determined by cytokines. For example, the development of both Tregs and Th17 cells is promoted by TGF-β. Treg cells require TGF-β plus retinoic acid and IL-2 whereas Th17 cells require TGF-β plus IL-6 and IL-23. The major transcription factors used by Tregs and Th17 are Foxp3 and RORγt respectively. These are co-expressed in naı¨ve and effector CD41 cells. It has recently been found that commensal SFB are the source of this retinoic acid in the intestine. The retinoic acid primes epithelial defenses and so promotes innate immunity against intestinal invasion [18].

5.5.4.9 Treg cells Intestinal Treg cells are a subset of CD41 Th cells required to maintain the body’s commensal relationship with its microbiota. Treg cells produce IL-10 as well as express high levels of CTLA-4 (cytotoxic T lymphocyte antigen 4)— the inhibitory ligand for CD80. Treg production responds to signals from both the microbiota and enterocytes. These Treg cells, when exposed to proinflammatory cytokines, can convert to IL-17- or IFN-γ -expressing effector cells and so “break” tolerance. Th17 cells can give rise to IFN-γ producers that functionally resemble Th1 cells, and it is likely that many intestinal Th1 cells develop through this pathway. Under other circumstances, the Treg cells may convert to helper cells and promote the B cell switch to IgA production. Indeed about 75% of the IgA directed against the microbiota is produced through a pathway controlled by Treg cells. Mucosal inflammation is therefore actively suppressed by the production of large numbers of IL-10-producing Foxp31 Tregs. Under stable conditions, the production of Treg cells is favored while that of Th17 cells is suppressed. In the absence of Treg cells, uncontrolled effector T cells will respond to microbial antigens and trigger inflammatory bowel disease [43]. For example, IL-10 deficient mice develop chronic unremitting colitis driven by IL-23 and the Th17 pathway. It is also clear however that this tolerance can only go so far. Should a potential pathogen seek to invade the body from the intestine, then the immune system must be prepared to act aggressively to prevent this.

5.5.4.10 Th17 cells Th17 cells are a subset of CD41 T cells that help to maintain the intestinal epithelial barrier (Fig. 5.6). Under the influence of IL-23, they produce the proinflammatory cytokines IL-17A, IL-17F, and IL-22. They also produce the antibacterial cytokine IL-26 in some species. (IL-26 is not produced by horses, rats, mice, or elephants!). Like B cells and Treg cells, the development of Th17 cells is regulated by signals from both the microbiota and enterocytes. Th17 cell FIGURE 5.6 The generation and functions of Th17 cells in the mammalian intestine. In general they serve a proinflammatory function.

Th0

Dendritic cell

TGF-E IL-6 IL-23 IL-21

Th17 IL-22

Suppress Th2 responses

IL-17

Endothelial cells and macrophages IL-8

GM-CSF

IL-6

Neutrophil accumulation

INFLAMMATION

The microbiota
immune system relationship Chapter | 5

65

development is specifically stimulated by the attachment of SFB to enterocytes. Enterocytes can sense this tight attachment [42]. The mechanism by which this affects enterocytes is unclear, but they cause enterocytes to produce the acutephase protein, serum amyloid A (SAA). SAA is a soluble pattern recognition receptor that stimulates IL-23 production by macrophages leading to ILC3 secretion of IL-22 and -17. This in turn triggers the development of Th17 cells. Lamina propria macrophages and dendritic cells can also detect SFB-derived molecules through their TLRs and respond by producing IL-23 and TGF-β so promoting additional Th17 cell differentiation. As a result, the T cells differentiate into RORγt - expressing Th17 cells. The Th17 cells, in turn, regulate the abundance of SFBs by promoting the production of antibacterial peptides such as β-defensins, lipocalins, and calprotectin by enterocytes. SFB (Candidatus savagella) are unique spore-forming, long filamentous Gram-positive anaerobic commensals found in the small intestine of some mammals. They have a unique ability to stimulate the maturation of T and B cells. They can also upregulate host innate defense genes, inflammatory cytokines, and lymphokines. They attach very strongly to the enterocytes of the terminal ileum and the cells overlying Peyer’s patches where they are in a good position to be sampled by dendritic cells. (Most other bacteria remain within the mucus layer.) SFB induce the development of germinal centers in Peyer’s patches and other intestinal lymphoid organs and specifically increases the production of IgA and Th17 cells [44]. In the absence of SFB, mice mount weaker IgA responses and poorer intestinal T cell responses and recruitment of intraepithelial lymphocytes [45]. It is not believed that the Th17 cells are directed against specific antigens in the SFBs, they appear to develop polyclonally through bystander activation.

5.5.4.11 γ/δ T cells When intestinal epithelial cells are exposed to a high carbohydrate environment, they change the frequency of specialized epithelial subsets suggesting that functional specialization occurs among this cell population. Interestingly, this carbohydrate response requires the presence of γ/δ T cells in species such as mice. γ/δ T cell production is also altered by the diet. This induces changes in their transcriptome, tissue localization, and behavior. These changes appear to result from the suppression of the activities of IL-22 by ILC3 cells. Thus some gut lymphocytes clearly play a role in nutrient sensing [46].

5.5.4.12 Retinoic acid Retinoic acid, the active metabolite of vitamin A is a central regulator of mucosal immunity. In association with TGFβ, it enhances T cell proliferation and cytotoxicity and is especially important in promoting Th2 and Treg differentiation and in the homing of IgA1 B cells to mucosal surfaces. It is essential for maintaining the stability of Th1 cells and preventing their transition to Th17 cells. Retinoic acid normally suppresses Th17 responses and favors tolerance to food antigens. SFB can secrete it and thus prime T cells [18].

5.6

Dysbiosis

It is clear from the above discussion that the gut microbiota exerts a significant influence on the mammalian systemic immune response. If the microbiota becomes unstable or imbalanced dysbiosis may result. Dysbiosis is a direct cause of several inflammatory diseases such as equine laminitis, and ruminal acidosis, and has been implicated in the development of several immune-mediated diseases such as inflammatory bowel diseases. Antibiotic treatment is an important cause of dysbiosis [47]. In critically ill animals, breakdown of the intestinal epithelium and mucosal barriers may permit leakage of bacterial components into the body. Conversely, depletion of the gut microbiota, especially as a result of antibiotic treatment can make the mucosal defenses vulnerable and perhaps reduce the priming of the systemic immune responses.

5.7

Behaviors

5.7.1 Odors Another way by which the microbiota influence mammalian behavior and survival is by increasing the diversity of their communication signals. Thus it is postulated that bacteria in mammalian scent glands, as well as those in the mouth and the intestine, generate odorous metabolites such as SCFAs. Some of these odors are used for host communication [48]. Variations in these host chemical signals may be a result of underlying changes in the microbiota of the scent glands or intestine. Thus surveys of the microbiota of hyaena scent glands suggest that variations in their microbiota correlate

66

SECTION | 1 Mammalian immunology

with the volume and fatty acid profiles of scent secretions. These bacterial populations also vary with species, gender, pregnancy, and reproductive state. In other mammals, odors are used for kin identification and mate choice. Given that genes for many odor receptors are linked to the MHC it is not difficult to imagine how animals recognize others by the characteristic mixture of odors generated by their microbiota.

5.7.2 Hibernation Some mammals hibernate to reduce energy demands and avoid having to find food when none is available. During this prolonged fasting period, the absence of nitrogen intake could potentially cause a severe protein imbalance resulting in significant muscle loss. However, in thirteen-lined ground squirrels, (Ictidiomys tridecemlineatus), this problem is solved by the intestinal microbiota. These organisms recycle nitrogen from urea into their metabolites that are then absorbed by the hibernating host. These are then reincorporated into the squirrel’s protein pool and enabling them to maintain protein balance. This urea recycling is greatest after prolonged fasting in late winter. This may be a common pathway in other hibernating mammals as well [49].

5.7.3 The aryl hydrocarbon receptor The aryl hydrocarbon receptor (AhR) was originally identified as a transcription factor in the liver where it regulates the enzymes that metabolize xenobiotics such as cytochrome p450. Its endogenous ligands include derivatives of tryptophan, indoles, and lipoxins. Dietary tryptophan is metabolized by Lactobacilli to generate indole-3-aldehyde, tryptamine, and indole, all of which are AhR ligands [50]. These receptors have been cloned from several mammalian species, especially humans, rats, and mice. Their overall structure is conserved but differences in the sequences of their binding sites result in significant differences in antigen binding affinities. They segregate into three clades as a result of a gene duplication that occurred before the emergence of vertebrates [51]. The AhR plays an important role in regulating immune responses. Thus it is expressed at high levels on antigenpresenting cells, intraepithelial lymphocytes, Th17, and Treg cells. Ligand binding to the AhR stimulates IL-23 production and Th17 differentiation. IL-22 from these cells enhances the production of antimicrobial peptides. The development of ILC3 cells depends upon AhR expression and activation. AhRs are essential for the formation of intraepithelial lymphocytes, cryptopatches, and isolated lymphoid follicles. Defective AhR signaling results in severe intestinal inflammation and the development of allergies. Thus AhR knockout mice mount enhanced Th2 responses and higher levels of IgE and IgG1. Their dendritic cells express higher levels of CD86 and MHCII molecules. AhR signaling negatively affects the type 1 interferon response. AhR ligands such as indoles and flavonoids are naturally found in cruciferous vegetables (cabbage, cauliflower, and lettuce) [52]. These act through AhRs to maintain IELs and promote normal intestinal lymphoid function. Thus the AhR may have evolved in herbivores enabling them to exploit otherwise toxic plants.

5.8

Environmental microbiota

Under some circumstances, animals may share their microbiota with others in a community. Thus in addition to the eating of scats by wild animals (coprophagia), gut microbes readily spread in aquatic environments. An example of this has been reported in the hippopotamus. Hippos live in rivers, lakes, and ponds and excrete large quantities of fecal matter into the water. Other members in the herd will drink this water and as a result, share their intestinal microbes [53]. Analysis has demonstrated that many hippo gut bacteria also thrive in hippo wallows, ponds, and lakes. Thus the microbiota is shared with the environment. This appears to be especially important where mammals congregate in high densities, especially in aquatic environments. A similar situation may well occur around large walrus and sea lion herds where they gather in dense colonies on beaches to mate and deliver their pups. Beaches that may be contaminated with large quantities of fecal material.

References [1] Davenport ER, Sanders JG, Song SJ, Amato KR, et al. The human microbiome in evolution. BMC Biol 2017;15:127. Available from: https://doi. org/10.1186/s12915-017-0454-7. [2] Ley RE, Hamady M, Lozupone C, Turnbaugh P, et al. Evolution of mammals and their gut microbes. Science 2008;320(5883):1647
51. [3] Lee YK, Mazmanian SK. Has the microbiota played a critical role in the evolution of the adaptive immune system? Science 2010;330:1769
72.

The microbiota
immune system relationship Chapter | 5

67

[4] Sender R, Fuchs S, Milo R. Are we really vastly outnumbered? Revisiting the ratio of bacterial to host cells in humans. Cell 2016;164 (3):337
40. [5] Levin D, Raab N, Pinto Y, Rothschild D, et al. Diversity and functional landscape in the microbiota of animals in the wild. Science 2021;. Available from: https://doi.org/10.1126/science.abb5352. [6] Vijay-Kumar M, Aitken JD, Carvalho FA, Cullender TC, et al. Metabolic syndrome and altered gut microbiota in mice lacking Toll-like receptor 5. Science 2010;328:228
331. [7] Mulder IE, Schmidt B, Stokes CR, et al. Environmentally-acquired bacteria influence microbial diversity and natural innate immune responses at gut surfaces. BMC Biol 2009. Available from: https://doi.org/10.1186/1741-7007-7-79. [8] Song SJ, Sanders JG, Delsuc F, Metcalf J, et al. Comparative analysis of vertebrate gut microbiomes reveal convergence between birds and bats. mBio 2020. Available from: https://doi.org/10.1128/mBio.02901-19. [9] Nishida AH, Ochman H. Rates of gut microbiome divergence in mammals. Mol Ecol 2018;27:1884
97. [10] Xue Z, Zhang W, Wang L, Hou R, et al. The bamboo-eating giant panda harbors a carnivore-like gut microbiota, with excessive seasonal variations. mBio 2015. Available from: https://doi.org/10.1128/mBio.00022-15. [11] Delsuc F, Metcalf JL, Wegener Parfrey L, Song SJ, et al. Convergence of gut microbiomes in myrmecophagous mammals. Mol Ecol 2014;23:1301
17. [12] Hecker N, Sharma V, Hillwer M. Convergent gene losses illuminate the metabolic and phtsiological changes in herbivores and carnivores. Proc Natl Acad Sci USA 2019. Available from: https://doi.org/10.1073/pnas.1818504116. [13] Altman MO, Gagneaux P. Absence of Neu5Gc and presence of anti-Neu5Gc antibodies in humans—an evolutionary perspective. Front Immunol 2019. Available from: https://doi.org/10.3389/fimmu.2019.00789. [14] Rodrigues Hoffmann A, Patterson AP, Diesel A, Lawhon SD, et al. The skin microbiome in healthy and allergic dogs. PLoS One 2014;9:e83197. [15] Ross AA, Rodrigues Hoffmann A, Neufeld JD. The skin microbiome of vertebrates. BMC Microbiome 2019;. Available from: https://doi.org/ 10.1186/s40168-019-0694-6. [16] Hong HH. Skin microbiota: genomics-based insights into the diversity and role of skin microbes. Trends Mol Med 2011;17:320
8. [17] O’Dwyer DN, Dickson RP, Moore BB. The lung microbiome, immunity, and the pathogenesis of chronic lung disease. J Immunol 2016; 196:4839
47. [18] Woo V, Eshleman EM, Hashimoto-Hill S, Whitt J, et al. Commensal segmented filamentous bacteria-derived retinoic acid primes host defense to intestinal infection. Cell Host Microbe 2021;. Available from: https://doi.org/10.1016/j.chom.2021.09.010. [19] Hooda S, Minamoto Y, Suchodolski JS, Swanson KS. Current state of knowledge: the canine gastrointestinal microbiota. Anim Health Res Rev 2013;13:78
88. [20] Costa MC, Silva G, Ramos RV, Staempfli HR, et al. Characterization and comparison of the bacterial microbiota in different gastrointestinal tract compartments in horses. Vet J 2015;205:74
80. [21] Kamada N, Chen GY, Inohara N, Nunez G. Control of pathogens and pathobionts by the gut microbiota. Nat Immunol 2013;14:685
90. [22] Gensollen T, Iyer SS, Kasper DL, Blumberg RS. How colonization by microbiota in early life shapes the immune system. Science 2016; 352:539
644. [23] Kamada N, Nunez G. Role of the gut microbiota in the development and function of lymphoid cells. J Immunol 2013;190:1389
95. [24] Belkaid Y, Hand TW. Role of the microbiota in immunity and inflammation. Cell 2014;157:121
41. [25] Van der Hee B, Wells JM. Microbial regulation of host physiology by short chain fatty acids. Trends Microbiol 2021;29(8):700
12. [26] Takeuchi T, Miyauchi E, Kanaya T, Kato T, et al. Acetate differentially regulates IgA reactivity to commensal bacteria. Nature 2021; 595:560
4. [27] Sharon G, Sampson TR, Geschwind DH, Mazmanian SK. The central nervous system and the gut microbiome. Cell 2016;167:915
32. [28] Arpaia N, Rudensky AY. Microbial metabolites control gut inflammatory responses. Proc Natl Acad Sci USA 2014;111:2058
9. [29] Peterson LW, Artis D. Intestinal epithelial cells: regulators of barrier function and immune homeostasis. Nat Rev Immunol 2014;14:141
53. [30] Artis D. Epithelial-cell recognition of commensal bacteria and maintenance of immune homeostasis in the gut. Nat Rev Immunol 2008; 8:411
20. [31] Ramanan D, Cadwell K. Intrinsic defense mechanisms of the intestinal epithelium. Cell Host Microbe 2016;19:434
41. [32] Gury-Benari M, Thaiss CA, Serafini N, Winter DR, et al. The spectrum and regulatory landscape of intestinal innate lymphoid cell are shaped microbiome. Cell 2016;166:1231
46 e13. [33] Hepworth MR, Fung TC, Masur SH, Kelsen JR, et al. Immune tolerance. Group 3 innate lymphoid cells mediate intestinal selection of commensal bacteria-specific CD4(1) T cells. Science 2015;348:1031
5. [34] Hepworth MR, Monticelli LA, Fung TC, Ziegler CG, et al. Innate lymphoid cells regulate CD41 T-cell responses to intestinal commensal bacteria. Nature 2013;498:113
17. [35] Kim S-H, Cho B-H, Kiyono H, ZJang Y-S. Microbiota-derived butyrate suppresses group 3 innate lymphoid cells in terminal ileal Peyer’s patches. Sci Rep 2017;. Available from: https://doi.org/10.1038/s41598-017-02729-6. [36] Goguyer-Dechaumes R, Waekel L, Killian M, et al. Metabolites and secretory immunoglobulins: messengers and effectors of the hostmicrobiota intestineal equilibrium. Trends Immunol 2022;. Available from: https://doi.org/10.1016/j.it.2021.11.005. [37] Rollenske T, Burkhalter S, Muerner L, von Gunten S, et al. Parallelism of intestinal secretory IgA shapes functional microbial fitness. Nature 2021;598:657
61. [38] Huus KE, Petersen C, Finlay B. Diversity and dynamism of IgA-microbiota interactions. Nat Rev Immunol 2021;. Available from: https://doi. org/10.1038/s41577-021-00506-1.

68

SECTION | 1 Mammalian immunology

[39] Mu Q, Swartwout BK, Edwards M, Zhu J, et al. Regulation of neonatal IgA production by the maternal microbiota. Proc Natl Acad Sci USA 2021;118(9) e2015691118. [40] Usami K, Niimi K, Matsuo A, Suyama Y, et al. The gut microbiota induces Peyer’s patch-dependent secretion of maternal IgA into milk. Cell Rep 2021. Available from: https://doi.org/10.1016/j.celrep.2021.109655. [41] Choi JH, et al. IgD class switching is initiated by microbiota and limited to mucosa-associated lymphoid tissue in mice. Proc Natl Acad Sci USA 2017;114:E1196
204. [42] Ivanov II, Atarashi K, Manel N, et al. Induction of intestinal Th17 cells by segmented filamentous bacteria. Cell 2009;139:485
98. [43] Bollrath J, Powrie F. Feed your Tregs more fiber. Science 2013;341:463
4. [44] Lecuyer E, Rakotobe S, Lengline-Garnier H, Lebreton C, et al. Segmented filamentous bacterium uses secondary and tertiary lymphoid tissues to induce gut IgA and specific T helper 17 cell responses. Immunity 2014;40:608
20. [45] Gaboriau-Routhiau V, Rakotobe S, Lecuyer E, Mulder I, et al. The key role of segmented filamentous bacteria in the coordinated maturation of gut helper T cell responses. Immunity 2009;31:677
89. [46] Sullivan ZA, Khoury-Hanold W, Lim J, Smillie C, et al. γδ T cells regulate the intestinal response to nutrient sensing. Science 2021;371: eaba8310. [47] Becattini S, Taur Y, Pamer EG. Antibiotic-induced changes in the intestinal microbiota and disease. Trends Mol Med 2016. Available from: https://doi.org/10.1016/j.molmed.2016.04.003. [48] Tizard IR, Skow L. The olfactory system: the remote sensing arm of the immune system. Anim Hlth Res Revs 2021. Available from: https:// doi.org/10.1017/s1466252320000262. [49] Regan MD, Liy CY, Tonelli M, Verdoorn KM, et al. Nitrogen recycling in gut symbionts increases in ground squirrels over the hibernation season. Science 2022. Available from: https://doi.org/10.1126/science.abh2950. [50] Barroso A, Gualdron-Lopez M, Esper L, Brant F, et al. The aryl hydrocarbon receptor modulates production of cytokines and reactive oxygen species and development of myocarditis during trypanosoma cruzi infection. Infect Immun 2016;84:3071
82. [51] Hahn ME. Aryl hydrocarbon receptors: diversity and evolution. Chem-Biol Interact 2002;141:131
60. [52] Lawrence BP, Sherr DH. You AhR what you eat? Nat Immunol 2012;13:117
19. [53] Dutton CL, Subalusky AL, Sanchez A, Estrela S, et al. The meta gut: community coalescence of animal gut and environmental microbiomes. Sci Rep 2021. Available from: https://doi.org/10.1038/s41598-021-02349-1.

Chapter 6

Innate immunity: basic features Bacteria and viruses multiply incredibly rapidly. A single bacterium with a doubling time of 45 minutes can produce 500 million offspring within 24 hours. When these organisms invade the body, they must be destroyed before they can overwhelm its defenses. Time is of the essence, and delay can be fatal. The body must therefore employ rapid response mechanisms as its first line of defense against invaders. These mechanisms need to be on constant standby and respond to the first signs of microbial invasion. They constitute the innate immune system (Fig. 6.1). Infectious diseases exert major selective pressure on all mammals. A failure to successfully defend oneself against a bacterial and viral infection results in death and elimination from the gene pool. This constant battle results in an arms race between the invading pathogens and the defending hosts. This is the basis of the “Red Queen” scenario. The name is derived from a character in Lewis Carroll’s Through the Looking Glass, who said. “It takes all the running you can do to keep in the same place.” All mammals need to detect and eliminate microbial invaders as fast and as effectively as possible. Many different innate defense mechanisms have evolved over the millennia, and they act to destroy invaders while minimizing PAMPS from INVADERS

DAMPS from CELL DAMAGE

RECEPTORS (PRRs)

INFLAMMASOME FORMATION

Cytokine release

Vasoactive Factors

Antimicrobial Molecules

Changes in blood flow Activation of adaptive immunity

Emigration of White cells

Phagocytosis

DESTRUCTION OF INVADER FIGURE 6.1 The basic features of the Innate Immune system. It is a method of focusing defensive cells as well as antimicrobial molecules at sites of microbial invasion and tissue damage. It relies in large part on the activation of inflammasomes and the subsequent release of proinflammatory cytokines. Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00026-5 © 2023 Elsevier Inc. All rights reserved.

69

70

SECTION | 1 Mammalian immunology

collateral damage. Innate immune responses are activated when cell surface pattern recognition receptors (PRRs) detect either microbial invasion or tissue damage. For example, cells can sense the presence of invading microbes by detecting their characteristic conserved molecules called “Pathogen-Associated Molecular Patterns” or PAMPs. These cells can also sense tissue damage by detecting the characteristic molecules released from damaged or broken cells. These molecules are called “Damage Associated Molecular Patterns” or DAMPs (Fig. 6.2) [1]. The body employs sentinel cells whose job is to detect PAMPs and DAMPs and once activated, emit signals to attract white blood cells. The white cells converge on the invaders and destroy them in the process we call inflammation. In addition, mammals make many different antimicrobial proteins such as complement, defensins, and cytokines, that can either kill invaders directly or promote their destruction by defensive cells. Some of these antimicrobial molecules are constitutively present in normal tissues while others are produced only in response to the presence of PAMPs or DAMPs. The innate immune system lacks specific memory and as a result, each episode of infection tends to be treated identically. ("Trained" immunity is an exception; Chapter 10). The intensity and duration of innate responses such as inflammation, therefore, tend to remain unchanged no matter how often a specific invader is encountered. These responses also come at a price, - the collateral damage and pain of inflammation as well as the characteristic features of “sickness” largely result from the activation of innate immune processes. Importantly, however, the innate immune responses serve as one of the triggers that stimulate antigen-presenting cells to initiate the adaptive immune responses and eventually result in strong long-term protection. An uncontrolled innate immune response is potentially damaging to the body and must be carefully regulated to avoid severe damage. Studies on the transcriptome of cells exposed to infectious agents and other stimuli have suggested that the genes from many different species of mammals that are directly involved in the effector arms of the innate response are highly conserved. Examples include the genes that encode transcription factors and kinases that play a regulatory role and show limited variability of expression [2]. These transcription factors have however, to interact with many other proteins and are limited in the degree of change that can be tolerated. As a result, the process of inflammation is broadly similar irrespective of the species of mammal involved.

6.1

Constitutive innate immunity

6.1.1 Antibacterial peptides Among the body’s constitutive defenses are many potent antimicrobial peptides. The most important of these are the defensins [3]. Defensins are peptides containing 28
42 amino acids arranged in a β-sheet crosslinked by three or four disulfide bonds. More than 50 different mammalian defensins have been identified. The defensins are classified as α, β, or θ based on their origin and the number and position of their disulfide bonds. The α-defensins account for about 15% of the total protein in human neutrophil granules. In cattle, at least thirteen different α-defensins are produced by neutrophils alone. They are also found in the granules of Paneth cells in the small intestine. α-Defensins are produced by the most basal of the mammalian superorders including elephants and tenrecs from the Afrotheria, and the nine-banded armadillo from the Xenarthra as well as the marsupial opossums. They are EXTRACELLULAR DAMPs INTRACELLULAR DAMPs HMGB1 Uric acid Chromatin Heat Shock Proteins Adenosine Galectins S100 proteins Cathelicidins Defensins N-formyl peptides Lactoferrin Mitochondrial Heat-shock proteins DNA

Hyaluronic acid Heparan sulfate Fibrinogen Collagen-derived peptides Fibronectin Laminin Elastin

FIGURE 6.2 Damage-associated molecular patterns. Damage-associated molecular patterns consist of selected molecules released by dead, damaged, and dying cells. They are among the most important triggers of both local inflammation and sickness behavior. Both of these can adversely affect mammalian survival.

Innate immunity: basic features Chapter | 6

71

also found in the Euarchontoglires (rabbits, rodents, and primates) [4]. The β-defensins are much more broadly expressed across mammalian orders. They are expressed in the epithelial cells that line the airways, skin, salivary gland, reproductive and urinary systems [5
7]. Theta defensin is a circular peptide found only in primate neutrophils. Defensins may be produced at a constant rate (constitutively) or in response to microbial infection. Some defensins attract monocytes, immature dendritic cells, and T cells to invasion sites. All defensins identified so far can kill or inactivate bacteria, fungi, or enveloped viruses. They act by disrupting the microbial cell membrane or viral envelope. Some defensins can neutralize microbial toxins such as those from Bacillus anthracis, Corynebacterium diphtheriae, and Staphylococcus aureus. Although present in normal tissues, defensin concentrations increase in response to infections. The equine defensin DEFA1 is an enteric defensin exclusively produced in Paneth cells. It has potent activity against the major horse pathogens, especially Rhodococcus equi and Streptococcus equi [8,9]. A second major family of antibacterial peptides found in neutrophil granules are the cathelicidins [10]. These are 12
80 amino acids in size. They are stored in an inactive form attached to a precursor protein and released following cleavage of the precursor. Humans and mice have only a single cathelicidin gene, whereas the pig, cow, and horse have multiple cathelicidins. Porcine cathelicidin PR-39 has been shown to promote wound repair, angiogenesis, and neutrophil chemotaxis. Because of their high cationic charge, porcine cathelicidins efficiently bind bacterial nucleic acids. They can then deliver these nucleic acids to dendritic cells and as a result, trigger a very strong IFN-α response [11]. The bovine cathelicidin BMAP-28 induces apoptosis and may serve to get rid of unwanted cells. Canine cathelicidin K9CATH has broad-spectrum activity against both Gram-positive and Gram-negative bacteria [12]. Many cathelicidins have been given species-specific names such as protegrins from pigs, and ovispirins from sheep. Granulysins are peptides produced by cytotoxic T cells and natural killer (NK) cells. In addition to their antibacterial functions, granulysins attract and activate macrophages. Two other important antibacterial proteins are bactericidal permeability-increasing protein (BPI) and calprotectin. BPI is a major constituent of the primary granules of human and rabbit neutrophils. It kills Gram-negative bacteria by binding to lipopolysaccharides (LPS) and damaging their inner membrane. Calprotectin is found in human neutrophils, monocytes, macrophages, and epidermal cells. It forms about 60% of neutrophil cytosolic protein and is released in large amounts into blood and tissue fluid in inflammation. It belongs to the S100 family of antimicrobial proteins. Calprotectin sequesters zinc and manganese during bacterial infections and thus makes them unavailable for bacterial growth. Serprocidins are a family of antimicrobial serine proteases also found in human neutrophil granules. The production of antimicrobial peptides by epithelial cells is regulated by cytokines. In particular, the two cytokines produced by Th17 cells, IL-17, and IL-22, are crucial regulators of antimicrobial peptide production in the intestine and lungs. Likewise, IL-1 stimulates epithelial cells to produce antimicrobial peptides. Antimicrobial peptides may regulate cytokine production and can serve as immunomodulators. For example, some cathelicidins stimulate the production of IL-6, IL-8, and IL-10.

6.1.2 The complement system As is typical of the innate immune systems, the complement system evolved long before the emergence of mammals. As a result, the differences between complement systems in mammalian species tend to be relatively minor. Three complement pathways are recognized in mammals: the lectin, alternative, and classical pathways [13,14]. Cartilaginous and bony fish also possess all three complement pathways. Their lytic pathway generates a terminal complement complex like that formed in mammals, although it works at a lower temperature (B25 C). Amphibians also possess a complement system that, although similar to that of mammals, is more effective at 16 C. The mammalian complement system like other antimicrobial systems has evolved under intense selective pressure [15]. In general, this selection has primarily acted on the protein residues that come into direct contact with the pathogens themselves such as the binding of C3b in the alternate pathway or the recognition of polysaccharides in the lectin pathway [16]. Thus microbes evolve to avoid complement binding while complement components do the opposite, This has been well examined in primates [16]. It is a classic example of the Red Queen race Some organisms have evolved mechanisms to evade the complement system. Thus leptospirosis is a widespread zoonosis, but Hedgehog tenrecs (Afroscoricida) and bats appear to act as healthy carriers. Some Leptospires produce an immunoglobulin-like protein that binds to factor H and C4BP and inactivates the complement components bound to the bacterial surface. Some human-specific bacteria such as the two Neisseria (gonorrhea and meningitis) bind complement factor H from humans (but not other primates) and inactivate it.

72

SECTION | 1 Mammalian immunology

6.1.3 Ribonucleic acid interference The intracellular ribonucleic acid (RNA) interference pathway (RNAi) is a gene-silencing system that prevents virus replication. RNA normally occurs only in a single-stranded (ss) form. Long segments of double-stranded (ds) RNA (dsRNA) only occur in cells infected by RNA viruses. Any dsRNA formed is therefore recognized and rapidly degraded into many short fragments by host endonucleases called dicers. These small fragments, 21
23 nucleotides in length, are called small-interfering RNAs (siRNA). siRNAs are bound and stabilized by a protein complex called the RNAinduced silencing complex (RISC). Half of these siRNAs are complementary to the original viral messenger RNAs (mRNAs) and as a result, will bind them tightly and “silence” them. Once these viral mRNAs have been bound by a RISC, they are rapidly degraded so that viral replication is effectively blocked. Extracellular RNAs also play a communication role within the immune system [17].

6.2

Induced innate immunity

6.2.1 Pattern recognition receptors The innate immune system is activated when the body senses that it is under attack. Alarm signals are generated either by the presence of invading microorganisms or by dead, damaged, and dying cells. Together, the released PAMPs and DAMPs bind to PRRs found on sentinel cells located throughout the body. This binding triggers inflammation. Microbes not only grow very fast but also are highly diverse and can alter their surface molecules very rapidly. For this reason, the PRRs of the innate immune system cannot recognize all possible microbial molecules. Rather, the PRRs recognize abundant, essential microbial molecules. Because they are essential, these molecules are structurally conserved and may be shared by entire classes of pathogens. They form, in effect, conserved molecular patterns. For example, the walls of Gram-positive bacteria are largely composed of peptidoglycans and lipoteichoic acids. Likewise, the walls of Gram-negative bacteria consist of peptidoglycans covered by a layer of LPS. Acid-fast bacteria are covered in glycolipids. Yeasts have a mannan-or β-glucan-rich cell wall. Viruses have unique nucleic acids. PRRs can recognize all of these conserved molecules. Mammals use many different PRRs to ensure the detection of as many PAMPs as possible. Most PRRs are cell-associated and found on cell membranes, within the cytosol, and within cytoplasmic vesicles. However, other, soluble PRRs circulate in the bloodstream (Fig. 6.3).

6.2.2 Toll-like receptors The most significant family of PRRs are the Toll-like receptors (TLRs). Mammals possess 10 or 12 different functional TLRs (TLRs 1
10 in humans, sheep, and cattle or thirteen TLRs (1
9 and 11
13) in mice) (Table 6.1) [18]. A single ancestral TLR gene probably originated about 500 mya. TLR10 diverged from the precursors of TLR1 and TLR6 at about 300 mya. Mammalian TLR4 diverged B180 mya while TLRs 3, 5, 7, and 8 diverged B150 mya [19]. The name of this receptor family alludes to the original discovery of a protein called “Toll” in fruit flies (Drosophila). This protein was necessary for proper embryological development. Toll protein was subsequently found to be necessary for antifungal immunity in Drosophila. When the first PRRs were identified in mammals, they were found to be similar in sequence and structure to the Drosophila toll protein, hence “toll-like receptors.” FIGURE 6.3 Pattern recognition molecules. The body employs a highly diverse array of pattern recognition molecules. Their function is to instantly recognize the presence of microbial invaders and trigger an innate immune response.

Pattern-recognition receptors

Soluble Collectins Ficolins Complement Pentraxins

Within vesicles

Cytoplasmic

Membrane -bound

TLR 3,7,8,9

Rig-1 NOD-like Peptidoglycan receptors DNA receptors MDA5 AIM2

TLR 1, 2, 4, 5, 6 Lectins Mannose receptor Langerin Dectins Integrins Scavenger receptors

Innate immunity: basic features Chapter | 6

73

TABLE 6.1 Pathogen-associated molecular patterns and the functions of the Mammalian toll-like receptors. TLR

Location

Ligand

Pathogens recognized

TLR1

Cell surface

Triacylated lipopeptides

Bacteria

TLR2

Cell surface

Lipoproteins, peptidoglycans

Bacteria, viruses, parasites

TLR3

Endosomal

dsRNA

Viruses

TLR4

Cell surface

LPS

Bacteria, viruses

TLR5

Cell surface

Flagellin

Bacteria

TLR6

Cell surface

Diacylated lipopeptides

Bacteria, viruses

TLR7

Endosomal

ssRNA, guanosine

Viruses, bacteria

TLR8

Endosomal

ssRNA

Viruses, bacteria

TLR9

Endosomal

Bacterial (CpG) DNA, dsDNA

Viruses, bacteria, protozoa

TLR10

Endosomal

Regulates TLR2 responses

Suppresses inflammation

TLR 11- Mice

Cell surface

Profilin and flagellin

Protozoa, bacteria

TLR12 - Mice

Cell surface

Profilin

Protozoa

TLR 13
Mice

Endosomal

Unmethylated bacterial RNA

Bacteria

CpG, cytosine-guanosine; LPS, lipopolysaccharide; TLR, toll-like receptor.

TLRs are mainly expressed in the cells most likely to encounter invaders. These include neutrophils, macrophages, and dendritic cells. However dendritic cells and macrophages include multiple cell types and their TLR expression may differ between subpopulations and depend on the degree of cellular activation. TLRs are also expressed on T and B cells, as well on non-immune cells such as the epithelial cells that line the respiratory and intestinal tracts. TLR11 differs from the others in that it is found only on dendritic cells, macrophages, and epithelial cells in the mouse urinary tract, where it binds PAMPs from bacteria and protozoan parasites. TLRs are also expressed on bone marrow stem cells, the source of blood leukocytes. Bacterial LPS binding to TLR4 on these stem cells can stimulate them to divide, so increasing leukocyte production. An increase in leukocyte numbers in the blood (the white cell count) is therefore a consistent feature of mammalian infectious diseases [20]. All TLRs are transmembrane glycoproteins [21]. Most are homodimers formed by two identical paired peptide chains. They may also form heterodimers by using two different chains. For example, TLR2 can associate with TLR6, and this dimer can then bind bacterial diacylated lipopeptides. TLR2 can also associate with TLR1 to recognize mycobacterial triacylated lipopeptides. Given the number of possible TLR chain pairings, the presently known TLRs can collectively bind almost all known PAMPs. Many TLRs are located on cell surfaces, where they can bind PAMPs from extracellular invaders such as bacteria and fungi. Other TLRs are expressed within cells, where they are optimally situated to bind nucleic acids from intracellular invaders such as viruses (Fig. 6.4).

74

SECTION | 1 Mammalian immunology

Triacetylated Diacetylated lipoproteins lipoproteins LPS

TLR 1/2

TLR 2/6

TLR4

Flagellin

TLR5

E-Glucans HMGB1

Dectin-1

FIGURE 6.4 Toll-like receptors and their ligands. Toll-like receptors can be expressed on the outer cell membrane where they are in a position to bind circulating bacterial molecules. They are also expressed in intracellular endosomes where they can bind viral and intracellular bacterial nucleic acids.

RAGE

TLR3

ENDOSOME dsRNA guanosine ssRNA

TLR7

CpG DNA

TLR9

TLR8

6.2.3 Cell surface toll-like receptors The ligands of the cell surface TLRs (TLR1, 2, 4, 5, 6, and 11) are bacterial and fungal proteins, lipoproteins, and LPS. For example, TLR4 binds LPS from Gram-negative bacteria. TLR2 binds peptidoglycans and lipoproteins from Grampositive bacteria, and a glycolipid called lipoarabinomannan from Mycobacterium tuberculosis. TLR5 binds flagellin, the major protein of bacterial flagella. Most vertebrates have only a single gene ortholog for each TLR family. They are clearly under strong selection for maintenance of function and as a result, mutations are suppressed. However, the TLR1 gene family appears to have more species adaptations than the others [22]. The two TLRs that are most distant from the putative ancestor belong to a murine lineage, TLR11, and TLR12. TLR11 is present in some mammals but in humans is only a pseudogene. Negative selection for maintenance of function has dominated TLR2, TLR4, and TLR5-binding specificities. The strongest selective pressure has been on TLR4. TLR4 has evolved actively in response to Gram-negative bacteria, and these may account for many of its species-specific sequence differences [23]. TLR2 is functional in all mammals and is highly effective in detecting both bacterial and viral ligands. However phylogenetic analysis shows that TLR2 has a longer branch length in the Lagomorphs (rabbits, hares, and pikas) [24]. Further analysis shows that these TLR2 genes exhibit a higher nucleotide diversity and thus accelerated evolutionary rate and positive selection. This suggests that lagomorphs at an early stage in their evolution were under strong selective pressure from an unknown pathogen. All mammals appear to have only a single functional TLR2 gene but there is an upstream TLR2 pseudogene in the opossum, dog, and humans. This gene duplication event must have occurred before the divergence of the marsupials. Most TLR genes show evidence of positive selection. Results are however inconsistent since some investigators suggest that there has been greater positive selection in the antibacterial TLRs (1, 2, and 6) than in the antiviral TLRs [25]. Likewise, several positively selected sites on other TLRs are located in the PAMP binding regions perhaps reflecting host-pathogen coevolution. Analysis of TLR substitution rates demonstrates that they have not been uniform among all mammals [26]. For example, flagellin-binding TLR5 in rabbits, rodents, carnivores, and bats has a higher substitution rate than in other orders. There is also evidence of significant adaptive evolution of TLR4 during the transition of marine mammals from a terrestrial to an aquatic environment [27]. It is also relevant to note that TLR5 is convergently lost in four unrelated mammalian lineages, guinea pigs, Yangtse river dolphins, pinnipeds, and pangolins. In addition, TLR11 which acts as a second extracellular flagellin receptor is also absent from these four lineages. In addition to its total loss, TLR5 is inactivated in some vertebrates and TLR5 stop codon polymorphism is widespread in human populations [28]. The ability to sense the presence of flagellin is not considered essential in some species. As with other PRRs, this may be due to positive selection pressure, neofunctionalization, or even the beginning of TLR pseudogenization.

6.2.4 Intracellular toll-like receptors The intracellular TLRs (TLR3, 7, 8, 9, and 10), bind viral and bacterial nucleic acids [29]. TLR9 for example is an intracellular sensor of bacterial DNA and is therefore triggered by intracellular bacteria. Other intracellular receptors,

Innate immunity: basic features Chapter | 6

75

such as TLR3, bind viral dsRNA, whereas TLR7 and TLR8 bind viral ssRNA. Mouse TLR10 is disrupted by two retroelements. One, B1 is unlikely to be older than 10 million years while the second one, an LTR, is likely younger than 100,000 years [30].

6.2.5 Toll-like receptor signaling When their ligands bind, the TLRs become activated and in turn activate genes encoding proinflammatory enzymes. Thus TLRs trigger inflammation immediately after they sense the presence of invaders or tissue damage. The TLR signaling pathways are conserved across mammals. Their differences are generally associated with sites of TLR expression and ligand-specific binding. When a PAMP binds to its corresponding TLR, signals are passed to the cell. As a result, multiprotein signaling complexes form, signal transduction cascades are initiated, and proinflammatory molecules are produced by the cell. (Fig. 6.5) Each step in the process involves multiple biochemical reactions involving many different proteins. Additionally, the cell surface TLRs use different signaling pathways than do the intracellular TLRs. All extracellular TLRs except TLR3 use an adaptor protein called MyD88 to activate the transcription factors, nuclear factor kappa-B (NF-κB), and IRF3. NF-κB activates the genes for three proteins, interleukin-1 (IL-1), interleukin-6 (IL-6), and tumor necrosis factor-alpha (TNF-α). IRF3 activates the gene for interferon-β (IFN-β) [31]. Unlike humans, mice are often quite unresponsive to TLR signaling. Thus they are tolerant to the TLR4 ligand, bacterial LPS, and the TLR2 agonist peptidoglycan-associated glycoproteins, or the staphylococcal superantigen enterotoxin B. As a result, a dose of about 15 μg/kg of LPS can cause lethal shock in humans. In contrast, the median lethal dose of endotoxin in mice is 10 mg/kg. This appears to be due to a much lower level of cytokine release in mice. A similar phenomenon is seen in non-human primates (Chapter 24). The mediators produced in response to TLR-ligation on sentinel cells are generated as inactive precursors and then activated by caspase-1. The production of caspase-1 is initiated by a protein complex called an inflammasome. Caspases are proteases (cysteinyl aspartate-specific proteinases). Many, such as caspase-1, -4, -5, and -12, are activated by signals generated by TLRs. Caspase-1 is most important because it acts on inactive precursors to generate active cytokines. Different TLRs trigger the production of different cytokine mixtures, and different PAMPs trigger distinctly different responses even within a single cell type. For example, TLRs that bind bacterial PAMPs tend to trigger the production of cytokines optimized to combat bacteria while those that bind viral PAMPs produce antiviral cytokines.

Bacterial PAMP

TLR4

MyD-88 NF-NB

IRF3

GENE ACTIVATION Interleukin-1 Interleukin-6 Tumor Necrosis factor-D

INFLAMMATION

Type 1 Interferons

VIRUS INHIBITION

FIGURE 6.5 Toll-like receptors trigger cellular responses when bound by their ligands. Acting through the MyD88 pathway they activate genes encoding multiple proinflammatory and antiviral cytokines. These cytokines mediate inflammation and antiviral resistance.

76

SECTION | 1 Mammalian immunology

TLRs not only trigger innate responses such as inflammation but also begin the process of “turning on” the adaptive immune system. For example, DAMPs binding to TLR4 trigger macrophages and their close relatives, the dendritic cells, to produce cytokines that are potent stimulators of lymphocytes. The intracellular TLRs detect the presence of viral nucleic acids. When triggered, they synthesize antiviral cytokines, the type I interferons (IFNs). These interferons turn on antiviral pathways and so “interfere” with viral growth.

6.2.6 Retinoic acid-inducible gene-1-like receptors Retinoic acid-inducible gene (RIG)-like receptors (RLRs) are another family of PRRs expressed within cells and are important sensors of viral RNA [32]. They detect viral dsRNA molecules. Because dsRNA molecules do not occur in uninfected cells, their detection by RLRs activates caspases and triggers the production of type I interferons. Positive selection has been observed in the genes encoding the RIG1, MDA5, and LGP2 RNA helicases.

6.2.7 Nucleotide-binding oligomerization domain-like receptors Nucleotide-binding oligomerization domain (NOD), and leucine-rich repeat receptors (NLRs) are a family of PRRs that also detect intracellular PAMPs (Table 6.2). Although TLRs and NLRs differ in their location and function, they both react to microbial PAMPs and trigger innate responses to invaders. NOD1 binds bacterial peptidoglycans, whereas NOD2 binds muramyl dipeptide and serves as a general sensor of intracellular bacteria. Ligand binding to either NLR activates the NF-κB pathway and triggers the production of proinflammatory cytokines. NOD2 binding also triggers the production of defensins. NOD3 binds diverse ligands including many viral nucleic acids and inorganic matter such as silica, asbestos, and alum. Macrophages, mast cells, and dendritic cells express many other receptors that can recognize microbial molecules and trigger innate responses. These include C-type lectin receptors (CLRs) that bind carbohydrates, a molecule called CD36 that binds lipoproteins, and CD1 that binds glycolipids. CD1 is an important ligand binding molecule that is used for the presentation of selected lipids to invariant γ/δ T cells (Chapter 8).

TABLE 6.2 Other Mammalian pattern-recognition molecules. Receptor

Location

Ligand

Source of ligand

Short dsRNA

RNA viruses

Retinoic acid-inducible gene-like receptors (RLRs) RIG-1

Endosomal

Nucleotide-binding oligomerization domain (NOD), leucine-rich repeat receptors (NLRs) NOD1

Cytosol

Peptidoglycans

Bacteria

NOD2

Cytosol

Muramyl dipeptide

Bacteria

Dectins

Cell surface

Glucans

Fungi

Mannose receptor (CD206)

Cell surface

Glycoproteins

Bacteria

AIM2 receptors

Cytosol

dsRNA

Bacterial

CD14

Cell surface

LPS

Bacteria

CD1

Cell surface

Glycolipids

Bacteria

CD36

Cell surface

Lipoproteins

Bacteria

CD48

Cell surface

Fimbria

Bacteria

C-type lectin receptors (CLRs)

Others

Innate immunity: basic features Chapter | 6

77

6.2.8 AIM2 receptors Interferon-inducible protein AIM2 is a cytoplasmic sensor of microbial dsDNA. AIM2 binds to both strands of the DNA molecule. This displaces a pyrin domain that in turn activates an activator protein that recruits procaspase-1 to form an AIM2 inflammasome. The activation of this caspase results in cleavage of pro-IL-1, pro-IL-18, and gasdermin D. The gasdermin D kills cells by pyroptosis and the release of the inflammatory cytokines. AIM2-like receptors have been remarkably diversified among mammals [33]. Multiple members of the AIM2 family have now been identified including four in humans, 13 in mice but none yet, in bats.

6.2.9 Cyclic GMP-AMP synthase A third important DNA binding sensor is cyclic GMP-AMP synthase. This initiates a response mediated by a downstream adaptor molecule, known as a stimulator of interferon genes (STING). The engagement of this pathway by bacterial and viral DNA leads to the induction of large quantities of type I interferons and the NF-κB-mediated expression of inflammatory cytokines [34]. This pathway is suppressed in bats, a feature that may help explain their unusual responses to some viruses [35].

6.2.10 Pathogen-associated molecular patterns As described earlier, PAMPs are common, essential, conserved molecular structures that are consistently present in diverse microbial invaders. They include structural proteins such as bacterial LPS, peptidoglycans, and their critical genetic material, nucleic acids.

6.2.10.1 Bacterial lipopolysaccharides LPS are structural components of the cell walls of many bacteria, especially Gram-negative ones. They are recognized by TLR4 [36]. TLR4 does not bind LPS directly but only when linked to three other proteins. These proteins are, MD-2 (myeloid differentiation factor-2), LPS-binding protein (LBP), and CD14. The CD14 interacts with TLR4 in such a way that it decreases the specificity of these reactions and enables both rough and smooth strains of bacteria to be recognized. The binding of LPS to the CD14/TLR4/MD-2 complex activates macrophages and triggers cytokine production. The LPS can dissociate from CD14 and binds to lipoproteins, where its toxic activities are lost. CD14 also binds many other bacterial molecules, including lipoarabinomannans from mycobacteria, mannuronic acid polymers from Pseudomonas, and peptidoglycans from Staphylococcus aureus [36].

6.2.10.2 Bacterial peptidoglycans Peptidoglycans are polymers of alternating N-acetyl glucosamine and N-acetyl muraminic acid and are major constituents of the cell walls of both Gram-positive and -negative bacteria. PRRs that can bind these peptidoglycans include some TLRs, NODs, and CD14. Peptidoglycan recognition proteins (PGRPs) are PRRs that induce the production of proinflammatory and antimicrobial peptides [37]. They are found in humans, mice, cattle, and pigs. In pigs, PGRPs are expressed constitutively in the skin, bone marrow, intestine, liver, kidney, and spleen. One specific form, bovine PGRPS can kill microorganisms in which the peptidoglycan is either buried (Gram-negative bacteria) or absent (Cryptococcus), raising questions about its precise ligand. PGRP-S also binds bacterial LPS and lipoteichoic acids [38]. It is found in bovine neutrophil granules, and these neutrophils release PGRP-S when exposed to bacteria. Thus PGRPS probably plays a significant role in the resistance of cattle to bacterial infections.

6.2.11 Bacterial DNA Bacterial DNA can stimulate innate immunity because it is structurally different from eukaryotic DNA. Much of it contains large amounts of the dinucleotide, unmethylated cytosine-guanosine (CpG). (The cytosine in eukaryotic DNA is normally methylated, but not in bacteria.) Unmethylated CpG di-nucleotides bind and activate TLR9. Bacterial DNA also contains deoxyguanosine (dG) nucleotides. These dG nucleotides form molecular structures that differ from the usual mammalian DNA double helix. They also bind to TLR9 and trigger the production of cytokines such as TNF-α, IL-6, and IL-12.

78

SECTION | 1 Mammalian immunology

6.2.12 Viral nucleic acids Viruses have few characteristic molecular signatures. However, their nucleic acids are structurally different from those in animals, so they can be recognized by intracellular PRRs. TLR9 binds DNA from viruses and intracellular bacteria, whereas TLR7 and TLR8 bind ssRNA from viruses. TLR3, in contrast, mainly binds viral dsRNA, but it can also recognize some ssRNA and some dsDNA viruses. Intracellular RLRs also bind and respond to viral dsRNA. TLR7 and TLR9 activate the MyD88-mediated signaling pathways and trigger the production of inflammatory cytokines and type I IFNs. TLR3 uses another signaling molecule, the “TIR-domain-containing adaptor protein inducing IFN-β (TRIF).” The TRIF pathway activates the transcription factor IRF3 which then activates the genes encoding inflammatory cytokines and IFN-β.

6.2.13 Damage-associated molecular patterns Inflammation and sickness can be triggered not only by microbial infection but also by physical trauma and tissue damage. Thus the TLRs recognize not only PAMPs from invading microorganisms but also molecules escaping from dead, dying, and damaged tissues. These molecules, collectively called DAMPs or “alarmins,” may be released when cells die (intracellular) or generated when the connective tissue is damaged (extracellular). Other DAMPs may be produced by stimulated sentinel cells. Some of these DAMPs have potent antimicrobial properties. Others recruit and activate cells of the innate immune system and so promote adaptive immunity (Fig. 6.2). Mitochondria provide a link between PAMPs and DAMPs. Mitochondria are cytoplasmic organelles that generate energy for cells. They evolved from intracellular bacteria and retain many of their original bacterial features. Indeed, in many respects, they act like intracellular bacteria. For example, they have DNA that is rich in unmethylated CpG. When cells die, broken mitochondria may release large amounts of this DNA that can bind TLR9 and trigger inflammation. Mitochondria, like bacteria, also contain proteins with a formyl group at their amino termini. When these formylated proteins escape, they bind and activate neutrophil chemotactic receptors. These neutrophils leave blood vessels and migrate into tissues where they release their proteases and cause damage. One of the most important intracellular DAMPs is the high mobility group box protein-1 (HMGB1). HMGB1 normally binds DNA molecules and ensures that they fold correctly. However, HMGB1 is also a potent trigger of inflammation [39]. It is secreted by macrophages that have been activated by LPS or by cytokines such as IFN-γ. HMGB1 also escapes from broken cells. HMGB1 binds both TLR2 and TLR4 and so sustains and prolongs inflammation. It stimulates the secretion of inflammatory cytokines from macrophages, monocytes, neutrophils, and endothelial cells. Administration of HMGB1 causes fever, weight loss, anorexia, acute lung injury, arthritis, and even death. HMGB1 stimulates the growth of new blood vessels and tissue repair. It also has potent antimicrobial activity [39]. The cytokine IL-33 is also stored within the nucleus and released when cells die. It too is a potent DAMP as well as a regulator of cellular energy metabolism and thermogenesis (Chapter 4). Many other molecules released from broken cells act as intracellular DAMPs. These include adenosine and adenosine triphosphate, uric acid, S100 proteins (a family of calcium-binding proteins involved in cell growth and tissue injury), and heat-shock proteins. An important extracellular DAMP is heparan sulfate. This molecule is normally found in cell membranes and the extracellular matrix but is shed into tissue fluids following injury. Heparan sulfate binds and triggers TLR4 as well as some NK cell receptors. Other extracellular DAMPs include hyaluronic acid, fibronectin, and peptides from collagen and elastin.

6.2.14 Soluble pattern-recognition molecules Although the TLRs, NLRs, and RLRs are expressed on cell surfaces, soluble pattern recognition molecules also bind PAMPs. Because these molecules function in the extracellular fluid, they promote cellular destruction (phagocytosis) of any organisms they encounter. In general, they do not induce the expression of inflammatory cytokines. That task is accomplished by the cell-surface PRRs (Fig. 6.3) [40]. Since many bacterial PAMPs are glycoproteins and polysaccharides, circulating carbohydrate-binding lectins play important roles in innate immunity. Three extracellular lectin families, the S-, C-, and P-type lectins, are involved in innate immunity. C-type lectins (CTLs) are a large family of carbohydrate-binding proteins with many different roles. (At least 1000 have been identified.) All require calcium to bind to their carbohydrate ligands. Each end of a CTL peptide chain has a distinct function; the C-terminal domain binds to carbohydrates, whereas the N-terminal domain interacts with cells or

Innate immunity: basic features Chapter | 6

79

complement components, thereby exerting their biological effect. CTLs may be both soluble and membrane-bound. The most important of the soluble CTLs is mannose-binding lectin (MBL). MBL is found in high levels in the blood. It has multiple sites that bind oligosaccharides, such as N-acetylglucosamine, mannose, glucose, galactose, and N-acetylgalactosamine. MBL thus binds strongly to bacteria such as Salmonella enterica and Listeria monocytogenes. It binds to Escherichia coli with moderate affinity. It binds strongly to yeasts such as C. albicans and Cryptococcus neoformans. It can also bind viruses such as influenza A as well as parasites such as Leishmania. Bacteria coated by MBL are readily ingested by phagocytic cells. MBL plays an important role in activating the complement system. There are two forms of MBL in the pig: MBL-A and MBL-C. These can bind to Actinobacillus suis and Haemophilus parasuis. Some European pig breeds may express very low levels of MBL-C and hence suffer from increased disease susceptibility. Multiple CTLs such as the surfactant proteins SP-A and SP-D are produced in the lungs. Six different soluble CTLs (conglutinin, MBL, pulmonary surfactant proteins [SP-A, SP-D], and collectins-46 [CL-46 and CL-43]) have been identified in mammals. However, conglutinin, CL-46, and CL-43 have been identified primarily in Bovidae (Chapter 17). Some CTLs are expressed on cell surfaces where they can bind bacteria, fungi, and some viruses. The most important are the dectins. The dectins bind β-glucans on fungal cell walls and play an important role in antifungal defense by promoting their intracellular destruction. Dectin-1 is expressed by bovine macrophages, monocytes, and dendritic cells. Bovine dectin-2 is expressed on Langerhans cells in the skin. DEC-205 is expressed on bovine dendritic cells. Selectins are CTLs expressed on vascular endothelial cells and play a key role in the emigration of leukocytes from the bloodstream into the tissues in inflammation. Galectins are extracellular S-type lectins. Their name derives from their specificity for galactosides. They play a role in inflammation by binding leukocytes to the extracellular matrix. Other soluble collectins include the ficolins, (H-, L-, and M-ficolins) produced by the liver and some lung cells. These too, can bind bacterial carbohydrates and activate the complement system. There are also many cell-associated lectins. DC-SIGN, for example, is a lectin expressed on macrophages and dendritic cells. Not only does DC-SIGN recognize bacterial carbohydrates, but it also recognizes T cells and so is used by dendritic cells to interact with T cells. The collectins are especially important in the defense of young animals whose immature adaptive immune system is not capable of mounting an efficient adaptive response. P-type lectins are also called pentraxins. Pentraxins are formed by five protein subunits arranged in a flat disk. They serve as pattern recognition molecules. One face of the disk recognizes diverse molecular patterns. The opposite face interacts with complement component C1q in addition to several immunoglobulin Fc receptors. Pentraxins have multiple biological functions, including activation of complement and stimulation of leukocytes. They bind bacterial LPS in a calcium-dependent manner and activate the classical complement pathway by interacting with C1q. They also interact with Fc receptors on neutrophils, monocyte/macrophages, and NK cells and augment their activities [41]. Two of the most important pentraxins are C-reactive protein and serum amyloid P. C-reactive protein (CRP) is the major acute-phase protein produced in primates, rabbits, hamsters, and dogs and is important in pigs. CRP has a pentameric structure (five 20-kDa units arranged in a disk) with two faces. One face binds phosphocholine, a common side chain found in all cell membranes and many bacteria and protozoa. The other face of the disk binds to the antibody receptors FcγRI and FcγRIIa on neutrophils. CRP thus acts as an opsonin and promotes the phagocytosis and removal of damaged, dying, or dead cells in addition to microorganisms. CRP can bind to bacterial polysaccharides and glycolipids and to necrotic cells, where it activates complement C1q. (Its name derives from its ability to bind and precipitate the C-polysaccharide of Streptococcus pneumoniae.) CRP also has an anti-inflammatory role since it inhibits neutrophil superoxide production and degranulation and blocks platelet aggregation. CRP stimulates fibrosis and may promote healing by reducing damage and enhancing the repair of damaged tissue. In lactating cows, serum CRP rises two- to fivefold. The reasons for this are unknown. Serum amyloid P (SAP) is also a pentraxin and the major acute-phase protein in rodents [42]. Like CRP, it is a PRR, where one face of the disk can bind nuclear constituents such as DNA, chromatin, and histones as well as cell membrane phospholipids. The other face binds and activates C1q and thus triggers the classical complement pathway. A major function of SAP is to regulate innate immune responses. It interacts with macrophage Fc receptors, reduces the binding of neutrophils to the extracellular matrix, reduces the differentiation of macrophages into fibroblasts so inhibiting fibrosis, and promotes phagocytosis of cell debris. Other soluble PRRs that act as acute-phase proteins include LPSbinding protein in cattle, humans, and rabbits, CD14 in humans, horses, and mice, and C-type lectins such as mannosebinding lectin and conglutinin in other species. LPS-binding protein is an acute-phase protein that presents LPS to CD14 and TLR4 on phagocytic cells and enhances their proinflammatory activity up to 1000-fold. It can also bind lipoteichoic acids on the cell wall of Gram-positive bacteria and trigger inflammation through TLR2 activation. Haptoglobin is another major acute-phase protein in ruminants, horses, and cats. It can rise from virtually undetectable levels in normal calves to as high as 1 mg/mL in calves with acute respiratory disease. Haptoglobin binds

80

SECTION | 1 Mammalian immunology

to iron and makes it unavailable to invading bacteria. Mammals may also capture iron by stealing bacterial siderophores.

6.3

Inflammasomes

When PAMPs and DAMPs bind to NOD-like receptors, they initiate the assembly of large multiprotein complexes called inflammasomes (Fig. 6.6) [43]. These complexes often require TLR- or NLR-mediated activation. The inflammasomes then activate two proteolytic enzymes, caspase-1, and caspase 11. Caspase 11 triggers cell death by a process called pyroptosis. Caspase 1 acts on pro-IL-1 and pro-IL-18, to generate the active forms of these cytokines [44]. Four different types of inflammasome have been characterized, each generated by different PAMPs and DAMPs, and each containing slightly different components and thus presumably generating different cytokines and proinflammatory molecules. Inflammasome-mediated responses are important in controlling microbial infections as well as in regulating some metabolic processes and immune responses, especially in the intestinal mucosa. Caspase-1 deficient mice are more susceptible to bee and snake venoms suggesting that inflammasomes also play a role in defense against envenomation. Although some inflammasomes require TLR-mediated activation and downstream NF-κB mediated transcription of pro-IL-1 and IL-18, others do not require this transcriptional activation step [45].

6.4

Inflammatory cytokines

6.4.1 Tumor necrosis factor-alpha TNF-α is a 17 kDa protein produced by sentinel cells in response to TLR stimulation. TNF-α can also be produced by endothelial cells, T cells, B cells, and fibroblasts. It is initially membrane-bound but is cleaved from the cell surface by TNF-α convertase. TNF-α is a member of a family of related molecules whose other members include TNF-β, CD40L, and FasL (CD95L). Soluble TNF-α triggers the release of chemokines and cytokines from nearby cells and promotes leukocyte adherence, migration, attraction, and activation. TNF-α is an essential mediator of inflammation because, in combination with IL1, it triggers changes in small blood vessels. A local increase in TNF-α causes the classic signs of inflammation, including heat, swelling, pain, and redness. TNF-α can depress cardiac output, induce microvascular thrombosis, and cause capillary leakage. TNF-α acts on neutrophils to enhance their ability to kill microbes. It attracts neutrophils to sites of tissue damage and increases their adherence to vascular endothelium. It stimulates macrophage phagocytosis and oxidant production. It amplifies and prolongs inflammation by promoting macrophage synthesis of enzymes such as nitric oxide synthase (NOS) and cyclooxygenase. TNF-α induces macrophages to increase their synthesis together with that of IL-1.

6.4.2 Interleukin-1 When stimulated through CD14 and TLR4, sentinel cells such as macrophages also synthesize cytokines belonging to the interleukin-1 family. The most important of these are IL-1α and IL-1β. IL-1β is produced as a large precursor LIGAND

TLR

LIGAND

INFLAMMASOME

NLR

MAPK NF-NB

Caspase-11 Gene transcription

Pro-Cytokines

Caspase-1

Pyroptosis

Active cytokines

SECRETED FIGURE 6.6 The activation of Inflammasomes. Inflammasomes are star-shaped multienzyme complexes that upon activation, mediate the production of inflammatory cytokines as well as their activating caspases.

Innate immunity: basic features Chapter | 6

81

protein that is cleaved by caspase-1 to form the active 17.5-kDa molecule. Ten- to 50-fold more IL-1β is produced than IL-1α, and whereas IL-1β is secreted, IL-1α remains attached to the cell surface. As a result, IL-1α only acts on cells in direct contact with macrophages. Transcription of IL-1β messenger RNA (mRNA) occurs within 15 minutes of ligand binding. It reaches a peak three to four hours later and levels off for several hours before declining. Like TNF-α, IL-1β acts on nearby cells to initiate and amplify inflammation. For example, it acts on vascular endothelial cells to make them adhesive to neutrophils. IL-1 also acts on macrophages to stimulate their synthesis of NOS2 and COX-2. During severe infections, IL-1β circulates in the bloodstream where, in association with TNF-α, it is responsible for sickness behavior. It is an endogenous pyrogen that acts on the brain to cause fever, lethargy, and malaise. It acts on muscle cells to mobilize amino acids, causing pain and fatigue. It acts on liver cells to induce the production of acutephase proteins. The most important IL-1 receptors are CD121a and CD121b. CD121a is a signaling receptor, whereas CD121b is not. CD121b thus binds IL-1 but nothing more happens. If soluble CD121b binds IL-1 it, therefore, acts as a blocking antagonist. IL-1 activity is also regulated by IL-1 receptor antagonist (IL-1RA), a protein that binds and blocks CD121a. IL-1RA is therefore an important regulator of IL-1 activity and inflammation. IL-1 is a member of a large family of cytokines that regulate innate immune responses. Other important family members include IL-1RA, IL-18, IL-33, IL-36, IL-37, and IL-38. All these cytokines signal through closely related receptors. Some, like IL-36, have a proinflammatory effect, while others such as IL-37, have anti-inflammatory effects.

6.4.3 Interleukin-6 Interleukin-6 is a glycoprotein of about 20 kDa produced by macrophages, T cells, and mast cells. Its production is triggered by bacterial LPS, as well as by IL-1, and TNF-α. IL-6 affects both inflammation and adaptive immunity. It promotes inflammation and is a major mediator of the acute-phase reaction and septic shock. IL-6 is also produced by muscles during exercise. IL-6 may have an anti-inflammatory role in that it inhibits some activities of TNF-α and IL-1 and promotes the production of IL-1RA as well as the suppressive cytokine IL-10.

6.4.4 Chemokines Chemokines are a family of at least 50, relatively small, 8
10 kDa cytokines. They coordinate the migration of leukocytes and hence dictate the course of many inflammatory and immune responses [46]. Chemokines are produced by sentinel cells, including macrophages and mast cells. They are classified into four subfamilies based on their amino acid sequences. For example, the CXCchemokines have two cysteine (C) residues separated by another amino acid (X), whereas the CC chemokines have two contiguous cysteine residues. (Chemokine nomenclature is based on this classification, each molecule or receptor receiving a numerical designation. In addition, ligands have the suffix “L” [e.g., CXCL8], whereas receptors have the suffix “R” [e.g., CXCR1].) Inflammatory chemokines are mainly involved in leukocyte recruitment. Others promote new blood vessel growth. Some are homeostatic and are found in normal tissues where they regulate cellular migration and homing, and many have overlapping functions [47]. One of the most important proinflammatory chemokines is CXCL8 (also called interleukin-8). CXCL8 attracts and activates neutrophils, releasing their granule contents and stimulating their respiratory burst. Another important chemokine is CXCL2 (otherwise called macrophage inflammatory protein-2, MIP-2), which is secreted by macrophages and also attracts neutrophils. CC chemokines act predominantly on lymphocytes. For example, CCL3 and CCL4 (MIP-1α and -1β) are produced by macrophages and mast cells. CCL4 attracts CD41 T cells whereas CCL3 attracts B cells, eosinophils, and cytotoxic T cells. CCL2 (monocyte chemotactic protein-1, MCP-1) is produced by macrophages, T cells, fibroblasts, keratinocytes, and endothelial cells. It attracts and activates monocytes, stimulating their respiratory burst and lysosomal enzyme release. CCL5 is produced by T cells and macrophages. It attracts monocytes, eosinophils, and some T cells. It activates eosinophils and stimulates histamine release from basophils [48]. Two chemokines fall outside the CC and CXC families. Lymphotactin (XCL1) is a C (only one cysteine residue) chemokine, that is chemotactic for lymphocytes. Its receptor is XCR1. Fractalkine (CX3CL1) is a CXXXC (two cysteines separated by three amino acids) chemokine, that triggers adhesion by T cells and monocytes. Its receptor is CX3CR1. Most chemokines are produced in infected or damaged tissues and attract other cells to sites of inflammation or microbial invasion. The complex chemokine mixture produced by damaged or infected tissues likely regulates the

82

SECTION | 1 Mammalian immunology

precise composition of incoming inflammatory cell populations. In this way, the body can adjust the inflammatory response to optimize the destruction of different microbial invaders. Many chemokines, such as CXCL4, CCL20, and CCL5, are structurally similar to antimicrobial defensins. Chemokines play a major role in infections and inflammation in mammals. Cattle possess fewer cytokines than humans but also have others not found in humans [49]. For example, regakine-1 is a CC chemokine found in bovine serum that acts with CXCL8 and C5a to attract neutrophils and enhance inflammation. Impaired neutrophil migration is associated with certain specific CXCR2 genotypes, and this may result in increased susceptibility to bacterial mastitis in dairy cattle [50].

6.4.5 Interferons Among the most important innate defenses against viral invasion is the interferon system. There are three types of interferons, type I that include, most importantly IFN-α and IFN-β, type II IFNs that include IFN-γ, and Type III IFNs that include the IFN-λs. The interferons are generated in response to signaling through PRRs and their associated inflammasomes. When interferons act on cellular targets they upregulate hundreds of interferon-stimulated genes (ISGs) [51]. This interferome is complex and most importantly, differs significantly between different mammalian species. As might be expected, the antagonism between the interferons and invading viruses results in an “arms race.” Comparative studies on the composition of the interferome of different species have rarely been investigated. However, such studies on the type I interferome have now revealed that as expected, although the number of IFN upregulated genes varies between species, their distribution pattern is generally similar [51]. In one study the interferomes of humans, rats, cows, sheep, pigs, horses, dogs, Myotis lucifugus (little brown bat), and Pteropus vampyrus (large flying fox) were investigated [51]. Some such as cattle and sheep are similar, as expected, however, the human and pig interferomes are also closely related. In general, there is a core set of 71 ISGs that are upregulated in all nine mammal species studied. These have likely been conserved for over 310 my and are presumably ancestral in origin. They include genes encoding proteins involved in antigen presentation such as MHC genes; IFN induction and response proteins including the key adaptor molecule MyD88; IFN suppression, ubiquitination, and protein degradation; cell signaling and apoptosis as well as antiviral responses. Except for MyD88, there is limited upregulation of nucleic acid-sensing genes, and all are linked to RNAbinding rather than DNA. This suggests that the type I IFN responses are biased towards RNA viruses. There is also a link between IFN and the synthesis of early components of the complement system. C1r and C1s are upregulated by IFN in all species examined except cattle. This is also the case with the negative regulator of C1r and C1s, the C1inhibitor. Very few genes are downregulated among the ISGs in response to virus invasion. While each species uses some specific upregulated genes, the overall pattern in the two bat species examined is unremarkable with similar numbers of upregulated and downregulated genes. However, the basal transcription level of the type I interferome is higher than in other species, suggesting that bats might live in a constitutive antiviral state [51]. (Chapter 18). A closer analysis of the ISGs upregulated in primates has found that the greatest evolutionary pressures are not on the type 1 interferon genes, but on the ISGs. About a third of the ISGs examined show evidence of positive selection [52]. There are also significant species differences in the responses to IFN-γ. Thus in mice, IFN-γ induces the expression of eighteen different GTPases. These immunity-related GTPases (IRG) in turn enhance immunity to both bacteria and protozoa. For example, the IRG proteins in mice induce autophagy in pathogen-loaded macrophages thus ensuring pathogen destruction. This autophagy pathway enables mouse macrophages to kill mycobacteria and overcome the inhibition of their phagolysosomes. These IFN-γ -activated macrophages have a similar ability to destroy intracellular Toxoplasma gondii [53]. In contrast to mice, humans express only two IRG proteins and these are not induced by IFNγ. They participate in autophagy but do not have antimicrobial activity. Mx proteins are large GTPases with potent activity against negative-stranded RNA viruses. Their expression is controlled by type I and type III interferons. Mx genes are present in all mammals except for the opossum (Monodelphis domestica). Mammals have two Mx genes, MX1 and MX2. The Mx locus in rodents is rearranged in comparison to other mammals so that their MX2 gene has been lost while MX1 is duplicated. In rats, three MX genes are present while in the guinea pig, MX2 is 4.3 mb away on the same chromosome. The two platypus MX genes resemble the arrangement in eutherian mammals although they are more closely related to each other than to other mammals [54].

Innate immunity: basic features Chapter | 6

6.5

83

Leukocytes

6.5.1 Sentinel cells The cells whose primary function is to recognize and respond to invading microbes are considered sentinel cells. The major sentinel cell types, namely macrophages, dendritic cells, and mast cells, are scattered throughout the body but are found in the highest numbers just below body surfaces where invading microorganisms are most likely to be encountered. All these cells are equipped with multiple, diverse PRRs, so they can detect and then respond rapidly to both PAMPs and DAMPs. Other cells, such as epithelial cells, endothelial cells, neutrophils, and fibroblasts, may serve as sentinel cells when the opportunity arises.

6.5.2 Blood cells There are great differences among mammals in the relative proportions of their white blood cell types. For example, humans have 50%
70% neutrophils (3500
7000 cells/μL) and 20%
40% lymphocytes (1400
4000 cells/μL). In contrast, mice have only 10%
25% neutrophils but they have 75%
90% lymphocytes. In most mammals, the normal white cell count varies from less than 1000 /μL in deer to 15,000 /μL in some carnivores and non-human primates [55]. The neutrophil:lymphocyte ratio is also highly variable. Neutrophils tend to predominate in carnivores whereas, in ruminants, lymphocytes generally predominate. Changes in these populations also occur throughout life. For example, piglets are born with predominantly neutrophils but by weaning, lymphocytes predominate in their bloodstream [55].

6.5.3 Neutrophils Neutrophils are the major phagocytic cells that participate in acute inflammation. They are generally similar across mammals but may vary somewhat based on the size and specific staining properties of their granules (Fig. 6.7) [56]. In rabbits and hystricomorph rodents (guinea pigs, porcupines, and capybaras) these cells tend to stain more with acid dyes such as eosin and contain prominent eosinophilic granules. They are therefore called heterophils or pseudoeosinophils [57]. Despite this, rabbit neutrophils are the functional equivalents of those in other mammals. Likewise, the degree of nuclear lobulation varies among mammals. The nuclei of mice and some rodents may be Ushaped without obvious lobules. Conversely, the neutrophils of non-human primates tend to be hyperlobulated. In the case of humans, they are primarily attracted to inflammatory sites by the chemokine IL-8. On the other hand, in mice, their major chemoattractant is CXCL15 (lungkine). Neutrophils are also attracted to infection sites by N-formyl peptides from Gram-negative bacteria. Humans have two high-affinity receptors for formyl peptides while mice have three low-affinity ones. Neutrophil L-selectin from humans binds to E-selectin on vascular endothelial cells but this is not the case in mice. The gene encoding another chemokine, CCL16, has been pseudogenized in some species of leporids to generate a premature stop codon [58]. Defensins are expressed in large amounts in human neutrophils where they amount to 30%
50% of the protein content of the azurophilic granules. Mouse neutrophils, on the other hand, do not express any defensins! Conversely, Paneth cells located in the small intestine of mice express more than twenty different defensins while human Paneth cells express only two [59]. These defensins also differ in the way they are processed. Myeloperoxidase is a key component of the neutrophil respiratory burst. The amount of this enzyme in mouse neutrophils is 10%
20% of the level in human cells. Likewise, mouse neutrophils have lower levels of β-glucuronidase, lysozyme, and alkaline phosphatase. FIGURE 6.7 Images of typical blood neutrophils from three different mammalian species. In other species such as rabbits, their staining properties may be different. (See Figure 22
1). Courtesy of Dr. Mark Johnson.

Horse

Cat

Dog

84

SECTION | 1 Mammalian immunology

6.5.4 Eosinophils The proportion of eosinophils among the blood leukocytes varies greatly since their numbers are affected by the presence of parasites, especially by migrating helminth larvae. In general, the numbers and morphology of blood eosinophils in domestic mammals are similar. The shape and size of eosinophil granules in contrast, are highly variable among mammals. Some species such as bottlenose dolphins (Tursiops truncatus) and owl monkeys ( Aotus trivirgatus) appear to have a high eosinophil count in the absence of parasites [55]. In the dog (beagles) there are about 300
500 eosinophils/μL. Normal values range from about 2% of all leukocytes in dogs to about 10% in cattle. In unparasitized humans, there are less than 400 /μL. In cats, there are from 200 to 600 eosinophils/μL although some publications report numbers as high as 1500/μL. These higher numbers likely reflect the presence of parasites. Cat eosinophil granules may be rod-shaped. In horses, eosinophil numbers range from 0 to 1000/μL. Their granules are large and may obscure the bilobed nucleus. In cattle, eosinophils are packed with small granules. As in other species, their numbers fluctuate, and they may show seasonal eosinophilia reflecting changes in parasite load. In general, their absolute counts range from 100 to 1200 eosinophils/ μL. These numbers and fluctuations are similar to those seen in sheep and pigs. The eosinophils of humans and laboratory rodents as well as cats and goats contain specific granules with an electron-dense core surrounded by a less dense matrix. They form biconcave disks with a crystalloid core consisting of nanocrystals containing inert deposits of major basic protein and eosinophil peroxidase. The core is surrounded by a matrix containing two ribonucleases, eosinophil cationic protein, and eosinophil-derived neurotoxin. Specific granules are also loaded with a mixture of cytokines and growth factors as well as multiple enzymes. Cats show much diversity in their granule size and structure. Only about 10% of canine eosinophil granules possess a distinct core. Horse eosinophil granules have a dense core that varies in position within each granule and does not appear to be crystalline. Cattle and mink have granules that lack a dense core and appear to be homogeneous. While rare in most mammals, basophils can be frequently found in rabbit blood smears [55].

6.5.5 Macrophages The most important sentinel cells are macrophages. Macrophages scattered throughout the body can capture, kill, and destroy microbial invaders. When activated they can differentiate into two populations. Proinflammatory M1 cells and anti-inflammatory M2 cells. There are species differences in the metabolic changes induced during these states. The most important difference is in arginine metabolism. Thus in mouse macrophages, when exposed to IFN-γ or LPS, nitric oxide synthase (iNOS) produces large quantities of NO and L-citrulline. However, in human macrophages, iNOS is not expressed and NO is induced by a different set of stimulants such as IFN-α/β and some chemokines. Thus IFN-α and -β can induce nitric oxide expression in human macrophages but IFN-α has an opposite effect in mice. Macrophages also differ in their arginase expression. In mice, cytoplasmic arginase is upregulated in M2 cells, and it antagonizes the iNOS in M1 cells. In humans, arginase is present in neutrophil azurophilic granules. Humans do not make ascorbic acid, but mice do, where it is an important antioxidant [60]. One significant difference between mammals is the presence or absence of pulmonary intravascular macrophages [61]. Historically, based on studies in laboratory rodents and rabbits it was believed that in mammals, bacteria, immune complexes, and other particulate materials were removed from the bloodstream by the large populations of macrophages found in the liver, spleen, and bone marrow.
the mononuclear phagocytic system. However subsequent studies have revealed that other species such as sheep, cattle, horses, cats, and pigs, capture circulating particles in their pulmonary capillaries. These particles are removed by macrophages that line the endothelium of lung capillaries (pulmonary intravascular macrophages) (Figure 15
4). These are large mature macrophages 20-80μm in diameter that are anchored to the pulmonary capillary endothelium. Intravenous inoculation of colloidal gold particles has shown that up to 60% of these particles are taken up by pulmonary intravascular macrophages in the lungs of sheep, calves, pigs, and cats. However, less than 1.1% are taken up in the lungs of macaques (M. fasciata), hyraxes, rabbits, guinea pigs, rats, or mice [61]. In all these species, any remaining particles were captured by liver macrophages.

6.5.6 Dendritic cells The second major population of sentinel cells consists of dendritic cells, so-called because many possess long, thin cytoplasmic processes called dendrites that can trap invaders. Dendritic cells encompass several distinct cell types, many of which are closely related to, or derived from, macrophages. DCs are subdivided into classical cDCs and

Innate immunity: basic features Chapter | 6

85

plasmacytoid pDCs. cDCs are specialized antigen-presenting cells. pDCs sense nucleic acids and as a result are the major producers of type I interferons. cDCs can be further subdivided into those that induce type 1 immune responses (cDC1s), and those that induce type 2 immune responses (cDC2s). The DC system is relatively well conserved within mammals. Thus bovine DCs are found in afferent lymph where two subsets are found, a major subset CD1bhi CDllaCD132CD262 and CD1721. as well as a minor subset of CD1blo CD11a1 CD131 CD261 and CD172a2. The conserved expression of CD26 and CD172a suggest that these are DC1 and DC2 cells, respectively. A similar pattern exists in sheep and pigs [62].

6.5.7 Mast cells A third population of professional sentinel cells consists of mast cells. These cells, strategically located close to epithelial and endothelial surfaces, are among the first to detect pathogens and danger signals. They express multiple PRRs and are packed with cytosolic granules that store a complex mixture of inflammatory mediators. When released in response to appropriate stimuli, these mediators promote the clearance of pathogens. Long known to play a key role in allergies, it has now been recognized that they also trigger inflammation in conventional situations. In humans, mast cells mainly produce IL-5 while in mice they produce TNF-α. Mouse mast cells produce large amounts of IL-4 whereas human mast cells do not.

6.6

The costs of innate immunity

Compared to adaptive immune responses, innate responses are costly and uncomfortable. As a result, as species thrive and expand their ranges, they tend to rely increasingly on adaptive immunity. Too high an energy cost may affect resource use, behavior, and eventually survival. Thus innate immune responses and their accompanying fevers, fatigue, depression, and inappetence are potentially lethal. Energy expenditure must be greatly increased in order to generate a fever; there is accelerated protein turnover, and nutrients are sequestered from both the microbiota and the victim. They result in decreased mobility, slower growth rates in the young, and impaired reproduction in adults. Thus innate responses are costly and often maladaptive. Vigorous innate responses are best avoided. In practice, this is readily observed in both humans and other mammals. The healthy and fit are the individuals that tend to take the lead in exploring new territories and environments [63].

References [1] Bianchi ME. DAMPS, PAMPs and alarmins: all we need to know about danger. J Leukoc Biol 2007;81:1
5. [2] Hagai T, Chen X, Miragaia RJ, Rostom R, et al. Gene expression variability across cells and species shapes innate immunity. Nature 2018;563:197
202. [3] Sahi H-G, Pag U, Bonness S, et al. Mammalian defensins: structures and mechanism of antibiotic activity. J Leukoc Biol 2005;77:466
75. [4] Lynn DJ, Bradley DG. Discovery of α-defensins in basal mammals. Dev Comp Immunol 2007;31:963
7. [5] Patil A, Hughes AL, Zhang G. Rapid evolution and diversification of mammalian a-defensins as revealed by comparative analysis of rodent and primate genes. Physiol Genomics 2004;20:1
11. [6] Sang Y, Ortega MT, Blecha F, et al. Molecular cloning and characterization of three β-defensins from canine testes. Infect Immunol 2005;73:2611
20. [7] Leonard BC, Affolter VK, Bevins CL. Antimicrobial peptides: agents of border protection for companion animals. Vet dermatol 2012;23(3) 177-e36. [8] Bruhn O, Cauchard J, Schlisselhuber M, et al. Antimicrobial prop equine α-defensin DEFA1 bacterial horse pathogens. Vet Immunol Immunopathol 2009;130:102
6. [9] Leonard BC, Marks SL, Outerbridge CA, Affolter VK, et al. Activity, expression and genetic variation of canine beta-defensin 103: a multifunctional antimicrobial peptide in the skin of domestic dogs. J Innate Immun 2012;4(3):248
59. [10] Zhu S, Gao B. Positive selection in cathelicidin host defense peptides: adaptation to exogenous pathogens or endogenous receptors. Heredity 2017;118:453
65. [11] Baumann A, Demoulins T, Python S, Summerfield A. Porcine cathelicidins efficiently complex and deliver nucleic acids to plasmacytoid dendritic cells and can thereby mediate bacteria-induced IFN-alpha responses. J Immunol 2014;193:364
71. [12] Sang Y, Teresa Ortega M, Rune K, et al. Canine cathelicidin (K9CATH): gene cloning, expression, and biochemical activity of a novel promyeloid antimicrobial peptide. Dev Comp Immunol 2007;31:1278
96. [13] Nonaka M, Miyazawa S. Evolution of the initiating enzymes of the complement system. Genome Biol 2001;3(1) 1001.1
1001.5. [14] Fujita T. Evolution of the lectin-complement pathway and its role in innate immunity. Nat Rev Immunol 2002;2:346
86.

86

SECTION | 1 Mammalian immunology

[15] Minton K. Innate immunity: the inside story on complement activation. Nat Rev Immunol 2014;. Available from: https://doi.org/10.1038/ nri3603. [16] Cagliani R, Forni D, Filippi G, Mozzi A, et al. The mammalian complement system as an epitome of host pathogen conflicts. Mol Ecol 2016;25:1324
39. [17] Nation GK, Saffold CE, Pua HH. Secret messengers: extracellular RNA communication in the immune system. Immunol Revs 2021;304:62
76. [18] Akira S. Mammalian toll-like receptors. Curr Opin Immunol 2003;15:5
11. [19] Du X, Poltorak A, Wei Y, Beutler B. Three novel mammalian toll-like receptors: gene structure expression, and evolution. Eur Cytok Network 2000;11:362
71. [20] Baldridge MT, King KY, Goodell MA. Inflammatory signals regulate hematopoietic stem cells. Trends Immunol 2011;32:57
65. [21] Jungi TW, Farhat K, Burgener IA, Werling D. Toll-like receptors in domestic animals. Cell Tissue Res 2011;343:107
20. [22] Liu G, Zhang H, Zhao C, Zhang H. Evolutionary history of the toll-like receptor gene family across vertebrates. Genome Biol Evol 2019;12 (1):3615
34. [23] Novak K. Functional polymorphisms in toll-like receptor genes for innate immunity in farm animals. Vet Immunol Immunopathol 2014;157:1
11. [24] Neves F, Agueda-Pinto A, Pinheiro A, Abrantes J, Esteves PJ. Strong selection of the TLR2 coding region among the Lagomorpha suggests an evolutionary history that differs from other mammals. Immunogenetics 2019;71:437
43. [25] Areal H, Abrantes J, Esteves PJ. Signatures of positive selection in toll-like receptor (TLR) genes in mammals. BMC Evol Biol 2011;. Available from: https://doi.org/10.1186/1471-2148-11-368. [26] Pinheiro A, Agueda-Pinto A, Melo-Ferreira J. Neves et al. analysis of substitution rates showed that TLR5 is evolving at different rates among mammalian groups. BMC Evol Biol 2019;. Available from: http://doi.org/10.1186/s12862-18-1547-4. [27] Shen T, Xu S, Wang X, Yu W, et al. Adaptive evolution and functional constraint at TLR4 during the secondary adaptation and diversification of cetaceans. BMC Evol Biol 2012;12:39. Available from: https://doi.org/10.1186/1471-2158-12-39. [28] Sharma V, Hecker N, Walther F, Stuckas H, Hiller M. Convergent losses of TLR5 suggest altered extracellular flagellin detection in four mammalian lineages. Mol Biol Evol. 2020;37(7):1847
54. [29] Blasius AL, Beutler B. Intracellular toll-like receptors. Immunity 2010;32:305
15. [30] Roach JC, Glusman G, Rowen L, Kaur A, et al. The evolution of vertebrate toll-like receptors. Proc Natl Acad Sci USA 2005;102 (27):9577
82. [31] Kawasaki T, Kawai T. Toll-like receptor signaling pathways. Front Immunol 2014;. Available from: https://doi.org/10.3389/fimmu.2014.00461. [32] Loo YM, Gale MJR. Immune signaling by RIG-I-like receptors. Immunity 2011;34:680
92. [33] Brunette RL, Young JM, Whitley DG, Brodsky IE, et al. Extensive evolutionary and functional diversity among AIM2-like receptors. J Exp Med 2012;209(11):1969
83. [34] Motwani M, Pesiridis S, Fitzgerald KA. DNA sensing by the cGAS-STING pathway in health and disease. Nature Rev Genetics 2019;. Available from: https://doi.org/10.1038/s41576.0151-1. [35] Xie J, Li Y, Shen X, Goh G, et al. Dampened STING-dependent interferon activation in bats. Cell Host Microbe 2018;23:297
301. [36] Vaure C, Liu Y. A comparative review of toll-like receptor 4 expression and functionality in different animal species. Front Immunol 2014;. Available from: https://doi.org/10.3389/fimmu.2014.00316. [37] Sang Y, Ramanathan B, Ross CR, Blecha F. Gene silencing and overexpression of porcine peptidoglycan recognition protein long isoforms: involvement in β-defensin-1 expression. Infect Immun 2005;73:7133
41. [38] Tydell CC, Yuan J, Tran P, Selsted ME. Bovine peptidoglycan recognition protein-S: antimicrobial activity, localization, secretion, and binding properties. J Immunol 2006;176:1154
62. [39] Yanai H, Ban T, Wang Z, et al. HMGB proteins function as universal sentinels for nucleic-acid-mediated innate immune responses. Nature 2009;462:99
103. [40] Werling D, Coffey TJ. Pattern recognition receptors in companion and farm animals: the key to unlocking the door to animal disease? Vet J 2007;174:240
51. [41] Yang R, Hu J, Zeng B, Yang D, et al. Structural characterization of immune receptor family short pentraxins., C-reactive protein and serum amyloid P component, in primates. Dev Comp Immunol 2022;. Available from: https://doi.org/10.1016/j.dci.2022.104371. [42] Cox N, Pilling D, Gomer RH. Serum amyloid P: a systemic regulator of the innate immune response. J Leukoc Biol 2014;96(5):739
43. [43] Lamkanfi M, Dixit VM. Mechanisms and functions of inflammasomes. Cell 2014;157:1013
22. [44] Bryant C, Fitzgerald KA. Molecular mechanisms involved in inflammasome activation. Trends Cell Biol 2009;19:455
64. [45] Vrentas CE, Schaut RG, Boggiatto PM, Olsen SC, et al. Inflammasomes in livestock and wildlife: insights into the intersection of pathogens and natural host species. Vet Immunol Immunopathol 2018;201:49
56. [46] Baggiolini M. Chemokines in pathology and medicine. J Intern Med 2001;250:91
104. [47] Zlotnik A, Yoshie O. The chemokine superfamily revisited. Immunity 2012;36:705
16. [48] Gangur V, Birmingham NP, Thanesvorakul S. Chemokines in health and disease. Vet Immunol Immunopathol 2002;86:127
36. [49] Widdison S, Coffey TJ. Cattle and chemokines: evidence for species-specific evolution of the bovine chemokine system. Anim Genet 2011;42:341
53. [50] Rambeaud M, Pighetti GM. Impaired neutrophil function associated with specific bovine CXCR2 genotypes. Infect Immun 2005;73:4955
9.

Innate immunity: basic features Chapter | 6

87

[51] Shaw AE, Hughes J, Gu Q, Behdenna A, et al. Fundamental properties of the mammalian innate immune system revealed by multispecies comparison of type 1 interferon responses. PLoS Biol 2017;. Available from: http://doi.org/10.1371/journal.pbio.2004086. [52] Judd EN, Gilchrist AR, Meyerson NR, Sawyer SL. Positive natural selection in primate genes of the type I interferon response. BMC Ecol Evo 2021;. Available from: http://doi.org/10.1186/s12862-021-01783-z. [53] Zschaler J, Schlorke D, Arnhold J. Differences innate immune responses man mouse. Crit Rev Immunol 2014;34:433
54. [54] Verhelst J, Hulpiau P, Saelens X. Mx proteins: antiviral gatekeepers that restrain the uninvited. Microbiol Mol Biol Revs 2013;77(4):551
66. [55] Hawkey CM. Comparative Mammalian Hematology. Cellular components and blood coagulation of captive wild animals. London: Heinemann Medical books; 1975. [56] Bertram TA. Neutrophil leukocyte structure and function in domestic animals. Adv Vet Sci Comp Med 1985;30:91
129. [57] Montali RJ. Comparative pathology of inflammation in the higher vertebrates (Reptiles, birds, and mammals). J Comp Path 1988;99:1
26. [58] Neves F, Abrantes J, Lopes AM, Fusinatto LA, et al. Evolution of CCL16 in Glires (Rodentia and lagomorpha) shows an unusual random pseudogenization pattern. BMC Evol Biol 2019;. Available from: http://doi.org/10.1186/s12862-019-1390-7. [59] Mestas J, Hughes CCW. Of mice and men: differences between mouse and human immunology. J Immunol 2004;172:2731
8. [60] Ginhoux F, Jung S. Monocytes and macrophages: developmental pathways and tissue homeostasis. Nat Rev Immunol 2014;14(6):392
404. [61] Brain JD, Molina RM, DeCamp MM, Warner AE. Pulmonary intravascular macrophages: their contribution to the mononuclear phagocytic system in 13 species. Am J Physol 1999;276(1):L145
54. [62] Summerfield A, Auray G, Ricklin M. Comparative dendritic cell biology of veterinary mammals. Ann Rev Anim Biosci 2014;. Available from: https://doi.org/10.1146/annurev-animal-022114-111009. [63] Lee KA, Klasing KC. A role for immunology in invasion biology. Trends Ecol Evol 2004;19(10):523
9.

This page intentionally left blank

Chapter 7

The mammalian major histocompatibility complex Immunity to infectious agents is a strongly inherited trait. After all, a lethal infection effectively removes susceptible individuals from the gene pool. “Survival of the fittest” applies most strongly to those individuals able to defend themselves successfully against infectious agents. As a result, many genes positively influence susceptibility and resistance to infection. Among the most important of these are those encoding cell-surface receptors that present antigens to the effector cells of the immune system,—T and B cells. In order to trigger adaptive immunity, antigen molecules must first be captured and processed by antigen-presenting cells. The antigenic proteins are broken up inside cells, and the resulting peptide fragments are bound to appropriate antigen receptors. The most important of these antigen receptors are cell-surface glycoproteins encoded by genes clustered together to form the major histocompatibility complex (MHC). These receptors are therefore called MHC molecules. Protein antigens can only trigger adaptive immune responses if their peptides first bind to MHC molecules. The peptide-MHC complexes are then presented to T cell antigen receptors and, if recognized, trigger immune responses. Since each MHC molecule acts as an antigen receptor, the genes encoding them effectively determine which antigens can or cannot trigger adaptive immunity. Thus the MHC can be considered a complex of genes that control antigen presentation and so determine resistance or susceptibility to infections. The ability of pathogens to evade, escape, or subvert immune defenses places strong selection pressures on MHC genes and has resulted in their rapid evolution. All jawed vertebrates possess multiple MHC genes, and their overall organization and functions are relatively conserved. However, these genes are not unchanged. They are in fact greatly modified as a result of the extreme selection pressure placed on them by diverse microbiological challenges in the form of infectious and parasitic diseases.

7.1

Major histocompatibility complex structure

All mammals possess an MHC. Each MHC contains about 200 expressed genes arranged on a chromosome in three distinct regions (I, II, and III). [1] The class I region contains genes coding for MHC molecules expressed on most nucleated cells. Class I genes can be subdivided into those that are highly polymorphic (class Ia genes) and those that show very little polymorphism (class Ib genes). (Polymorphism refers to structural variations between proteins). Additional class Ib genes may be located outside the MHC on a different chromosome. Some mammals also have a greatly extended class I region located outside the main MHC. The role of the class Ia genes is to encode proteins that can present T cells with peptides derived from endogenous antigens. (Newly synthesized proteins generated by viruses within an infected cell) (Fig. 7.1). MHC genes clustered in the class II regions encode a different set of polymorphic protein heterodimeric receptors. Their expression is usually restricted to professional antigen-presenting cells (dendritic cells, macrophages, and B cells). The function of these class II gene products is to bind and present foreign peptides that have been captured and processed by the antigen-presenting cells. Genes within the MHC class III region in contrast are functionally diverse. They code for many different proteins, many of which are important in innate immunity such as some complement components. Although each MHC contains all three regions, their gene content, and arrangement vary greatly among different mammalian species. The collective name given to the proteins encoded by MHC genes depends on the species. In humans these molecules are called human leukocyte antigens (HLAs); in dogs, they are called DLA; in rabbits, RLA, etc. In other species they are denoted by a Latin name prefix, for example in the chimpanzee (Pan troglodytes) the prefix is Patr-, in the sheep (Ovis aries) it is Ovar-, and so forth. [2] In some species, MHC encoded proteins were identified as transplantation antigens before their true function was recognized and their nomenclature is anomalous. Thus in the mouse the Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00011-3 © 2023 Elsevier Inc. All rights reserved.

89

90

SECTION | 1 Mammalian immunology

ANTIGEN

Antigenprocessing cell

MHC molecules

Processed antigen

T cell

Antigen receptor

Immune response FIGURE 7.1 The role of the major histocompatibility complex in antigen recognition. Class I molecules are generally expressed on most nucleated cells. They present “endogenous” antigen fragments generated as a result of viral infections to cytotoxic T cells. Class II molecules are usually expressed on specialized antigen-processing cells and present “exogenous” antigen fragments generated by the processing of endocytosed foreign antigens to helper T cells.

MHC is called H-2, in the rat, it is called RT-1. In humans the functional class I polymorphic genes are called A, B, and C. In mice, they are called K and D (and in some strains, L). In other mammals, they are usually numbered. Mammals use two distinct strategies for maintaining high levels of diversity in their MHC class I molecules. In some species a relatively few MHC class I genes exhibit extreme allelic polymorphism. In others, MHC diversity is generated by variations in the number and combinations of expressed MHC class I genes. Since some MHC genes are expressed in some individuals but not in others, the effect is to generate more diversity in their products. Gene content variation and allelic polymorphism can therefore be considered as two alternative strategies to diversify MHC haplotypes. Non-human primates, rats, horses, pigs, and ruminants tend to rely on variations in MHC gene content. [3,4] In contrast, humans, mice, dogs, and cats have relatively few functional MHC class I genes and so rely on diversity generated by a high level of allelic polymorphism. [5
7]

7.2

Major histocompatibility complex class Ia molecules

Classical MHC class Ia genes are expressed on many different cell types. Their function is to encode receptor molecules that can present endogenous antigens derived from invading viruses to cytotoxic CD81 T cells. The non-classical class Ib molecules have less polymorphism, are expressed in specific tissues, and perform diverse immune or non-immune functions. [8,9] Because MHC class Ia molecules bind endogenous peptides to form complexes that are then presented on the surface of nucleated cells to be recognized by cytotoxic CD81 T cells, they are expressed on most nucleated cells. In pigs, for example, class I molecules are found on lymphocytes, platelets, granulocytes, hepatocytes, kidney cells, and sperm. They are not usually expressed on mammalian red cells, gametes, neurons, or on trophoblast cells. Some cell types such as myocardium and skeletal muscle may express very few class Ia molecules. [10,11]

7.2.1 Structure Class Ia molecules consist of two linked glycoprotein chains. An α-chain of 45 kDa is associated with a much smaller chain called ß2-microglobulin (ß2M) of 12 kDa. (Fig. 7.2). The α-chain is expressed on the presenting cell membrane. It consists of five domains: three extracellular domains called α1, α2, and α3, each about 100 amino acids long; a transmembrane domain; and a cytoplasmic domain. The antigen-binding site is formed by the α1 and α2 domains. β2M consists of a single domain and its function is to stabilize the structure.

The mammalian major histocompatibility complex Chapter | 7

91

Antigenic peptide

D1 E2-microglobulin

D2

D3

D chain

Antigen-presenting cell FIGURE 7.2 The structure of major histocompatibility complex class I molecules. The antigen-binding domains, α1 and α2 have great variability in shape in order to bind to diverse antigens. These account for the class I polymorphism whereas the rest of the receptor is relatively conserved.

7.2.2 Gene arrangement The size and arrangement of the MHC class I region vary greatly between eutherian species. Each MHC class I region has a common framework of non-MHC genes, and regional size differences between species are mainly due to variations in these framework genes.

7.2.2.1 Gene numbers The number of MHC class 1a genes ranges from three in the orca (O. orca) to more than 60 in rats. [12] 51 class I genes have been identified in the rhesus macaque (Macaca mulatta) and seven in the domestic dog. [13] However, because of the presence of pseudogenes and ORFs, the number of functional MHC genes is usually much less and ranges from 30 in the rhesus macaque to about four in the dog. In mice, only two or three class I genes are expressed. The remainder are pseudogenes. The pangolin has only four, apparently functional MHC class I genes. [14] Pseudogenes are present in all species examined except the orca. These range from one in the pangolin to 12 in the African elephant. The number of class I genes in the orca is similar to that found in other dolphins and the North Atlantic right whale. [12] Presumably, this low number in cetaceans reflects their adaptation to an aquatic environment with a lower pathogen selection pressure and a reduced need for MHC diversity. The reason for the small number of these genes in pangolins is unclear but it may be associated with their solitary lifestyle, their thick scaly skin, or their specialized diet of ants. In four bat species examined, the number of functional MHC class I genes ranges from 1 to 13 while the number of their pseudogenes ranges from 1 to 6. [14,15] Bats of course are important reservoir hosts for many viruses but show little clinical disease. One might expect them to have large numbers of diverse class Ia proteins. Nevertheless, their numbers are smaller than in rodents, rhesus macaques, and even elephants. Humans have high allelic variation among relatively few MHC loci. Conversely, the large number of genes found in rhesus macaques reflects low allelic variation among many different loci. However, rhesus macaques also show many geographic regional variations in their expressed MHC alleles. [16]

7.2.2.2 Recombination While the effect of gene recombination on the MHC is generally considered to be minor it can be detected in many species. For example, the elephant class I MHC shows evidence of eight recombination events. Two each have been detected in the genes of the pangolin, pig, and rhesus macaque. Three each are found in the horse and tenrec, while between three and five such genes have been found in bats. These breakpoints are randomly distributed. [14]

7.2.3 Evolution Sequence analysis has shown that the classical class I region consists of three duplicated blocks of genes separated by two framework regions (Fig. 7.3). These duplication blocks are called alpha, kappa, and beta. It is possible that the

92

SECTION | 1 Mammalian immunology

Class III

MHC class Ia

Beta

Kappa

Extended Class I

FIGURE 7.3 The class I region of the major histocompatibility complex appears to be constructed of up to three duplication blocks each separated by a framework region. Some mammals have retained all three blocks whereas others have lost some.

Alpha

Primates

Mice

Horses

Pigs

Dogs

genes found in the κ and β duplication blocks originated in a common ancestor of the eutherians. Genomic studies have shown great variations in their arrangement and sizes. For example, in the extended class I region, the κ block and the β block of the elephant contain four, one, and eight functional genes. In the tenrec there is only one, located in the κ duplication block. In bats, there are only one or two functional genes located in the β duplication block. [14] Both class I and class II MHC gene families have evolved by a birth and death process in which new genes are created by repeated duplication (birth). [17] Some persist, whereas others are deleted or become pseudogenes (death). As a result, the mammalian MHC contains large numbers of related genes and pseudogenes. In the class II gene loci, this rate of birth and death is relatively slow, and it is possible to determine divergence times. [18] The orthologous relationships of the class II genes can also be determined. However, the rate of birth and death in the MHC class I loci is very much faster. After all, they have to be able to bind to any invading viral antigen and the “red-queen” problem is extreme. As a result, there do not appear to be any obvious orthologous relationships between the class I loci in the different mammalian orders! In addition, these are accompanied by dozens of class I gene remnants, pseudogenes, and gene fragments indicating an extensive and complex life and death history. Orthologous genes may be identified within, but not between the mammalian orders. For example, orthologous genes can be identified within the primates or within the rodents but not between the primates and rodents. The degree of MHC class I divergence is so rapid and ongoing that even humans and New-World monkeys that diverged only about 33
35 mya do not share functional genes! Likewise, two different class I genes from two marsupial species that separated about 48 mya show no orthologous relationships. Class I genes present in marsupials and monotremes have been lost from eutherians. [19,20] They clearly have a very high turnover rate. As a result of this rapid life and death process, it has not been possible to develop a phylogeny of the class Ia genes. It has also been suggested that the putative class I ancestral antigen receptor in mammals may have been encoded by three types of genes, namely an MHC class Ia gene, a MIC gene, and a class IIb gene (M1/M10). [21] The class I genes underwent repeated duplication in both species. MICs are the ligands of the natural killer (NK) cell receptor NKG2D. Mice have lost the MIC lineage leaving behind a few fragments. There are also four pseudogenes located in the H-2 M region of mice that belong to two subfamilies, M1 and M10. Humans have lost this M1/M10 lineage while M1/M10 gene products in the mouse are only expressed in their vomeronasal organ. [22] Novel MHC alleles may confer better resistance to local parasites. These alleles may differ by orders of magnitude in the range of antigens they can bind. Promiscuous alleles and species with more MHC receptors appear to be more common in pathogen-rich populations and locations where they are needed, as opposed to solitary or marine mammals that need fewer. There is also a need to optimize the identification of protective antigens from microbial invaders while at the same time minimizing the opportunities for the immune system to respond to normal body components—as a result, a balance needs to be maintained between infectious disease resistance and potential autoimmunity. [23] The number of MHC class I alleles correlates negatively with the size of the T cell receptor repertoire reflecting these constraints. (Thus you can have lots of diverse MHC alleles or lots of diverse T cell receptors but not both). [24]

The mammalian major histocompatibility complex Chapter | 7

93

7.2.4 Polymorphism In mice and humans, MHC class Ia gene products are highly polymorphic as a result of variations in the amino acid sequences in their α1 and α2 domains. As of September 2021, 23,002 different class I alleles had been assigned to humans. These are two orders of magnitude greater than the genomic average. [25] The most extreme polymorphism is restricted to three or four small regions located within the α1 and α2 domains. This is the region of the MHC heterodimer that binds antigenic peptides. In these variable regions, two or three alternative amino acids can occur at each position. The other domains of MHC class Ia molecules usually show little variation. The α1 and α2 domains of MHC class I molecules fold together to form an open-ended groove. A flat β sheet forms the floor of this groove, and its walls are formed by two α helices. (Fig. 7.4). This groove can bind antigenic peptides that are 8
10 amino acids long. The variable regions located along the walls of this groove determine its shape. The shape of the groove in turn determines which peptides can fit and thus trigger T cell responses. Polymorphism in the α1 and α2 domains results from variations in the nucleotide sequences encoding these MHC alleles. These gene sequence variations result from point mutations, reciprocal recombination, and gene conversion. Point mutations are simply changes in individual nucleotides. Reciprocal recombination involves crossing over between two chromosomes. In gene conversion, small blocks of DNA are exchanged between different class I genes in a nonreciprocal fashion. The donated DNA blocks may come from nearby nonpolymorphic class I genes, non-functional pseudogenes, or other polymorphic class I genes. Class I MHC genes have the highest mutation rate of any germline genes yet studied (1023 mutations per gene per generation in mice). This high mutation rate implies that there are significant advantages to be gained by having very polymorphic MHC genes. In bats, for example, three- or five- amino acid insertions are found in the MHC1 α1 domain of their expressed proteins. These insertions result in a larger and wider antigen-binding groove and thus increase their ability to bind to a diverse antigen repertoire. [14]

7.2.5 Nonpolymorphic major histocompatibility complex class I molecules Mammalian cells also express many nonpolymorphic class Ib molecules. Some are encoded by genes within the extended MHC class I region, others by genes on other chromosomes. They are classified according to their evolutionary origin. Class Ib molecules show reduced expression and tissue distribution compared with class Ia molecules and they have limited receptor site polymorphism. They probably originated from class Ia precursors by gene duplication. For example, the class Ib genes in mice are clustered in three large loci called Q, T, and M. They code for proteins expressed on the surface of regulatory and immature lymphocytes and hematopoietic cells. These proteins also consist of a membrane-bound α-chain associated with ß2-microglobulin, so their overall shape and antigen-binding groove are similar to those in MHC class Ia molecules. Since they are not polymorphic, however, MHC class Ib molecules only bind a

FIGURE 7.4 A ribbon diagram showing the three-dimensional structure of the major histocompatibility complex class I antigen-binding groove from a Florida manatee (Trichechus manatus). The view is from directly above. The floor of the groove is formed by a beta pleated sheet. The walls are formed by parallel alpha helices. The α1 domain is colored blue while the α2 domain is yellow. Courtesy of Dr. B. Breaux.

94

SECTION | 1 Mammalian immunology

limited range of ligands. They effectively act as pattern recognition receptors for commonly encountered, microbial PAMPs. [26] Other class Ib genes have limited polymorphism and are also found within the MHC. Their products include MICA and MICB, specialized proteins that are involved in signaling to NK cells but do not bind antigenic peptides. Orthologous non-classical MHC genes have been identified in primates. Old-World monkeys diverged from humans about 27
30 mya. Thus rhesus monkeys do not have a MICA gene but possess many repeats of their MICB gene. A third such gene, MICC, must have diverged after that. [27] Some nonpolymorphic class I
related genes are not located on the MHC chromosome. Many of their products contribute to innate immunity since they bind PAMPs. For example, CD1 molecules serve as antigen-presenting receptors that bind lipid antigens. The neonatal antibody receptor, FcRn, is a nonpolymorphic class I MHC molecule that acts as an antibody (Fc) receptor on epithelial cells. It is expressed on mammary gland epithelium and on the enterocytes of newborn mammals where it serves to transport immunoglobulins across epithelial surfaces and the placenta (Chapter 2).

7.3

Major histocompatibility complex class II molecules

Mammals differ in their expression of MHC class II molecules. In rodents, these receptors are restricted to the professional antigen-presenting cells (dendritic cells, macrophages, and B cells) but can be induced on T cells, keratinocytes, and vascular endothelial cells. In pigs, dogs, cats, mink, and horses, MHC class II molecules are constitutively expressed on nearly all resting adult T cells. [28] In cattle, MHC class II molecules are expressed only on B cells and activated T cells. [29] In pigs, resting T cells express MHC class II molecules at about the same level as macrophages. In humans and pigs, MHC class II molecules are expressed on renal vascular endothelium and glomeruli. The expression of class II molecules is enhanced in rapidly dividing cells and in cells treated with interferon-γ.

7.3.1 Structure MHC class II molecules consist of two peptide chains of about 30 kDa each, called α and β. Each chain has two extracellular domains (one constant and one variable), a connecting peptide, a transmembrane domain, and a cytoplasmic domain. A third chain, called the Ii or γ-chain, is associated with the assembly of class II molecules within cells. (Fig. 7.5) They present antigens to helper T cells when intimately linked by CD4 peptide chains. [30]

7.3.2 Gene arrangement A “complete” MHC class II region such as that in humans, contains three paired gene loci. In primates, these polymorphic loci are DPA and DPB, DQA and DQB, and DRA and DRB. (The genes that encode α- chains are designated A, and the genes encoding the β-chains are called B.) Mammals may also have additional nonpolymorphic class II genes such as DM and DO in humans. The DM and DO gene products regulate the loading of antigen fragments into the MHC groove. Not all mammals possess a complete set of polymorphic class II genes. For example, non-primates lack Antigen-binding site

D1

E1 D2

D chain

E2 E chain

FIGURE 7.5 The structure of an major histocompatibility complex class II heterodimer. The antigen-binding site is formed by the variable α1 and β1 domains.

The mammalian major histocompatibility complex Chapter | 7

95

DPA and DPB. Not all loci contain genes for both chains, and some contain many pseudogenes. These pseudogenes may serve as DNA donors that can be used to generate class II polymorphism by gene conversion. Because of their life and death evolutionary history, different mammals inherit different sets of class II genes. Thus in humans, there are three functional gene pairs, DP, DQ, and DR. On the other hand, in mice, only the DQ (IA) genes are functional in all strains. The DR genes (IE) are functional in about half the cases, while the DP genes are all pseudogenes. Cats in contrast have lost all their functional DQ genes but have amplified the number of their functional DR genes. [5] Mice have contracted their class II loci but increased their number of class I genes to 30. However, nonorthologous genes can have the same function as a result of convergent evolution. [21]

7.3.3 Polymorphism MHC class II proteins have an antigen-binding groove formed by their paired α1 and β1 domains. Its walls are formed by two parallel α-helices, and its floor consists of a β-sheet. Polymorphism results from variations in the amino acids forming the sides of the groove. These variations are generated in the same way as in class Ia molecules. Other genes usually located within the class II region code for molecules involved in antigen-processing. These include the transporter proteins TAP1 and TAP2 and some proteasome components. [31]

7.3.4 Evolution Like the MHC class I genes, the class II MHC genes have been subject to evolution by a birth and death process. This is a situation in which new genes are created by repeated gene duplication. Some of these new genes persist for a very long time. Others are soon deleted or become pseudogenes. [32] However, the rate of change in MHC class II genes is much slower than in the class I gene region and thus their longevity is much greater. As a result, many orthologous class II genes are shared by diverse mammalian orders. The class II region can be subdivided into different orthologous subregions that have undergone species-specific duplications. It is apparent, for example, that both class II α-chain and β-chain gene clusters are shared by both eutherians and marsupials. It appears that most class II gene clusters probably originated 170
200 mya. One exception is the DO β-chain, whose genes diverged much earlier, probably about 210
260 mya. This was long after the mammals diverged from the other vertebrates. These are however earlier than the marsupial divergence times, suggesting that they may be ancestral orthologs. Presumably, some have also been lost. As of September 2021, 8673 human class II alleles had been assigned The evolutionary trees of the α- and β-chain genes are not congruent. The α-chain phylogenetic tree shows that the sequences belonging to the DR, DP, DQ, and DN clusters are monophyletic although DP is closer to DR than to DQ. The wallaby DRA and DNA genes cluster with the eutherian DRA and DQA genes. Some mammals have three internally duplicated class II regions. In mice, for example, two of these regions contain class II genes but the third contains butyrophilin genes. The beta chain phylogenetic tree shows that there are four clusters of β-chain genes in mammals corresponding to DRB, DPB, DQB, and DOB. In addition, wallaby DAB and DDB are basal to DRB and the DRB/DPB/DQB clusters. DI/DY cluster genes are found only in certain ruminants where their expression appears to be restricted to a subpopulation of dendritic cells and suggesting that they play a role in antigen presentation. [33]

7.4

Major histocompatibility complex class III molecules

Most of the remaining genes located within the MHC are found in the class III region. This is usually located centrally between the class I and II regions. This is a longstanding association. Even in Xenopus toads, the class III region is linked to MHC class I and II. The MHC class III is the most gene-dense region within the human genome with 62 genes containing more than 500 exons over 706 kb and one gene per 11,387 bp. These genes code for proteins with many diverse functions. Some are important in the defense of the body such as the genes encoding the complement components C4, factor B (FB), and C2. (Fig. 7.6)They also include genes that encode tumor necrosis factor-α (TNF-α), several lymphotoxins, and some NK cell receptors. Thus one subregion contains several cytokine genes such as lymphotoxin-beta LTB, and TNF-α, as well as regulatory genes such as IFN-β. This subregion has been referred to as the Class IV region or the “Inflammatory region.” This cluster is highly conserved. For example, it is present in the class III region of the tammar wallaby (N. eugeneii) and the opossum (Malus domestica). This is a longstanding association. Comparative analysis suggests that the genes in this cluster have remained together for over 450 mya and predate the emergence of mammals. [34] The class III genes are unrelated to either MHC class I or II or even often to each other.

96

SECTION | 1 Mammalian immunology

C4

C2 Liver macrophages

Function

B

21-OH TNF-D TNF-E HSP Adrenal

Steroid Innate and acquired immunity metabolism

Macrophages

Many cells

Induce apoptosis

Damage protection

Pig

II

Bovine

IIb

Horse

II

III

Dog

II

III

II

III

Mouse

I

II

Primate

II

III

Cat

Production site

III

IIa

I

III

FIGURE 7.6 Some of the immunologically relevant genes typically contained within an major histocompatibility complex class III region. They are expressed in many different cells and tissues and have very diverse functions. TNF; Tumor necrosis factor; HSP; Heat shock protein, C4, C2, B; Complement components, 21-OH; 21-hydroxylase.

7

I

23

20

I I

12

I B2

I

III

I

17

6

FIGURE 7.7 Variations in the overall structure of the major histocompatibility complex in some selected mammals. The yellow dot is the position of the centromere. The numbers denote the chromosome involved.

7.5

Mammalian variations

Every mammalian MHC contains class I, class II, and class III regions. When the MHCs of different mammals are compared, some regions such as class III are conserved, whereas others are highly diverse. Likewise, the precise arrangement and a number of loci vary among species (Fig. 7.7). In general, genes within the class II and class III regions possess obvious orthologs in all species. That is, they are clearly derived from a single ancestor and have not usually been subjected to major rearrangements during evolution (ruminant class II genes are an exception). Class I genes, in contrast, have been reorganized so many different times by deletion and duplication that their amino acid sequences differ widely, and it is very difficult to compare class I genes in different species. They are said to be paralogous.

7.5.1 Major histocompatibility complex molecules and disease Since the function of MHC molecules is to present antigens to the cells of the immune system, MHC gene products regulate immune responses. A foreign molecule that cannot be bound to at least one MHC molecule will not trigger an adaptive immune response. Thus the expression of specific MHC alleles determines resistance to infectious and autoimmune diseases. Because class Ia and class II MHC molecules are structurally diverse, each MHC allele can bind and present a different set of antigenic peptides. The more diversity within an animal’s MHC, the more antigens it can respond to. Thus an MHC heterozygous animal will express many more alleles and respond to a greater diversity of antigens than can a homozygous animal. MHC polymorphism is maintained in populations by a process called overdominant selection or heterozygote advantage. Simply put, MHC heterozygotes are at an advantage because they make more diverse receptors and can respond to a greater range of microbial antigens and so are best fitted to survive infectious diseases. The antigen-binding sites of MHC class Ia or II molecules are also very non-specific, and it has been estimated that an average MHC molecule can

The mammalian major histocompatibility complex Chapter | 7

97

bind about 2500 different peptides. This is because the MHC groove binds to the peptide backbone rather than to its amino acid side chains. Nevertheless, structural constraints limit the efficiency of binding each allele. As a result, only one or two peptides from an average antigenic protein can likely bind to any given MHC molecule. The ability of MHC molecules to bind antigens is a limiting factor in generating adaptive immunity and resistance to infectious agents. Increasing the diversity of expressed MHC molecules increases the diversity of antigens that can be bound and so increases resistance to infectious diseases. Because most individuals are MHC heterozygotes, each individual normally expresses at most six different class Ia molecules (in humans, for example, two each are coded for by the HLA-A, -B, and C loci). The number of expressed MHC molecules is not greater because that would increase the risk that the MHC molecules could bind and present more self-antigens. This would require the elimination of many more self-reactive T cells in the thymus during development. Thus the presence of six different MHC class Ia molecules appears to be a reasonable compromise between maximizing the recognition of foreign antigens while at the same time minimizing the chances of recognizing self-antigens,—at least in humans. MHC class Ia loci encode very polymorphic genes. For example, the H-2K locus in the mouse codes for more than 100 alleles. Since there can never be more than two alleles per locus (one on each chromosome), in any individual animal, it appears that this number of alleles has evolved to maximize receptor polymorphism. Its ultimate effect is to protect the population as a whole from complete destruction. Because of MHC polymorphism, most individuals in a population carry a unique set of class Ia alleles, and each individual can therefore respond to a unique mixture of antigens. When a new infectious disease strikes such a population, it is likely that at least some individuals will possess MHC molecules that can bind the new antigens and trigger immunity. Those that can respond will mount an immune response and live. Those that lack these molecules cannot respond and will die. Even in mass die-offs such as those affecting saiga antelope in Kazakhstan, “only” 88% of the population died from hemorrhagic septicemia. The survivors likely had a genetic make-up that conferred resistance. [35] When large populations of humans or mice are examined, no single MHC haplotype predominates. In other words, no single set of MHC receptors confers major survival advantages on individual animals. This reflects the futility of the host attempting to bind all the antigens in a population of invading microorganisms. Microbes will always be able to mutate and evade the immune response faster than mammals can develop new receptors. Any changes in an MHC allele may increase resistance to one organism but at the same time decrease resistance to another. It is more advantageous therefore for the members of a population to possess many highly diverse MHC alleles so that any pathogen spreading through a population will have to adapt anew to each individual. Highly adaptable social animals, such as humans or mice, with large populations through which disease can spread rapidly, usually show extensive MHC polymorphism. In contrast, low-density solitary species such as marine mammals (whales and elephant seals), moose, or Tasmanian devils have much less polymorphism. It is also of interest to note the case of the cheetah, where some wild populations have reduced MHC class II polymorphism as a result of recent population bottlenecks. Because of this low MHC diversity, some cheetahs will accept allografts from other, unrelated cheetahs. Likewise, an infectious disease such as feline infectious peritonitis causes 60% mortality in captive cheetahs compared with 1%
2% mortality in domestic cats. However, there is little evidence to suggest that wild cheetah populations have reduced immune competence. [36] There are many examples of links between MHC haplotype and resistance to infectious disease in domestic mammals. For example, in cattle, there is an association between possession of certain BoLA alleles and resistance to bovine leukosis, squamous cell eye carcinoma, trypanosomiasis; responsiveness to foot-and-mouth disease virus; paratuberculosis; and susceptibility to the tick Boophilus microplus. [37
39] In sheep, there is an association of the class I allele SY1 with resistance to Trichostrongylus colubriformis. Resistance to scrapie and caseous lymphadenitis is also associated with certain MHC class I alleles. In goats, the class I allele Be7 is associated with resistance, and Be1 and Be14 are associated with susceptibility, to caprine arthritis-encephalitis. In horses, an allergic response to the bites of Culicoides midges is linked to ELA-Aw7. In pigs, the SLA complex has an influence on major reproduction parameters such as ovulation rate, litter size, and piglet viability. [40]

7.6

Major histocompatibility complex and body odors

Mammals use odors to detect information about another individual’s gender, status, and individuality. The molecules that carry this information are small volatile peptides found in urine. These peptides can bind to the antigen-binding grooves of MHC class I molecules. Thus peptides known to bind to two mouse MHC class I molecules of different haplotypes were shown to induce responses (field potentials) in mouse vomeronasal organs. The responses were not

98

SECTION | 1 Mammalian immunology

haplotype-specific, but different peptides induced different activation patterns. This finding may well explain how mammals such as mice can recognize the MHC of other mice by smell. [40] The class I region of mice, cattle, and pigs contains numerous genes coding for pheromone olfactory receptors. As a result, their MHC haplotype affects the recognition of peptide ligands causing individual odors in an allele-specific fashion and thus influencing the mating preferences of mammals. Under controlled conditions, mice prefer to mate with MHC-incompatible individuals. Such pairings preferentially generate heterozygote advantage, resulting in optimized disease resistance. However, this type of mating can also prevent genome-wide inbreeding. Inbreeding avoidance may possibly be the most important function of MHC-based mating preferences and therefore the fundamental selective force diversifying MHC genes in species with such mating patterns. Despite not having a vomeronasal organ, humans also have the ability to sense MHC peptides in body odor and this may influence human mate choice (or at least friendships). The genital odors of ring-tailed lemurs provide honest information about an individual’s absolute and relative MHC quality, specifically their MHC-DRB diversity in sex and season-dependent manner. [41] In a more commonplace setting, the mutual sniffing that occurs when two dogs meet likely provides them with a sense of the relationship, social status, nutritional status, and the health of the other dog. Much of this odorous information is likely generated by volatiles generated by the gut microbiota

7.6.1 Odorant receptor-major histocompatibility complex linkage All mammals appear to have one or more clusters of genes encoding odorant receptors (OR) in close physical linkage to the MHC class I region. [40] OR genes constitute the largest gene family in vertebrates. In humans, several OR genes or pseudogenes are located within the class I region of the HLA. At least one of these genes is intact, appears to encode an mRNA, and is homologous to a previously reported murine OR. [42] A comparison of DNA and protein sequences of ORs from the genomes of human, chimpanzee, gorilla, orangutan, rhesus macaque, mouse, rat, dog, cat, cow, pig, horse, elephant, opossum, frog, and zebrafish reveal pan-vertebrate conservation of the evolutionarily conserved MHC-OR linkage. Each of the taxa studied showed a typical architecture of MHC-linked OR genes. This conserved linkage between distinct OR genes and the MHC supports the concept that some alleles may function in a concerted fashion during mate selection. [43]

References [1] Trowsdale J. “Both man and bird and beast”: comparative organization of MHC genes. Immunogenetics 1995;41:1
17. [2] Gao J, Liu K, Blair HT, Li G, et al. A complete DNA sequence map of the ovine major histocompatibility complex. BMC Genomics 2010;. Available from: https://doi.org/10.1186/1471-2164-11-466. [3] Amills M, Ramiya V, Norimine J, Lewin HA. The major histocompatibility complex of ruminants. Rev Sci Tech Int Epiz 1998;17:108
20. [4] Janova E, Matiasovic J, Vahala J, et al. Polymorphism and selection in the major histocompatibility complex DRA and DQA genes in the family Equidae. Immunogenetics 2009;61:513
27. [5] Holmes JC, Holmer SG, Ross P, et al. Polymorphisms and tissue expression of the feline leukocyte antigen class I loci FLAI-E, FLAI-H, and FLAI-K. Immunogenetics 2013;65:675
89. [6] Kennedy LJ, Barnes A, Happ GM, et al. Extensive interbreed, but minimal intrabreed, variation of DLA class II alleles and haplotypes in dogs. Tissue Antigens 2002;59:194
9. [7] Kennedy LJ, Barnes A, Happ GM, et al. Evidence for extensive DLA polymorphism in different dog populations. Tissue Antigens 2002;60:43
52. [8] Birch J, Codner G, Guzman E, Ellis SA. Genomic location and characterization of nonclassical MHC class I genes in cattle. Immunogenetics 2008;60:267
73. [9] Birch J, Sanjuan CDJ, Guzman E, Ellis SA. Genomic localization and characterization of MHC genes in cattle. Immunogenetics 2008;60:477
83. [10] Ho CS, Lunney JK, Franzo-Romain MH, et al. Molecular characterization of swine leucocyte antigen class I genes in outbred pig populations. Anim Genet 2009;40:468
78. [11] Ho CS, Lunney JK, Lee JH, et al. Molecular characterization of swine leucocyte antigen class II genes in outbred pig populations. Anim Genet 2010;41:428
32. [12] Gillett RM, Murray BW, White BN. Characterization of class I
and class II-like major histocompatibility complex loci in pedigrees of North Atlantic right whales. J Heredity 2014;105(2):188
202. [13] Wagner JL, Sarmiento UM, Storb R. Cellular, serological, and molecular polymorphism of the class I and class II loci of the canine major histocompatibility complex. Tissue Antigens 2002;59:205
10. [14] Abduriyim S, Zou D-H, Zhao H. Origin and evolution of the major histocompatibility complex class I region in eutherian mammals. Ecol Evol 2019;9:7861
74.

The mammalian major histocompatibility complex Chapter | 7

99

[15] Ng JHJ, Tachedjian M, Deakin J, Wynne JW, et al. Evolution and comparative analysis of the bat MHC-1 region. Sci Rep 2016;. Available from: https://doi.org/10.1038/srep21256. [16] Heijmans CMC, de Groot NG, Bontrop RE. Comparative genetics of the major histocompatibility complex in humans and nonhuman primates. Int J Immunogenet 2020;47:243
60. [17] Nei M, Gu X, Sitnikova T. Evolution by the birth-and-death process in multigene families of the vertebrate immune system. Proc Natl Acad Sci USA 1997;94:7799
806. [18] Takahashi K, Rooney AP, Nei M. Origins and divergence times of mammalian Class II MHC gene clusters. J Heredity 2000;91(3):198
204. [19] Papenfuss AT, Feng Z-P, Krasnec K, Deakin JE, et al. Marsupials and monotremes possess a novel family of MHC class I genes that is lost from the eutherian lineage. BMC Genomics 2015;. Available from: https://doi.org/10.1186/s12864-015-1745-4. [20] Houlden BA, Greville WD, Sherwin WB. Evolution of MHC class I loci in marsupials: Characterization of sequences from Koala (Phascoarctos cinereus). Mol Biol Evol 1996;13(8):1119
27. [21] Kumanovics A, Takada T, Fischer Lindahl K. Genomic organization of the mammalian MHC. Annu Rev Immunol 2003;21:629
57. [22] Leinders-Zufall T, Brennan P, Widmayer P, et al. MHC class I peptides as chemosensory signals in the vomeronasal organ. Science 2004;306:1033
7. [23] Radwan J, Babik W, Kaufman J, Lenz TL, Winternitz J. Advances in the evolutionary understanding of MHC polymorphism. Trends Genet 2020;36:298
311. [24] Codner GF, Birch J, Hammond JA, Ellis SA. Constraints on haplotype structure and variable gene frequencies suggest a functional hierarchy within cattle MHC class I. Immunogenetics. 2012;64(6):435
45. [25] Piertney SB, Oliver MK. The evolutionary ecology of the major histocompatibility complex. Heredity 2006;96:7
21. [26] Hulpke S, Tampe R. The MHC I loading complex: a multitasking machinery in adaptive immunity. Trends Biochem sci 2013;38(8):412
20. [27] Fukami-Kobayashi K, Shiina T, Anzai T, Sano K, et al. Genomic evolution of MHC class I region in primates. Proc Natl Acad Sci USA 2005;102:9230
4. [28] Barbis DP, Bainbridge D, Crump AL, Zhang CH, et al. Variation in expression of MHC class II antigens on horse lymphocytes determined by MHC haplotype. Vet Immunol Immunopathol 1994;42(1):103
14. [29] Lewin HA, Russell GC, Glass EJ. Comparative organization and function of the major histocompatibility complex of domesticated cattle. Immunol Rev 1999;167:145
58. [30] Konig R, Huang LY, Germain RN. MHC class II interaction with CD4 mediated by a region analogous to the MHC class I binding site for CD8. Nature. 1992;356(6372):796
8. [31] Gojanovich GS, Ross P, Holmer SG, Holmes JC, Hess PR. Characterization and allelic variation of the transporters associated with antigen processing (TAP) genes in the domestic dog (Canis lupus familiaris). Dev Comp Immunol 2013;41(4):578
86. [32] Schaschl H, Wanderer P, Suchentrunk F, Obexer-Ruff G, Goodman SJ. Selection and recombination drive the evolution of MHC class II diversity in ungulates. Heredity 2006;97:427
37. [33] Ballingall KT, MacHugh ND, Taracha ELN, Mertens B, et al. Transcription of the unique ruminant class of major histocompatibility complex DYA and DIB genes in dendritic cells. Eur J Immunol 2001;31:82
6. [34] Deakin JE, Papenfuss AT, Belov K, Cross JCR, et al. Evolution and comparative analysis of the MHC class III inflammatory region. BMC Genomics 2006;. Available from: https://doi.org/10.1186/1471-2164/7/281. [35] Fereidouni S, Freimanis GL, Orynbayev M, Ribeca P, et al. Mass die-off of Saiga antelopes, Kazakhstan, 2015. Emerg Inf Dis 2019;25 (6):1169
76. [36] Castro-Prieto A, Wachter B, Sommer S. Cheetah paradigm revisited: MHC diversity in the world’s largest free-ranging population. Mol Biol Evol 2011;28(4):1455
68. [37] Xu A, van Eijk MJT, Park C, Lewin HA. Polymorphism in BoLA-DRB3 exon 2 correlates with resistance to persistent lymphocytosis caused by bovine leukemia virus. J Immunol 1993;151:6977
85. [38] Rastislav M, Mangesh B. BoLA-DRB3 exon 2 mutations associated with paratuberculosis in cattle. Vet J 1997;192(3):517
19. [39] Baxter R, Craigmile SC, Haley C, et al. BoLA-DR peptide binding pockets are fundamental for foot-and-mouth disease virus vaccine design in cattle. Vaccine 2009;28:28
37. [40] Tizard IR, Skow L. The olfactory system: The remote sensing arm of the immune system. Anim Hlth Res Revs 2021;. Available from: https:// doi.org/10.1017/s1466252320000262. [41] Grogan KE, Harris RL, Boulet M, Drea CM. Genetic variation at MHC class II loci influences both olfactory signals and scent discrimination in ring-tailed lemurs. BMC Evol Biol 2019; doi.org/10.1186/s12862-019-1486-0. [42] Fan W, Liu YC, Parimoo S, Weissman SM. Olfactory receptor-like genes are located in the human major histocompatibility complex. Genomics 1995;27(1):119
23. [43] Santos PSC, Mezger M, Kolar M, Michler F-U, Sommer S. The best smellers make the best choosers: Mate choice is affected by female chemosensory receptor gene diversity in a mammal. Proc Roy Soc B 2018;. Available from: https://doi.org/10.1098/rspb.2018.2426.

This page intentionally left blank

Chapter 8

T Cells and their receptors Before the evolution of the vertebrates, there existed many different innate defenses that were effective in protecting invertebrates from microbial invasion. As pointed out previously, however, innate immunity is insufficient to ensure the survival of much larger vertebrates. It also comes at a price, the discomfort and behavioral issues associated with local inflammation and the systemic sickness responses. Innate immunity alone is not sufficient to ensure the long-term survival of an individual. The vertebrates required an immune system that worked quietly in the background, protecting them efficiently without the drawbacks of chronic inflammation and the vulnerability and debility of sickness. The evolution of the adaptive immune system fulfilled this need.

8.1

Flexible immunity

The adaptive immune system must have the ability to respond appropriately to an enormous diversity of microbial pathogens. Thus it must be able to assume different functional configurations depending upon the nature of the microbial challenge [1]. This is largely achieved by employing differentiated T cell subpopulations. For example, mammals can mount a cell-mediated, type I immune response by activating Th1 cells, destroying virus-infected cells, and hence effectively combatting viral invasion. Alternatively, they can turn on an antibody-mediated immune response by activating Th2 cells and so combat invasion by extracellular bacteria. In other situations, Th17 cells can be activated and promote local inflammation or generate antiviral interferons and many other defensive processes [2]. If required, they can generate regulatory T (Treg) cells and control inappropriate responses [3]. T cells can also change their phenotype very rapidly if the situation demands it [4]. This extreme immune flexibility clearly originated and diversified during the evolution of the early mammals as described in Chapter 1. It results from the use of multiple polymorphic gene families that can be selected based on the nature of the invader, fine-tuned to identify specific invaders, activate the appropriate defenses, and so maximize an animal’s chances of survival. These multigene families have been optimized over hundreds of millions of years under constant selection pressure in the evolutionary arms race. Thus functional diversity and flexibility in using the available immune responses are significant assets. One obvious result of this flexibility is the emergence of multiple T cell subsets that, when correctly selected, will yield the best results, namely survival. This selection process involves the use of pattern recognition receptors, antigen-binding by major histocompatibility complex (MHC) encoded receptors and the use of very specific antigen-binding receptors by T and B cells (Fig. 8.1). The responding immune system makes a choice between different T cell subsets. These subsets are triggered by different stimuli depending on the nature of the invaders and their routes of invasion (Fig. 8.2). The responding T cells employ different transcription factors to release different cytokine mixtures. These mixtures, in turn, trigger different forms of immune defense, ideally optimized to rapidly eliminate the invaders with a minimum of damage to the defender.

8.2

T cell evolution

The first vertebrates to utilize T cells for defense were the cartilaginous fish. Fish have T cells and make rearranged T cell antigen receptors (TCRs) whose overall structure is similar to that observed in mammals. Both helper and cytotoxic T cells can also be detected in fish. Fish also have both MHC class I and class II genes that encode antigen-presenting receptors. Their function is to present antigen fragments to the appropriate T cell population. As a result, the basic structure of each MHC molecule is conserved, as is the organization of the MHC class I and class II loci. Many homologs of mammalian cytokines have been identified in fish including proinflammatory interleukins such as IL-1β, IL-18, IL-6, and IL-11; interleukin-2 subfamily members, IL-2, -4, -7, -15, and -21; interleukin 10 subfamily members, IL-10, -20-like, and -22/26; and five interleukin 17 members. TNF-α, TGF-β, IFN-β, and IFN-γ are also expressed by fish [5]. Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00015-0 © 2023 Elsevier Inc. All rights reserved.

101

102

SECTION | 1 Mammalian immunology

T cells

Antigen receptors

JG

DE

CD4

Helper 1 Helper 2 Helper 17 Regulatory

CD8

Memory

Cytotoxic Regulatory

Co-receptors

Memory

FIGURE 8.1 A simple classification of T cells. They may express one of two types of antigen receptors, depending on the paired peptide chains involved. They may use their CD4 or CD8 co-receptors to determine which specific type of MHC they interact with. This receptor usage determines their precise role in the defense of the body.

Naive CD4+ T cell

Th-polarizing cytokines

IL-12 IFN-J

IL-4 IL-33 TSLP

IL-6 IL-21 IL-23 TGF-E

IL-10 TGFE

Transcription factors

T-bet

GATA3

ROR-Jt

FoxP3

Effector cytokines

IFN-J TNFD IL-2

IL-4 IL-5 IL-9 IL-13

IL-17 IL-21 IL-22

IL-10 IL-35 TGFE

Th1

Th2

Th17

Cell name Functions

FIGURE 8.2 T cells can be subdivided into four major functional phenotypes. They differentiate into these phenotypes based upon their exposure to specific polarizing cytokine mixtures generated by antigen-presenting cells. Once differentiated, each T cell phenotype is characterized by the mixture of effector cytokines that it secretes and so determines its functional role in the defense of the body. This is the basis of the great flexibility of the adaptive immune responses and appears to be a feature of all mammals.

Treg

B cell Increased T cell Decreased responses responses inflammation inflammation

Amphibians also have T cells with functional TCRs, and anurans such as bullfrogs and toads can reject allografts. Delayed hypersensitivity skin reactions have been described in the axolotl (Ambystoma) and Xenopus in response to mycobacterial sensitization. As in other ectotherms, reptile allograft rejection is also temperature-dependent. Other evidence of cell-mediated immune responses such as delayed hypersensitivity reactions have been demonstrated in reptiles [5].

8.3

T cell antigen receptors

The ability of lymphocyte receptors to bind antigens is determined by the structure of their ligand binding sites. This structure is determined by the folding of its peptide chains, which are governed, in turn, by their amino acid sequences.

T Cells and their receptors Chapter | 8

103

These sequences are determined by the sequence of nucleotides in the DNA encoding the receptor protein. The diversity of antigen-binding receptors implies either a corresponding diversity in the genes coding for these receptors or a mechanism that generates diversity from a limited pool of receptor genes. This second mechanism is the method employed by the T and B cells of the mammalian adaptive immune system. The key players in the development of many receptors, especially those that bind antigenic peptides are proteins belonging to the immunoglobulin superfamily. The members of this superfamily all contain at least one immunoglobulin domain. In a typical immunoglobulin domain, the peptide chains weave back and forth to form a pleated sheet that folds into a sandwich-like structure. Immunoglobulin domains were first identified in antibody molecules (immunoglobulins) but they are used by many other immune system proteins. Important proteins with multiple immunoglobulin domains include the B cell antigen receptors (BCRs), the TCRs, as well as the MHC class I and II molecules. All of the members of this superfamily are cell surface receptors. In practice, immune responses are triggered by interactions between two different members of the superfamily, for example, between TCR and MHC molecules.

8.3.1 The antigen-binding chains Each T cell expresses about 30,000 identical antigen receptors (TCRs) on its surface. Each TCR is constructed from two components, an antigen-binding component, and a signal transduction component. The antigen-binding component consists of two peptide chains that form a heterodimer with the antigen-binding site located between two chains (Fig. 8.3). Based on their paired antigen-binding chains there are two major classes of TCRs. One class employs γ and δ (γ/δ) chains. The other employs α and β (α/β) chains. In humans, mice, and most nonruminants, 90%
99% of T cells use α/β chains. In ruminants and pigs up to 66% of T cells use γ/δ chains. Thus mammals can be classed as γ/δ-low or γ/δ-high species. The biological significance of these differences remains unclear [5]. The four T cell antigen-binding chains (α, β, γ, and δ) are similar in structure across the mammals, although they differ in size as a result of variations in glycosylation. For example, the human α chain is 43
49 kDa, the β chain is 38
44 kDa, the γ chain is 36
46 kDa, and the δ chain is 40 kDa. Each chain is constructed from four domains. The Nterminal domains contain about 100 amino acids and their sequence varies greatly among T cells. This is a variable (V) domain and its function is to bind peptides presented by MHC molecules. The second domain contains about 150 amino acids. Its amino acid sequence does not vary much, so it is a constant (C) domain. A third, very small domain consists of 20 hydrophobic amino acids passing through the T cell membrane. This transmembrane domain contains a very conserved sequence, the conserved antigen receptor transmembrane motif (CART) which plays a key role in TCR signal transduction. The C-terminal domain within the cytoplasm of the T cell is only 5
15 amino acids in length. The paired receptor chains are linked by a disulfide bond between the C domains to form a stable heterodimer. As a result, a groove is present between the two V domains that serves as the antigen-binding site. The precise shape of this antigenbinding groove varies among different TCRs because of variations in the amino acid sequences of their V domains. The specificity of the binding between a TCR and an antigen is determined by the shape of this groove [5]. Within each V domain is a region where the amino acid sequences are very highly variable. This is the region that comes into contact with the antigen. For this reason, it is called the complementarity-determining region (CDR). The antigen-binding site of the TCR is formed by the CDRs from the two chains that form the walls of the groove. The rest Antigen binding groove D-chain E-chain

N Variable domain

Constant domain

Cell membrane

Transmembrane domain

C

Cytoplasmic domain

FIGURE 8.3 The structure of the antigen-binding component of the T cell antigen receptor. It consists of two linked peptide chains α and β, with the antigen-binding site located between the chains.

104

SECTION | 1 Mammalian immunology

Antigen-binding site

Variable domains

Beta chain

Alpha chain

Constant domains CELL MEMBRANE FIGURE 8.4 A ribbon diagram showing the three-dimensional structure of a T cell antigen receptor. In this case, this is a bovine α/β T cell receptor. Courtesy of Dr. B. Breaux.

of the V domain outside the CDRs has a relatively constant sequence, it provides stability and is called the framework region (Fig. 8.4). Each of these TCR peptide chains is encoded by a sequence of three or four gene segments called, V, D, joining (J), and C segments. The V domains of the TCRs are encoded either by either linked V, D, and J gene segments or by just V and J gene segments. For example, the α and γ chains are encoded by three gene segments V, J, and C. In contrast, the β and δ chains use an additional D gene segment. Thus there are two major subtypes of T cells. Those that use paired α and β (α/β) chains, and those that use paired γ and δ (γ/δ) chains. Different species favor different dimers. In general, γ/δ-high T cell subfamilies have much less receptor diversity than α/β-high subfamilies [6]. Each TCR V domain has seven distinct subregions. Three hypervariable CDR loops and four framework regions. Two of the CDR loops, CDR1 and CDR2 are encoded solely by the germline sequences of the V gene segment. The third CDR loop, in contrast, is generated by VDJ or VJ combinations. As a result, it is generally larger, more variable, and plays the major role in antigen binding.

8.3.2 The signal transduction components 8.3.2.1 CD3 complex The ability of the immune system to mount a rapid and appropriate response to a specific invader centers on the responding T cells. It is they that make the decision that will result in survival or death [7]. Millions of years of evolution have enabled them to get their act together and ensure that, under most circumstances, their decision is the correct one. Much of this decision-making depends upon the signals sent to the T cell by newly captured antigen together with multiple costimulatory signals from a mixture of cytokines. A key part of this pathway is the signal-transducing component of the TCR. When antigen binds to its TCR, a signal is generated that triggers the T cell response. The paired antigen-binding chains of each TCR are associated with a cluster of signal-transducing proteins called the CD3 complex (Fig. 8.5). The CD3 complex consists of five peptide chains (γ, δ, ε, ζ, and η) arranged as three dimers γ
ε, δ
ε, and either ζ
ζ or ζ
η. The TCR β chain is linked to the γ
ε dimer and the TCR α chain is linked to the δ
ε dimer [8]. The CD3 genes are located in a 50-kb region of chromosome 11 in humans and chromosome 9 in mice. Within the receptor complex, CD3δ is centrally located with oppositely transcribed CD3γ and CD3ε on either side [9]. CD3δ is encoded by five exons while CD3γ has seven and CD3ε has seven or eight. Sequence analysis suggests that CD3γ and CD3δ originated from a duplication event about 250 mya. These two proteins each have unique, nonredundant functions. Thus CD3γ is required for early T cell development. The structure of this signal transduction component also differs between the α/β and γ/δ TCRs. For example, γ/δ T cells can develop in the absence of CD3δ but not in the absence of CD3γ [9] (Box 8.1).

T Cells and their receptors Chapter | 8

105

MHC-Antigen binding

D

H

G

E J H

Signal

Antigen binding

J

G

J H

G

H

Signal FIGURE 8.5 The CD3. Complex. The signal-transducing components of the T cell antigen receptors. These differ between the α/β T cells and the γ/δ T cells. They employ the same peptide chains but these are paired differently. The significance of these differences is unclear.

BOX 8.1 Gene duplication and the immune system. Gene duplication is an important mechanism of evolution. It is readily apparent in the genes encoding the immune system where the repeated duplication of V-region genes in both T and B cells has greatly expanded their numbers as well as their ability to bind multiple antigens and to maintain their place in the antimicrobial arms race. Thus gene duplication is an important contributor to immune complexity. Genes may duplicate as a result of small-scale duplication events occurring as a result of “errors” in DNA replication. It is estimated that about 5% of human genes have originated from this small-scale duplication. More importantly, about 26% of human genes have originated from a whole-genome duplication. As a result, copy number variations are the usual situation in TCR and Immunoglobulin V-region genes. In the normal course of events, most of these duplicated genes eventually become nonfunctional but they may be retained in the genome as they assume new functions. They can serve as dose amplifiers, or as reserves. Like V-region genes that may have some degree of functional redundancy. Small-scale duplications may result from tandem duplication between sister chromatids, and transposition to a different region on the same chromosome. If a duplicated gene has a function that becomes advantageous for survival, perhaps by binding a pathogen epitope, then gene duplication can offer a selective advantage. If a gene serves two functions then duplication may permit them to evolve (Continued )

106

SECTION | 1 Mammalian immunology

BOX 8.1 (Continued) independently to the advantage of the body’s defenses. Some genes may be retained since they simply provide a marginal advantage and in effect act as a strategic reserve. In examining mammalian V gene content, it is obvious that the ratio of functional V genes to V pseudogenes varies greatly between species. However, it must not be assumed that the pseudogenes represent wasted genetic material. In many cases, they provide sequences important in the gene conversion process. They in effect provide a necessary backup (or buffer) in an environment where microbial pathogens can change their antigenic structure frequently and unpredictably. Coronaviruses provide a prime example of this. An additional advantage of some gene duplications is dose amplification and a resulting increase in the production of an important protein. This may be important in the production of functional heterodimers such as the MHC antigens and the TCRs and BCRs where the amount of each peptide chain needs to be matched to the production of the other. If the dose is imbalanced however, one of the components may eventually diverge functionally. Kuzmin E, Taylor JS, Boone C. Retention of duplicated genes in evolution. Trends Genet 2022;38(1):59
71.

8.3.2.2 CD4 and CD8 Two additional glycoproteins are closely associated with the TCR. These are CD4 and CD8. CD4 is a single peptide chain of 55 kDa, while CD8 is a dimer of 68 kDa. (One chain of CD8 is called α, the other is β. In humans, pigs, mice, and cats, CD8 is an α-β heterodimer or, less commonly, an α
α homodimer.) CD4 and CD8 determine the class of MHC molecule that can be recognized by the T cell. For example, CD4 found only on helper T cells, binds MHC class II molecules on antigen-presenting cells. CD8, in contrast, is found only on cytotoxic T cells and binds MHC class Ia molecules on virus-infected or other abnormal cells. CD4 and CD8 binding enhance TCR signal transduction by holding the T cell in contact with an antigen-presenting cell through the MHC for up to several hours while activating signals are generated and exchanged.

8.4

T cell antigen receptor functions

8.4.1 Receptor-antigen binding T cell receptors cannot simply bind free antigen. The antigen must first be processed within an antigen processing cell such as a dendritic cell [10]. The processed antigen fragments are then bound to the antigen-binding groove of an MHC molecule. The antigen-MHC complex is then presented to a T cell. If an antigen-MHC complex and its T cell receptor bind, they interact through the side-chains on the antigen and the MHC molecule on the CDRs of the TCR. The binding of an antigenic peptide to a TCR is exclusively noncovalent, so the strongest binding occurs when the shape of the foreign peptide and the shape of the receptor V domain perfectly match. This requirement for a close conformational fit has been likened to the specificity of a key for its lock. The major bonds formed between an antigen and its receptor are hydrophobic. When antigen and receptor molecules come together, they exclude water from the area of contact. This frees some water molecules from constraints imposed by the proteins and is therefore energetically stable. A second type of binding between an antigen and its receptor is through hydrogen bonds. When a hydrogen atom covalently bound to one electronegative atom, for example, an 2 OH group, approaches another electronegative atom such as an O 5 C 2 group, the hydrogen is shared between the two electronegative atoms. This situation generates a hydrogen bond. The major hydrogen bonds formed in antigen-receptor binding are O 2 H 2 O, N 2 H 2 N, and O 2 H 2 N. Van der Waals forces also function when the antigen and its receptor interlock. This force, although very weak, may become collectively important when two large molecules come into very close contact. It can therefore also contribute to antigen-receptor binding. The binding of a T cell receptor to its antigen as well as the MHC is therefore mediated by multiple noncovalent bonds. Each bond is relatively weak in itself, but collectively they may have significant binding strength. All these bonds act only across short distances and weaken rapidly as that distance increases. Electrostatic bond and hydrogen bond strengths are inversely proportional to the square of the distance between the interacting molecules; the van der Waals forces and hydrophobic bonds are inversely proportional to the seventh power of that distance. Thus the strongest binding between an antigen and its receptors occurs when their shapes match perfectly and multiple noncovalent bonds form. Antigens can bind to receptors when they fit less than perfectly, although their binding affinity and hence the signal intensity generated will be much reduced.

T Cells and their receptors Chapter | 8

8.5

107

Antigen receptor diversity

The information needed to make all proteins, including TCRs, is encoded in an animal’s genome. Once the appropriate genes are activated, they can be transcribed into RNA and translated into the appropriate receptor protein on B or T cells. It has been estimated that mammals can produce up to 1015 different antigen receptors to be expressed on B and T cells. To produce this enormous diversity of receptors, they use fewer than 500 genes! In general, the TCR gene organization is more conserved among mammals than the immunoglobulin genes. This is because T cell receptors must also be able to bind to the presenting MHC molecule,
a major constraint. Multiple V gene segments code for each V region, whereas only one codes for a C-region. As a result, the single C-region gene segment can be combined with any one of the available V gene segments to make a complete receptor chain gene. Instead of having genes for all possible receptor chains, it is only necessary to have a collection of gene segments for all the V regions and to join one of these to an appropriate C-region segment as required. In addition, antigen receptor heterodimers may be paired in different combinations to yield even greater diversity, a process called combinatorial association. There are three hypervariable regions (CDRs) within each TCR V region. The first two, located within the V genes themselves, have probably arisen through selection in the germline. The third CDR is by far the most variable and is generated when the V, D, and J gene segments recombine. These gene segments are separate in the germline but are brought together by DNA rearrangement and are then modified by base insertion or deletion as the T cells mature and differentiate [11].

8.5.1 Gene rearrangement TCR α and γ chains are constructed using only V, J, and C gene segments. TCR β and δ chains use V, D, J, and C segments. They may also insert more than one D segment into the complete gene. As a result, V-D-D-J- or longer constructs can be formed. This amount of gene recombination also means that the reading frame of the D genes may change and still yield productive rearrangements. Looping-out and deletion account for more than 75% of these TCR rearrangements. The remaining rearrangements result from either unequal sister chromatid exchange or inversion. That is, moving an inverted segment of one gene into a position beside a segment in the opposite orientation.

8.5.2 Base insertion and deletion Although in general, TCRs are constructed from a smaller V-, D-, and J- gene segment pool than immunoglobulins, their antigen-binding diversity tends to be similar as a result of junctional diversity mechanisms. Random N-nucleotides may be inserted at the V, D, and J junctions using the enzyme terminal deoxynucleotidyl transferase. Up to five nucleotides may be added between V and D and four between D and J genes. Likewise, random nucleotides may be removed by nucleases. This insertion of N-nucleotides and base deletion is probably the most significant contributor to TCR junctional diversity.

8.5.3 Somatic mutation Somatic mutation does not occur in TCR V genes, with the notable exception of the γ and δ V genes of camels and llamas (Chapter 14) [12]. As pointed out above, T cells can only recognize a foreign antigenic peptide in association with a presenting MHC molecule. If random somatic mutation were to occur, it would carry the unacceptable risk of preventing MHC binding and rendering the foreign antigen unrecognizable. It might also lead to the production of TCRs able to bind self-antigens and thus trigger autoimmunity. The reasons why camels and llamas are different are unknown.

8.6

T cell receptor diversity

8.6.1 Gene structure α/β T cells and γ/δ T cells arise from a common precursor and the TCR class switch is mediated by signals within the thymus [13]. Cells with α/β TCRs rearrange and express TRA and TRB genes whereas γ/δ T cells express TRG and TRD genes (Fig. 8.6). This pattern of TCR expression is conserved throughout the mammals. Thus the four TCR peptide chains are each encoded by genes located within three loci. The TRA/D locus codes for both α and δ chains since the TRD genes are embedded within the TRA locus. Developing T cells committed to the α/β TCR lineage delete their TRD genes by looping out and so switch to using their TRA genes. Some of the V genes in the TRA/D locus may be used in either α or δ TCR chains (Fig. 8.7).

108

SECTION | 1 Mammalian immunology

DC

DJ

DV DD

5'

TRA/D

AV

3'

AJ

AC

TRB

TRG

V

D

J

C

FIGURE 8.6 There are three T cell antigen receptor gene loci in all mammals. One locus contains the T cell antigen receptor alpha chain genes. The delta chain genes are always embedded within them. The beta and gamma chain loci are usually located on different chromosomes. In most mammals, the gamma chain locus is relatively small.

Delete

Germline DNA

TRB cluster V

D1

J1

C1

J2

D2

C2

DNA rearrangement T cell DNA VD

J

Transcription RNA transcript VD

J

RNA splicing mRNA VDJ

Translation NH2

COOH

C

V

TCR E chain

V

D

J

C

FIGURE 8.7 The mechanism by which V region diversity is generated by recombination within a T cell antigen receptor locus. This example shows the process in the TRB locus encoding a beta chain.

T Cells and their receptors Chapter | 8

109

The TRB locus encodes only β chains, while the TRG locus contains only γ chain genes. All three TCR loci contain V, J, and C genes, while the TRB and TRD loci also contain D genes. Each of the three TCR loci may contain two or more C region genes as well. In the TRA/D locus one C gene codes for TRAC and the other for TRDC. The TRB and TRG loci, in contrast, may contain multiple constant region genes. The number of these C genes varies among mammals. For example, there are eight TRGC genes in dogs (Chapter 20) (Table 8.1).

8.6.1.1 TRA/D The TRA and TRD genes are joined in a single chromosomal region with TRD embedded within the TRA locus
the TRA/D locus. Despite this, each locus is independently controlled. There is great variability in the numbers and arrangement of both TRAV and TRDV gene segments across the mammalian spectrum. The number of TRAV gene segments varies ranging from five (so far) in the horse, to 56 in the dog, and 48 in the pig, to 183 in cattle, and more than 300 in sheep [14]. Likewise, TRAJ gene segment numbers range from five in the horse to 61 in mice, and humans. Only one TRAC gene has been identified in the mammals investigated so far. The TRD locus contains between 5 and 55 V gene segments depending on the species, two to six J segments, one to nine D segments, and a single C segment in all mammals examined. As mentioned previously, D gene use is optional. Pigs have at least six TRDD gene segments compared to three and two in humans and mice respectively [15]. These pig genes may form transcripts in a chain that encodes up to four connected TRDD domains [16].

8.6.1.2 TRB The TRB locus usually contains a cluster of V gene segments located upstream of two D-J-C cassettes, each containing several functional J genes. The D genes are all similar in sequence and length and their use is optional. Any of the TRBV genes may be joined to either of the two D-J-C cassettes, and a V gene may be linked to either a D or a J gene [17]. A single TRBV gene in an inverted configuration is usually located at the 3' end of the gene cluster. In most mammals including humans, mice, chimpanzees, rhesus macaque, dogs, rabbits, ferrets, and cats there are two TRBD-JC clusters, whereas, in artiodactyls, there has been a duplication event, so they have a third TRBD-J-C cluster thus permitting more somatic rearrangements [18
20]. Dogs have about 39 TRBV genes, but about one-third of these are used to make up 90% of the beta chain repertoire. TRBV gene use may be restricted to a single V gene family in the dog [21]. In the pig, 43 TRBV genes and three D-JC cassettes have been identified. Other mammals may have very diversified TRBV genes, ranging from 16 in the horse to 134 in cattle [22].

8.6.1.3 T cell receptor gamma In contrast to the TRB and TRA/D loci, the TRG locus is structurally highly variable. It generally consists of multiple TRGV segments linked to TRGJ and TRGC segments arranged in multiple J-C clusters. The typical human TRG locus contains about 160 kb, but its orientation is reversed. it consists of 12
15 TRGV segments upstream of a duplicated JC cluster. The first contains three J segments and a single TRGC1 segment. The second cluster contains two TRGJ segments and the TRGC2 segment. The TRG locus contains 8
22 V genes, from 2 to 16 J genes, and from 1 to 8 C gene segments depending on species. The TRGV segments in humans belong to six different subfamilies with more than 75% identity at the nucleotide level. TRGV9 is expressed in 85%
90% of human γ/δ blood T cells. The TRGC1 segment contains three exons and encodes a C region of 173 amino acids while the TRGC2 gene has 4 or 5 exons (Duplicated or triplicated exon 2) so the C region of the γ chain may contain 189 or 205 amino acids. The total number of TRGV genes in a human haploid genome is 19
22 of which 11
13 are functional. In the dog, the TRG locus is organized into eight cassettes, each containing a basic V-J-J-C unit, except for a J-J-C cassette at the 30 end. It contains a total of 40 gene segments (16V, 16J, and 8C). Eight of the 16 canine TRGV genes, seven of 16 TRGJ genes, and six of eight TRGC genes are functional. The existence of these multiple TRGC genes suggests that the TCRs they generate may have diverse biological properties. Variations in gamma chain sequence length and amino acid residues have been reported in many other mammals [23].

8.6.1.4 TRM A fifth gene locus encoding TCR μ chains is present in marsupials [24]. The TCRμ chain consists of constant domains related to TCRδ but it is transcribed in such a way that it encodes two variable domains (Chapters 12 and 13). Thus it

TABLE 8.1 T cell antigen receptor diversity among mammal germlines.a TRA

TRD

TRB

TRG

Species

V

J

C

V

J

D

C

V

D

J

C

Horses

5

5

1

8

6

1

1

16

1

14

2

Bovine

183

60

1

55

3

9

1

134

3

21

3

18

8

6

Sheep

307

79

40

4

9

1

94

5

23

3

18

8

6
8

Rabbit

61

61

1

4

3

2

1

77

2

12

2

22

2

1

Pigs

48

50

1

28

4

6

1

43

3

20

3

8

6

3
4

Dogs

56

59

1

5

4

2

1

39

2

12

2

16

16

8

Cats

62

64

1

11

5

2

1

33

2

12

2

12

12

6

Mouse

98

61

1

6

2

2

1

34

2

14

2

7

4

4

Humans

54

61

1

8

4

3

1

68

2

14

2

15

5

2

a

V

J

C .2

The numbers in this table have been drawn from multiple references. They represent the number of germline genes reported at each locus. Not all these genes will be expressed, and many are pseudogenes. When sources differ, I have chosen the highest number of genes reported since as genetic analysis proceeds, more and more genes are being identified, and these numbers may be expected to rise.

T Cells and their receptors Chapter | 8

111

may generate a V-V-C sequence rather than the conventional two domain V-C structure. TCRμ does not substitute for TCRδ in marsupials. The genes encoding TCRδ in these mammals are also conserved and expressed. TCRμ is also present in a monotreme, the duck-billed platypus (Ornithorhynchus anatinus), suggesting that it was also present in the last common ancestor of the mammals over 200 mya. [24]. TCR μ chains normally pair with γ chains.

8.6.2 Possible combinations In the human TRA locus, there are at least 61 TRAJ genes and 54 TRAV genes, giving more than 3000 possible combinations. N-region addition and base deletion also occur resulting in great junctional diversity. After correction for codon redundancy and correct reading frame, the number of potentially different TCR α chains is about 106. In the human TRB locus, there are at least 68 TRBV genes, two TRBD genes, and 14 TRBJ genes that can generate 1904 possible combinations. In addition, there are junctional diversity mechanisms and the use of many TRBD combinations. After corrections there appear to be about 5 3 109 possible VDJ ß sequences. Thus the number of possible TCR α/β combinations in humans is about 5 3 1015.

8.6.3 TRC genes and habitat As pointed out throughout this book mammals inhabit a vast variety of ecological niches and climate zones. Each of these niches will have, to a greater or lesser degree, a different microbial population and as a result, will place different selective pressures on its mammalian inhabitants and their immune systems. When the T cell receptor peptide chain repertoire, TRAC, TRBC, TRGC, and TRDC in 37 mammalian species from different environmental niches was analyzed and related to the habitat, diet, and sociality, it is clear that habitat is the major factor shaping constant region repertoires. There is a clear trade-off between TRGC numbers and positive selection for TRAC [25]. This is especially obvious in wide-ranging cetaceans and bats when they are compared to the other groups.

8.7

γ/δ T cells

There are major differences between species in the number of γ/δ T cells employed by the immune system. In addition, the “γ/δ-low” species such as humans and mice possess relatively few V genes in their TRD and TRG loci and as a result, their combinational repertoire is small. The T cells bearing these receptors tend to use only a few V gene combinations. In contrast, human α/ß T cells have much more diverse TRA and TRB V genes and as a result, possess a much wider range of binding specificities. Thus there are marked differences between the size of the α/ß and γ/δ TCR repertoires. Unlike most α/β T cells, γ/δ T cell ligands are not limited to peptides presented by MHC proteins. Some can recognize lipids presented by CD1, or metabolites presented by MR1. Specific γ/δ ligands identified include MHC, CD1, and haptens that may modify target cell surface proteins [26]. Unlike most T cells, γ/δ T cells can respond very rapidly to infections since, in addition, to their TCR, they can be triggered through their pattern recognition receptors [27] (Fig. 8.8). γ/δ T cells express PRRs such as TLRs and NLRs. γ/δ T cells may also express the NK cell receptor NKG2D [28]. As a result, they acquire their effector functions before they even leave the thymus. They can home to specific tissues, secrete both regulatory and pro-inflammatory cytokines, act as memory cells, and in addition, can function as antigenpresenting cells [29]. Unlike α/β T cells, γ/δ T cells infrequently express CD4. Most γ/δ T cells express very little CD8 as well although they do express CD2, CD5, and CD6. Despite having similar structures, α/β and γ/δ TCRs also differ in the way in which they interact with the signaltransducing complex, CD3 [30]. They differ in their glycosylation patterns, their plasma membrane organization, and the accessibility of their signaling motifs on the intracellular domains of CD3. Thus in α/β T cells, the responding CD3 molecule exposes a proline-rich sequence on CD3ε and this is required to mediate α/β TCR signaling. Signaling through γ/δ TCR does not induce this CD3ε exposure [30] (Fig. 8.5).

8.7.1 γ/δ-high species As many as 66% of T cells in young calves and lambs, and 85% of T cells in neonatal piglets express γ/δ TCRs. This proportion rapidly decreases as the animals age but remains relatively high in adults. In other γ/δ -high species, the proportion ranges from 5% to 20% in goats; 11%
40% in adult cattle; deer and elk at 17%; and adult pigs at 30% [28]. In pigs, cattle, and sheep, γ/δ T cells can also be distinguished by their expression of a co-receptor/PRR called WC1 (Chapter 15).

112

SECTION | 1 Mammalian immunology

Ruminants and pigs INNATE

REGULATORY

Skin, mammary gland Intestine Blood WC1+ Antimicrobial

FIGURE 8.8 The different functions of γ/δ T cells. These functions, while broadly similar, differ between γ/δ-high and γ/δ low species.

JG-low species

JG-high species

Humans and mice

INNATE

ADAPTIVE

Skin Genital tract

Secondary lymphoid organs

WC1IL-12 IFN-J IL-10

Th1-like Treg-like

PAMPs Lipids MHC

Th1 IFN-J

Th2 IL-4

8.7.1.1 Bovine Up to two-thirds of T cells in young calves are γ/δ positive. This proportion decreases with age but is still in the range of 8%
18% in adult cattle. Bovine γ/δ T cells colonize the skin, mammary gland, reproductive organs, tonsils, and the intestinal wall where they form the major T cell population. They are polyclonal at birth, but their diversity decreases with age. They fall into two major subsets, innate cells, and regulatory cells. Between 50% and 99% of ruminant blood γ/δ T cells express WC1 and are engaged in innate immunity, whereas WC1-negative γ/δ cells are regulatory cells [31
33].

8.7.1.2 Innate cells Bovine innate γ/δ T cells respond to microbial PAMPs by increasing their expression of lymphotactin (XCL1), MIP-1β, TNF-α, and granulocyte-macrophage colony-stimulating factor (GM-CSF) [34]. They express TLR3, TLR9, mannosebinding lectin, and CD36. These γ/δ T cells may well be major contributors to innate immunity especially since many of them have nonpolymorphic TCRs. Cattle do not possess functional CD1 genes and so they probably lack true NKT cells that recognize glycolipid antigens [35]. They may recognize microbial glycolipids presented by other signaling molecules and release cytokines and lyse target cells just like conventional α/β NKT cells. IFN-γ-producing bovine γ/δ WC11 cells are also involved in recall responses to infections caused by Leptospira and Anaplasma [36]. While bovine γ/δ T cells do not express CD4, CD8, or immunoglobulins, they can use receptors in addition, to the γ/δ TCR [37]. Thus they can be stimulated not only through their TCRs but also through NLRs and NKG2D. This is in addition, to responding to antigens, cytokines, and pathogen-infected host cells. Ruminant γ/δ T cells show a considerably greater receptor diversity than those in γ/δ-low species.

8.7.1.3 Regulatory cells Bovine WC11 and WC12 cells have a different tissue distributions. WC12 cells predominate in the spleen and uterus. WC11 γ/δ T cells are found in granulomas surrounding schistosomes and Mycobacteria. In these cases, the initial T cell infiltration is dominated by γ/δ T cells, and this is followed by α/β T cells. A second wave of γ/δ T cells may terminate the response. These WC11 cells may secrete IL-12, IL-17, and IFN-γ and may promote a Th1 bias in the immune response. Other WC11 γ/δ T cells can produce IL-10 and TGF-β and presumably have a regulatory function [38].

8.7.1.4 Sheep In young lambs, two-thirds of their circulating T cells are γ/δ positive. Their γ/δ V-region diversity results from the use of 40 TRDV genes and 18 TRGV genes that contain two distinct hypervariable segments similar to the CDRs seen in immunoglobulin V genes [39]. In addition, there are multiple TCR γ/δ isoforms generated by the pairing of a single Cδ chain with one of up to eight Cγ chains. All this suggests that γ/δ T cells in these domestic artiodactyls recognize a very wide diversity of antigens and mount adaptive rather than innate responses [6]. Vaccine-induced γ/δ T cell responses have been reported in pigs and lambs.

T Cells and their receptors Chapter | 8

113

8.7.1.5 Pigs In most species, excluding pigs, γ/δ cells accumulate in large numbers in epithelial tissues and at inflammatory sites [40]. In pigs, γ/δ T cells migrate between blood, skin, intestine, and lymphoid tissues [41]. Pigs also have multiple phenotypically distinct subpopulations of γ/δ T cells [42]. Some may produce IFN-γ alone, TNF-α alone, or both IFN-γ and TNF-α. Other subpopulations can produce IL-17. Some have differing levels of TCR γ/δ expression.

8.7.1.6 Rabbits In rabbits, γ/δ TCRs may account for 20%
50% of circulating T cells [43]. The rabbit has two subfamilies of TRA/DV genes and two subfamilies of TRGV genes. The expression of these does however, differ between different tissues. For example, the rabbit TRDV1 gene is preferentially used in adult peripheral blood lymphocytes as a part of the γ/δ heterodimer.

8.7.2 γδ-low species γ/δ -low mammals are those that have less than 10% γ/δ T cells in their blood (Fig. 8.9). These species include humans and mice with 0.5%
10%; rats, 1%
5%; dogs with 2.5%; and guinea pigs 8.6%. Thus γ/δ T cells constitute a very small portion of their entire T cell population. The human TRV repertoire is similar to that of the mouse. Both have representatives of all subfamilies except that humans lack some TRGV sequences that are present in mice and sheep. Human and mouse TRDV are dispersed throughout the phylogenetic tree but genes from the cattle, sheep, and rabbits are confined to a few clusters [15]. Interestingly γ/δ-low species such as mice and humans have relatively few γ/δ cassettes. In addition, when different TCR-V gene subfamilies are analyzed, it has been found that both humans and mice have a very high degree of V gene diversity with representatives of most V gene subfamilies. In contrast, γ/δ-high species have much more limited V-gene segment diversity [6]. In γ/δ-low species there are two major subsets of γ/δ cells. One subset is engaged in innate immunity, has limited γ/δ receptor diversity, and is mainly found in the skin and genital tract. The other subset is engaged in adaptive immunity, has extensive receptor diversity, and is found in secondary lymphoid organs and the digestive tract. The innate γ/δ cells preferentially bind microbial PAMPs, especially heat-shock proteins and phospholigands (carbohydrates or nucleotides with a phosphate group). They also respond to the MHC class Ib molecules, MICA and MICB, produced by stressed cells, cancer cells, and virus-infected cells. These innate γ/δ T cells also respond to lipid antigens presented by CD1 molecules. When stimulated, they secrete IL-17 and IFN-γ. Like Th17 cells, innate γ/δ T cells are activated by IL23. Ligands for γ/δ T cells in mice include T22 and MICA in humans, and CD1 in mice and humans. T22 and MICA are MHC class I cell surface proteins. Their expression indicates that the cell is stressed. T22 binds to the antigenbinding groove of the γ/δ T cell receptor [44]. CD1 presents lipid antigens to the γ/δ cells. γ/δ cells can also recognize HSP60, mitochondrial ATPase, and prenyl phosphates. In contrast, the adaptive subset of γ/δ T cells in these species can be further subdivided into helper and effector cells. These effector cells can destroy target cells infected with mycobacteria and some leukemic cells. Many γ/δ T cells mature prior to α/β T cells. Opossums are an exception to this rule (Chapter 13). These cells also respond very rapidly (like an innate response) to fill the gap prior to the appearance of α/β T cells and B cells [26]. Data suggests that mouse and human γ/δ T cells have three functional roles. The first is a role in adaptive immunity by responding clonally to microbial antigens in a conventional manner, Many of these T cells may be self-reactive. Some proliferate and release IL-17 in response to microbial antigens. The second role is a quasi-adaptive one when in FIGURE 8.9 γ/δ T cell antigen receptor -high and -low mammals. The ranges reported are in the median range. Individuals, especially the very young may possess much higher numbers. Much depends on their age.

Guinea pig 9 Dromedary 35 Pigs 30-50 Beluga 31

Human 0.5-10 Mouse 0.5-10

I

I

I

I

I

0

10

20

30

40 PERCENT 50 GAMMA/DELTA T CELLS

Rat 1-5 Dog 1.7-2.5

Deer 17 Sheep 17 Goat 5-20

Rabbit Bovine 20-40

I

114

SECTION | 1 Mammalian immunology

BOX 8.2 γ/δ T cells and butyrophilins. Butyrophylins are cell surface glycoprotein receptors that are members of the immunoglobulin superfamily. They were originally found in cow’s milk
hence their name. They belong to the B7 family of costimulatory T cell receptors. This family includes such molecules as CD80, CD86, PD-L1 (CD274), and B7-H33 (CD276). They all modulate T cell responses to antigens presented by MHC molecules. Some influence the responses of α/β T cells while others regulate γ/δ responses. The BTN genes mainly cluster in the extended MHC class I region on human chromosome 6 and mouse chromosome 12. They are constructed of two immunoglobulin domains, IgV and IgC2 closely related to the B7 stimulatory receptors CD80 and CD86. The BTN genes are mainly expressed on epithelial cells. Some members of the family guide the thymic selection and tissue homing of Vγ5Vδ1 T cells. As a result, these T cells are found predominantly in the skin where they bind specifically to phosphoantigens. Butyrophilin-like (Btnl1) receptors expressed on gut epithelial cells shape the γ/δ compartment in the intestinal wall influencing their maturation and proliferation. Some Btnl heterodimers can specifically activate γ/δ T cells. In mice, some members of the butyrophilin family downregulate T cell function and induce Foxp31 Treg cells. Other butyrophilins are found in milk where they promote lipid secretion by binding to xanthine oxidoreductase. Rhodes DA, Reith W, Trowsdale J. Regulation of immunity by butyrophilins. Ann Review Immunol. 2016; doi:10.1146/ annurev-immunol-041015-055435. Di Marco Barros R, Roberts NA, Dart RJ, et al. Epithelia use butyrophilin-like molecules to shape organ-specific γδ T cell compartments. Cell 2016; 167:203
218.

response to pathogens such as Listeria or Plasmodia. In humans, clones with very limited diversity are produced and do not recirculate. They are major producers of IL-17A. In the third role, human γ/δ T cells respond rapidly to phosphoantigens by using cells expressing a specific γ/δ receptor pair
Vγ9Vδ2. This response is dependent on the presence of butyrophilin 3A proteins that act as T cell regulators (Box 8.2). Thus γ/δ T cells can participate in both innate and adaptive responses and bridge any remaining gaps in the defenses.

8.7.3 Invariant T cells Many mammals possess populations of T cells that express invariant antigen receptors. These include NKT cells and mucosal-associated invariant T cells. They are discussed in Chapter 10.

8.8

Memory T cells

As described above, when naı¨ve T cells encounter antigens with appropriate co-stimulation, they differentiate into multiple effector T cell populations. These effector cells are usually short-lived because they are eliminated by apoptosis. However, some resist apoptosis and develop into long-lived memory cells. These can be thought of as “antigen-experienced” stem cells. Memory T cells can be the most abundant T cell population in the body, especially in older animals since they accumulate throughout life. Compared to naı¨ve T cells, memory cells are easier to activate, live longer, and have enhanced effector activity. As a result, they mount a strong rapid cytokine response the next time they encounter the antigen and can provide life-long protection against pathogens. The differences in behavior between naı¨ve and memory T cells likely result from epigenetic modifications that alter gene transcription (mainly histone methylation) and hence cell functions [45]. This development of effector and memory T cell populations results from asymmetrical T cell division [46]. Naı¨ve T cells interact with antigen-presenting cells for several hours through a surface structure termed an immunological synapse [47]. Once it receives sufficient stimulation a T cell begins to divide even before it separates from the APC. The dividing T cell is polarized since one pole of the cell contains the antigen receptors within the immunological synapse. The other pole contains molecules excluded from the synapse. Thus when the cell divides, it forms two distinctly different daughter cells. The daughter cell adjacent to the synapse is the precursor of the effector T cells. The daughter cell formed at the opposite pole is the precursor of the memory T cells. Three types of memory T cells have been characterized in mice and humans. These are central memory cells, tissueresident memory T cells, and effector memory T cells [48]. Central memory T cells circulate through secondary lymphoid tissues, such as lymph nodes, awaiting the arrival of invaders. They lack immediate effector function but have very rapid recall responses. Effector memory T cells, in contrast, have receptors enabling them to home to inflamed tissues, where they immediately attack invaders without the need to differentiate further. Tissue-resident memory T cells occupy tissues and provide a first response to pathogens invading through body surfaces. They rapidly produce

T Cells and their receptors Chapter | 8

115

cytokines after infection. They do not circulate in peripheral blood. All memory T cell populations express either CD4 or CD8 and persist in the absence of antigen. CD81 memory cells tend to accumulate under epithelial surfaces while CD41 memory cells are scattered through the tissues in memory lymphocyte clusters [49]. These cells slowly divide and replenish their numbers. They can be thought of as adult stem cells. IL-7 and IL-15 are required for the survival of memory CD81 T cells, whereas only IL-7 is needed for the survival of CD41 T cells. These maintain the cells in a state of slow proliferation. In humans, memory CD41 T cells have a half-life of 8
12 years, whereas memory CD81 T cells have a half-life of 8
15 years [50]. The size of the immune system is fixed somewhat, but the effector memory CD81 T cell pool can double without loss of preexisting memory cells. Human memory T cells express TLR2. If exposed to its ligand lipopeptide in the presence of either IL-2 or IL-15, they will proliferate. Thus it is possible that bacterial PAMPs such as the lipopeptides may promote the long-term survival of memory T cells even in the absence of persistent antigen.

References [1] Wan YY. Multi-tasking of helper T cells. Immunology 2010;130:166
71. [2] Crome SQ, Wang AY, Levings MK. Translational mini-review series on Th17 cells: function and regulation of human T helper 17 cells in health and disease. Clin Exp Immunol 2010;159:109
19. [3] Annunziato F, Romagnani S. The transient nature of the Th17 phenotype. Eur J Immunol 2010;40:3312
16. [4] Gagliani N, Amezcua Vesely MC, Iseppon A, Brockmann L, Xu H, et al. Th17 cells transdifferentiate into regulatory T cells during resolution of inflammation. Nature 2015;523(7559):221
5. [5] Tizard IR. Veterinary immunology: an introduction. 10e. St Louis: Elsevier; 2017. [6] Su C, Jakpbsen I, Gu X, Nei M. Diversity and evolution of T-cell receptor variable region genes in mammals and birds. Immunogenetics 1999;50:301
8. [7] Brodovitch A, Bongrand P, Pierres A. T lymphocytes sense antigens within seconds and make a decision within one minute. J Immunol 2013;191:2064
71. [8] Nel AE. T-cell activation through the antigen receptor. 1: signaling components, signaling pathways, and signal integration at the T-cell antigen receptor synapse. J Allergy Clin Immunol 2002;109:758
70. [9] Go¨bel TWF, Dangy J-P. Evidence for a stepwise evolution of the CD3 family. J Immunol 2000;164:879
83. [10] Walsh KP, Mills KH. Dendritic cells and other innate determinants of T helper cell polarisation. Trends Immunol 2013;34(11):521
30. [11] Olivieri DN, Gambon-Cerda S, Gambon-Deza F. Evolution of V genes from the TRV loci of mammals. Immunogenetics 2015;67(7):371
84. [12] Ciccarese S, Vaccarelli G, Lefranc MP, Tasco G, et al. Characteristics of the somatic hypermutation in the Camelus dromedarius T cell receptor gamma (TRG) and delta (TRD) variable domains. Dev Comp Immunol 2014;46:300
13. [13] Kreslavsky T, Gleimer M, von Boehmer H. αβ vs γδ lineage choice at the first TCR-controlled checkpoint. Curr Opin Immunol 2010;22:185
92. [14] Reinink P, Van Rhijn I. The bovine T cell receptor alpha/delta locus contains over 400 V genes and encodes V genes without CDR2. Immunogenetics 2009;61:541
9. [15] Glusman G, Rowen L, Lee I, et al. Comparative genomics of the human and mouse T cell receptor loci. Immunity 2001;15:337
49. [16] Uenishi H, Eguchi-Ogawa T, Toki D, et al. Genomic sequence encoding diversity segments of the pig TCR delta chain gene demonstrates productivity of highly diversified repertoire. Mol Immunol 2009;46:1212
21. [17] Eguchi-Ogawa T, Toki D, Uenishi H. Genomic structure of the whole D-J-C clusters and the upstream region coding V segments of the TRB locus in pig. Dev Comp Immunol 2009;33:1111
19. [18] Antonacci R, Massari S, Linguiti G, Jambrenghi AC, et al. Evolution of the T-cell receptor (TR) loci in the adaptive immune response: the tale of the TRG locus in mammals. Genes 2020;11:624. Available from: https://doi.org/10.3390/genes11060624. [19] Weiss ATA, Hecht W, Henrich M, Reinacher M. Characterization of C-, J-, and V-region genes of the feline T-cell receptor γ. Vet Immunol Immunopathol 2008;124(63-74):2008. [20] Conrad ML, Pettman R, Whitehead J, et al. Genomic sequencing of the bovine T cell receptor beta locus. Vet Immunol Immunopathol 2002;87:439
41. [21] Massari S, Bellahcene F, Vaccarelli G, et al. The deduced structure of the T cell receptor gamma locus in Canis lupus familiaris. Mol Immunol 2009;46:2728
36. [22] Connelly T, Aerts J, Law A, Morrison WI. Genomic analysis reveals extensive gene duplication within the bovine TRB locus. BMC Genomics 2009;10:192. [23] Conrad ML, Mawer MA, Lefranc M-P, et al. The genomic sequence of the bovine T cell receptor gamma TRG loci and localization of the TRGC cassette. Vet Immunol Immunopathol 2007;115:346
56. [24] Wang X, Parra ZE, Miller RD. Platypus TCRm provides insight into the origins and evolution of a uniquely mammalian TCR locus. J Immunol 2011;187:5246
54. [25] Zhang Z, Mu Y, Shan L, Sun D, et al. Divergent evolution of TRC genes in mammalian niche adaptation. Front Immunol 2019;10:871. Available from: https://doi.org/10.3389/fimmu.2019.00871.

116

SECTION | 1 Mammalian immunology

[26] Hayday AC. γδ T cell update: adaptate orchestrators of immune surveillance. J Immunol 2018;203:311
20. [27] Hedges JF, Lubick KJ, Jutila MA. γδ T cells respond directly to pathogen-associated molecular patterns. J Immunol 2005;174:6045
53. [28] Holderness J, Hedges JF, Ramstead A, Jutila MA. Comparative biology of γδ T cell function in humans, mice and domestic animals. Ann Rev Animal Biosci 2013;1:99
124. [29] Brandes M, Willimann K, Moser B. Professional antigen-presentation by human γδ T cells. Science 2005;309:264
7. [30] Morath A, Schamel WW. αβ and γδ T cell receptors: similar but different. J Leukoc Biol 2020;107(6):1045
55. Available from: https://doi. org/10.1002/JLB.2MR1219-233R. [31] Baldwin CL, Hsu H, Chen C, Palmer M, et al. The role of bovine γδ T cells and their WC1 co-receptor in response to bacterial pathogens and promoting vaccine efficacy: a model for cattle and humans. Vet Immunol Immunopathol 2014;159:144
55. [32] Baldwin CL, Telfer JC. The bovine model for elucidating the role of gammadelta T cells in controlling infectious diseases of importance to cattle and humans. Mol Immunol 2015;66(1):35
47. [33] Hsu H, Chen C, Nenninger A, Holz L, Baldwin CL, Telfer JC. WC1 is a hybrid gammadelta TCR coreceptor and pattern recognition receptor for pathogenic bacteria. J Immunol 2015;194(5):2280
8. [34] Wilson E, Hedges JF, Butcher EC, et al. Bovine γ/δ T cell subsets express distinct patterns of chemokine responsiveness and adhesion molecules: a mechanism for tissue-specific γ/δ T cell subset accumulation. J Immunol 2002;169:4970
5. [35] Van Beeck FAL, Reinink P, Hermsen R, Zajonc D, et al. Functional CD1d and/or NKT invariant chain transcript in horse, pig, African elephant and Guinea pig, but not in ruminants. Mol Immunol 2009;46:1424
31. [36] Wang F, Herzig CT, Chen C, et al. Scavenger receptor WC1 contributes to the γ/δ T cell response to Leptospira. Mol Immunol 2011;48:801
9. [37] Baldwin CL, Damani-Yokota P, Yirsaw A, Loonie K, et al. Special features of γδ T cells in ruminants. Mol Immunol 2021;134:161
9. [38] Chen C, Herzig CT, Telfer JC, Baldwin CL. Antigenic basis of diversity in the gamma/delta T cell co-receptor WC1 family. Mol Immunol 2009;46:2565
75. [39] Vaccarelli G, Miccoli MC, Antonacci R, et al. Genomic organization and recombinational unit duplication-driven evolution of ovine and bovine T cell receptor gamma loci. BMC Genomics 2008;9:81. [40] Thome M, Hirt W, Pfaff E, et al. Porcine T-cell receptors: molecular and biochemical characterization. Vet Immunol Immunopathol 1994;43:13
18. [41] Sedlak C, Patzl M, Saalmuller A, Gerner W. CD2 and CD8alpha define porcine gammadelta T cells with distinct cytokine production profiles. Dev Comp Immunol 2014;45(1):97
106. [42] Stepanova K, Sinkora M. Porcine γδ T lymphocytes can be categorized into two functionally and developmentally distinct subsets according to expression of CD2 and level of TCR. J Immunol 2013;190:2111
20. [43] Porcelli SA, Morita CT, Modlin RL. T-cell recognition of non-peptide antigens. Current Opin Immunol 1996;8(4):510
16. [44] Sandstrom A, Scharf L, McRae G, Hawk AJ, et al. γδ T cell receptors recognize the non-classical major histocompatibility complex (MHC) molecule T22 via conserved anchor residues in a MHC peptide-like fashion. J Biol Chem 2012;287(8):6035
43. [45] Weng NP, Araki Y, Subedi K. The molecular basis of the memory T cell response: differential gene expression and its epigenetic regulation. Nat Rev Immunol 2012;12(4):306
15. [46] Ciocca ML, Barnett BE, Burkhardt JK, Chang JT, Reiner SL. Cutting edge: asymmetric memory T cell division in response to rechallenge. J Immunol 2012;188(9):4145
8. [47] Padhan K, Varma R. Immunological synapse: a multi-protein signaling cellular apparatus for controlling gene expression. Immunology 2010;129:322
8. [48] Schenkel JM, Masopust D. Tissue-resident memory T cells. Immunity 2014;41(6):886
97. [49] Vezys V, Yates A, Casey KA, et al. Memory CD8 T-cell compartment grows in size with immunological experience. Nature 2009;457:196
200. [50] Hale JS, Ahmed R. Memory T follicular helper CD4 T cells. Front immunol. 2015;. Available from: https://doi.org/10.3389/fimmu.2015.00016.

Chapter 9

Mammalian B cells The cells of the adaptive immune system must be able to detect and respond immediately to foreign antigens. This requires the use of two key antigen receptor systems, T cell antigen receptors (TCRs) and B cell antigen receptors (BCRs). Both require the rearrangement of V, D, and J gene segments to encode their specific antigen-binding sites. Sometime during the 100 million years between the divergence of the jawless and jawed vertebrates and the emergence of cartilaginous and bony fish, about 450 mya, the enzymes needed for this rearrangement of V gene segments appeared. The mechanism of this sudden appearance—“the immunological big bang”—is unknown. It has been suggested, however, that a transposon carrying the precursors of the recombinase-activating genes RAG1 and RAG2 (most likely a bacterial integrase) was successfully inserted into an immunoglobulin superfamily V-like gene within the germline of early jawed vertebrates. As a result, the V-like gene could be expressed after splicing by the RAG enzymes. Thus emerged the ability to generate diverse antigen-binding sites and functional immunoglobulins. The advantages of this new “improved” system were such that it is now a feature of all jawed vertebrates. It is important to point out, however, that these evolutionary innovations did not confer invincibility to infectious agents. They simply made life more difficult for potential pathogens and conferred an incremental selective advantage on animals with such defenses. The pathogens adapted and the Red-Queen kept running.

9.1

Before the mammals

9.1.1 Fish Fish B cells can be found in their thymus, anterior kidney, spleen, Leydig organ, and blood, and their cell surface immunoglobulins serve as antigen receptors. These B cells mature into plasma cells. Unlike most mammalian B cells, however, teleost B cells can phagocytose particles, generate phagolysosomes, and kill ingested microbes. This supports the theory that B cells may have evolved from an ancestral phagocytic cell and may account for some of the functional similarities between macrophages and mammalian B-1 cells (Chapter 23). Fish differ from mammals in the organization of immunoglobulin gene segments within the genome. For example, elasmobranch fish have clustered immunoglobulin genes, where V, D, J, and C segments form clusters that are duplicated many times; thus: -VDJC-VDJC-VDJC-. There are 100
500 of these VDJC clusters in sharks; each about 16 kilobases in size. Teleost fish, in contrast, have an immunoglobulin gene arrangement like that of mammals (the translocon pattern), with multiple V genes arranged thus: -V-V-V-V-V-D-D-J-J-J-C. IgM is found in both bony and cartilaginous fish and their antibody responses are characterized by its predominance [1]. This IgM is mainly produced by plasma-like cells in the anterior kidney. Fish IgM plus complement can lyse target cells but there is no evidence that these antibodies can act as opsonins, nor have Fc receptors been detected on their phagocytic cells. Blood vessel walls in fish are permeable to IgM. As a result, antibodies can be found in plasma, lymph, and skin mucus. Immunoglobulin D (IgD) and secretory IgD are also present in fish. These have some similarities to mammalian IgD, including co-expression with IgM on B cell surfaces. Fish do not produce IgA, but they secrete a mucosal IgM together with IgD. IgT is the major mucosal and skin immunoglobulin in teleosts. In serum, it is monomeric, in the gut and skin mucus it is multimeric. Mucosal IgT and IgM both can bind to a polymeric Ig receptor for transport into the gut lumen. Three additional immunoglobulin isotypes have been identified in elasmobranchs. These are IgW in the sandbar shark, IgNAR in the nurse shark, and IgR in the skate. IgNAR is a homodimeric heavy chain, an antibody with no associated light chains. This is a structure that subsequently reemerged in the immunoglobulins of camels (Chapter 14) [2]. Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00013-7 © 2023 Elsevier Inc. All rights reserved.

117

118

SECTION 1 | Mammalian immunology

9.1.2 Amphibians Anuran amphibians make five immunoglobulin classes, IgM, IgD, IgX, IgY, and IgF. The two most important are IgM and IgY. Anurans are also the least evolved vertebrates to show class switching. Frogs immunized with bacteria or foreign erythrocytes produce only IgM. Soluble foreign proteins can induce both IgM and IgY production. Amphibians do not mount a secondary response to erythrocytes and bacteria, but immunologic memory develops in response to those antigens that can subsequently stimulate an IgY response. The gene encoding the IgD heavy chain has been identified in the toad Xenopus tropicalis where it is expressed in mature B cells. As in mammals, it is located immediately 30 to the IGHM gene. Xenopus IgD is orthologous to IgW found in cartilaginous fish and lungfish implying that IgD/W was present in the ancestors of all living jawed vertebrates. Amphibian IgD is structurally very variable. Its gene shows many duplications, deletions of domains, the presence of multiple splice forms, or even the loss of the entire gene in some species [3]. An additional isotype, IgF with a φ heavy chain and only two constant domains, has been identified in Xenopus. It is unique in having a hinge region, the earliest example of such a structure in vertebrates. Urodeles also produce a monomeric IgM and can mount a good but slow antibody response against bacterial antigens. They do not respond to soluble protein antigens such as serum albumin or ferritin. Other immunoglobulin isotypes found in urodeles include IgY and IgD.

9.1.3 Reptiles The reptiles that have been studied possess IgM, IgD, and IgY. The IgM of turtles is comparable to mammalian IgM in size, chain structure, and carbohydrate content. They also produce several isoforms similar to IgD, as well as a hybrid IgM/IgD molecule. IgY is found in both full-sized and truncated isoforms. Some reptiles such as the Leopard gecko (Eublepharis macularius) produce IgA but anoles (Anolis sp.), turtles, and snakes do not. Analysis of gecko IgA shows that while its CH1 and CH2 domains are homologous to Xenopus IgY, its CH3 and CH4 domains are homologous to Xenopus IgM! It is probable therefore that recombination between IgY and IgM genes gave rise to this IgA [4]. Turtles in general have a single IgM gene, one IgD gene, several IgY genes, and several IgD2 genes. (IgD2 is a variant of IgD with its first four CH domains similar to IgD and the last two CH domains similar to the CH3 and CH4 domains of IgA.) Alligators have four IgM genes, one IgD, three IgA, three IgY, and two IgD2 genes [1].

9.1.4 Antibodies in mammals Given the presence of diverse immunoglobulins of multiple classes in ancestral vertebrates, it is clear that every mammalian species that has ever existed could make antibodies in response to foreign antigens. Over the B300 million years since their emergence, however, natural selection has acted to modify mammalian immunoglobulins in many ways. To make as many different antibody molecules as possible, it is necessary to diversify the amino acid sequences of the immunoglobulin antigen-binding sites. Since these amino acid sequences are determined by the nucleotide sequences, mechanisms must exist for generating this nucleotide sequence diversity within mammalian genomes. In practice, gene diversity is generated through three distinct mechanisms: gene recombination, somatic mutation, and gene conversion. The relative importance of each of these mechanisms differs among species, and the diversitygenerating mechanisms that operate in humans and mice are not necessarily the same as those that predominate in other mammals. The assembly of these diverse antigen receptors is carefully controlled during lymphocyte development by such epigenetic factors as DNA methylation, chromatin structure, and location within the nucleus. Unlike TCRs, BCRs can recognize antigens of diverse shapes and sizes in solution and independent of their location. This antigen-binding does not require the presence of antigen-presenting cells nor presentation by MHC molecules.

9.2

B cell antigen receptor structure

Each B cell is covered with about 200,000
500,000 identical antigen receptors; many more than the B30,000 antigen receptors expressed on each T cell. Each BCR is constructed from four peptide chains and these, like the TCR, can be divided into antigen-binding and signaling components. Antibodies are simply soluble BCRs released into body fluids; they all belong to the immunoglobulin superfamily [1].

Mammalian B cells Chapter | 9

119

Antigen binding-site

Antigen binding-site

NH2

Light chain

Hinge region

Disulfide bonds

Heavy chain

COOH

FIGURE 9.1 The basic structure of an immunoglobulin. In this chapter, constant immunoglobulin domains are colored green, the hinge regions are brown, and the variable regions are yellow. Similar color coding is also applied to immunoglobulin genes.

The basic structure of the antigen-binding component of the BCR (or immunoglobulin) is a glycoprotein of 160
180 kDa consisting of four linked peptide chains (Fig. 9.1). These chains consist of two identical heavy chains, each 60 kDa in size, and two light chains, about 25 kDa each. The light chains are linked by disulfide bonds to the heavy chains so that the complete molecule forms the letter Y. The tail of the Y (the Fc region) is formed from the paired heavy chains and attaches to the B cell surface. The arms of the Y (the Fab regions) are formed by paired light and heavy chains, and they bind antigens. The antigen-binding sites are located in the grooves between the light and heavy chains. Thus, each BCR has two identical antigen-binding sites. The amino acid sequences in the C-terminal domains in the BCRs from different B cells are largely identical and form the constant domains (C). In contrast, the sequences in the N-terminal domains differ in each B cell receptor and so form variable domains (V) [1].

9.2.1 Light chains Light chains are constructed from two domains, one variable and one constant, each containing about 110 amino acids. Mammals make two types of light chains, called κ (kappa) and λ (lambda). Although their amino acid sequences are different, they are functionally almost identical. The ratio of κ to λ chains in BCRs varies among mammals, ranging from mice and rats, which have more than 95% κ chains, to cattle and horses, which have 95% λ chains. Primates such as the rhesus monkey and the baboon have 50% of each, whereas humans have 70% κ chains. Carnivores such as cats and dogs use 70%
90% λ chains.

9.2.2 Heavy chains Immunoglobulin heavy chains are constructed from four or five domains each of about 110 amino acids. The Nterminal domain is a variable domain (VH). The remaining three or four domains show few sequence differences and thus are constant domains (CH) (Fig. 9.2). Mammalian B cells make four or five classes of heavy chains that differ in their sequence and domain structure. As a result, each class has a different biological activity. The five different immunoglobulin heavy chains are called α, γ, δ, ε, and μ. These heavy chains determine the immunoglobulin class. Thus, immunoglobulin molecules that use α heavy chains are called immunoglobulin A (IgA), and those that use γ chains are called IgG; μ chains are used in IgM, δ chains in IgD, and ε chains in Immunoglobulin E (IgE). Not all mammals make δ chains. A distinctly different immunoglobulin class called IgO has been described in the platypus (Chapter 12). The o gene encodes a heavy chain with four constant domains and a hinge. It is structurally different from all the other immunoglobulin subclasses. Phylogenetically this heavy chain is related to IgE and IgG [5].

9.2.2.1 Variable regions When the V domain sequences from light and heavy chains are examined in detail, two features become apparent. First, their sequence variation is largely confined to three smaller regions each containing 6
15 amino acids, within the

120

SECTION 1 | Mammalian immunology

Antigen binding

Antigen binding

Hinge region Fab Light chain

Heavy chains

Fc

FIGURE 9.2 A ribbon diagram showing the three-dimensional structure of an immunoglobulin molecule—in this example, bovine IgG. It is clear that it is formed by three globular regions linked by a hinge region. In this figure blue denotes constant domains, orange/yellow denotes variable domains while red denotes the antigen binding sites. Courtesy of Dr. B. Breaux.

Complementaritydetermining regions

CDR1 CDR2

CDR3 Light chain

CL

Antigen binding site

FIGURE 9.3 Each V region contains three complementarity determining regions (CDR) separated by relatively constant framework regions. While all three CDRs contribute to antigen binding, by far the most important and largest is CDR3. It is also the most variable in size depending upon the species of mammal.

CH1 Heavy chain

FR1 FR2 FR3 Framework regions

variable domain. These regions are said to be hypervariable. Between these three hypervariable regions are relatively constant sequences called framework regions (Fig. 9.3). The hypervariable regions in the paired light and heavy chains determine the shape of the antigen-binding site and thus the specificity of antigen binding. Since the shape of the antibody-binding site is complementary to the conformation of the antigenic determinant, the hypervariable sequences are also called complementarity determining regions (CDRs). Each V-domain is folded in such a way that its three CDRs come into close contact with the antigen. In practice, however, the most important and largest CDR is CDR3. It contributes most to antigen binding.

9.2.2.2 Constant regions The number of heavy chain constant domains differs between immunoglobulin classes [6]. There are three constant domains in a γ heavy chain; they are labeled, from the N-terminal end, CH1, CH2, and CH3. Three constant domains are also found in α and most δ chains, whereas μ and ε chains have an additional constant domain called CH4. Since heavy chains are paired, the domains in each chain come together to form dimeric structures by which antibody molecules can exert their biological functions. Thus, VH and VL together form an antigen-binding site, and CH1 and CL together stabilize the antigen-binding site. The paired CH2 domains of IgG contain a site that activates the classical complement pathway and a site that binds to Fc receptors on phagocytic cells. The heavy chain also determines the transfer of IgG into colostrum through binding to FcRn as well as antibody-mediated cellular cytotoxicity. When immunoglobulin molecules act as BCRs, their Fc region is embedded in the B cell surface membrane. These cell-bound immunoglobulins differ from the secreted form in that they have a small transmembrane domain located at their C-terminus. This contains a

Mammalian B cells Chapter | 9

121

sequence of hydrophobic amino acids, the conserved antigen receptor transmembrane (CART) motif, that associates with the cell membrane lipids and plays a role in receptor signaling.

9.2.2.3 Hinge region An important feature of immunoglobulin molecules is that their Fab regions can swing freely around the center of the molecule as if hinged. This hinge consists of a short domain of about twelve amino acids located between the CH1 and CH2 domains. The hinge region contains multiple hydrophilic and proline residues that cause the peptide chain to unfold and make this region readily accessible to proteases (Fig. 9.2). This region also contains the interchain disulfide bonds that bind the four peptide chains together. Because amino acids can rotate around peptide bonds, the effect of closely spaced proline residues is to produce a universal joint around which the immunoglobulin chains can swing freely. The μ chains of IgM do not possess a hinge region.

9.2.2.4 Signal transducing component BCR immunoglobulins cannot signal directly to their B cell since their cytoplasmic domains contain only three amino acids. However, their CH4 and transmembrane domains are associated with two glycoprotein heterodimers formed by pairing CD79a (Ig-α) with CD79b (Ig-β). These CD79 heterodimers are the signal transducers of the BCR (Fig. 9.4). The CD79β chains are identical in all BCRs. The CD79α chains differ depending on their associated heavy chains and employ different signaling pathways [7]. Antigen-BCR binding and cross-linking of two receptors exposes immunoreceptor tyrosine-based activation motifs (ITAMs) on CD79a and CD79b. Phosphorylation of these ITAMs by Src kinases leads to phosphorylation of phospholipase C and G-proteins. Subsequent hydrolysis of phosphatidylinositol and calcium mobilization generates protein kinase C and calcineurin and activates the transcription factors NF-κB and NF-AT. This eventually results in B cell mitosis and immunoglobulin production—providing the B cell also receives appropriate costimulatory signals from other sources [8,9].

9.3

B cell antigen receptor diversity

Three gene loci code for immunoglobulin peptide chains and each is found on a different chromosome. One locus, called IGL, uses three gene segments to code for λ light chains; one, called IGK, uses three segments to encode κ light chains; and one, called IGH, uses four gene segments to encode heavy chains. As in the TCRs, immunoglobulin diversity is generated by the use of multiple gene segments to produce a complete heavy or light chain. Gene recombination results from the random selection of one gene segment from each of several groups of segments followed by recombining these selected segments to generate sequence diversity in the CDR3 region of each V domain.

CD79

CD79

E

D D

E

FIGURE 9.4 Immunoglobulins are also B cell surface receptors. They lack signaling domains but they bind to their signal transducing component CD79. CD79 has a relatively simple structure consisting of paired α and β chains.

122

SECTION 1 | Mammalian immunology

9.3.1 IGH locus Four different gene segments, IGHV, IGHD, IGHJ, and IGHC, are required to form a complete gene for an immunoglobulin-heavy chain. The IGH locus contains multiple IGHV segments as well as multiple IGHJ segments. These are situated 30 to the IGHV genes. Several short segments, called IGHD genes (D for diversity), are positioned between the IGHV and IGHJ regions. The IGHC genes consist of a series of constant-region genes, one for each heavy chain class and subclass, usually arranged in the order 50 -M-D-G-E-A-30 along the chromosome. There is great variation in the number of heavy chain V genes among mammals. These range from 26 in opossums to 372 in rabbits (Table 9.1). Likewise, the number of heavy chain IGHG genes varies between species ranging from one in the opossum and rabbit to seven in the horse and eight in the African elephant. Each constant gene contains multiple exons encoding the sequences for each of the heavy chain domains including one or two for the hinge region, as well as exons for the transmembrane and intracellular domains expressed in cell-bound immunoglobulins (Fig. 9.5).

9.3.2 IGL locus Each λ light chain is encoded by three genes—IGLV, IGLJ, and IGLC. Mammals may also possess multiple IGLC genes. For example, dogs have nine, opossums have eight, and horses have four [10]. The IGLV gene codes for the variable region from the N-terminus to position 95. The IGLC gene codes for the constant region starting at position 110. The amino acids between 95 and 110 are encoded by IGLJ. The number of each of these IGLV region genes varies among domestic species. However, not all of these genes are functional—many are pseudogenes.

9.3.3 IGK locus Kappa light chains are also encoded by three genes, IGKV, IGKJ, and IGKC. In general, mammals possess only a single IGKC gene. In the rabbit, there are two IGKC genes as a result of lineage-specific duplication [11]. In rodents, IGKV genes are much more abundant than IGLV genes. In humans, they are present in similar numbers. The IGK locus appears to be lost in Myotis lucifugus, a microbat [10].

9.4

Evolution

As in the other antigen receptor genes, the evolution of immunoglobulin V-region genes is a result of two processes. First, new V genes are created by a process of gene duplication. Some of these new genes may acquire new functions, act beneficially, and be conserved within the genome. Others become nonfunctional pseudogenes and are eventually eliminated from the genome. The second process is diversifying selection whereby rates of non-synonymous mutation exceed those of synonymous mutations and so permit progressive minor positive changes over time. For example, new V genes can gradually emerge as a result of diversifying selection in their complementarity determining regions. In most mammalian species, germline VH and VL gene segments can be grouped into multigene subfamilies based on their nucleotide sequence similarity [12]. Thus, eleven V gene subfamilies have been described in the mouse, each containing between three to several hundred V gene segments. These subfamilies have expanded and contracted over time, and it is difficult to determine their evolutionary relationships. However, they are seen in different rodent species that diverged over a million years ago. These V segments can also acquire changes through somatic mutation at a rapid rate. Thus, while the adult V gene repertoire is very close to random, the situation differs in the fetus and newborn. There is also a high level of sequence similarity between mice and humans. It turns out that VH genes are well conserved in the germline largely based on the selection in favor of specific framework sequences. Many of these germline sequences are able to bind autoantigens [13]. Human VH segments are classified into seven subfamilies (IGHV1 to IGHV7). with different members of each subfamily being at least 80% homologous in nucleotide sequence. The total number of VH segments in the human genome range is about 129 on chromosome 14. Twenty-four more are found on chromosomes 15 and 16. A recombination signal sequence (RSS) is located immediately downstream of the V coding region. It is removed by the process of V-D joining so it will only work in the germline genome. The seven subfamily members are interspersed throughout the locus [14].

TABLE 9.1 The Immunoglobulin heavy chain genes of mammals. Species

IGHV

Platypus (O. anatinus)

44

Opossum (M. domestica)

26

Camel (C. dromedarius)

IGHD

IGHJ

IgM

11 9

50

Pig (S. scrofa)

30

Dolphin (T. truncatus)

47

Bovine (B. taurus)

42

Goat (C. hircus)

34

Bats (E. roussetus)

IgD

IgG

IgE

IgA

1

2

1

2

6

1

1

1

1

7

7

1

1

4

1

1

5

5

1

1

4
6

1

1

2

1

1

2

1

2

23

6

2

1

3

1

1

4

6

1

1

3

1

1

66

8

9

1

4

2

1

Cat (F. domesticus)

64

7

6

1

2

2

2

Dog (C. familiarus)

89

6

6

1

1

4

2

1

Horse (E. caballus)

52

40

8

1

1

7

1

1

Rabbit (O. cuniculus)

372

11

6

1

1

1

13

Mouse (M. musculus)

152

20

4

1

1

4
5

1

1

Rat (R. rattus)

195

21

5

1

1

4

1

1

Human (H. sapiens)

129

27

9

1

1

4

1

2

Elephant (L. africanus)

112

87

6

1

8

1

1

124

SECTION 1 | Mammalian immunology

5’

3’ V

CH1

H1 H2

CH2

CH3

TM

M1 M2

FIGURE 9.5 Within a typical constant region gene are sequences encoding not only the constant domains, but also one or two hinge regions (H1 and H2) as well as the transmembrane (TM) and intracellular domains (M1 and M2) required for B cell-bound immunoglobulins. This is typical of ruminants, equids, carnivores, rodents, and primates.

Opossum Pig Bovine Dog Mouse Human

Clan I

Clan II

Clan III

Relatively uncommon

Only clan in cattle and sheep

Only clan in pigs

0 0 0 1 58 11

0 0 11 1 11 11

26 30 0 41 27 22

FIGURE 9.6 The framework regions within V gene segments are relatively conserved. As a result, it is possible to determine their phylogenetic relationships. They fall into three distinct clans. Of these, clan I is least common while class III is present in most mammals. The numbers denote the number of V gene segments in each species.

9.4.1 V region clans There is a positive correlation between the number of functional and nonfunctional IGHV genes among mammals. Thus, the more gene duplications occur, the more IGHV pseudogenes are generated. These pseudogenes appear to have evolved much faster than functional genes and there is no evidence of concerted evolution to homogenize these genes [15]. As with the other multigene families involved in the mammalian immune responses such as the TCRs and the MHC families, it is clear that immunoglobulin V gene segments evolved through diversifying selection by a birth-anddeath process. Thus, the V segments duplicate, they then may mutate to form functional genes, or they may mutate to become dysfunctional pseudogenes [16]. The final result is a V region containing a mixture of positively selected V segments and scattered pseudogenes. However, sequence analysis suggests that these are not random as demonstrated by the existence of V-region clans. A clan is a set of V region subfamilies, originating in different species that, based on amino acid sequence diversity in their framework regions, appear to be phylogenetically related. For example, it is possible to classify immunoglobulin variable regions into three distinct phylogenetic families or clans based upon the amino acid sequence of two of their three framework regions [17]. Thus, in heavy chain variable regions (IGHV), the variable residues are clustered together into three complementarity determining regions that bind the specific epitope. These are separated by three relatively stable framework regions. These conserved framework sequences can be used to determine phylogenetic relationships. On this basis, it is believed that mammalian VH gene segments are probably derived from three distinct progenitors whose descendants form three clans. These clans are defined based on their sequence homology between families in the 6
24 codon interval in framework region 1 and in the 67
85 interval in framework region 3 (Fig. 9.6). The three IGHV clans are found in many diverse tetrapods suggesting that they originated before 400 mya, long before the major mammalian radiations [18]. All three clans have many conserved nucleotide sequences and the most conservation is seen in clan III. It is of interest to note that clan III members tend to be preferentially expressed during fetal life. Among the mammals, clan III and or class II genes are relatively abundant whereas clan I genes are either present in small numbers or absent in some species. The number of IGHV genes present in each of these clans also varies between different mammal species [10]. In addition, there is also a marked heterogeneity in the intraspecies sequence divergence. For example, artiodactyls have a low level of sequence divergence in their IGHV genes. Conversely, primates and rodents show a high degree of sequence variation. This may reflect differences in the history of their exposure to infectious agents. Clan I include the human IGHV1, 5, and 7 subfamilies as well as the mouse IGHV1, 9, 14, and families [19]. Clan II split into two before the divergence of rodents and primates. Thus, there is a VHII subclan (consisting of human IGHV2, mouse IGHV2, and 3) and a VHIV subclan (containing human 4, and 6 and mouse IGHV8, and 12). In cows and sheep, only clan II genes are present. In contrast, pigs have only clan III genes. Presumably, their common ancestor had both clans II and III. The pigs lost clan II while cattle and sheep lost clan III. Clan III contains the human IGHV3 subfamily and the mouse IGHV4, 5, 6, 7, 10, 11, 13, and 16 subfamilies. Clan III has a broader taxonomic distribution than either clans I or II.

Mammalian B cells Chapter | 9

125

The conservation of V region clans appears to reflect protein sequence selection. The conserved regions are localized at the solvent-exposed face of the V region frameworks at some distance from the CDR. A subfamily-specific region is also found within the recombinase recognition sequences. A clan structure is also apparent in immunoglobulin light chain V regions. Thus, there are three kappa chain clans. Clan 1 contains human IGKV1 and mouse IGKV 9, 10, 11, 12, 13, 14, 15, 16, and 19 subfamily genes. Kappa chain clan II contains human IGKV2, 3, 4, and 6, as well as mouse IGKV1, 2, 5, 6, 7, 8, and 18. Kappa chain clan III contains human IGKV5 and 7. It also contains mouse IGKV3, 4, and 17 subfamily genes. There are also seven distinct clans among the lambda light chains. For example, clan I includes the human 1, 2, 6, and 10 subfamily IGLV genes as well as seven pseudogenes. It also includes the mouse IGLV subfamily [19].

9.4.2 Immunoglobulin D IgD is found in all classes of jawed vertebrates, but not all species. These include fish, amphibians, birds, and mammals [3]. In contrast to other immunoglobulin classes however it shows great structural plasticity, and its functions remain unclear (Fig. 9.7) [20]. B cells that express IgD as their surface receptor may survive in peripheral lymphoid tissues despite, in many cases, being autoreactive. IgD may also play a role on mucosal surfaces since it reacts with antigens from the commensal microbiota. It may maintain mucosal integrity, and it may assist in optimizing certain type 2 immune responses. It is clearly not absolutely necessary in all species and has probably played different roles in different vertebrate taxa over evolutionary time [3]. Thus, IgD is present in humans, artiodactyls, and mice. It is absent from all marsupials examined to date. The gene has been pseudogenized in camelids (alpaca), some rodents (guinea pig), lagomorphs (rabbit), and Afrotheria (elephants and the Florida manatee).

9.4.3 Immunoglobulin E Both IgG and IgE appear to have originated as a result of duplication of the IgY gene in an early mammal. This likely occurred 220
310 mya. Specific receptors for IgE must have emerged at around the same time. Mast cells and their complex contents, especially serine proteases such as the chymases and tryptases likely developed around the same time as well [21]. The function of IgE is to trigger an acute inflammatory process that serves to eliminate parasites, especially helminth parasites from the tissues and the gastrointestinal tract. IgE is a typical Y-shaped, four-chain immunoglobulin with four constant domains in each of its ε heavy chains and a molecular weight of 190 kDa. IgE is readily destroyed by mild heat treatment. It does not activate complement and its ability to cross the placenta in primates is limited. It is however found in colostrum.

Pig

Mouse

Long hinge (2 exons)

Hinge region

Hinge region

Only two Ch domains Cytoplasmic domain

Transmembrane domain

Human

Hinge region

GCH1 or PCH1 domains

FIGURE 9.7 Some of the structural features of immunoglobulin D from different mammalian taxa. Similar differences are not seen in other immunoglobulin mammalian classes.

126

SECTION 1 | Mammalian immunology

TABLE 9.2 Serum IgE levels in domestic mammals [22]. SPECIES

“NORMAL” IgE (range)

Range in allergic or parasitized animals

Humana

5
513 ng/mL

,240 ng/mL

Dog

182 6 112 μg/mL (24
410)

195 6 108 μg/mL (85
550)

Catc

46 6 19 μg/mL

328 6 124 μg/mL

Horsed

84 6 90.9 μg/mL

109 6 69 μg/mL

Sheepe

1.8 6 1.3 μg/mL

15
30 μg/mL

2 μg/mL

B5 μg/mL

b

f

Pig a

Martins TB, Bandhauer ME, Bunker AM, et al. New childhood and adult reference intervals for total IgE. J Allergy Clin Immunol. 2013;33(2):589
591. Nimmo Wilkie JS, Yager JA, Eyre P, Parker WM. Morphometric analysis of the skin of dogs with atopic dermatitis and correlations with cutaneous and plasma histamine and total serum IgE. Vet Pathol. 1990;27(3):179
186. c Delgado C, Lee-Fowler TM, DeClue AE, Reinero CR. Feline-specific serum total IgE quantitation in normal, asthmatic and parasitized cats. J Feline Med Surg. 2010;12:991
994. d Wagner B. IgE in horses: occurrence in health and disease. Vet Immunol Immunopathol. 2009;132:21
30. e Shaw RJ, McNeill MM, Gatehouse TK, Douch PG. Quantitation of total sheep IgE concentration using anti-ovine IgE monoclonal antibodies in an enzyme immunoassay. Vet Immunol Immunopathol. 1997;57:253
265. f Wu JJ, Cao CM, Meng TT, Zhang Y, et al. Induction of immune responses and allergic reactions in piglets by injecting glycinin. Ital J Anim Sci. 2016;15 (1):166
173. Source: Tizard IR. Allergies and hypersensitivity diseases in domestic animals. St Louis: Elsevier, 2022. b

IgE does not act simply by binding and coating antigens, as do the other immunoglobulins. IgE acts as a signaltransducing molecule. When two-mast cell-bound IgE molecules are cross-linked by an allergen, they trigger receptor activation leading to the release of inflammatory molecules. The resulting acute inflammation enhances local defenses and helps eliminate invaders. IgE is found in exquisitely small quantities in serum, especially in humans but its concentration varies greatly within species. In healthy humans, its concentration is age-dependent, but it only accounts for 0.0005% of the total serum immunoglobulins (Table 9.2). In domestic, and especially wild mammals, IgE levels are greatly influenced by the presence of parasites [21]. IgE also has a very short half-life of 2
3 days in humans and about 12 hours in mice. As a result, it has to be produced continuously to maintain its serum levels. Most of the body’s IgE is firmly bound to highaffinity FcεRI receptors on the surface of tissue mast cells. Consequently, the IgE serum concentration does not reflect the total amount in the body [22].

9.5

Generation of immunoglobulin diversity

Immunoglobulin V domain diversity is generated in different ways depending on the species [23]. Some mammals rely on gene recombination followed by somatic mutation. In these species, immunoglobulin diversity is continuously generated from B cell precursors throughout an animal’s life. Other mammals, in contrast, use gene conversion for a short period early in life. After initial B cell diversity is generated, this pool of B cells expands by a self-renewing mechanism with little somatic mutation mediated by signals from the intestinal microbiota.

9.5.1 Recombination signal sequences Immunoglobulin and TCR genes are unique since they are assembled by the recombination of V, D, and J gene segments in developing lymphocytes. As discussed elsewhere, the immunoglobulin V, D, and J gene segments are believed to evolve by the “birth and death” model. Thus, genes duplicate, some are retained while others become pseudogenes or are eliminated as a result of mutations. When this happens, the V gene segments must contain, in addition to the coding sequence, the adjacent 5’ regulatory region and the adjacent RSS. The question arises whether their RSS and the gene segments have evolved as a unit or separately. The RSS is located immediately adjacent to the DNA recombination site. It is composed of conserved hexamer and nonamer motifs separated by a spacer that is either 12 or 23 base pairs in length [24]. VDJ recombination is mediated by the

Mammalian B cells Chapter | 9

127

recombination activating gene (RAG) proteins RAG-1 and RAG-2. These enzymes cause the double-stranded DNA breaks at the borders between the RSS and the coding gene segments. Phylogenetic analysis of the RSS evolutionary tree has indicated that they have a different evolutionary history than the VH coding genes. For example, their history does not correlate with the three mammalian V region clans. Analysis of the VH RSS and their adjacent gene segments have identified four types of RSS in mammals. These are called H1, H2, H3, and H5. (H is human, the number is frequency). H3 is the largest RSS type with 26 members that correspond to VH3, -4, and -6. (VH3 belongs to clan III while -4 and-6 belongs to clan II.) H2 associates with clan II. H1 associates with the clan 1 VH family 1. H5 is also associated with some members of clan I. Thus, when compared to the VH coding sequence, it is clear that RSS type H2 has been replaced by type H3 during evolution sometime between 115 and 65 mya. This may be due to unequal crossing over or perhaps activation of VDJ recombination within the germline. Presumably, the VH-H3 product has been positively selected because it makes for more efficient recombination. In examining other species, H2 appears to be human-specific. It has not been detected in other mammals. Type H1 is found in mice while type H3 is found in cows, sheep, camels, pigs, rabbits, chimpanzees, gorillas, rhesus macaques, and mice [24].

9.5.2 Gene rearrangement The mechanism of antibody diversification in artiodactyls differs from that in primates and rodents in that somatic gene conversion and hypermutation play a much more important role. Within these V regions, some conserved nucleotide sequences have been maintained for millions of years across many different species suggesting that these subdomains have an important functional role to play. The beta loops that define the FR1 and FR3 intervals are solvent-exposed and can diverge without affecting the basic structure of the immunoglobulin fold. Thus, the FR3 β loop is adjacent to the CDR1 and CDR2 domains of the heavy chain. Residues within this loop may interact with the bound antigen. Each class contains a characteristic FR1 interval that is solvent-exposed and structurally separated from the antigen-binding site. Families within a clan have an FR3 interval that is capable of either influencing the conformation of the antigenbinding site or even interacting directly with the bound epitope. During B cell development, to produce a complete peptide chain, the intervening or unused gene segments must first be removed and discarded. The first step in this process is to identify the sites where the DNA has to be cut. Thus, V and J genes have sites called switch regions at each end that guides the process. Cutting is the function of an “activation-induced” cytidine deaminase (AID). When a B cell receives the appropriate signals instructing it to eliminate unwanted genes, AID deaminates the cytidines in the specific switch regions involved. As a result, these cytidines are converted to uracils. This conversion results in DNA “damage” and as a result, both strands of the DNA are cut by an endonuclease at these points. The looped-out genes are chopped off, and the free ends of the DNA are “rejoined” by a DNA ligase so that the V and J genes form a continuous sequence. Two sets of enzymes are used in this process; endonucleases cut the DNA at two points, thus excising unwanted genes. Following this, DNA ligases join the free ends to form a continuous sequence. Light chain assembly requires the combination of one V, one J, and one C gene segment (Fig. 9.8). Light chain gene recombination also occurs in two steps. Randomly selected V and J gene segments are first joined to form a complete V-region gene. The joined V-J segments remain separated from the C gene segment until messenger RNA (mRNA) is generated. At that time the unwanted J segments are excised, and the complete V-J-C mRNA is then translated to form a light chain. When a heavy chain V region is assembled, its construction requires the splicing together of V, D, and J segments (Fig. 9.9). The use of three randomly selected segments enormously increases the amount of variability. The recombination of these segments also occurs in a specific order. Thus, IGHD is first joined to IGHJ, and then IGHV is attached to make a complete V-region gene. After transcription, any unwanted J genes are deleted, the IGHC gene mRNA is attached, and the assembled V-D-J-C mRNA is translated to form a heavy chain.

9.5.3 Base deletion and insertion Although random recombination of two or three gene segments generates much V-region diversity, additional mechanisms increase this diversity still further. For example, endonucleases may remove nucleotides randomly from the cut ends of the genes. As a result, the precise nucleotide at which V and J genes join varies, leading to changes in the nucleotide sequence at the splice site and variations in the amino acid sequence in the V region.

128

SECTION 1 | Mammalian immunology

Delete Germline DNA V1 V2 V3 V4 V5 V6 V7 V8

Vn

J1 J2 J3 J4 J5

DNA rearrangement B cell DNA V1 V2 V3 V4J3 J4 J5

Transcription RNA transcript V4J3 J4 J5

RNA splicing

mRNA J4 J5

Translation Light chain NH 2

C

V

COOH

FIGURE 9.8 The mechanism of immunoglobulin light chain gene assembly. It occurs in discrete steps beginning with the splicing out of unwanted V and J gene segments and so joining the selected V to the selected J segment. This is then transcribed and the unwanted J genes removed so that only the three required mRNA transcripts remain.

IGH gene cluster

Delete Germ-line DNA

5'

3'

V

V

D

CP

J

Delete

CG

DNA rearrangement I Pre-B cell

V

V

D

CP

J

CG

DNA rearrangement II B-cell DNA VD

CP

J

CG

Transcription RNA transcript VD

J

CP RNA splicing mRNA Translation

NH 2

COOH P chain

FIGURE 9.9 The mechanism of heavy chain V gene assembly. The first step is the joining of the required D and J segments by deletion of the intervening segments. In the second step the joined DJ segment is attached to the required V gene segment. Once V, D, and J are joined then the sequence is transcribed. The RNA transcript is further edited to get rid of unwanted segments and join the VDJ sequence with the constant region mRNA.

Mammalian B cells Chapter | 9

129

In immunoglobulin heavy chain gene processing, additional nucleotides may also be inserted at the V-D and D-J splice sites. Some of these nucleotides (N-nucleotides) are added randomly by an enzyme called terminal deoxynucleotidyl transferase (TdT). Up to ten N-nucleotides may be inserted between V and D and between D and J. Although the random selection of gene segments from two or three pools generates a huge number of combinations, not all of these combinations produce usable antibodies. Some combinations may generate non-productive rearrangements. For example, if nucleotides are inserted or deleted so that the codon reading frame is changed, the resulting gene may code for a different amino acid sequence. If this “frameshift” results in inappropriate splicing, translation is prematurely terminated. Non-productive rearrangements are probably produced in two out of three attempts during B cell development. When this happens, the B cell has several additional opportunities to produce a functional antibody. In the case of light chains in mice, for example, immature B cells initially rearrange one of the IGK genes. If this fails to produce a functional light chain, they switch to the second IGK allele for another attempt. If this does not work, the B cell will use one of the IGL alleles, and if this fails, the second IGL allele represents the last resort. If all these efforts fail to produce a functional light chain, the B cell cannot make a functional immunoglobulin. It will then undergo apoptosis. The sequence of events described above has been demonstrated in mice and humans; the details differ in domestic mammals. One obvious difference lies in the use of κ and λ light chains. In mice, rabbits, pigs, and humans, κ chains are preferentially used. In the other domestic species, λ light chains predominate. The reasons for these differences are unknown. It should also be pointed out that immunoglobulin gene rearrangement is not entirely random. For example, rabbits, mice, and humans tend to use the most 30 IGHV genes most often. This preferential use of certain genes results from a combination of factors, including the RSSs, the accessibility of the genes to the recombinase enzyme, sequences at the splicing sites, and how DNA folds.

9.5.4 Receptor editing Although each new B cell expresses a specific antigen receptor, developing B cells continue to rearrange their V, D, and J gene segments even after exposure to antigen. Thus, a B cell expressing a specific κ chain may restart V gene rearrangement by switching to the other IGKV genes or even switching to either of the IGLV genes. The cell may also continue to rearrange upstream non-rearranged V segments or downstream non-rearranged J genes. This receptor editing, which occurs within germinal centers, maybe a method of eliminating receptors that react with self-antigens.

9.5.5 Somatic hypermutation Recombination does not account for all the sequence variability found in immunoglobulin V regions. For example, there are three hypervariable areas (CDRs) within a V region. The largest and most important of these, CDR3, is located around position 96 and is generated by recombination between V, (D), and J genes. However, the other two hypervariable regions, CDR1 and CDR2, are located far from the V-J or V-D-J splice sites. Other mechanisms of generating antibody variability must therefore exist. In fact, gene recombination is only the first step in generating antibody diversity. It is followed by somatic hypermutation that alters gene sequences and generates antibodies that bind much more strongly and specifically to antigens [25]. In some mammals, such as pigs, up to 10% of their V regions are hybrids that use the CDR1 from one VH gene and CDR2 from another. These hybrid V regions are present in the genome [26]. Following initial exposure to an antigen, B cells proliferate and undergo antigen-driven selection within germinal centers [27]. The mutations in immunoglobulin V genes are generated by the same enzymes used for class switch recombination. They are triggered by antigen cross-linking of two BCRs, by the binding of CD40 to CD154, and by the binding of CD80 to CD28. These signals activate cytidine deaminase (AID) which deaminates the cytidines in V gene DNA and triggers repair processes. Any gaps are repaired by DNA polymerases using randomly selected nucleotide sequences [28]. As a result of this “repair,” the V gene sequences gradually change as the B cells respond to antigens. On average, one amino acid changes each time a B cell divides. The degree to which a B cell responds to an antigen is directly related to the strength (affinity) with which its receptors bind that antigen. The better the fit between antigen and receptor, the greater will be the stimulus received by the B cell. If its receptors cannot bind an antigen, the B cell will not be stimulated and will die. In contrast, those B cells whose receptors bind antigen with a high affinity survive and proliferate. Thus, as B cells respond to an antigen, successive cycles of mutation and selection of the highest affinity receptors eventually generate populations of B cells producing very high-affinity antibodies [29].

130

SECTION 1 | Mammalian immunology

Repeated somatic mutation and selection do not begin until after B cells have switched from making immunoglobulin M (IgM) to making either IgG or IgA. As a result, the affinity of IgM antibodies does not increase during an immune response, whereas the affinity of IgG antibodies does.

9.5.6 Gene conversion Mammals differ in the manner in which somatic hypermutation is generated. In general mice and humans rely on AID as described above to generate random point mutations and diverse V regions. Other mammals such as sheep, cattle, and dolphins may have relatively few V genes and random gene recombination alone cannot explain all their immunoglobulin diversification. In these species, V region diversity is also generated by gene conversion [30]. Species that employ gene conversion must have available a supply of short homologous DNA sequences used as a template to repair a “damaged” region [31]. These may be multiple V genes or pseudogenes and are usually plentiful in the V regions of the IGH, IGL, and IGK loci. During the conversion process, the B cell cytidine deaminase inserts uracil, which is then removed, leaving a gap in the nearest V gene. This gap is then “repaired” by selected short segments of DNA obtained from an upstream V-region gene or pseudogene. The “repaired” V gene will therefore have a different sequence than its precursor. Some of these gene conversion events may not generate a functional V region. In these cases, the defective B cells are eliminated.

9.5.7 Receptor assembly When BCRs are generated, the first chain to be assembled is the heavy chain. This chain is capable of generating much more junctional and combinatorial diversity than the light chain and is the major contributor to antigen binding. This heavy chain is linked to signal transduction molecules, and a surrogate partner chain is provided so that the pre-B cell can respond in a limited way to antigens. As a result, a small clone of B cells expressing only the heavy chain is formed. Signaling through this pre-receptor triggers limited proliferation. This is followed by the assembly of the light chain. The light chain uses only V and J genes and thus contributes much less diversity to the antigen receptor although it tends to “fine-tune” its antigen-binding abilities. Once a complete first chain is formed using V, D, and J genes, further recombination, and rearrangement of its genes are stopped thus preventing assembly of the second heavy chain allele. As discussed in Chapter 15, the order in which BCRs are generated in pigs is the opposite of most mammals with their light chains being assembled on the B cell surface before their heavy chains [32].

9.5.8 Intestinal bacteria and the B cell repertoire Mammals can be divided into two groups based on how their B cell receptor repertoire develops [33]. Thus, pigs, rodents, and primates provide the immune system with a constant supply of new B cells simply by generating them in the bone marrow throughout life [34]. Other mammals, including the large domestic herbivores, develop their B cell antibody repertoire in two stages. The first diversification stage involves rearrangements of a small number of V, D, and J gene segments within the bone marrow [35]. These early B cells then migrate to the intestinal lymphoid tissues such as Peyer’s patches, where they greatly increase their numbers of B cells as well as the diversity of their B cell repertoire. This second phase of B cell diversification generally takes place in intestinal lymphoid organs in direct content with the intestinal microbiota. The importance of the microbiota as a source of B cell stimulants is supported by the failure of germ-free pigs to develop significant B cell diversity [36]. Intestinal bacteria play an especially critical role in this process. For example, in rabbits, normal intestinal lymphoid tissue development can take place in the presence of both Bacteroides fragilis and Bacillus subtilis but not with either alone. Other bacterial combinations are also effective, suggesting that some form of bacterial interaction is needed for optimal effect. Analysis of the diversification of intestinal B cells by commensal bacteria also shows that this tends to affect B cells with certain VH domains. This expansion is not simply a specific B cell response to microbial antigens but rather a polyclonal, non
antigen-specific response. It may be directed through the toll-like receptors or be a result of microbial superantigen binding to the BCR, or some combination thereof.

Mammalian B cells Chapter | 9

131

9.5.9 Epigenetics While every cell in an animal body contains a complete copy of the genome, they only employ the genes needed for their specific function. Thus, a hepatocyte, for example, only uses those genes required for liver cell function. Processes that collectively control gene expression determine which genes are transcribed in each cell. This is due to epigenetic regulation and has been likened to the software in a computer. Like software, a cell selects the most appropriate components (genes) for the task at hand. The importance of epigenetic regulation is well seen in the regulation of immunoglobulin production [37]. There are three major mechanisms of epigenetic regulation: DNA methylation is one. By adding a methyl group to the 5’-position in certain cytosines, a gene can be turned off. Conversely, demethylation results in gene activation. A second epigenetic mechanism involves histone modification. DNA in the nucleus is closely bound to the nuclear histones. These histones can be subjected to many chemical modifications such as methylation, phosphorylation, or acetylation. These histone modifications are introduced or removed by histone-modifying enzymes. The modifications influence the interactions between the histones and DNA and as a result may activate or inhibit gene transcription. The third major epigenetic mechanism involves the production of microRNAs (miRNAs). These are small, non-coding RNAs that regulate gene expression by binding to mRNAs and influencing their functions. Collectively, these three epigenetic mechanisms determine which genes are active in any given cell type. Thus, B cell activation and differentiation are associated with genome-wide hypomethylation, an increase in histone acetylation, and the appearance of a specific set of miRNAs. In the case of immunoglobulin synthesis, class switch recombination and somatic hypermutation are regulated by all three of these epigenetic processes. In effect, the production of the appropriate recombinases, polymerases, AID, TdT, and repair enzymes is turned on by activating histone modifications and DNA hypomethylation triggered by exposure to antigen and the presence of appropriate cytokines. These epigenetic pathways also regulate immunoglobulin somatic hypermutation, and class switch DNA recombination as well as the differentiation of B cells into plasma cells, and long-lived memory B cells. Histone modifications affect the class switch and possibly somatic hypermutation. DNA methylation and miRNAs influence the activities of cytidine deaminase as well as plasma cell differentiation. These epigenetic pathways influence the normal immune response and, if dysregulated, also influence the development of abnormal B cell responses and autoimmunity.

9.5.10 Fc receptors While antibodies can bind antigens, their biological activities and the subsequent fate of the antigen are determined by where they bind. Thus, their Fc region binds to specific antibody (Fc) receptors. Fc receptors are a conserved family of glycoproteins belonging to the immunoglobulin superfamily. Humans generally possess four for IgG (one for each IgG subclass), one for IgE, IgM, and IgA, and one for both IgM and IgA. They are all structurally related. They all possess a similar α chain and they likely have arisen from a common ancestor by repeated gene duplication [38]. It appears that the first of these receptors to appear were the polymeric Ig receptors that are present in fish. Conversely, the latest to appear were the IgA receptors that are found only in placental mammals. The IgM and IgA/M receptors first appeared in the monotremes such as the platypus. The classical IgG and IgE receptors first appeared in marsupials [38]. The biological functions of these receptors are all mediated by immunoreceptor tyrosine-based activation motifs (ITAMs), or inhibitory motifs (ITIMs). Signaling through the ITAMs results in cell activation, opsonization, and phagocytosis. ITIMs have the opposite effect. Related immunoglobulin-domain-containing receptors include the natural killer Ig-like receptors, the leukocyte Iglike receptors, and the leukocyte-associated Ig-like receptors, expressed on natural killer cells. In primates, all these receptors plus the IgA receptor are expressed on chromosome 1. Structurally, the IgA receptor appears to be more closely related to the KIRs than to the other Fc receptors. It is perhaps no coincidence that the genes for some signal transduction molecules such as those for DAP10, DAP12, and the TCR zeta chain are also found in both the FcR locus and in the leukocyte receptor complex [38].

9.5.11 Fc receptor-like molecules In addition to the classical Fc receptors described above, B cells may also express members of a large family of type 1 transmembrane receptors that are homologous to the functional cell surface Fc receptors. Eight of these have been identified in the human genome [38]. These Fc receptor-like (FCRL) molecules contain a variable number of extracellular immunoglobulin-like domains as well as cytoplasmic domains with either inhibitory ITIMs or activating

132

SECTION 1 | Mammalian immunology

ITAMs. In humans, FCRL1
5 are expressed at different stages of B cell development while FCRL3 and -6 are expressed on subsets of T and NK cells. The FCRLs mainly use ITIMs in their cytoplasmic domains, suggesting that they serve an inhibitory function and regulate different stages of B cell development. For example, FCRL2 is expressed by memory B cells. However, FCRLs do not have the same binding specificities as classical FcR, and their ligands are not always known. FCRL3 binds secretory IgA, FCRL4 binds IgA, FCRL5 binds IgG while FcRL6 binds MHC class II molecules [39]. All of the classical FCRs and FCRL are encoded by genes on primate chromosome 1 suggesting that they originated from a common ancestor. However not all FCRL is expressed in every mammalian species. They appear to be abundant in humans, dogs, and elephants but only five orthologs have been identified in mice. An atypical chimeric gene designated FCRLS is absent in humans but present in rodents, some Carnivora, bats, artiodactyls, and elephants. Nor is it present in monotremes or marsupials [39]. It has the properties of an Fc receptor but contains a scavenger receptor cysteine-rich domain. Its function is currently unknown.

References [1] Tizard IR. Veterinary immunology: an introduction. St. Louis: Elsevier; 2017. [2] Griffin LM, Snowden JR, Lawson ADG, Wernery U, et al. Analysis of heavy and light chain sequences of conventional camelid antibodies from Camelus dromedarius and Camelus bactrianus species. J Immunol Meth 2014;405:35
46. [3] Ohta Y, Flajnik M. IgD, like IgM, is a primordial immunoglobulin class perpetuated in most jawed vertebrates. Proc Natl Acad Sci USA 2006;103(28):10723
8. [4] Wi Z, Wu Q, Ren L, et al. Expression of IgM, IgD and IgY in a reptile anole carolinensis. J Immunol 2009;183:3858
64. [5] Gambo´n-Deza F, Sa´nchez-Espinel C, Magada´n-Mompo´ S. The immunoglobulin heavy chain locus in the platypus (Oynchurnithorhs anatinus). Mol Immunol 2009. Available from: https://doi.org/10.1016/j.molimm.2009.05.025. [6] Eguchi-Ogawa T, Toki D, Wertz N, Butler JE, Uenishi H. Structure of the genomic sequence comprising the immunoglobulin heavy constant (IGHC) genes from Sus scrofa. Mol Immunol 2012;52:97
107. [7] Clark MR, Campbell KS, Kazlauskas A, et al. The B cell antigen receptor complex: association of Ig-α and Ig-β with distinct cytoplasmic effectors. Science 1992;258:123
5. [8] Gauld SB, Dal Porto JM, Cambier JC. B cell antigen receptor signaling: roles in cell development and disease. Science 2002;296:1641
2. [9] Wakabayashi C, Adachi T, Wienands J, Tsubata T. A distinct signaling pathway used by the IgG-containing B cell antigen receptor. Science 2002;298:2392
5. [10] Das S, Nozawa M, Klein J, Nei M. Evolutionary dynamics of the immunoglobulin heavy chain variable region genes in vertebrates. Immunogenetics 2008;60(1):47
55. [11] Das S, Hirano M, Tako R, McCallister C, Nikolaidis N. Evolutionary genomics of immunoglobulin encoding loci in vertebrates. Curr Genomics 2012;13:95
102. [12] Olivieri D, von Haeften B, Sanchez-Espinel C, Gambon-Deza F. The immunologic V-gene repertoire in mammals. bioRxiv 2014. Available from: https://doi.org/10.1101/0022667. [13] Watson CT, Breden F. The immunoglobulin heavy chain locus: genetic variation, missing data, and implications for human disease. Genes Immun 2012;13:363
73. [14] Cook GP, Tomlinson IM. The human immunoglobulin VH repertoire. Immunol Today 1993;16(5):237
42. [15] Ota T, Nei M. Divergent evolution and evolution by the birth-and-death process in the immunoglobulin VH gene family. Mol Biol Evol 1994;11(3):469
82. [16] Nei M, Gu X, Sitnikova T. Evolution by the birth-and-death process in multigene families of the vertebrate immune system. Proc Natl Acad Sci USA 1997;94:7799
806. [17] Kirkham PM, Mortari F, Newton JA, Schroeder HW. Immunoglobulin VH clan and family identity predicts variable domain structure and may influence antigen binding. EMBO J 1992;11(2):603
9. [18] Schroeder HW, Hillson JL, Perlmutter RM. Structure and evolution of mammalian VH families. Intern Immunol 1989;2(1):41
50. [19] International Immunogenetics Information System. http://www.imgt.org/IMGTindex/clan.php. [20] Wan Z, Zhao Y, Sun Y. Immunoglobulin D and its encoding genes: an updated review. Dev Comp Immunol 2021;124:104198. [21] Hellman LT, Akula S, Thorpe M, Fu Z. Tracing the origins of IgE, mast cells, and allergies by studies of wild animals. Front Immunol 2017. Available from: https://doi.org/10.3389/fimmu.2017.01749. [22] Tizard IR. Allergies and hypersensitivity diseases in domestic animals. St Louis: Elsevier; 2022. [23] Murre C. A common mechanism that underpins antibody diversification. Cell. 2015;163(5):1055
6. [24] Hassanin A, Golub R, Lewis SM, Wu GE. Evolution of the recombination signal sequences in the Ig heavy-chain variable locus of mammals. Proc Natl Acad Sci USA 2000;97(21):11415
20. [25] Papavasiliou FN, Schatz DG. Somatic hypermutation of immunoglobulin genes: merging mechanisms for genetic diversity. Cell 2002;109: S35
44.

Mammalian B cells Chapter | 9

133

[26] Butler JE, Weber P, Wertz N. Antibody repertoire development in fetal and newborn piglets, XIII. Hybrid VH genes and the preimmune repertoire revisited. J Immunol 2006;177:5459
70. [27] Berek C. The development of B cells and the B-cell repertoire in the microenvironment of the germinal center. Immunol Rev 1992;126:5
19. [28] Faili A, Aoufouchi S, Flatter E, et al. Induction of somatic hypermutation in immunoglobulin genes is dependent on DNA polymerase iota. Nature 2002;419:944
6. [29] Barnett BE, Ciocca ML, Goenka R, Barnett LG, Wu J, Laufer TM, et al. Asymmetric B cell division in the germinal center reaction. Science 2012;335(6066):342
4. [30] Schwartz JC, Lefranc MP, Murtaugh MP. Evolution of the porcine (Sus scrofa domestica) immunoglobulin kappa locus through germline gene conversion. Immunogenetics 2012;64(4):303
11. [31] Dale GA, Wilkins DJ, Bohannon CD, Dilernia D, et al. Clustered mutations at the murine and human IgH locus exhibit significant linkage consistent with templated mutagenesis. J Immunol 2019;203:1252
64. [32] Sinkora M, Stepanova K, Sinkorova J. Immunoglobulin light chain κ precedes λ rearrangement in swine but a majority of λ1 B cells are generated earlier. Dev Comp Immunol 2020;111:103751. [33] Parra D, Takizawa F, Sunyer JO. Evolution of B cell immunity. Annu Rev Anim Biosci 2013;1:65
97. [34] Sinkora M, Sinkorova J. B cell lymphogenesis in swine is located in the bone marrow. J Immunol 2014;193(10):5023
32. [35] Liljavirta J, Niku M, Pessa-Morikawa T, Ekman A, Iivanaienen A. Expansion of the preimmune antibody repertoire by junctional diversity in Bos Taurus. PLoS One 2014;9(6):e99808. Available from: https://doi.org/10.1371/journal.pone.0099808. [36] Butler JE, Sun J, Weber P, et al. Antibody repertoire development in fetal and newborn piglets, III. Colonization of the gastrointestinal tract selectively diversifies the preimmune repertoire in mucosal lymphoid tissues. Immunology 2000;100:119
30. [37] Li G, Zan H, Xu Z, Casali P. Epigenetics of the antibody response. Trends Immunol 2013;34:460
9. [38] Akula S, Mohammadamin S, Hellman L. Fc receptors for immunoglobulins and their appearance during vertebrate evolution. Plos One 2015. Available from: https://doi.org/10.1371/journal.pone.0124530. [39] Matos MC, Pinhiero A, Melo-Ferreira J, Davis RS, et al. Evolution of Fc receptor-like scavenger in mammals. Front Immunol 2021. Available from: https://doi.org/10.3389/fimmu.2020.590280.

This page intentionally left blank

Chapter 10

Mammalian innate lymphoid cells Cells with receptors that could recognize foreign molecules first appeared in the jawless vertebrates such as the lamprey and hagfish. These first cells used receptors encoded by germline genes. As a result, they could only bind a relatively conserved mixture of common microbial molecules. Some of these receptors were probably also able to bind and respond to molecules generated by cells in distress, especially modified MHC class I molecules. The origins of these early innate cells are unclear but they may have coevolved with T and B cells in the earliest vertebrates [1]. It is therefore probable that some innate lymphocytes have been assisting vertebrate immune responses for over 500 million years. Most phylogenetic evidence relates to natural killer (NK) cells. Their cytotoxic effector molecules such as the perforins and granzymes as well as regulatory transcription factors are of ancient origin. Innate cell receptors related to leukocyte immunoglobulin-like receptors and killer cell immunoglobulin-like receptors have been identified in bony fish. Cells that are functionally similar to NK cells have been identified in amphibians, reptiles, and birds in addition, to mammals [2].

10.1

Innate helper cells

Three major populations of innate lymphoid cells (ILC) with helper functions have been characterized in humans and mice and are presumably present in other mammalian species [3,4]. These cells perform functions that were once thought to be performed only by helper T cells. Thus there are innate counterparts of Th2 cells, Th22 cells, and Th17 cells. Each of these populations is characterized by its surface antigens, especially their mixture of receptors and the signals that trigger their production, by their transcription factors, their secreted cytokines as well as their functions. These innate lymphoid cells play important roles in the very early stages of antimicrobial immune responses. They also contribute to tissue repair and defense and the maintenance of epithelial integrity [5]. They are classified into three groups. Group 1 defends against viruses, intracellular bacteria, and parasites. Group 2 defends against helminths, and Group 3 promotes immunity to intracellular bacteria [6,7].

10.1.1 Group 1 innate lymphoid cells Group 1 ILCs are located under body surfaces such as the intestinal mucosal epithelium where they are scattered throughout the lamina propria (Fig. 10.1) [8]. They develop from lymphoid stem cells through the use of the transcription factor T-bet. Group 1 ILCs produce large amounts of Th1-associated cytokines such as interferon-γ and TNF-α in response to activation by IL-12, IL-15, and IL-18 from stimulated dendritic cells (cDC1 cells). As a result, ILC1 cells activate macrophages. They differ from NK cells in that they cannot produce perforins and therefore are not cytotoxic. ILC1 cells play a critical role in the defense against viruses, intracellular bacteria, and protozoa as well as some cancers through their production of IFN-γ and TNF-α. As a result, they also antagonize type 2 immune responses. ILC1 cells can be converted into ILC3 cells by exposure to IL-23, IL-1β, and retinoic acid. Conversely, ILC3 cells may be converted into ILC1 cells under the influence of IL-12, IL-18, and IL-15. NK cells are closely related to group 1 ILCs since they too produce IFN-γ and rely on T-bet for gene transcription [2].

10.1.2 Group 2 innate lymphoid cells Group 2 ILCs are found in the lung, skin, bone marrow, liver, mesenteric fat, and the lamina propria of the small intestine [8]. They arise from lymphoid stem cells and use two unique transcription factors, GATA-binding protein-3 (GATA3), and retinoic acid receptor-related orphan receptor-α (RORα). ILC2s produce large amounts of the Th2associated cytokines IL-5 and IL-13 and smaller amounts of IL-4, IL-5, and IL-9 in response to thymic stromal Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00004-6 © 2023 Elsevier Inc. All rights reserved.

135

136

SECTION | 1 Mammalian immunology

LYMPHOID PRECURSOR

Innate cells T and B cells Cytotoxic

Helper

NK cells

ILC1 ILC2

Immunity to intracellular pathogens

Immunity to helminth parasites

ILC3 Immunity to intracellular bacteria

FIGURE 10.1 A classification of the major types of innate lymphoid cells. These have primarily been investigated in laboratory mice. It is assumed that they are also present in other mammals and possibly, in other vertebrates.

lymphopoietin (TSLP), IL-25, and IL-33 from epithelial cells. ILC2 cells are required for the development of early innate immunity to parasitic helminths since they are the major source of IL-13 produced during helminth infections. They also regulate some type 2 inflammatory responses such as asthma and allergic disease [9]. They play a key role in regulating type 2 immunity by acting on macrophages, mast cells, and eosinophils [10]. Thus they control eosinophil production and induce an eosinophilia. They also induce mucus production by goblet cells, alternative activation of macrophages, and tissue repair. ILC2s can be readily converted into ILC1s by IL-12, and this can be reversed by IL-4 [11].

10.1.3 Group 3 innate lymphoid cells Group 3 ILCs are found in the lamina propria of the gastrointestinal tract, tonsils, Peyer’s patches, and appendix, as well as in the lung [8]. They use RORγt as their transcription factor and their maintenance and function depend on signals from the aryl hydrocarbon receptor. ILC3s produce IL-17 and IL-22 in response to stimulation by TSLP and IL-23 and they thus resemble Th17 cells in their cytokine profile. They play a central role in immunity on mucosal surfaces by resisting extracellular bacteria and fungi. They do this by producing IL-22 [12]. On exposure to IL-22, enterocytes produce antimicrobial peptides that protect the intestinal epithelium. ILC3-derived IL-22 also inhibits T cell-mediated intestinal inflammation caused by antigens from commensal bacteria. ILC3s activate dendritic cells (cDC2 cells) and so influence IgA class switching on mucosal surfaces. In the spleen, ILC3s express the B cell growth factors BAFF, APRIL, and CD40L and stimulate IgM production. ILC3 cells are closely related to the lymphoid tissue inducer (LTi) cells that are required for the development of lymph nodes, Peyer’s patches, and isolated lymphoid follicles. LTi cells regulate the differentiation of B cells and appear to be essential for the formation of lymph nodes and Peyer’s patches. It is believed that LTi cells emerged later than the other ILCs, about 400
500 mya [13].

10.2

Natural killer cells

The first innate lymphocytes to be identified were called natural killer (NK) cells because they can kill virus-infected and tumor cells without requiring prior activation. NK cells are, in effect, cytotoxic ILC1s. NK cells can recognize and kill cells infected by viruses or are otherwise abnormal, such as those that have undergone malignant transformation. They play an important role in mammalian resistance to viral infections. NK cells are present in all mammals investigated so far (With the possible (but unlikely) exception of rabbits. Chapter 22) [14,15]. In most of these species, NK cells are large, granular lymphocytes. In cattle, NK cells are large cells, but may not contain large cytoplasmic granules. There is debate about NK cell morphology in the pig. Some

Mammalian innate lymphoid cells Chapter | 10

137

investigators claim that they are large granular lymphocytes, whereas others believe that they are small lymphocytes without obvious cytoplasmic granules. NK cells are produced by bone marrow stem cells and are found in peripheral blood, lymph nodes, spleen, and bone marrow but not in the thymus. They range from 2% of the lymphocytes in mouse spleen to 15% of the lymphocytes in human blood. Their characteristic cell surface phenotype is CD32, CD561, and NKp461. In some species such as pigs, not all NK cells express NKp46 [16].

10.3

Nature killer cell receptors

NK cells do not make the enormous diversity of variable antigen receptors employed by T and B cells. Instead, they use multiple conserved receptors [17]. These receptors enable NK cells to distinguish normal from abnormal cells by monitoring their expression of MHC class I antigens (Figs. 10.2 and 10.3). As described in Chapter 7, virus-infected cells present processed viral antigens on MHC class I molecules to cytotoxic T cells, and in response, the T cell kills the virus-infected target (Fig. 10.2). An obvious defensive response on the part of the virus would be to prevent MHC class I expression. NK cells foil this because they detect MHC class I expression on a cell surface. If the MHC is absent (or abnormal), then the NK cell will kill it! NK cells use diverse receptors to detect MHC antigens on target cells. These receptors may be encoded by singlecopy genes or by complex multigene families encoding large numbers of highly polymorphic receptors [18]. Interestingly, some single-copy NK receptor genes are present in some mammalian species whereas they form complex multigene families in others. These are among the most rapidly evolving gene families in mammals and reflect adjusting priorities in the face of attack by many diverse viral pathogens. NK cells recognize abnormal cells using two different strategies. One, as described above, is the “missing-self” strategy by which NK receptors bind MHC class I molecules expressed on healthy cells, and as a result generate inhibitory signals that prevent NK cell killing. If, however, a cell fails to express MHC class I, then these inhibitory signals are not generated, activation signals will predominate, and the target cell will be killed. This occurs for example, when a virus suppresses cellular MHC class I expression to avoid destruction by cytotoxic T cells. Likewise, tumor cells that fail to express MHC class I molecules are also killed by NK cells. The second recognition strategy employs NK receptors that can recognize stress-induced proteins expressed on cell surfaces. When the NK cells detect these stress proteins, an activating signal is generated, and the NK cells kill the MHC class I Ly49

INHIBITION

NO LYSIS

MHC class I is missing Ly49

Stress ligands

NO INHIBITION

CELL DESTRUCTION

FIGURE 10.2 The missing-self hypothesis. Natural killer (NK) cells check for the presence of MHC class I antigens expressed on nucleated cells. If they are absent or cannot otherwise be detected, the NK cell is activated and the target cell is killed by the NK cell. If the MHC antigens are present, the receptors transmit an inhibitory signal, and the cell is usually ignored. Thus if viruses downregulate MHC class I expression to avoid cytotoxic T cells, they become vulnerable to attack by NK cells.

138

SECTION | 1 Mammalian immunology

Ly49

Target cell lysis MICA or MICB

OR NKG2 family

FIGURE 10.3 Examples of Natural killer cell receptors and their ligands. The use of these recepors differs significantly between mammalian species.

MHC class I

Inhibition of target cell lysis

MHC class I

KIR stressed cells. Cells infected by viruses and intracellular bacteria, as well as some cancer cells express these stress proteins. NK cell receptors are encoded in two different gene complexes. The leukocyte receptor complex (LRC) and the natural killer complex (NKC). The LRC encodes immunoglobulin superfamily molecules, while the NKC, located on human chromosome 12 encodes C-type lectin molecules (Fig. 10.4). The relative importance of the receptors encoded by these two complexes differs greatly between different species reflecting their rapid, lineage-specific, expansions and contractions. Mice and horses predominantly use the NKC to generate members of the C-type lectin-related family called Ly49. Humans and most other mammals, in contrast, primarily use the LRC to produce immunoglobulin superfamily members known as killer immunoglobulin-like receptors (KIRs). Both of these receptor types recognize the presence, or absence of MHC class I molecules. They can also recognize the molecules generated by stressed cells. As a result, combinations of these receptors and their ligands the MHC class I molecules, determine the response of NK cells to virus infections, cancers, allografts, and pregnancy and have significant effects on disease susceptibility.

10.3.1 The leukocyte receptor complex The human LRC is located on chromosome 19q13.4. It contains about 30 Ig superfamily genes encoding KIRs, leukocyte Ig-like receptors (LILRs), and leukocyte-associated Ig-like receptors (LAIRs). These all belong to the immunoglobulin superfamily (IgSF). Of these 14 are KIRs, 13 are LILRs and two are LAIRs. Of their products, KIRs have two or three Ig domains, LILRs have two or four, and LAIRs have one Ig domain [19]. In addition, multiple other structurally and functionally related genes have been located within the LRC. These include genes such as natural cytotoxicity triggering receptor 1 (NCR1), TARM1, and OSCAR. These LRC genes all appear to have descended from a single precursor [20]. Except for the LAIR2 protein, which is secreted, these genes encode type I cell surface receptors with up to four C2-type immunoglobulin domains.

10.3.1.1 Killer cell immunoglobulin-like receptors The most important of the genes located within the LRC encode the family of killer cell Immunoglobulin-like receptors
KIRs. KIRs are a family of closely related type 1 transmembrane glycoproteins expressed by natural killer cells and The genes of the LRC

13 LILR

LAIR

14 KIR

Human LAIR

10 LILR (PIR)

Mouse

The genes of the NKC CD69

1 Ly49 4 KLR 4 NKG2 pseudogene

Human CD94 NKG2 Mouse

23 Ly49

FCAR NCR1

FIGURE 10.4 The two major families of natural killer receptor genes are located within either the lymphocyte receptor complex or the natural killer complex. These are very different among mammals as exemplified by this simple diagram demonstrating how KIR genes predominate in humans while Ly49 genes (KLRA) predominate in mice.

Mammalian innate lymphoid cells Chapter | 10

139

TARGET CELL

MHC class I

KIR

(2 or 3 domains)

NK CELL

FIGURE 10.5 The major proteins encoded by genes within the leukocyte receptor complex are the killer immunoglobulin-like receptors. These are highly polymorphic receptors that bind MHC class I molecules. Their long cytoplasmic domain contains an immunoreceptor tyrosine-based inhibitory motif and as a result inhibits natural killer cytotoxicity.

some T cells (CD81 memory cells). They each consist of a single peptide chain with either two (KIR2D), or three (KIR3D) extracellular domains and either a long or short cytoplasmic domain (Fig. 10.5). The long forms are inhibitory and prevent NK cells from killing their targets. Conversely, the short forms lack ITIMs and are mostly activating. In humans, there are about 14 functional KIR genes arranged in tandem in the human LRC. The human KIR gene locus is extremely polymorphic. (As are its ligands, MHC class I molecules) [21]. Indeed, allelic polymorphism is so extensive in this locus that it is difficult to find unrelated individuals with identical KIR haplotypes. KIR gene polymorphism is focused on the MHC class I binding sites of these receptors. As a result, it determines their binding specificity. KIR gene expression patterns also vary clonally, so that individual NK cells may express random combinations of KIR receptors. This extreme diversity at the primate KIR locus results from selective pressure analogous to that seen in MHC loci. In other words, resistance to specific infections conferred by the KIR locus will depend on an animal’s haplotype [22,23]. KIR gene expression is clonally distributed and only a fraction of NK and T cells express any given KIR. A specialized population of uterine NK cells also employs KIRs to recognize MHC class I antigens on invasive trophoblast cells in the placenta (Chapter 2). The structures of the human KIR family point to their common origin. They evolved from a single precursor. This may have happened independently several times during evolution. KIR diversification is a result of multiple mechanisms including chromosomal recombination, high mutation rates, alternative splicing, and variable expression [24]. The expansion of the KIR family appears to be specific for primates since most other mammals have only one or few KIR genes. Although primate KIR genes vary in their number and diversity, four are present in virtually all haplotypes. Of the 14 human KIR proteins, seven are inhibitory while six are activating and one has a dual function. The total number of KIR genes expressed by a single individual range between seven and twelve, depending on the presence or absence of activating KIR loci. Thus human genotypes show great variation in their content of activating KIRs. While there are both inhibitory and activating KIRs, phylogenetic analysis indicates that the inhibitory receptor genes are ancestral [25]. Thus the activating receptors evolved from them as a result of mutations. This process probably occurred at different times in different species. In general, activating receptor genes are short-lived since they are subjected to conflicting selective pressures and eventually to negative selection. KIR genes have also been found to belong to two ancient lineages, 3DX and 3DL that arose 110
95 mya by duplication of a common ancestor before the K-Pg event and the subsequent radiations of the placental mammals. In primates, the 3DX lineage has remained as a single-copy gene (KIR3DX1) whereas the 3DL lineage has expanded and diversified. Conversely, in cattle, the opposite happened so that the 3DX lineage expanded while the 3DL did not. This second lineage in cattle consists of a single KIR gene and a related gene fragment [26]. 10.3.1.1.1 Functions KIRs regulate the cytotoxic activities of NK cells by interacting with their ligands, the MHC class I molecules expressed on most nucleated cells. They can distinguish between MHC allelic variants and identify virus-infected or transformed cells. Most KIRs play an inhibitory role. That is, when they bind to a class I MHC molecule they inhibit

140

SECTION | 1 Mammalian immunology

NK cytotoxicity. As a result, NK cells generally leave normal healthy cells alone. If, however, they encounter a cell that is not expressing MHC class I, in the absence of KIR-mediated inhibition, they will kill it. KIR inhibitory receptors signal through an immunoreceptor tyrosine-based inhibitory motif (ITIM) on their cytoplasmic domain [27]. A small number of KIRs have an opposite function and activate NK cells when they bind their ligands. Their ligands are expressed in stressed or otherwise abnormal target cells. These ligands include MICA, MICB, and ULBP, all of which are MHC class Ib molecules. The KIR activating receptors tend to have low binding affinities compared to the inhibitory receptors, so this contributes to the dominance of inhibition.

10.3.1.2 Leukocyte Ig-like receptor glycoproteins A second important group of receptors encoded within the LRC consist of the leukocyte Ig-like receptors (LILRs). LILR are structurally related to KIRs and may in fact be ancestral to them. The human LILR genes are located centromeric to the KIR cluster in the LRC complex (Fig. 10.4). They consist of two inverted clusters of six and seven loci [28]. All encode membrane-bound receptors except for LILRA3 which is soluble. They may be either inhibitory or activating. LILRs show less allelic variation than the KIRs. LILRs are designated Paired immunoglobulin-activating receptors (PIRs) in the mouse. LILR expression is not restricted to NK cells. They are present on diverse leukocytes such as neutrophils as well as osteoclasts, endothelial cells, and neurons [29].

10.3.1.3 Other leukocyte receptor complex receptors Other receptor genes found within the LRC include NCR1 encoding a receptor protein called NKp46, (CD335). NCR1 is only expressed on NK cells. They also include FCAR which encodes the IgA receptor. Between the two LILR clusters there are also two leukocyte-associated immunoglobulin-like receptors (LAIR) loci, LAIR1 (CD305) and LAIR2 as well as an osteoclast-associated receptor (OSCAR). These also encode receptors that bind to collagen [30]. LAIR-1 is a high-affinity receptor for extracellular matrix collagens. When cells bind to collagen. they directly inhibit immune cell activation.

10.3.2 The natural killer complex The products of a second gene complex, the NKC also play a critical in regulating NK cell functions (Fig. 10.6). In humans, the NKC genes are clustered in a single region on chromosome 12p13. It spans 2.8 mb. It contains genes encoding glycoproteins belonging to the C-type lectin superfamily. Eight of these appear to be functional [31]. The most important of these belong to the killer cell lectin-like receptor (KLR) family. In mice, these glycoproteins are called Ly49. In humans, there is a single member of the KLR gene family, KLRA1. It is transcribed but is nonfunctional as a result of a point mutation. In rodents and horses, in contrast to humans and cattle, the predominant NK cell MHCbinding receptors belong to the KLR family. They have the same function as KIR receptors in that they bind MHC class I molecules. They are exceptionally polymorphic. Ly49 proteins are also encoded by the KLRA gene cluster. They are all homodimeric type II transmembrane C-type lectins. KLR haplotypes also contain variable numbers of inhibitory and activating genes, some of whose products can recognize MHC class I molecules. Target cell MHC class I

MICA/B

Ly49

NKG2D

NK cell DAP 10 in humans DAP 12 in mouse

FIGURE 10.6 The natural killer complex contains genes encoding two types of MHC receptor. One type consists of Ly49 molecules that bind to MHC class I molecules. The other consists of NKG2 receptors that bind nonclassical MHC class I molecules such as Major histocompatibility complex, class I Chain-related A (MICA) and Major histocompatibility complex class I Chain-related B (MICB) on target cells.

Mammalian innate lymphoid cells Chapter | 10

141

The NKC encoded receptors also play a key role in protection against virus infections and tumors [31]. For example, four lineages KLRA1, KLRB1, CLEC2D, and CLEC4A/B/C, have greatly expanded in rodents so that mice express 23 members of the KLR gene family while rats express 36. These receptors are primarily expressed on NK cells but they are also present on some T cells, NKT cells, macrophages, and dendritic cells. Their primary function is to serve as receptors for MHC class I molecules and so distinguish between healthy and infected or otherwise abnormal cells. They, therefore have an analogous function to KIRs. Like the KIR family, KLRs include both activating and inhibiting NK receptors [32]. Since Ly49 and KIRs are unrelated and bind to different sites on the MHC, this suggests that they have been subjected to convergent evolution. Horses and donkeys have at least five highly conserved KLR (Ly49) genes [33]. Humans, pigs, dogs, cats, and cattle possess only a single KLR-like gene. This may be functional in cattle. NK1.1 is a member of the Ly49 family that acts as an activating receptor on mouse NK cells.

10.3.2.1 NKG2 receptors The NKC also contains genes belonging to the NKG2 (KLRC) family of receptors. These encode homodimeric C-type lectin-like type II transmembrane receptors (Fig. 10.6). NKG2D (NK group 2D) receptors bind to nonclassical MHC class I proteins produced by stressed cells. Thus their ligands act as danger signals alerting the immune system to the presence of abnormal cells. The NKG2 receptor family, like the others described above, contains inhibitory receptors (NKG2A, -B, and KLRL1) as well as activating receptors (NKG2C, -E, and -H). When both types are expressed on the NK cell surface the inhibitory ones predominate. The NKG2 family members are transmembrane glycoproteins that associate with an invariant CD94 molecule on the cell surface to form a heterodimeric receptor. Some of the NKG2 family, specifically NKG2A (KLRC1) and NKG2C (KLRC2) /CD94 complexes bind specifically to the nonclassical MHC1 molecule HLA-E. This enables the NK cells to monitor the expression of HLA-E on healthy cells and maintain self-tolerance. NKG2D (KLRC3) shows less homology with the other members of the family and is expressed on NK cells, macrophages, and T cells as a homodimer. It can also serve as a costimulatory signal on CD81 α/β T cells. The ligands for human and mouse NKG2D are also different. In humans, it binds to MICA and MICB, and the UL16 protein-binding family. (UL-16 binding proteins are a family of MHC class-I related molecules that can bind to viruses such as cytomegalovirus and MICB) [34]. In mice it binds to two other class I related proteins, RAE1 (a homolog of ULBP), and H-60 [35]. Mice completely lack the MIC gene family [36,37]. NKG2D interacts with a signal-transducing adapter molecule called DAP10 to form a functional receptor complex and sends an activating signal to the NK cell resulting in the killing of the target cell. 10.3.2.1.1 NKG2 ligands Two of the most important NKG2 ligands in humans are the polymorphic MHC class I-like molecules called MICA (Major histocompatibility complex, class I Chain-related A) and MICB coded for by MHC class Ic genes [38]. Humans have seven MIC genes named MICA to MICG but only two, MICA and MICB, are functional. Their polymorphism is not restricted to the antigen-binding sites and unlike conventional class I molecules, their products do not bind antigenic peptides [39]. While minimally expressed on normal, healthy cells they are expressed in large amounts by stressed cells. These stresses may include DNA damage due to ionizing radiation or alkylating agents, heat shock, and oxidative stress. MICA and MICB are overexpressed in tumor cells and virus-infected cells. When these ligands bind to NKG2D, it overrides the inhibitory effects of conventional MHC class I molecules and triggers NK cytotoxicity. NKG2D is also expressed on activated γ/δ and α/β T cells, suggesting that they too have a role in innate immunity. It may be that the combination of γ/δ T cells and NK cells kills tumors on body surfaces, whereas a combination of α/β T cells and NK cells is most effective within the body. NKG2 receptors can recognize the MIC genes of chimpanzees, gorillas, and Cynomolgus macaques [37]. However, chimpanzees and gorillas each have a single MICA/B fusion gene. In addition to rats and mice, other rodents such as kangaroo rats (Dipodomys ordii) and guinea pigs lack MIC genes but thirteen-lined ground squirrels (Icidomys tridecemlineatus) have them! Lagomorphs including rabbits and pikas (Ochotona princeps) do not. Cows have four, horses three, and pigs two MIC genes.

10.3.3 Other natural killer cell receptors 10.3.3.1 Natural cytotoxicity receptors Natural cytotoxicity receptors are type I transmembrane receptors belonging to the immunoglobulin superfamily C. The most important are NKp46, NKp44, and NKp30 encoded by genes NCR1, NCR2, and NCR3 respectively. NKp46 and

142

SECTION | 1 Mammalian immunology

NKp30 are expressed on resting NK cells. NKp44 in humans is only expressed following NK cell activation. NKp46 is expressed only on NK cells while the other two are also expressed on T cell subsets. Their ligands are expressed on virus-infected and tumor cells. The NCR1 gene is located within the LRC. Its product, NKp46 recognizes membrane-associated heparan sulfate proteoglycans and can act as a receptor for the influenza virus hemagglutinin [20]. NCR1 is known to be expressed on a subset of canine NK cells [40]. The NCR2 gene is located on human chromosome 6. Its product, NKp44 is not expressed in the mouse but is in primates. NKp44 is a 44 kDa glycoprotein containing a single variable domain. It signals through the adapter protein DAP12 and its ITAM motif. When crosslinked, it activates NK cells triggering cytotoxicity and the release of IFN-γ and TNF-α. The ligands of NKp44 include heparan sulfate proteoglycans and platelet-derived growth factors. It can also bind some viral hemagglutinins as well as some bacterial surface components. The NCR3 gene is located in the MHC class III region [41]. One known activating ligand of NKp30 is B7-H6, a protein selectively expressed on many tumor cells. Other cellular and viral ligands including viral hemagglutinins have been reported [27]. NK cells may also recognize and kill target cells using an antibody-dependent pathway employing the Fc receptor, CD16 (FcγRIII). CD16 is a 38-kDa transmembrane protein linked to either the γ chain of FcγRI (in macrophages) or to the zeta chain of CD3 (in NK cells). When antibodies bind to target cells, the bound antibody links to NK cell CD16 triggers cytotoxicity and the target cells are killed. NK cells can spontaneously release their CD16 so that the NK cell can detach from an antibody-coated target after it has delivered its lethal hit. Other important human NK cell receptors include CD2, CD178 (CD95L or Fas ligand), CD40L (CD154), toll-like receptors (TLR3 and TLR9), and leukocyte function-associated antigen-1 (LFA-1).

10.3.3.2 Species differences Once the KIR and Ly49 genes were discovered in humans and mice, studies on other species showed that some such as rats and horses have diverse Ly49 genes but not KIRs [27]. On the other hand, primates and artiodactyls have diverse KIR genes but not Ly49 genes. Most KIR and Ly49 families appear to be species-specific, reflecting their unique histories of infectious disease exposure and the benefits of specific allelic forms [42]. Mammalian NK cell expression of KIR and Ly49 receptors tends to be mutually exclusive. A species may have either a diverse Ly49 or a diverse KIR gene family, but not both. One notable example is the horse that has at least six Ly49 genes but no functional KIR genes. Five of the equine Ly49 molecules have a functional ITIM motif and one has arginine in its transmembrane region. (Horses have four KIR-Ig-like transcripts (ILT) that contain premature stop codons and/or frameshift mutations and are probably pseudogenes as well as two KIR3DL sequences with mutated ITIMs) [33]. The use of multiple Ly49 genes by rodents and horses is not typical of mammals in general. Cattle have six KIR genes and a single functional Ly49 gene [43]. Cattle do however, possess a novel T cell subset that has features of both T and NK cells (NKT cells). Thus these T cells express both NCR1 and CD3. They can be activated by either the NKR or TCR pathways and can kill Theileria parva-infected cells. Cattle and goat NKT cell receptors have ITIMs while a novel KLRH gene product has acquired an activating tail. Marine mammals do not require either KIR or Ly49 receptors. Thus in pinnipeds, both seals and sea lions, Ly49 and KIR are each represented by a single gene [42]. KIR genes among the different primate species are very diverse, and consistent with rapid, species-specific gene expansion [28]. Thus chimpanzees have seven; gorillas have 11; the Rhesus macaque has five and orangutans have 17! [28] Catarrhine and platyrrhine monkeys but not prosimians also have them. Platyrrhine KIRS and MHC class I highly diverge. In the Old-World monkeys, diversification focuses on MICA and MICB and lineage II KIR. As primates evolved and the orangutan switched to expressing MHC-C its KIR also switched to lineage III. Now primates have distinctive KIRA and KIRB haplotypes. These include multiple subfamilies of C-type lectin receptors that vary among species.

10.3.3.3 Inhibition vs activation NK cell receptors of both superfamilies come in two forms, inhibitory and activating. Whether the NK cell receptor is inhibitory or activating depends upon the amino acid sequence of its transmembrane and cytoplasmic domains and the nature of its signal-transducing motifs. Signals from their inhibitory receptors prevent NK cell activation and block NK cytotoxicity when target cell MHC class I molecules are correctly expressed. The inhibitory receptors are monomeric peptides with long cytoplasmic tails

Mammalian innate lymphoid cells Chapter | 10

143

carrying a six amino acid immunoreceptor tyrosine-based inhibitory motif (ITIM) on their long cytoplasmic domain. When activated, they recruit phosphatases. As a result, these inhibitory tyrosine residues become phosphorylated. The phosphate comes from dephosphorylating cell activation pathways thus turning them off and so preventing NK cell cytotoxicity. Once inactivated, an NK cell may remain in that state for an indefinite period. Activating receptors, some of which bind ligands other than MHC class I, trigger NK cytotoxicity [28]. The activating forms have short cytoplasmic domains but have an immunoreceptor tyrosine-based motif (ITAM) in their transmembrane subunits. The ITAM consists of a conserved four amino acid motif repeated twice. The tyrosine residues in these motifs are phosphorylated by Src family kinases. This induces a signaling cascade leading to cell activation. For example, NK cells trigger actin polymerization leading to cytotoxicity and cytokine release. This ITIM/ITAM dichotomy applies to most of the NK receptors. DAP 12 is used as a signal-transducing molecule in inhibiting KIRs while the FcRγ chain is used in activating LILRs, NCR1, FCAR, OSCAR, and TARM1. TARM1 is an LRC encoded receptor that stimulates proinflammatory cytokine secretion by macrophages and neutrophils [44].

10.3.3.4 Effector mechanisms NK cell activities are also regulated by cytokines and pathogen-associated molecular patterns (PAMPs). For example, IL-2 and IL-4 enhance their cytotoxicity, whereas IL-3 promotes NK survival. Although NK cells are active in nonimmune animals, virus infections or interferon inducers will enhance their activity. When macrophages phagocytose invading organisms and produce tumor necrosis factor-α (TNF-α) and IL-12, these cytokines then induce interferon-γ (IFN-γ) production by NK cells. The IFN-γ enhances NK activity further by promoting the rapid differentiation of preNK cells. Once activated, NK cells kill target cells through either a perforin/granulysin/NK-lysin pathway or through the death domain pathway involving CD95L. NK cell granules are stored preformed in resting NK cells (in contrast to cytotoxic T cells, which only produce theirs on demand). Once an NK cell encounters a target cell, a synapse forms at the contact site. Activating KIRs induces MHC molecules to form a ring around a cluster of adhesion molecules. At the center of the synapse, there is a molecular activation cluster through which NK cell granule contents can pass [45]. Receptors and signaling molecules also segregate in the center, whereas integrins and talin accumulate around the outside of the synapse. Inhibitory KIRs in contrast, cluster within the synapse and block tyrosine phosphorylation of signaling molecules. Perforins, granulysin, and NK-lysin are found in NK cell granules, and their expression is increased by exposure to IL-2 and IL-12. NK cell perforin is a protein of 70 to 72 kDa (slightly larger than that produced by T cells). It produces small (5 to 7 nm dia) channels in target cell surfaces. Presumably, granzymes are injected into the target cells through these perforin channels. Unlike T and B cells that circulate as resting cells and so require several days to become fully activated, NK cells are “on-call” and can be rapidly activated by IFNs released from virus-infected cells or by IL-12 from stimulated macrophages. As a result, NK cells promptly attack tumors and virus-infected cells. They participate in innate defenses long before antigen-specific primary adaptive responses can be generated [46]. NK cells kill some tumor cells, xenografts, and virus-infected cells. Thus they are active against herpesviruses, influenza, and poxviruses. Some Ly49 molecules on mouse NK cells can also recognize viruses directly so that, for example, they can kill cytomegalovirus-infected cells. NK cells can also kill bacteria such as Staphylococcus aureus, Mycobacteria, and Salmonella typhimurium, protozoa such as Neospora caninum, and some fungi. NK cells can destroy some cultured tumor cells, and there is a positive correlation between this activity in vitro and resistance to tumor cells in vivo. Experimentally, it is possible to increase resistance to tumor growth in vivo by passive transfer of NK cells from a resistant animal. NK cells destroy human leukemia, myeloma, and some sarcoma and carcinoma cells in vitro, and this activity is enhanced by IFN-γ [47,48].

10.4

“Trained” immunity

NK cells increase in numbers in response to stimulation and are removed once invaders have been eliminated. They are adaptable cells however, and some may survive and develop a “memory” [49]. As a result, they can mount a form of secondary response to some antigens. For example, NK cells bearing a Ly49 specific for cytomegalovirus can expand their numbers in response to viral antigens. Thereafter these NK “memory” cells persist in both lymphoid and nonlymphoid tissues for several months. These self-renewing memory cells are reactivated on re-exposure to the viral antigen. Adoptive transfer of these reactivated NK cells leads to a rapid expansion of their numbers and protective immunity to cytomegalovirus. Because NK cells employ multiple activating receptors with different specificities, an NK cell activated through one receptor may well be reactivated through a different receptor. For example, mouse NK cells initially

144

SECTION | 1 Mammalian immunology

activated through Ly49 may be reactivated through NKG2D. This NK “memory” response could be more accurately described as training rather than memory. Thus the innate immune system can adapt and develop what is termed “trained immunity” [50]. This “trained” immunity results from developmental, transcriptional, and transient epigenetic changes that develop in the survivors of infections. Studies have also demonstrated that the progeny of these “trained” mice show enhanced resistance to endotoxin toxicity and protection against some bacterial infections
an example of Lamarckian inheritance! This resistance is associated with differences in DNA methylation and histone modifications in germ cells resulting in inherited improvements in innate lymphoid cell function [51,52]. This trained immunity may play an important role in protecting mammals such as rats and mice that live in relatively dense colonies and who otherwise would be very vulnerable to epizootics.

10.5

Natural killer cell subsets

NK cell subsets have been identified in many species. Some of these subsets may simply represent cells at different stages of development. Some NK cell diversity probably reflects organ-specific populations such as those found in the uterus, liver, or thymus. NK phenotypes may also change with age. In mice, some NK cells express Ly49, whereas others do not. In humans, some subsets differentially express CD56 and CD16. Cells that express both are primarily cytotoxic, whereas those that express CD56 in the absence of CD16 are mainly cytokine producers [53]. This second population predominates in secondary lymphoid organs. In humans, there is also evidence of two NK subsets based on cytokine secretion. NK1 cells produce IFN-γ but almost no IL-4, IL-10, or IL-13. NK2 cells do not secrete IFN-γ but produce IL-13. Another subset of NK cells has a regulatory function, secretes IL-10, and dampens immune responses. It has been suggested that NK cells exposed to moderate levels of IL-12 secrete IFN-γ, but if exposed to very high levels of IL-12, they produce IL-10 that can then suppress T cell activities. In effect, overstimulation turns on a suppressive function. These regulatory NK cells may reduce the severity of virus-induced immunopathology.

10.6

Natural killer T cells

Natural killer T cells (NKT cells) are innate-like T cells that express both NK cell markers and a TCR of limited diversity [54]. NKT cells are found not only in humans but also in horses and pigs, elephants, and guinea pigs. They are not present in ruminants [55]. There are two functionally distinct NKT cell subsets. Type I NKT cells are T cells that express a semi-invariant TCR consisting of an invariant α-chain associated with diverse β-chains. This TCR recognizes lipid, lipopeptide, and glycolipid antigens presented by the MHC class I-like molecule CD1 [56]. CD1 proteins are nonpolymorphic molecules found on antigen-presenting cells. The CD1-lipid presentation system allows the immune system to sense and respond to lipid antigens. It thus complements the MHC-peptide presentation system (Fig. 10.7). Some CD1 proteins present lipid antigens to conventional T cells while others present lipids to NKT cells. Type I NKT cells can promote immunity in response to lipid antigens from infectious agents and tumor cells. They are found in the liver, spleen, blood, and adipose tissue. Most are found in the sinusoids of the liver. They make up about 1% of the mononuclear cells in human blood [57]. In the absence of prior antigenic stimulation, NKT cells respond more rapidly than conventional T cells. They produce proinflammatory cytokines such as IFN-γ and TNF-α as well as anti-inflammatory cytokines such as IL-4 and IL10. As a result, they can modulate immunity to a broad spectrum of infections. NKT cells trigger chemokine and cytokine release, enhance NK function, and promote dendritic cell maturation and B cell responses. NKT cells inhibit the development of Th17 cells and regulate IL-17 production [58]. NKT cells secrete IL-12 that acts on neutrophils to decrease their production of IL-10. They play a role in allergies, antitumor immunity, autoimmunity, and antimicrobial immunity, especially to mycobacteria. Thus they link the T-cell system with the innate NK cell system. Type I NKT cells also express receptors for inflammatory cytokines produced by antigen-presenting cells. They can be activated either in combination with TCR-mediated signals or even without such signals [59]. In pigs, type I NKT cells recognize α-galactosylceramide. Their production is enhanced by IL-2, -15, and -33. Type II NKT cells, in contrast, use an oligoclonal TCR where both α- and β-chains contribute equally to the recognition of a distinct group of lipid antigens. Their functions are predominantly anti-inflammatory, and they recognize diverse hydrophobic antigens. These type II cells also suppress the type I cells, suggesting that they have an immunoregulatory role.

Mammalian innate lymphoid cells Chapter | 10

145

Dendritic cell CD1d Lipid antigen D-GalCer)

IL-12 IL-18

Invariant TCR

IFN-D NKT cell

IFN-J + IL-4 FIGURE 10.7 The Natural killer T/CD1 system. These cells use an invariant T cell receptor to bind lipid antigens presented by the invariant MHC class I molecule CD1.

Two additional NKT subsets have been characterized in mice. They include NKT17 cells that produce IL-17 and NKT10 cells that produce IL-10 [60]. Thus just as in conventional T cells and innate lymphoid cells, NKT cells can differentiate into different functional subclasses based on their cytokine environment.

10.6.1 The CD1 system In mammals, the CD1 and MR1 genes are not part of the MHC locus. They are located on chromosome 1 in humans whereas the MHC is on chromosome 6 [61]. CD1 genes encode a family of glycolipid presenting proteins encoded by multiple gene clusters [61]. In humans, there are five different CD1 isoforms (CD1a
CD1e). As described above, CD1 is a member of the MHC class I family that presents lipid antigens to T cells and NKT cells. They are expressed on dendritic cells, thymocytes, B cells, and Langerhans cells [62]. The number of CD1 genes varies among mammals. It ranges from one in the bottlenose dolphin (Turciops truncatus) and the rat, to five in humans, to ten in rabbits, to 16 in horses, to 26 in the microbat Myotis lucifugis [61]. Rabbits have two CD1A, two CD1B, and one each of CD1D and CD1E. Guinea pigs have four functional CD1B, three for CD1C, one for CD1E, and at least five CD1 pseudogenes [61]. Cattle possess 12 CD1 genes. Most of these are nonfunctional pseudogenes [55]. However, two CD1A and three CD1B genes appear to be functional [63]. Cattle also express a CD1D gene that uses an alternative start codon. They also possess the genes required to generate functional NKT cells. However, α-GalCer immunization of cattle does not trigger an immune response against this glycolipid which raises the question as to whether bovine NKT cells are functional [62]. Horses are also unresponsive to α-GalCer [64]. Equine T cells recognize the cell wall lipids of Rhodococcus equi when they bind to CD1. The equine genome contains 13 expressed CD1 genes, which makes it the largest CD1 family yet recognized in mammals. All but one of their products are expressed on equine antigen-presenting cells. The polymorphic sites on these proteins are located in their antigen-binding sites suggesting that horse NKT cells can recognize many different lipid antigens. It is possible that this unusually large number of CD1 molecules reflects their key role in protection against R. equi a major pathogen of foals. Other equine cells that express CD1 include dendritic cells, Kupffer cells, endothelial cells, and hepatocytes [64]. In marsupials such as the bandicoot (Isoodon macrourus) or the opossum (Monodelphis domestica) CD1 is encoded by a single-copy gene. It is not orthologous with the eutherian CD1 isoforms (These separated by duplication 170
180 mya long after the emergence of the marsupials) [65]. The bandicoot gene is actively transcribed, but the opossum gene is a pseudogene and so not expressed anywhere.

146

SECTION | 1 Mammalian immunology

10.6.2 The MR1 system Mucosal-associated invariant T cells (MAIT) express an invariant TCR that recognizes nonpeptide microbial antigens presented by a nonpolymorphic MHC molecule, MR1. These cells specifically recognize 5-amino-ribityl-uracil a microbial precursor of riboflavin. The antigen-binding site is small and surrounded by aromatic and basic residues. As a result, so are its ligands. This precursor molecule is a common feature of many bacteria. MAIT cells are activated by most, but not all bacteria. MR1and its ligand are recognized by a TCR with an invariant α chain. This TRAV gene (TRAV41), is highly conserved and is absent from those species that lack MR1 [66]. MR1 is expressed ubiquitously on multiple types of cells. In some cells such as B cells, it is expressed within the cytosol rather than on the cell surface. Unlike MHC and TCR genes that appeared simultaneously during evolution and are present in all existing jawed vertebrates, CD1 and MR1 have much more limited distribution. The riboflavinderived ligands bound to MR1 are responsible for the activation of MAIT cells (Chapter 2) [67]. They thus express invariant T cell receptors. The MR1 gene is located within the MHC on chromosome 1 in humans and on chromosome 6 in mice. It appears to be functional in all mammals investigated so far. MR1 likely appeared about 170 mya and has evolved much more slowly than other MHC class Ib molecules [66]. Studies in other eutherians as well as in marsupials showed that the MR1 gene is exceptionally well conserved in the α1 and α2 domains resulting in a conserved antigenbinding groove. Presumably, this is directed against a highly conserved microbial antigen [68]. However, the MR1 gene has been pseudogenized or otherwise lost in carnivores, (both feliformes and caniformes), lagomorphs, and the armadillo but not in sloths [66,68,69]. In the rabbit, which has few NKT cells and no MR1, there appears to have evolved another TRA invariant alpha chain and an alternative ligand, called MH1.

References [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21]

Vivier E, van de Pavert SA, Cooper MD, Belz GT. The evolution of innate lymphoid cells. Nature Immunol 2016;17(7):790
4. Tizard IR. Veterinary Immunology: An Introduction. St Louis: Elsevier; 2017. Artis D, Spits H. The biology of innate lymphoid cells. Nature 2015;517:293
301. Eberl G, Colonna M, Di Santo JP, Mckenzie AN. Innate lymphoid cells: a new paradigm in immunology. Science 2015;34:aaa6566. Spits H, Di, Santo JP. The expanding family of innate lymphoid cells: regulators and effectors of immunity and tissue remodeling. Nat Immunol 2011;12:21
7. Annunziato F, Romagnani C, Romagagni S. The 3 major types of innate and adaptive cell-mediated effector immunity. J Allergy Clin Immunol 2015;135:626
35. Cording S, Medvedovic J, Aychek T, Eberl G. Innate lymphoid cells in defense, immunopathology and immunotherapy. Nat Immunol 2006;17:755
7. Simoni Y, Fehlings M, Kløverpris HN, McGovern N, et al. Human innate lymphoid cell subsets possess tissue-type based heterogeneity in phenotype and frequency. Immunity 2017;46:148
61. Gasteiger G, D’Osualdo A, Schubert DA, Weber A, et al. Cellular innate immunity: An old game with new players. J Innate Immunol 2017;9:111
25. Saenz SA, Noti M, Artis D. Innate immune cell populations function as initiators and effectors in Th2 cytokine responses. Trends Immunol 2010;31:407
13. Belz GT. ILC2s masquerade as ILC1s to drive chronic disease. Nat Immunol 2016;17:611
12. Colonna M. Interleukin-22-producing natural killer cells and lymphoid tissue inducer-like cells in mucosal immunity. Immunity 2009;31:15
23. Lane PJL, McConnell FM, Withers D, Gaspal F, et al. Lymphoid tissue inducer cells: bridges between the ancient innate and the modern adaptive immune systems. Nature Immunol 2009;2(6):472
7. Boysen P, Storset AK. Bovine natural killer cells. Vet Immunol Immunopathol 2009;130:163
77. Michael HT, Ito D, McCullar V, Zhang B, et al. Isolation and characterization of canine natural killer cells. Vet Immunol Immunopathol 2013;155:211
17. Mair KH, Stadler M, Saalmueller A, Gerner W. Porcine plasmacytoid dendritic cells are unique in their expression of a functional NKp46 receptor. Front Immunol 2022;. Available from: https://doi.org/10.3389/fimmu.2022.822258. Carrillo-Bustamante P, Kesmir C, De Boer RJ. The evolution of natural killer cell receptors. Immunogenetics 2016;68:3
18. Yoder JA, Litman GW. The phylogenetic origins of natural killer receptors and recognition: relationships, possibilities and realities. Immunogenetics 2011;63(3):123
41. Martin AM, Kulski JK, Witt C, Pontarotti P, Christiansen FT. Leukocyte Ig-like receptor complex (LRC) in mice and men. Trends Immunol 2002;23(2):81
8. Guselnikov SV, Taranin AV. Unravelling the LRC evolution in mammals:IGSF1 and A1BG provide the keys. Genome Biol Evol 2019;11 (6):1602
17. Middleton D, Gonzalez F. The extensive polymorphism of KIR genes. Immunology 2010;129:8
19.

Mammalian innate lymphoid cells Chapter | 10

147

[22] Parham P, Norman PJ, Abi-Rached L, Guethlein LA. Human-specific evolution of killer cell immunoglobulin-like receptor recognition of major histocompatibility complex class I molecules. Phil Trans R Soc B 2012;367:800
11. [23] Parham P. Immunogenetics of killer cell immunoglobulin-like receptors. Mol Immunol 2005;42:459
62. [24] Bruijnesteign J, de Groot NG, Bontrop RE. The genetic mechanisms driving diversification of the KIR gene cluster in primates. Front Immunol 2020;. Available from: https://doi.org/10.3389/fimmu.2020.5828094. [25] Abi-Rached L, Parham P. Natural selection drives recurrent formation of activating killer cell immunoglobulin-like receptor and Ly49 from inhibitory homologues. J Exp Med 2005;201(8):1319
32. [26] Guethlein LA, Abi-Rached L, Hammond JA, Parham P. The expanded cattle KIR genes are orthologous to the conserved single copy KIR3DX1 gene of primates. Immunogenetics 2007;59:517
22. [27] Kumar S. Natural killer cell cytotoxicity and its regulation by inhibitory receptors. Immunology 2018;154:383
93. [28] Kelley J, Walter L, Trowsdale J. Comparative genomics of natural killer cell receptor gene clusters. PLoS Genet 2005;. Available from: https:// doi.org/10.1371/journal.pgen.0010027. [29] Marffy ALL, McCarthy AJ. Leukocyte immunoglobulin-like receptors on human neutrophils: Modulators of infection and immunity. Front Immunol 2020;. Available from: https://doi.org/10.3389/fimmu.2020.00857. [30] Lebbink RJ, de Ruiter T, Adelmeijer J, Brenkman AB, et al. Collagens are functional, high affinity ligands for the inhibitory immune receptor LAIR-1. J Exp Med 2006;203(6):1419
25. [31] Hao L, Klein J, Nei M. Heterogeneous but conserved natural killer receptor gene complexes m four orders of mammals. Proc Natl Acad Sci USA 2006;103(9):3192
7. [32] Beziat V, Hilton HG, Norman PJ, Traherne JA. Deciphering the killer cell immunoglobulin-like receptor system at super-resolution for naturalkiller and T cell biology. Immunology 2016;150:248
64. [33] Takahashi T, Yawata M, Raudsepp T, Lear TH, et al. Natural killer cell receptors in the horse: evidence for the existence of multiple transcribed Ly49 genes. Eur J Immunol 2004;34(3):773
84. [34] Sutherland CL, Chalupny NJ, Cosman D. The UL16-binding proteins, a novel family of MHC class I-related ligands for NKG2D, activate natural killer cell functions. Immunol Reviews 2001;181:185
92. [35] Lanier LL. On guard
activating NK cell receptors. Nature Immunol 2001;2(1):23
7. [36] Kasahara M, Sutoh Y. Comparative genomics of the NKG2D ligand gene family. Immunol Revs 2015;267:72
87. [37] O’Callahan CA, Cerwenka A, Willcox BE, Lanier LL, Bjorkman PJ. Molecular competition for NKG2D: H60and RAE1 compete unequally for NKG2D with dominance of. Immunity 2001;15(2):201
211. [38] Gonza´lez S, Lo´pez-Soto A, Suarez-Alvarez B, et al. NKG2D ligands: key targets of the immune response. Trends Immunol 2008;29:397
403. [39] Choy MK, Phipps ME. MICA polymorphism: biology and importance in immunity and disease. Trends Mol Med 2010;16:97
106. [40] Grondahl-Rosado C, Boysen P, Johansen GM, Brun-Hansen H, et al. NCR1 is an activating receptor expressed on a subset of canine NK cells. Vet Immunol Immunopathol 2016;177:7
15. [41] Mair KH, Crossman AJ, Wagner B, Babasyan S, et al. The natural cytotoxicity receptor NKp44 (NCR2, CD336) is expressed on the majority of porcine NK cells Ex Vivo without stimulation. Front Immunol 2022;. Available from: https://doi.org/10.3389/fimmu.2022.767530. [42] Hammond JA, Guethlein LA, Abi-Rached L, Moesta K, Parham P. Evolution and survival of marine carnivores did not require a diversity of Killer cell Ig-like receptors or Ly49 NK cell receptors. J Immunol 2009;182:3618
27. [43] Dobromylskyj MJ, Connelley T, Hammpnd JA, Ellis SA. Cattle Ly49 is polymorphic. Immunogenetics 2009;61:789
95. [44] Radjabova V, Mastroeni P, Skjødt K, Zaccone P, et al. TARM1 is a novel LRC-encoded ITAM receptor that co-stimulates proinflammatory cytokine secretion by macrophages and neutrophils. J Immunol 2015;195(7):3149
59. [45] Orange JS. Formation and function of the lytic N[31]. K-cell immunological synapse. Nat Rev Immunol 2008;28:713
25. [46] Vivier E, Tomasello E, Baratin M, et al. Functions of natural killer cells. Nat Immunol 2008;9:503
10. [47] Guillerey C, Huntington ND, Smyth MJ. Targeting natural killer cells in cancer immunotherapy. Nat Immunol 2016;17:1025
36. [48] Zamai L, Ponti C, Mirandola P, et al. NK cells and cancer. J Immunol 2007;178:4011
16. [49] Ugolini S, Vivier E. Immunology: natural killer cells remember. Nature 2009;457:544
5. [50] Netea MG, Joosten LA, Latz E, Mills KH, et al. Trained immunity: a program of innate immune memory in health and disease. Science 2016;352:aaf1098. [51] Katzmarski N, Dominguez-Andres J, Cirovic B, Renieris G, et al. Transmission of trained immunity and heterologous resistance to infections across generations. Nature Immunol 2021;22(11):1382
90. [52] Netea MG, Joosten LA, Van Der Meer JW. Adaptation and memory in innate immunity. Semin Immunol 2006;28:317
18. [53] Kumar V, Delovitch TL. Different subsets of natural killer T cells may vary in their roles in health and disease. Immunology 2014;142:321
36. [54] Edholm E-S, Banach M, Robert J. Evolution of innate-like T cells and their selection by MHC class I-like molecules. Immunogenetics 2016;68:525
36. [55] Van Beeck FAL, Reinink P, Hemsen R, et al. Functional CD1d and/or NKT cell invariant chain transcript in horse, pig, African elephant and guinea pig, but not in ruminants. Mol Immunol 2009;46:1424
31. [56] Leadbetter EA, Brigl M, Illarionov P, et al. NKT cells provide lipid antigen-specific cognate help for B cells. Proc Natl Acad Sci U S A 2008;105:8339
44. [57] Godfrey DI, Stankovic S, Baxter AG. Raising the NKT cell family. Nat Immunol 2010;11:197
206.

148

SECTION | 1 Mammalian immunology

[58] Mars LT, Araujo L, Kerschen P, et al. Invariant NKT cells inhibit development of the Th17 lineage. Proc Natl Acad Sci U S A 2009;106:6238
43. [59] Zajonc DM, Flajnik MF. CD1, MR1, NKT and MAIT: evolution and origins of non-peptidic antigen recognition by T lymphocytes. Immunogenetics 2016;68(8):489
90. [60] Kumar S, Suryadevara N, Hill TM, Bezbradica JS, et al. Natural killer T cells: an ecological evolutionary developmental biology perspective. Front Immunol 2017;6. Available from: https://doi.org/10.3389/2017/01858. [61] Reinink P, van Rhijn. Mammalian CD1 and MR1 genes. Immunogenetics 2016;. Available from: https://doi.org/10.1007/s00251-016-0926-x. [62] Zajonc DM. The CD1 family: serving lipid antigens to T cells since the Mesozoic era. Immunogenetics 2016;68(8):561
76. [63] Nguyen TKH, Reinink P, El Messiaki C, Im JS, et al. Expression profiles of bovine CD1 in vivo and assessment of the specificities of the antibovine CD1 antibodies. PlosOne 2015;. Available from: https://doi.org/10.1371/journal.pone.0121923. [64] Dossa RG, Alperin DC, Garzon D, Mealey RH, et al. In contrast to other species, α-galactosylceramide (α-GalCer) is not an immunostimulatory NKT cell agonist in horses. Dev Comp Immunol 2015;49:49
58. [65] Baker ML, Miller RD. Evolution of mammalian CD1: marsupial CD1 is not orthologous to the eutherian isoforms and is a pseudogene in the opossum Monodelphis domestica. Immunology 2007;121:113
21. [66] Boudinot P, Mondot S, Jouneau L, Teyton L, et al. Restricting nonclassical MHC genes co evolve with TRAV genes used by innate-like T cells in mammals. Proc Natl Acad Sci USA 2016;. Available from: https://doi.org/10.1073/pnas.16006741132. [67] Krovi SH, Gapin L. Structure and function of the non-classical major histocompatibility complex molecule MR1. Immunogenetics 2016;68 (8):549
59. [68] Tsukamoto K, Deakin JE, Marshall Graves JA, Hashimoto K. Exceptionally high conservation of the MHC class I-related gene, MR1 among mammals. Immunogenetics 2013;65:115
24. [69] Mondot S, Boudinot P, Lantz O. MAIT, MR1, microbes and riboflavin: a paradigm for the co-evolution of invariant TCRs and restricting MHC1-like molecules? Immunogenetics 2016;68:537
48.

Chapter 11

The mammalian lymphoid system Immune responses must be carefully controlled. Lymphocytes must be selected so that their receptors will only bind foreign antigens, and the response of each lymphocyte must be regulated so that it is sufficient but not excessive for the body’s requirements. Mammalian lymphoid organs may therefore be classified on the basis of their roles in generating lymphocytes, regulating the production of lymphocytes, and providing an environment optimized for the capture of foreign invaders, processing them, and maximizing the opportunity for lymphocytes to respond to these foreign antigens.

11.1

Sources of lymphocytes

Lymphoid stem cells first appear in the fetal omentum, liver, and yolk sac of the developing mammal. In older fetuses and adults, these stem cells are mainly found within the bone marrow. The bone marrow has multiple functions in adult mammals. It is a hematopoietic organ containing the precursors of all blood cells, including lymphocytes. In some mammals, such as primates, it is also a primary lymphoid organ (a site where newly produced lymphocytes can mature). Like the spleen, liver, and lymph nodes, the bone marrow is also a secondary lymphoid organ. It contains many dendritic cells and macrophages and effectively removes circulating foreign material from the bloodstream. It contains large numbers of antibody-producing plasma cells. Because of these multiple functions, the bone marrow is divided into a hematopoietic compartment and a vascular compartment. The hematopoietic compartment contains stem cells for all the blood cells as well as macrophages, dendritic cells, and lymphocytes, and is enclosed by a layer of adventitial cells. In older animals, these adventitial cells may become loaded with fat. The vascular compartment, where antigens are mainly trapped, consists of blood sinuses lined by endothelial cells and crossed by a network of reticular cells and macrophages [1].

11.1.1 Lymphoid tissue inducer cells Lymphoid tissue inducer (LTi) cells are a type of innate lymphoid cell that plays an important role during the early stages of development of the immune system in fetal and adult mice. They have a similar phenotype to the Group 3 innate lymphocytes. That is, they use the transcription factor RORγT, and express OX40L, and the IL-7 receptor IL7Rα on their surface. They lack other lineage markers such as CD3 or CD1 although some are CD41. They are among the first cells to colonize the developing lymph nodes. These cells can induce Peyer’s patch and lymph node formation in mice. They also induce the expression of the autoimmune regulator AIRE in thymic epithelial cells. LTi cells enable memory cells to survive within the secondary lymphoid organs. They may also play a role in the development of tertiary lymphoid tissues [2]. It has been suggested that the LTis originated in cryptopatch-like structures in the intestinal mucosa but as the immune system evolved, so too did the LTi cells [3].

11.2

Primary lymphoid organs

The organs that regulate the development of lymphocytes are considered primary lymphoid organs. T cells mature in the thymus. B cells, in contrast, mature within several different organs depending on the species. These include the bone marrow in primates and rodents, and the intestinal lymphoid tissues in rabbits, and ruminants. These primary lymphoid organs develop early in fetal life. As the fetus develops, newly produced, immature lymphocytes migrate from the bone marrow to the primary lymphoid organs where they mature. Most primary lymphoid organs are not sites where lymphocytes encounter foreign antigens, and they usually do not enlarge in response to antigenic stimulation or microbial invasion. Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00014-9 © 2023 Elsevier Inc. All rights reserved.

149

150

11.3

SECTION |1 Mammalian immunology

Thymus

The thymus is the first lymphoid organ to form in the developing fetus. It grows rapidly immediately after birth in order to meet the demand for T cells. The thymus is located in the thoracic cavity in front of, and below the heart. In species such as horses, cattle, sheep, and pigs, it also extends up the neck as far as the thyroid gland. The size of the thymus varies; its relative size is greatest in the newborn animal and its absolute size is greatest before puberty. It may be very small and difficult to find in adult animals. The thoracic thymus arises from the third and fourth pharyngeal pouches from which the parathyroids also originate. Aberrant thymic tissue is often found in the ventral neck region of some mammals where it can be associated with the parathyroid glands. Likewise, accessory parathyroid glands can sometimes be found within the thoracic thymus and the mediastinum. This is probably a result of errors in cleavage within the developing third pharyngeal pouch. In humans, the aberrant parathyroid tissue is encapsulated and in rats, it is found in the connective tissue septa. These anatomical differences may be explained by a placode-like structure consisting of thymus-specific precursors that can organize and develop into a functionally complete thymopoietic unit. Depending on the extent of tissue rearrangements that occur in the developing thymus these can either be dispersed or aggregated into tissues [4].

11.3.1 Structure The thymus has the most consistent structure of all the lymphoid organs among mammals. It consists of lobules of loosely packed epithelial cells, each enclosed by a connective tissue capsule. It is covered by a thin connective tissue capsule that penetrates the thymic tissue and divides it into well-demarcated lobules. These are very obvious in larger mammals but less so in small rodents. The outer part of each lobule, the cortex, is densely infiltrated with dark staining lymphocytes (or thymocytes), but the paler inner medulla contains fewer lymphocytes, and the epithelial cells are clearly visible (Fig. 11.1). The cortex is continuous between adjacent lobules. Within the medulla are also found round, eosinophilic, layered bodies called Hassall’s corpuscles. They are named after Arthur Hill Hassall who described them in 1846. These are concentric whorls of flattened, terminally differentiated epithelioid reticular cells. They contain keratin, and the remains of a small blood vessel may be found at their center. They form from thymic epithelial cells after they cease expressing the autoimmune regulator, AIRE. The Hassall’s corpuscle cells show evidence of senescence [5]. They produce inflammatory cytokines and chemokines such as CXCL5 and as a result recruit and activate neutrophils to produce IL-23 within the thymic medulla. The IL-23 then acts on pDCs to produce significant quantities of IFN-α [6]. Thus the aging Hassall’s corpuscles play a key role in cell activation within the thymus. This interferon may also serve to exclude viruses and so prevent tolerance developing to these invaders. They are also a significant source of thymic stromal lymphopoietin (TSLP) that can convert thymic lymphocytes into Treg cells. In cattle, these corpuscles may contain immunoglobulin A. Keratinization of the Hassall’s corpuscles is especially evident in dogs, pigs, and nonhuman primates. Hassall’s corpuscles are rare in rats and mice when compared to humans and primates. In the mouse, they are very small and difficult to see without immunostaining [7].

FIGURE 11.1 A section of thymus obtained from a young rhesus macaque. Each lobule is divided into a cortex rich in lymphocytes, hence staining darkly, and a paler medulla consisting mainly of epithelial cells. Original magnification 3 10. The demarcation between the cells densely packed in the cortex and the much lower cell density in the medulla is obvious.

The mammalian lymphoid system Chapter | 11

151

An abnormally thick basement membrane and a continuous layer of epithelial cells surround the capillaries that supply the thymic cortex. This barrier prevents circulating foreign antigens from entering the cortex. No lymphatic vessels leave the thymus.

11.3.2 Function The results of thymectomy in mice indicate that the neonatal thymus is the source of most blood lymphocytes and that these lymphocytes—T cells, are mainly responsible for mounting cell-mediated immune responses. T-cell precursors originate in the bone marrow but then enter the thymus. Once within the thymus, the cells divide rapidly. Of the new cells produced, most die by apoptosis, whereas the survivors (about 5% of the total in rodents and about 25% in calves) remain in the thymus for four to five days before emigrating and colonizing the secondary lymphoid organs. T cells that enter the thymus have two conflicting tasks. They must recognize foreign antigens but at the same time must not respond strongly to normal body constituents (self-antigens). A two-stage selection process in the thymic medulla accomplishes this feat. Thus thymocytes with receptors that bind self-antigens strongly and that could therefore cause autoimmunity are programmed to die by apoptosis. Thymocytes with receptors that cannot bind any major histocompatibility complex (MHC) class II molecules and thus cannot react to any processed antigen are also eliminated [8]. On the other hand, those thymocytes that survive this “negative selection” process but can still recognize specific MHC class II
antigen complexes with moderate affinity are stimulated to grow—a process called positive selection. These surviving cells eventually leave the thymus as mature T cells, circulate in the bloodstream, and colonize the secondary lymphoid organs. Thymic epithelial cells are unusual since they each express hundreds of antigens normally expressed in other tissues [9]. In addition, these cells have a very high level of autophagy. As a result, their intracellular antigens are processed, bound to MHC class II molecules, and expressed in large amounts on the epithelial cell surfaces. This “promiscuous” antigen presentation ensures that developing thymocytes are exposed to an unusually diverse array of normal tissue antigens. Since T cells with receptors that bind and respond to these antigens will die, the system ensures that those T cells leaving the thymus are unreactive to most self-antigens and as a result, do not respond to normal body components.

11.3.3 Thymic hormones Within the thymus, cells are regulated by a complex mixture of cytokines and small peptides collectively known as thymic hormones. These peptides include thymosin, thymopoietin, thymic humoral factor, and thymulin. Thymulin is a zinc-containing peptide secreted by the thymic epithelial cells, and it can partially restore T cell function in thymectomized animals. Zinc is an essential mineral for the development of T cells. Consequently, zinc-deficient animals have defective cell-mediated immune responses. Hassall’s corpuscles play a functional role in regulating thymic activity since they produce TSLP. TSLP activates thymic dendritic cells that can stimulate regulatory T cells and so controls the positive selection process.

11.3.4 Thymic involution The thymus shrinks with age. While thymic involution is associated with increasing age, it is not a form of immunosenescence. Thus in many mammals, involution begins at a young age, and in humans may commence by one year of age. During the neonatal period, however, the thymus increases in absolute size as it supplies the developing animal with a fresh, carefully selected, supply of new T cells. In humans, the thymus epithelial space starts decreasing from the first year of life at a rate of 3% annually until around age 40. Thereafter it decreases by about 1% annually until death [10]. The thymus of humans over 40 cannot rebuild a new T cell compartment. As a result of this involution, naive T cell output drops steadily and by around age 65, there is a marked drop in T cell diversity. Acute involution can also be induced in many mammals by stresses including malnutrition, and infectious diseases. Involution during pregnancy is associated with progesterone-mediated downregulation of chemokine production [11]. It also decreases in size during hibernation [12]. The aging of the thymus is readily observed in larger mammals as it atrophies. By age 30 the human thymus mainly consists of adipose and connective tissues [13]. The functional thymic tissue shrinks to be replaced by adipocytes, stromal cells, epithelial cords, and tubules, especially in the medulla. In most species, this involution becomes most apparent at the onset of sexual maturity and is presumably driven at this stage by hormonal influences. As the thymic lymphocytes undergo apoptosis, phagocytic cells remove the cell debris, and increased numbers of macrophages appear

152

SECTION |1 Mammalian immunology

to invade the structure [14]. There are some species differences in this process with female rats developing more prominent epithelial cell infiltration than male rats. Likewise, in dogs, the process of thymic evolution is very variable. so that in dogs 9
12 months of age there is a great diversity of thymic sizes and weights. Likewise, nonhuman primates vary greatly depending on whether they are wild-caught (stressed) or bred in captivity. Cystic medullary remnants are frequently present in Cynomolgus macaques (Macaca fasicularis). Some thymic areas are free of epithelium where lymphocytes accumulate under the capsule. Other age-related changes include cystic dilation of Hassall’s corpuscles and the accumulation of cellular debris. This is commonly seen in rats and dogs. However, the process is progressive and inexorable [14].

11.3.5 Species differences In general, the histologic structure of the thymus is well conserved among mammals. The larger mammals such as humans, other primates, and dogs have a bilobed thymus located within the anterior thoracic cavity, cranial to the heart. Some kangaroos have two cervical lobes as well as a thoracic lobe. In smaller mammals such as rats, there may be an extension of one or both lobes out of the thoracic cavity into the cervical region. In guinea pigs, the thymus is located more anteriorly while in the mouse there are both cervical and thoracic lobes. The thymus in pigs is also located in the thorax above the pericardium but it too extends to the thoracic inlet. The two lobes however extend into the cervical region along the carotid artery to the pharynx—a cervical thymus [15].

11.4

Peyer’s patches

11.4.1 Structure Peyer’s patches (PPs) are ovoid lymphoid organs located in the mucosa along the antimesenteric walls of the small intestine from the jejunum to the ileum. They are named after Johann Conrad Peyer who wrote the first detailed description of them in 1677. Their structure and functions vary greatly. For example, mammals can be divided into two groups based on the timing of Peyer’s patch appearance. In group one, they develop well below birth. In group two they develop postnatally [16]. They can also be functionally divided by location with the ileal PP having a somewhat different function than the jejunal PP.

11.4.1.1 Group one species In group one mammals, the ileal PP develops and reaches maximal size and maturity well before birth and microbial colonization and at a time when it is still shielded from foreign antigens. In ruminants, pigs, horses, dogs, and humans, 80% to 90% of the PPs are found in the ileum, where they form a single continuous structure that extends forward from the ileocecal junction. In young ruminants and pigs, this large ileal PP may be as long as two meters. It is the largest lymphoid tissue in six-week-old lambs. It disappears by 15 months of age and cannot be detected in adult sheep. The ileal PP consists of densely packed lymphoid follicles, each separated by a connective tissue sheath, and contains only B cells. These B cells proliferate rapidly within this large lymphoid structure but the majority die as a result of apoptosis [17]. It is believed that they undergo negative selection just like developing T cells in the thymus. Only about 5% survive. These survivors emigrate from the ileal PP and colonize the other secondary lymphoid organs [18]. The group I species also have a second type of PP that consists of multiple discrete accumulations of follicles in the jejunum. While the ileal PP is involuted in adolescence, these jejunal PPs persist for the life of the animal. They consist of pear-shaped lymphoid follicles separated by extensive interfollicular tissue and contain mainly B cells with up to 30% T cells. These jejunal PP are clearly secondary lymphoid organs in sheep and cattle (Fig. 11.2). The ileal PPs of group I species such as the sheep, function in a manner similar to the avian bursa. Thus the prenatal maturation of PP in lambs occurs in the absence of any obvious antigenic stimulus. Primordial PP can be first detected in the small intestine of the fetal lamb at about 60 days gestation. Lymphoid follicles develop by 75 days gestation, and it is clear that significant lymphopoiesis is occurring by 100 days gestation. From then until birth at around 150 days, the PP follicles mature and have the greatest concentration of dividing lymphocytes in the body. They also greatly increase their antigen receptor diversity [19]. By the time of birth, lambs have developed 25
40 PP in their jejunum and proximal ileum together with the single continuous PP in the terminal ileum. Birth does not result in any apparent change in the rate of growth of the PP that could be related to the establishment of the microbiota [20]. These PP tissues in lambs are larger than any other single lymphoid tissue amounting to about 1.2% of their body weight. The large ileocecal PP extends 2.5 m along the terminal ileum and accounts for 90% of the total PP mass.

The mammalian lymphoid system Chapter | 11

153

FIGURE 11.2 Histological sections of the jejunal Peyer’s patches (upper panel) and ileal Peyer’s patch (lower panel) of a 7-day-old piglet (left panel) and a 21-day-old artificially reared piglet (right panel). At the antimesenteric side there are many primary and secondary lymphoid follicles within the submucosa. They may penetrate the mucosal layer to form a dome rather than a villus. Note that the ileal Peyer’s patches occupy more than half of the gut circumference. From Prims S, et al. The porcine tonsils and Peyer’s patches: a stereological morphometric analysis in conventionally and artificially reared piglets. Vet Immunol Immunopathol 2018;206:9
15. With Permission.

Lamb ileal PPs begin to involute about 12 weeks after birth and by 18 months only a few isolated follicles remain. The PP elsewhere in the intestine persist and remain functional. It appears therefore that the ileal PP of lambs is a primary lymphoid organ with a similar function to the avian bursa of Fabricius. Thus ileal PPs are sites of rapid B cell proliferation, although most cells then undergo apoptosis, and the selected survivors are released into the circulation [21]. If their ileal PPs are surgically removed, lambs become B cell-deficient and hypogammaglobulinemic. The bone marrow of lambs contains many fewer lymphocytes than the bone marrow of laboratory rodents, and the ileal PPs are therefore their most significant source of new B cells. The pig is also a group one species but its PPs do not appear to be primary lymphoid organs. Pigs have about 30 jejunal PPs of conventional structure in addition, to a single, large ileal PP. Their ileal PP lacks T cells and has a structure similar to that seen in lambs. It also regresses within the first year of life. However, surgical removal indicates that it is not required for B cell development. It appears to be a secondary lymphoid organ that plays a role in the immune response to the intestinal microbiota. The single ileal PP involutes early in life and contains predominantly immature B cells. Microbial exposure is not required for Peyer’s patch growth in the developing piglet. The number of follicles in piglets may reach as many as 75,000 by one month of age [14]. These are mainly found in the ileum. They regress in older animals eventually leaving just a few scattered follicles.

11.4.1.2 Group two species In primates, rabbits, and rodents (Group 2), the PPs are located at random intervals in the ileum and jejunum. In these mammals, PPs do not develop until two to four weeks after birth, and they persist into old age. The development of the PPs in some of these species appears to depend entirely on stimulation by the intestinal microbiota since they remain small and poorly developed in isolator-raised germ-free animals (Fig. 5.1) [14]. The proximal PP in dogs, humans, and rodents have a similar structure. In humans, they increase in size and number until puberty but then begin to regress. The terminal ileal PP in humans may contain 900
1000 individual follicles at maturity. The sacculus rotundus is a hypertrophied PP that encircles the terminal ileum in rabbits. Both humans and rabbits possess an appendix as well as an aggregation of lymphoid nodules at the ileocecal valve. There is no appendix in

154

SECTION |1 Mammalian immunology

rats and mice. About 50% of the cells overlying PPs in the rabbit are M cells but M cells constitute only 5%
10% of the epithelial cap in rats and humans. In most mammals, large numbers of lymphocytes are scattered throughout the lamina propria. In addition, there are intraepithelial lymphocytes, cryptopatches in the small intestine, and lymphoglandular complexes in the colon. The lamina propria lymphocytes have been calculated to amount to about the same mass as the spleen. The PP are a major source of IgA in rabbits and rodents. There are very few IgA-producing cells in the PP of humans and rats. The germinal centers in PPs develop public clonotypes based on signals from the microbiota, especially bacterial glycans, although others are independent of the gut bacteria [22]. Thus persistent gut antigens stimulate ongoing PP B cell responses. Histologically PPs are separated into a follicular area, an interfollicular area, and the follicle-associated epithelium. The follicular and interfollicular areas consist of lymphoid follicles with a germinal center containing proliferating B cells, follicular dendritic cells, and macrophages The follicle is surrounded by a corona or a subepithelial dome containing a mixed cell population, PPs are connected to the vascular system by endothelial venules and the lymphatic system by lymphatic vessels. On the serosal side of the PPs, these lymphatics connect to the mesenteric lymph nodes. The crosstalk between the PPs and the microbiota appears to be mediated through the NOD2 pattern recognition receptors that recognize the bacterial PAMPs. NOD2 regulates the number, size, and T-cell composition within PPs as a result of signals it receives from the gut microbiota [23]. It has been proposed that the rabbit appendix is also a primary lymphoid organ [24]. Thus the appendix in the young rabbit does not require the presence of the thymus to develop. The appendix B cells undergo IGHV gene rearrangements by gene conversion and somatic hypermutation. These newly generated B cells undergo both positive and negative selection. They then emigrate and populate secondary lymphoid organs. Rabbit appendix B cells, like those in their ileal PP, are exposed to diverse microbial antigens and superantigens that can promote both antigen-specific and nonspecific proliferation.

11.5

Bone marrow

The bone marrow has similar functions in different mammalian species. However, its location may differ. Thus rodents have plentiful marrow tissue in the femoral head, femoral shaft, sternum, and ribs. Dogs and nonhuman primates have little hematopoietic tissue in the femoral head and distal shaft. The specialized ileal PP is the primary lymphoid organ for B cells only in some group 1 mammals such as ruminants. In group 2 mammals the bone marrow probably serves this function. There is no exclusive B cell development site in the bone marrow, although it is suggested that precursor B cells develop at the outer edge of the marrow and migrate to the center as they mature and multiply. Negative selection occurs within the bone marrow so that, as in other primary lymphoid organs, any self-reactive pre-B cells will undergo apoptosis.

11.6

Secondary lymphoid organs

The cells of the immune system must be able to respond to a huge diversity of microbial invaders and potential pathogens. It is especially important that antigen-specific lymphocytes can encounter their target antigens. To maximize the probability of such encounters, the body employs secondary lymphoid organs. In contrast to the primary lymphoid organs, the secondary lymphoid organs arise late in fetal life and persist in adults. Unlike primary lymphoid organs, they enlarge in response to antigenic stimulation. Likewise, surgical removal of one of them does not significantly reduce immune capability. Examples of secondary lymphoid organs include the spleen, the lymph nodes, the tonsils, and other lymphoid tissues in the intestinal, respiratory, and urogenital tracts. These organs contain dendritic cells that trap and process antigens and lymphocytes that mediate the immune responses. The overall anatomical structure of these organs is therefore designed to facilitate antigen trapping and to provide the optimal environment for the initiation and coordination of adaptive immune responses. Secondary lymphoid organs are connected to both the blood and lymphatic systems, thus allowing them to capture and eliminate circulating particulate antigens [1].

11.7

Spleen

The spleen is the largest single secondary lymphoid organ. It contains two distinct forms of tissue—the red pulp used for blood storage and the white pulp that serves as a secondary lymphoid organ. It is enclosed in a fibromuscular connective tissue outer capsule, that extends into the spleen in the form of muscular trabeculae.

The mammalian lymphoid system Chapter | 11

155

Unlike lymph nodes, the spleen is present in all vertebrates whereas the lymph nodes are found only in mammals [4]. It is thus the most ancient of the lymphoid organs. The spleen has two roles. One is the removal of damaged or aged red blood cells which likely predates its immune functions. The other, in some mammals, is a hematopoietic organ and a source of red and white blood cells. It develops from condensation of mesenchyme within the dorsal mesentery (the mesogastrium). After a mesenchymal core forms, the spleen is sequentially colonized with populations of lymphoid cells that eventually develop into the white pulp. An additional advance occurred with subsequent colonization by macrophages to form an antigen-trapping marginal zone. One classification of spleens from different mammals is to rate them as defensive, or storage, or some combination of the two. This of course reflects the dual functions of this organ as both a filter for antigens in the bloodstream and as a blood storage organ. In some mammals, the spleen is also an important hematopoietic organ and a site for the production of B cells and monocytes [25]. In other mammals, it may also be a site where B cells complete their final maturation that initially begins in the bone marrow.

11.7.1 Red pulp In general, spleens can be considered either as primarily blood storage organs such as those in ungulates and carnivores or as primarily defensive spleens where the white pulp predominates as seen in primates and most rodents [26]. Mammalian spleens constitute a graded series between these two extreme types. At one extreme is the storage type while at the other extreme is the defense type. Most species have transitional spleens that fall somewhere in-between. Flight and stress are well recognized as triggers of splenic contraction. Thus it is possible that domestication, making flight less necessary may have affected spleen morphology. Just as lymph nodes filter antigens from lymph, so the spleen filters blood-borne pathogens and antigens [27]. Indeed, the spleen can be considered a specialized trap for blood-borne antigens. The filtering process removes antigenic particles such as bacteria, cellular debris, and aged blood cells. This filtering function, together with highly organized lymphoid tissue, makes the spleen an important component of the immune system. In addition, to its immune functions, the spleen also stores red cells and platelets, recycles iron, and undertakes red cell production in the fetus [28]. As a result, the spleen consists of two forms of tissue. The tissue used predominantly for blood filtering and red cell storage is the red pulp. It contains large numbers of antigen-presenting cells, lymphocytes, and plasma cells. Macrophages in the red pulp specialize in removing aged red blood cells and so regulate iron recycling. The splenic red pulp also consists of splenic cords and sinuses. It has major metabolic functions such as the storage of ferritin as well as intact red blood cells and as needed, the rapid release of these red cells [29]. To permit this rapid release, the muscular splenic capsule and trabeculae play a key role. Splenic cords consist of a network of reticular cells that support a population of macrophages. The function of these macrophages is to trap aged red blood cells as well as bloodborne invaders. Phylogenetically the most primitive form of the spleen is found in monotremes, insectivores, and tree shrews and it has a closed circulation. In this type of spleen, the blood passes from the vessels of the white pulp into the sheathed capillaries of the red pulp and then directly into the sinuses. In other words, the capillaries connect with the sinuses. In an open circulation spleen such as that seen in rodents, bats, and marsupials, the blood from the sheathed capillaries drains into the splenic cords and then enters the sinuses through slits in their walls. Humans and dogs have open systems. Ruminants, horses, pigs, and cats have closed systems.

11.7.2 White pulp The other splenic tissue, the white pulp, is rich in both B and T cells and is the site where immune responses occur (Fig. 11.3). The white pulp is separated from the red pulp by a region called the marginal zone. This zone contains numerous macrophages and dendritic cells as well as a large population of B cells. The spleen is not supplied with lymphatic fluid, although it does possess efferent lymphatics. The white pulp can be considered to consist of a series of lymph node-like structures embedded in the red pulp [27]. Arteries entering the spleen pass through muscular trabeculae before entering the white pulp and branching into arterioles. Immediately on leaving the trabeculae, each arteriole is surrounded by a layer of lymphoid tissue called the periarteriolar lymphoid sheath. The arteriole eventually leaves this sheath and branches into penicillary arterioles. In some mammals, these penicillary arterioles are surrounded by ellipsoids. Ellipsoids consist of arteriolar sheaths with a discontinuous endothelium surrounded by a meshwork of macrophages and reticular cells. They serve as highly efficient filters for blood-borne particulates and immune complexes [30]. These arterioles then open, either directly or indirectly, into venous sinuses that drain into the splenic venules. Ellipsoids are relatively large and prominent in pigs, mink, dogs,

156

SECTION |1 Mammalian immunology

Arteriole

Periarteriolar lymphoid sheath

Red pulp

B cell follicle

Marginal zone FIGURE 11.3 Histological section and diagram showing the structure of the bovine spleen. Original magnification 3 50. From a specimen provided by Dr. J.R. Duncan.

and cats. They are small and indistinct in horses and cattle; and are absent in rodents such as mice, rats, and guinea pigs, as well as rabbits. In species that lack ellipsoids, particles are trapped primarily in the marginal zone of the white pulp. The white pulp contains both B and T cells, that accumulate in their specific zones under the influence of chemokines. The inner layer of the periarteriolar lymphoid sheaths contains mainly CD41 T cells with smaller numbers of CD81 cells and many dendritic cells. Within the sheaths, the T cells interact with dendritic cells and passing B cells. The outer layers of the sheaths contain a mixture of T cells, B cells, macrophages, and plasma cells. The B cell areas, in contrast, consist of round primary lymphoid follicles scattered through the sheaths, especially where the central arteries branch. These follicles are sites where germinal center formation, clonal expansion, isotype switching, and somatic hypermutation occur [31]. The white pulp is separated from the red pulp by a marginal sinus, a reticulum sheath, and a marginal zone of cells. This marginal zone is a critically important transit area for cells moving between the blood and the white pulp. It is also rich in macrophages, dendritic cells, and B cells. Most of the blood that enters the spleen flows into the marginal sinus and through the marginal zone before returning to the circulation through venous sinuses. This flow pattern ensures that antigen-presenting cells encounter and capture any blood-borne antigens and deliver them to the B cells in the marginal zone. The white pulp is involved in adaptive immune responses, whereas cells of the marginal zone can participate in both innate and adaptive responses. White pulp does not contain high endothelial venules. Instead, lymphocytes enter the white pulp through the marginal zone, although the route by which they leave is unclear.

11.7.3 Function 11.7.3.1 Different types of spleens Mammalian spleens can be classified into four types based on their capsular structure: Thus there are defense spleens with a thin monolayer capsule and trabeculae consisting mainly of connective tissue. This form is found in humans, lagomorphs, and rodents. A second type consists of spleens with a monolayer capsule and trabeculae that consist primarily of smooth muscle. This type is found in Carnivora including dogs, cats, foxes, and lions. A third type consists of

The mammalian lymphoid system Chapter | 11

157

a thicker bilayered capsule and trabeculae rich in smooth muscle cells. It is a feature of perissodactyls,—horses and donkeys. In these species, the trabeculae can constitute as much as 25% of the splenic volume and these are very much storage spleens [29]. Cetaceans also possess this third type of spleen. Thus they are similar to terrestrial artiodactyls, but their spleens are small and cannot store a large volume of blood. This probably evolved in cetaceans from the artiodactyl blood storage spleen. Perhaps they have fewer natural enemies from which to flee! The fourth type is a spleen with a thick bilayered capsule rich in smooth muscle cells together with an additional network of smooth muscle cells in the red pulp. This type is found in elephants, some artiodactyls including the pig, camels, and deer as well as cattle, sheep, and goats [29]. Kangaroos also possess a spleen of this last type, and it is tempting to suggest that as grazers, they have occasion to flee rapidly from predators. This final type is likely to be the most efficient in contracting and rapidly expelling stored red cells. The spleen is a major hematopoietic organ in the mouse but much less so in rats. Humans and rabbits have essentially no hematopoietic activity in their healthy adult spleens. It is suggested that a primitive type of spleen with a closed circulation characteristic of monotremes, insectivores, and tree shrews evolved into a more typical type with an open circulation found in mice, gerbils, bats, and marsupials. It then diversified into storage spleens in ungulates and carnivores and defensive spleens in primates and most rodents [32].

11.7.3.2 Defensive type spleens Defensive type spleens are characterized by a predominance of white pulp consisting of large periarteriolar lymphoid sheaths, amounting to as much as 40% of splenic volume, with many lymphoid follicles and relatively few muscular trabeculae and a thin capsule [29]. This type of spleen is classically found in humans and other primates, rodents, and lagomorphs. The lymphoid follicles are more obvious in rats than in mice. Rats have a distinct marginal sinus that separates the marginal and mantle zones. In humans, the T cell zones are irregularly arranged around the central arteries. The human marginal zone does not contain specialized macrophages, unlike rodents. Defensive type spleens have numerous obvious primary and secondary lymphoid follicles. Nonhuman primates may have irregular coalescing lymphoid follicles, but this may simply reflect infection or other intense antigenic stimulation. Intravenously administered antigens are trapped in the spleen. Depending on the species, they are taken up by dendritic cells in the marginal zone or the periarteriolar macrophage sheaths. These dendritic cells and macrophages carry the antigen to the primary follicles of the white pulp, from which, after a few days, antibody-producing cells migrate. These antibody-producing cells (plasma cells and plasmablasts) colonize the marginal zone and move into the red pulp. Antibodies produced by these cells diffuse rapidly into the bloodstream. Germinal center formation also occurs in the primary follicles. In an animal possessing circulating antibodies, trapping by dendritic cells within these follicles becomes significant. As in a primary immune response, the antibody-producing cells migrate from these follicles into the red pulp and the marginal zone where antibody production occurs. The importance of defense type spleens tends to decrease with age as do other secondary lymphoid organs [25].

11.7.3.3 Storage spleens Storage spleens are characteristically found in dogs and other carnivores that require a reserve of red cells to be released while chasing their prey. They have a thick muscular outer capsule with many well-developed trabeculae made of smooth muscle penetrating the splenic parenchyma. This smooth muscle may contract the spleen so it can store enormous quantities of blood (up to 1/3 of blood volume) and be emptied rapidly. Dog red pulp has prominent venous sinuses. It has a relatively small periarteriolar lymphoid sheath and relatively few nodules [32]. The red pulp and marginal zone of the dog spleen contain ellipsoids and penicillary arteries that are poorly defined in humans [33].

11.7.3.4 Intermediate spleens Many spleens fall between the two extremes described above. This is the case in pigs, ruminants, and horses where they have a thick capsule with a moderate number of trabeculae [14]. The sheathed capillaries emerging from the central red pulp arteries have concentric ellipsoids and these may be both large and numerous. Ellipsoids are also found in dogs but not rats. Pig spleens have poorly developed or absent sinuses. Dogs and rats have a sinusoidal spleen, but mice lack true sinusoidal lining cells. In the pig, there are fewer smaller lymphoid follicles than in rodents but large periarteriolar lymphoid sheaths.

158

SECTION |1 Mammalian immunology

11.7.3.5 Accessory spleens Small, isolated, accessory spleens are occasionally found in the viscera of cynomolgus monkeys and humans and are common in some cetaceans (Fig. 16.3). They may be attached to the pancreas. In the pig, the gastrosplenic ligament may also contain accessory spleens.

11.8

Lymph nodes

11.8.1 Structure The size, shape, and numbers of mammalian lymph nodes are highly variable. Unlike the primary lymphoid organs such as the thymus that emerged with the appearance of the adaptive immune system, the peripheral secondary lymphoid tissues such as lymph nodes coincided, in large part, with the evolution of the mammals. The induction of lymph node tissues requires both IL-7 and the integrin α4β7. Lymph node development also requires the activities of both inducer cells (LTi cells), and organizer cells [34]. The least evolved mammals, the egg-laying monotremes, such as the duck-billed platypus (Ornithorhynchus anatinus) and the echidna (Tachyglossus aculeatus), diverged from the other mammals about 200 mya. They have a spleen, thymus, and gut-associated lymphoid tissues that are as well developed as those in marsupials and eutherian mammals. However, instead of typical mammalian lymph nodes, they possess structures that consist of multiple lymphoid nodules, each containing a germinal center, suspended by its blood vessels within the lumen of a lymphatic plexus. Thus each nodule is bathed in the lymph fluid. There is usually just one germinal center per nodule. While they lack the surrounding region densely packed with lymphocytes and supporting connective tissue they can trap and respond rapidly to any antigens they encounter (Figure 12.2) [35]. Newborn marsupials lack lymph nodes. Within a few days of entering the pouch, however, the lymph node anlagen appear and are concurrently invaded by lymphocytes. The first step in their development involves the production of a specialized reticulum. This is followed by infiltration with specialized supporting tissue—(this stage is absent in the monotremes). As the joey develops, the node is gradually populated with lymphocytes and develops the adult structure (Figure 13-6) [36]. The structure of eutherian lymph nodes depends upon the age of the animal as well as the amount of antigenic stimulation it has been subjected to. The number of lymph nodes is also highly variable [14]. Thus the mouse has approximately 22 nodes arranged in simple chains. However, more are needed in larger mammals and the human has about 450 nodes arranged in more complex chains. The rat lung is drained through just two mediastinal lymph nodes. In contrast, the dog has up to five tracheobronchial nodes while in humans there are 35 tracheobronchial nodes arranged in five separate groupings. These nodes are also connected to many more afferent lymphatic vessels [37]. In the mouse, lymph nodes are round or bean-shaped filters strategically placed on lymphatic vessels in such a way that they can sample antigens carried in the lymph. They are enclosed in a capsule beneath which is a reticular network filled with lymphocytes, macrophages, and dendritic cells and through which lymphatic sinuses penetrate. The lymph node thus acts as a filter for lymph fluid. A subcapsular sinus is located immediately under the connective tissue capsule. Other sinuses pass through the body of the node but are most prominent in the medulla. Afferent lymphatics enter the node around its circumference, and efferent lymphatics leave from a hilus on one side. The blood vessels supplying a lymph node also enter and leave through the hilus. The node itself is essentially a mass of fibrovascular tissue enclosed within a dilated lymphatic vessel [37]. In mammals other than the mouse, there are obvious links between the medullary sinus and the subcapsular sinus that effectively divide the node into lobules. In larger species, these lobules remain about the same size, but their numbers increase. All species develop age-related changes in their nodes. Thus nodes atrophy and are replaced by fibrous connective tissue and fat. Evidence of hematopoietic activity can be detected in rodents but not in primate or dog lymph nodes. The interior of lymph nodes is divided into three regions: a peripheral cortex, a central medulla, and an ill-defined region in between, called the paracortex. B cells predominate in the cortex, where they are arranged follicles. In lymph nodes that have been stimulated by antigen, many of these follicles will contain germinal centers (Fig. 11.4). Germinal centers are round, ovoid clusters of cells divided into a light and dark zone (Fig. 11.5). Germinal centers originate when antigen-specific B cells enter a follicle and then divide rapidly to become the centroblasts that form the dark zone. This is the site where B cells undergo somatic mutation. The centroblasts eventually produce non-dividing centrocytes that migrate to the light zone. The light zone is the site where immunoglobulin class switching, and memory B cell formation occurs. Light zones are rich in antigen-trapping follicular dendritic cells (fDCs) and CD41 T cells [38].

The mammalian lymphoid system Chapter | 11

159

FIGURE 11.4 A section of a bovine lymph node. Note the presence of several germinal centers in the cortex. Original magnification 3 12. From a specimen provided by Dr. W.E. Haensly.

FIGURE 11.5 A germinal center from the cortex of a cat’s lymph node. Note the obvious dark and light zones. B cells undergo somatic hypermutation in the dark zone, while memory B cells develop in the central light zone. Original magnification 12. From a specimen provided by Dr. W.E. Haensly.

T cells and dendritic cells predominate in the paracortex. The cells are arranged in cords between the lymphatic sinuses. At the center of each paracortical cord is a high endothelial venule (HEV). These vessels are lined with tall, rounded endothelial cells quite unlike the flattened endothelium found in other blood vessels (Fig. 11.6). HEVs are surrounded by concentric layers of fibroblastic reticular cells and a narrow space called the perivenular channel. The lymph node medulla contains lymph-draining sinuses separated by medullary cords containing many plasma cells, macrophages, and memory T cells. Lymph nodes are very busy places with cells coming and going in response to a multitude of signals. These signals are delivered through the reticular fibers that provide the structural scaffolding of the lymph node. The fibers are hollow and serve as conduits for the rapid transmission of signaling molecules [39]. The conduits consist of bundles of collagen fibers ensheathed by fibroreticular cells. The fibroreticular cell wall is not continuous so that follicular B cells and dendritic cells can insert cytoplasmic processes through the openings and sample any antigens within the lymphatic fluid [40]. A similar network of conduits occurs within the T cell zones where antigens are sampled by dendritic cells. The conduits provide for the rapid delivery of soluble antigens from the afferent lymph to the lumen of HEVs and enable these antigens to reach deep into a node [41].

11.8.2 Function The principal function of secondary lymphoid organs such as lymph nodes is to trap antigens and then facilitate the interactions between antigen-presenting cells and antigen-sensitive T and B cells. Each cell must be guided to its

160

SECTION |1 Mammalian immunology

appropriate contacts with great precision. A complex mixture of chemokines directs these cells. Thus chemokines drive the emigration of lymphocytes from HEVs into the lymph node. Once they enter the lymph node, the T and B cells are guided to their respective regions by chemokine gradients generated by stromal cells and follicular dendritic cells. Immature dendritic cells, once they encounter antigen, are also guided into lymph nodes by chemokines. For example, dendritic cells are attracted to the paracortex, where they present their antigen to T cells. Once this is accomplished, the dendritic cells change their chemokine receptors and then leave the node. Lymph nodes also contain innate lymphoid cells that are located close to sentinel macrophages lining the lymphatic sinuses. They thus are exposed to cytokines such as IL-18 released when the macrophages encounter invaders. The innate lymphocytes in turn rapidly increase IFN-γ secretion that activates the macrophages still further and enhances their antimicrobial activities. An interesting feature of secondary lymphoid organs is the fact that both B and T cells are highly active and motile. The T cells in the paracortex and B cells in the cortex are guided by follicular dendritic cells. Chemokine gradients control the relocation and recirculation of lymphocytes and ensure that they end up in the right place. For example, T cells are attracted to the perifollicular area of the cortex. B cells, on the other hand, are attracted to the interior of germinal centers. When T cells are activated, they too may enter germinal centers where they “help” B cells respond to antigens. Other secondary lymphoid organs employ different homing receptors. For example, MAdCAM-1 is a homing receptor found in blood vessels in PPs. Lymphocytes that recirculate to the intestine express high levels of the MAdCAM-1 ligand. Soluble antigens entering the node through its afferent lymphatics first pass into the subcapsular sinus. From there they enter the conduit network and are carried into the cortex. Antigenic particles such as viruses are first captured by macrophages within the subcapsular sinus. These macrophages carry the viral particles through the sinus floor and present them directly to B cells in the underlying follicles. The B cells then migrate to the T cell-B cell boundary, where they receive specific T cell help. B cells can also enter the paracortex directly from HEVs. There is a specialized population of follicular dendritic cells clustered around these blood vessels so that immigrating B cells can survey any antigens they may be carrying. This is also a perfect location to receive T cell help. When bacteria invade tissues, the resident dendritic cells are activated and migrate to the draining lymph node where they accumulate in the paracortex and cortex. These dendritic cells form a web through which the invaders must pass. Captured antigens are presented by the dendritic cells to T cells. T cells are initially activated in the paracortex, whereas the B cells remain randomly dispersed in the primary follicles. Both cell populations then migrate to the edges of the follicles where they interact. Once antibody production is stimulated, the progeny of these B cells move to the medulla and begin to secrete antibodies. Some of these antibody-producing cells may migrate into the efferent lymph and colonize downstream lymph nodes. Several days after antibody production is first observed in the medulla, germinal centers appear in the cortex. Some T cell-dendritic cell interactions are long-lasting, and in the presence of antigen, T cells and dendritic cells form stable complexes for many hours. However, before selecting its partner a dendritic cell might sample as many as 500 different T cells/hour and can interact with up to 10 T cells simultaneously. Binding by follicular dendritic cells is the predominant means of antigen trapping once an animal has been sensitized by previous exposure to an antigen. In a secondary response, the germinal centers become less obvious as activated memory cells emigrate into the efferent lymph. Once this stage is completed, the germinal centers redevelop. Antigen-stimulated lymph nodes also trap lymphocytes. Interactions between infectious agents and mast cells result in the production of tumor necrosis factor-α (TNF-α). The TNF-α blocks the passage of lymphocytes through these organs, the lymphocytes accumulate, and the lymph nodes swell. This trapping concentrates lymphocytes close to sites of antigen accumulation. After about 24 h, the lymph nodes release their trapped cells, and their cellular output is increased for several days.

11.8.3 Lymphocyte circulation In adult mammals, most cell types reside in stable tissues and do not move a lot. The cells of the immune system are in contrast, highly mobile. Cells move from the bone marrow to the thymus and secondary lymphoid organs; cells migrate around the body looking for invaders and also move from lymphoid organs to sites of microbial invasion. T cells, for example, constantly circulate in the body in the blood and tissue fluid and are the predominant lymphocytes in the blood. As they travel, they survey the body for foreign antigens and preferentially home to sites of microbial invasion and inflammation.

The mammalian lymphoid system Chapter | 11

161

Circulating T cells leave the bloodstream by two routes. T cells that have not previously encountered antigens (“naı¨ve” T cells) bind to HEV in lymph nodes. The high endothelial cells in these vessels are not joined by tight junctions but are linked by discontinuous “spot-welded” junctions (Fig. 11.6). This means that lymphocytes can pass easily between the high endothelial cells. Circulating lymphocytes can adhere to these high endothelial cells and then migrate into the paracortex. The emigration of lymphocytes from HEVs resembles that of neutrophils in inflamed blood vessels. Thus the cells first roll along the endothelial surface binding to selectins. As they roll, they become activated and express integrins. This results in their complete arrest and emigration. The number and length of HEVs are variable and controlled by local activity. Thus stimulation of a lymph node by the presence of antigens results in a rapid increase in the length of its HEVs. If, however, a lymph node is protected from antigens, its HEVs shorten [42]. Recognizable HEVs are not normally found in ruminant lymph nodes, but paracortical venules serve the same function in these species. In contrast to naı¨ve T cells, memory T cells leave the bloodstream through conventional blood vessels in tissues and are then carried to lymph nodes through afferent lymphatics. They leave the lymph nodes through the efferent lymphatics. Afferent lymph can be readily obtained by lymphatic cannulation in sheep. Typically, afferent lymph in sheep contains 85% T cells, 5% B cells, and 10% dendritic cells. Efferent lymph contains greater than 98% lymphocytes, of which 75% are T cells and 25% are B cells. The efferent lymphatics eventually join together to form large lymph vessels. The largest of these lymph vessels is the thoracic duct, which drains the lymph from the lower body and intestine and empties it into the anterior vena cava.

11.8.4 Species differences Domestic pigs and related swine, hippopotamuses, rhinoceroses, and some dolphins are different. Their lymph nodes consist of several inverted lymphoid “nodules” so that the cortex of each nodule is located toward the center of the node, whereas the medulla is at the periphery. Each nodule is served by a single afferent lymphatic that enters the central cortex as a lymph sinus (Fig. 11.7). Thus afferent lymph is carried deep into the node. A cortex surrounds the lymph sinus. Outside this region are a paracortex and a medulla. Lymph passes from the cortex at the center of the node to the medullary sinus at the periphery before leaving through the efferent vessels that drain the region between nodules. The cortex and paracortex have a similar structure to that seen in other mammals. Very few lymphocytes are found in pig lymph [43]. In marine mammals, lymph node structure is also highly variable. For example, all the lymph nodes of bottlenose dolphins (Tursiops truncatus) are of conventional structure. In striped dolphins, (Stenella coeruleoalba), on the other hand, some lymph nodes (mesenteric, for example) are of conventional structure, whereas others (mediastinal) have the inverted structure described previously. Thus both forms of lymph node may be present in a single dolphin [44] (Chapter 16). Shrews (Eulipotyphla) possess a unique lymphoid structure in their mesentery called the “pancreas of Aselli.” Its function is disputed. Some believe that is the equivalent of the avian Bursa of Fabricius. On the other hand, it may simply be a very large aggregate of mesenteric lymph nodes (Chapter 26).

FIGURE 11.6 A section of human tonsil showing a high endothelial venule with its characteristic high, rounded endothelial cells. Note the lymphocytes emigrating between the endothelial cells.

162

SECTION |1 Mammalian immunology

11.8.5 Hemolymph nodes Hemolymph nodes are structures similar to lymph nodes found in association with some blood vessels in ruminants such as cattle, sheep, and elephants, as well as in some rodents such as rats (Fig. 17.3) [45]. They are often located in the retroperitoneal fat. They are rare in humans [46]. Their function is unclear. They differ from conventional lymph nodes in that their lymphatic sinuses contain numerous red cells. Erythrocytes can gain access to these nodes by passing through the walls of blood vessels in the intermediate sinus area. The red cells pass between the expanded epithelial cells and an incomplete basement membrane. They do not appear to have lymphatic connections. Hemolymph nodes have a cortex containing germinal centers and B cells. T cells predominate at the center in association with lymphatic sinuses. These cells differ, however, from those found in conventional lymph nodes (more γ/δ1, WC11 T cells, fewer CD81 T cells) [47]. Intravenously injected carbon particles are trapped in the sinusoids of hemolymph nodes, suggesting that they may combine features of both the spleen and lymph nodes. However, their macrophages may also show evidence of erythrophagocytosis as well as blood pigment. These hemolymph nodes can produce antibodies in a manner similar to the spleen in response to an injection of foreign red cells.

Lymph flow

(A)

Paracortex

Cortex

Subcapsular sinus

Medulla Germinal center

High endothelial venule

Blood vessels

Afferent lymphatic

Efferent lymphatic Capsule

Lymph flow

(B)

Paracortex

Cortex

Germinal center

Medulla

Capsule Afferent lymphatic Blood vessels

Efferent lymphatic

FIGURE 11.7 A comparison of the structural features of A. A conventional lymph node with that of B. A pig lymph node.

The mammalian lymphoid system Chapter | 11

163

11.8.6 Other secondary lymphoid organs Secondary lymphoid organs include not only the spleen and lymph nodes but also the bone marrow, tonsils, and lymphoid tissues scattered throughout the body, most notably in the digestive, respiratory, and urogenital tracts. The lymphoid tissues of the intestinal tract constitute the largest pool of lymphocytes in the body, but the bone marrow also contains very large numbers of lymphocytes. If an antigen is administered intravenously, much will be trapped not only in the liver and spleen but also in the bone marrow and lungs. During a primary immune response, antibodies are mainly produced in the spleen and lymph nodes. Toward the end of that response, memory cells leave the spleen and colonize the bone marrow. When a second dose of an antigen is given, the bone marrow produces very large quantities of antibodies and is the major source of antibodies in adult rodents and rabbits. Up to 70% of the antibodies to some antigens may be produced by plasma cells in the bone marrow.

11.8.7 Mucosal-associated lymphoid tissues The mucosa-associated lymphoid tissues (MALTs) include lymphoid tissues in the eyelids, nasal mucosa, tonsils, pharynx, tongue, and palate (collectively called Waldeyer’s ring); PPs; solitary lymphoid nodules; the appendix if present; and numerous lymphoid nodules in the lung. These lymphoid tissues are known by their acronyms. Thus Gut-associated lymphoid tissue (GALT) is the collective term for all the lymphoid nodules, PPs, and individual lymphocytes found in the intestinal walls. MALT consists largely of aggregated non-encapsulated lymphoid tissues within the mucosa. It is found in multiple organs including the gastrointestinal tract, the bronchi (BALT), conjunctiva, larynx, salivary glands, and nasal mucosa. These aggregates of lymphoid cells are covered by a specialized overlying follicle-associated epithelium with or without microfold (M) cells. These organized lymphoid tissues, unlike lymph nodes, do not react to foreign antigens delivered through afferent lymph but rather sample them directly from the exterior surface.

11.8.8 Tonsils The tonsils are especially important in inducing immunity on mucosal surfaces. Some organisms, however, can overcome the defenses of the tonsils and use them as a portal of entry into the body. For example, pathogens such as bovine herpesvirus-1, Mannheimia hemolytica, Streptococcus suis, and Mycobacterium tuberculosis can persist indefinitely within the tonsils [48]. There are two types of tonsils, those with crypts and those without. A tonsillar crypt is a blind invagination of the surface epithelium surrounded by a mass of lymphoid tissue. A single crypt together with its associated lymphoid tissue constitutes a tonsillar follicle. They have no afferent lymphatics. Tonsils with crypts and follicles include the palatine tonsils of humans, horses, ruminants, and pigs (Fig. 11.8). Tonsils without crypts are formed by a single layer of lymphatic tissue that may bulge outward. These include the tonsils of Carnivora, the pharyngeal tonsils, and the tubal tonsils of ruminants. In general, tonsils are larger in young animals.

FIGURE 11.8 A section of pig tonsil showing a tonsillar crypt. Note how thin the epithelium is at the base of the crypt. This is an easy invasion route for many organisms. Original magnification 3 150. Courtesy Dr. S. Yamashiro.

164

SECTION |1 Mammalian immunology

Focal submucosal lymphoid aggregates are also present within the nasal cavity, especially in rodents. They have M cells on their epithelial surfaces. In rats, these are restricted to the ventral lateral walls, near the opening of the nasopharyngeal duct. In primates, they are found on the lateral and septal walls of the proximal nasopharynx.

11.8.9 Bronchus-associated lymphoid tissue Bronchus-associated lymphoid tissues (BALTs) are located within the lungs. The BALT consists of organized lymphoid follicles found in the bronchial submucosa. They are often located at the bifurcation of the airways. Definitive T and B cell areas have only been described in rabbit BALT (Fig. 11.9). The amount of BALT varies among species. Thus 100% of rabbits and rats have a BALT [14], 50% of guinea pigs have it, 33% of pigs, and none is normally present in cats or humans [49]. In humans, however, there are diffuse accumulations of lymphoid cells within the bronchial epithelium. It is clear that BALT is not constitutively present in all mammals and in many animals may be a tertiary lymphoid tissue that develops in response to inhaled antigens [49]. Many cells may be washed out of the airways of the lung with saline. In non-smoking human volunteers these consist of 85% macrophages, 11% lymphocytes, 1.3% neutrophils, 0.3% eosinophils [50]. Of these lymphocytes, 95% are CD31 (T cells). Fifty-four percent are CD41 and 36% are CD81. (In smokers these cell counts are roughly double, and the proportion of macrophages is much higher). In normal dogs, about 80% of bronchoalveolar cells obtained in this way are macrophages and 13% are lymphocytes, of which about half are T cells. In healthy horses, about 50% of the cells in bronchoalveolar washes are macrophages, 40% are lymphocytes, and 2% are neutrophils. In sheep, B cells form less than 10% of the lung lymphocyte population. Lung T cells can produce cytokines, and alveolar macrophages are activated following infection with Listeria monocytogenes. Cell-mediated immune reactions are therefore readily provoked among the T cells within the lower respiratory tract [1]. Alveolar macrophages reside on alveolar surfaces where they are in direct contact with the air and phagocytose inhaled particles. When they respond to invaders it is essential that they do not interfere with gas exchange. Thus a fullscale inflammatory response is to be avoided whenever possible. Thus in the absence of infection, alveolar macrophages are quiescent and tend to suppress local cytokine production. They are however highly phagocytic. They are also the major producers of type I interferons that in turn induce inflammation. The lungs of most domestic species (pigs, horses, sheep, goats, cattle, and cats) differ from rodent, human, or dog lungs in that they contain large numbers of pulmonary intravascular macrophages. It has been estimated that these macrophages cover 16% of the lung capillary surface in young pigs [1]. As a result, the lungs of these species can remove considerably more bacteria and particulate matter from the bloodstream than can the liver and spleen. A dense network of dendritic cells is also found within airway epithelium and alveoli. Macrophages are located throughout the body and so can detect and capture bacteria or fungi invading by many different routes. For example, bacteria injected intravenously are rapidly removed from the blood. Their precise fate depends on the species involved. In dogs, rodents,

FIGURE 11.9 Lymphoid follicle found in the bifurcation of an airway in a section of calf lung. This type of bronchus-associated lymphoid tissue is a key component of the defenses of the respiratory tract. From a specimen kindly provided by Drs. N.H. McArthur and L.C. Abbott.

The mammalian lymphoid system Chapter | 11

165

and humans, 80% to 90% are trapped and removed in the liver. The bacteria are removed by the macrophages (Kupffer cells) that line the sinusoids of the liver. The process occurs in two stages. Bacteria are first phagocytosed by blood neutrophils. These neutrophils are then ingested and destroyed by the Kupffer cells. These processes thus resemble acute inflammation in which neutrophils are primarily responsible for the destruction of invaders whereas the macrophages are responsible for preventing damage caused by apoptotic neutrophils.

11.8.10 Peyer’s patches As described above, the ileal PPs appear to be primary lymphoid organs in some species, especially sheep and possibly cattle [18]. However, their jejunal PP is clearly secondary lymphoid organs. PPs consist of masses of lymphocytes arranged in follicles and covered with an epithelium that contains M cells. Figure 11.2 M cells are specialized epithelial cells involved in antigen transportation. They have microfolds (M) rather than microvilli on their surface. The mucus layer tends to thin out over PPs so that the M cells protrude into the intestinal lumen. M cells endocytose the proteins and microbes they encounter, but rather than destroy them, they transport the processed antigens to their underlying lymphoid tissue. M cells may also transport soluble macromolecules such as IgA, small particles, and even whole organisms. (Some pathogens, such as Salmonellae, Yersinia, and Listeria species, M. tuberculosis, and the reoviruses may take advantage of the M cells and use them to gain access to the body.) The proportion of M cells in the follicleassociated epithelium varies from less than 10% in humans and mice to 50% in rabbits and 100% in the terminal ileum of pigs and calves. Ileal PPs do not function as primary lymphoid organs in all Cetartiodactyls. As described above, in the pig surgical removal of the ileal PP does not alter the level or phenotype of T or B cells following microbial colonization [51]. Removal of the piglet ileal Peyer’s patch does not result in a B cell deficiency such as would occur in a bursectomized bird. There are no differences in B cell diversity, distribution, or repertoire in piglets after this organ has been removed. In addition, there is no evidence that B cell diversification occurs within this organ. Evidence now suggests that in pigs at least, the ileal Peyer’s patch is a secondary mucosal lymphoid organ that regulates the initial microbial colonization of the lower bowel. There may be no discrete organ equivalent to the bursa in these mammals. Microbial colonization of the porcine ileal PP does however trigger a shift from a fetal type of lymphocyte to the appearance of large numbers of IgA1 B cells with a mature phenotype (CD22 CD211). Colonization also results in the appearance of CD41CD81 α/β T cells as well as CD21 CD82 γ/δ T cells [1]. In other mammals, the PPs can be considered to serve as immune sensors in the intestine [23]. Thus they interact with the intestinal microbiota, with food antigens, and with potential pathogens to either cause tolerance or defend the surfaces. Collectively they constitute one of the largest lymphoid organs, containing up to 70% of the body’s lymphocytes and plasma cells. They consist of aggregated lymphoid follicles covered by a specialized epithelial cap consisting of follicle-associated epithelium containing M cells. In humans the number of PPs peaks at ages 15
25 and declines thereafter. The area of the intestinal mucosa occupied by PPs reaches a maximum around the same time. In humans, the fetal intestine contains about 60 PPs before gestation week 30, and this steadily increases to reach about 240 PPs at puberty before declining. Distinct T and B cell clusters can be identified in the small intestine at 14
16 weeks gestation. By week 19 these mature into recognizable PPs although no germinal centers are present. Germinal centers rapidly develop after birth once the intestinal tract is colonized by its commensal bacteria. In mice, PP development begins at embryonic day 15.5. These attract LTi cells and organizer cells that in turn attract both T and B cells. A newborn calf normally has about 100 PP, and these may cover as much as half of its ileal mucosal surface. Although PPs are full of lymphocytes, most IgA is largely produced in diffuse lymphoid nodules and isolated plasma cells scattered throughout the walls of the intestine, in bronchi, salivary glands, and the gallbladder.

11.8.11 Lymphoglandular complexes Lymphoglandular complexes are structures present in the wall of the large intestine and cecum in horses, ruminants, dogs, and pigs. They consist of submucosal masses of lymphoid tissue penetrated by radially branching extensions of mucosal glands. These glands penetrate both the submucosa and the lymphoid nodule. They are lined by intestinal columnar epithelium containing goblet cells, intraepithelial lymphocytes, and M cells. Their function is unknown, but they contain many plasma cells, suggesting that they are sites of antibody production [1].

166

SECTION |1 Mammalian immunology

11.8.12 Cecal appendix The appendix of humans is a narrow tube located at the terminal end of the cecum. It is about 10 cm long with an external diameter of 7
8 mm and its internal lumen has a diameter of 1
3 mm. While once considered vestigial and hence useless, recent understanding suggests that it may play a role as a “safe house” for commensal microbes that enables the intestinal microbiota to be preserved during gastrointestinal diarrheal infections [52]. The intestinal immune system serves to support the growth of beneficial microbiota in the form of microbial biofilms. In addition, IgA and mucin play an important role in maintaining healthy populations of selected microbes in the intestine. The relative isolation of the appendix from the flow of digesta as well as its narrow lumen and the large quantity of lymphoid tissues in its walls suggest that it too can play a role in maintaining a healthy microbiota (Chapter 5). Even in the case of severe diarrhea with intestinal emptying, the appendix can serve as a sanctuary for the organisms required to repopulate the gut. Comparative analysis among mammals suggests that there are several types of cecal appendix as well as some appendix-like structures. It appears to have been maintained in mammalian evolution for at least 80 my. While no correlation has been found between the presence of an appendix and alterations in diet, it appears that the appendix has evolved at least 32 times and has been lost less than seven times suggesting that it has a positive fitness value [53]. Rabbits and pikas (Ochotonidae), mole rats (Bathyergidae), porcupines, flying squirrels (Pteromyini), meadow voles (Microtus), and wombats (Diprotodontia) all have cecal appendices as do humans and apes. Thus they are found in monotremes, some marsupials, lagomorphs, and several rodent species in addition, to primates. Besides the human, the only appendix to be extensively studied has been that of the rabbit. The appendices of mammals fall into three morphotypes. Thus humans have a distinct structure emerging from a rounded sac-like cecum; an appendix located at the end of a long, large cecum as in the rabbit, sugar glider (Petaurus breviceps), and the mole-rat; or an appendix that occurs in the absence of a cecum as in the wombat. There is a direct correlation between appendix and cecal morphology, and this suggests that they may be evolving as a module,—the cecoappendicular complex [54].

11.8.13 Cryptopatches Cryptopatches are very small, tightly packed, lymphoid aggregates found within the intestinal lamina propria of mice. They are considered to be distinctly different from isolated lymphoid follicles. There are more than a thousand cryptopatches in the mouse, scattered throughout both the large and small intestine. They appear to serve as production and maturation sites for α/β and γ/δ T cells and possibly function independently of the thymus. They are distinctly different from PPs and contain a totally different mixture of lymphocytes together with some dendritic cells [55]. These T cells are positive for the cell-surface marker c-kit and have IL-7 receptors. They do not express CD3 which is common on other intestinal T cells. They appear to act as thymus-independent T cell precursors, perhaps driven by the gut microbiota [56].

11.8.14 Anal tonsils The tonsils and adenoids in Waldeyer’s ring serve to protect the pharynx and anterior gastrointestinal tract from invaders. Anal tonsils serve to protect the posterior gastrointestinal tract. They are a feature of aquatic mammals since underwater defecation runs a risk of invasion by water-borne pathogens, especially at depths where water pressure is high (Chapter 16). However, some species of shrews also possess anal, and even vaginal tonsils (Chapter 26). Cattle also possess a ring of lymphoid tissue around the vaginal vestibule [57]. Their overall structure of clusters of lymphoid follicles located subepithelially is similar to that of the nasopharyngeal tonsils.

11.8.15 Tertiary lymphoid organs Within the body are found lymphoid structures that are structurally similar to secondary lymphoid organs but only develop in response to microbial colonization and chronic immune stimulation. These are organized masses of lymphocytes with clearly discrete T and B cell regions that contain germinal centers and other lymphoid tissue components. They are called tertiary lymphoid organs. They may develop in the intestinal wall in response to the intestinal microbiota. Other examples are the lymphoid nodules that develop in rheumatoid arthritis joints and atherosclerotic plaques in humans. The initial trigger for their development probably comes from stimulated fibroblasts that produce appropriate chemokines attracting T cells, B cells, and dendritic cells. Angiogenic factors trigger the production of lymphatics and HEVs [58].

The mammalian lymphoid system Chapter | 11

167

References [1] Tizard IR. Veterinary immunology: an introduction. 10th ed. St. Louis: Elsevier; 2017. [2] Withers DR. Lymphoid tissue Inducer cells. Curr Biol 2011;21(10):RR381 R382. [3] Lane PJL, McConnell FM, Withers D, Gaspal F, et al. Lymphoid tissue inducer cells: bridges between the ancient innate and the modern adaptive immune systems. Nat Immunol 2009;2(6):472
7. [4] Boehm T, Hess I, Swann JB. Evolution of lymphoid tissues. Cell 2012;33(6):315
21. [5] Wang J, Sekai M, Matsui T, Fujii Y, et al. Hassall’s corpuscles with cellular senescence features maintain IFNα production through neutrophils and pDC activation in the thymus. Int Immunol 2018;31(3):127
39. [6] Watanabe N, Wang Y-H, Lee HK, et al. Hassall’s corpuscles instruct dendritic cells to induce CD41CD251 regulatory T cells in human thymus. Nature 2005;436:1181
5. [7] Pearse G. Normal structure, function and histology of the thymus. Toxicol Pathol 2006;34:504
14. [8] Tizard IR. Autoimmune diseases of domestic animals. St. Louis: Elsevier; 2022. [9] Anderson G, Takahama Y. Thymic epithelial cells: working class heroes for T cell development and repertoire selection. Trends Immunol 2012;33(6):256
63. [10] Steinmann GG, Klaus B, Muller-Hermelin HK, et al. The aging of the human thymic epithelium is independent of puberty. A morphometric study. Scand J Immunol 1985;22(5):563
75. [11] Laan M, Haljasorg U, Kisand K, Salumets A, Peterson P. Pregnancy-induced thymic involution is associated with suppression of chemokines essential for T-lymphocyte progenitor homing. Eur J Immunol 2016;46:2008
17. [12] Galletti G, Cavallari A. The thymus of marmots: spontaneous natural seasonal thymectomy. Acta Anat 1972;83:593
605. [13] Cunningham CP, Kimpton WG, Holder JE, Cahill RN. Thymic export in aged sheep: a continuous role for the thymus throughout pre- and postnatal life. Eur J Immunol 2001;31:802
11. [14] Haley PJ. The lymphoid system: a review of species differences. J Toxicol Pathol 2017;30:111
23. [15] Sisson S, Grossman JD. The anatomy of the domestic animals. 4th ed. Philadelphia: Saunders; 1953. [16] Landsverk T, Halleraker M, Aleksandersen M, et al. The intestinal habitat for organized lymphoid tissues in ruminants: comparative aspects of structure, function and development. Vet Immunol Immunopathol 1991;28:1
16. [17] Motyka B, Reynolds JD. Apoptosis is associated with the extensive B cell death in the sheep ileal Peyer’s patch and the chicken bursa of Fabricius: a possible role in B cell selection. Eur J Immunol 1991;21:1951
8. [18] Yasuda M, Jenne CN, Kennedy LJ, Reynolds JD. The sheep and cattle Peyer’s patch as a site of B cell development. Vet Res 2006;37:401
15. [19] Reynaud C-A, Mackay CR, Mu¨ller RG, Weill J-C. Somatic generation of diversity in a mammalian primary lymphoid organ: the sheep ileal Peyer’s patches. Cell 1991;64:995
1005. [20] Reynolds JD, Morris B. The evolution and involution of Peyer’s patches in fetal and postnatal sheep. Eur J Immunol 1983;73:627
35. [21] Reynolds JD. Evidence of extensive lymphocyte death in sheep Peyer’s patches. I. A comparison of lymphocyte production and export. J Immunol 1986;136:2005
10. [22] Chen H, Zhang Y, Ye AY, Du Z, et al. BCR selection and affinity maturation in Peyer’s patch germinal centers. Nature 2020;582:421
5. [23] Jung C, Hugot J-P, Barreau F. Peyer’s patches: the immune sensors of the Intestine. Int J Inflamm 2010;. Available from: https://doi.org/ 10.4061/2010/823710. [24] Pospisil R, Mage RG. Rabbit appendix: a site for development and selection of the B cell repertoire. Curr Top Microbiol Immunol 1998;229:59
70. [25] Tischendorf F. On the evolution of the spleen. Experientia 1985;41:145
52. [26] Udroiu I. Evolution of sinusal and non-sinusal spleens of mammals. Hystrix It J Mamm 2006;17(2):99
116. [27] Lewis SM, Williams A, Eisenbarth SC. Structure and function of the immune system in the spleen. Sci Immunol 2019; eaau 6085. [28] Mebius RE, Kraal G. Structure and function of the spleen. Nat Rev Immunol 2005;5:606
16. [29] Hartwig H, Hartwig HG. Structural characteristics of the mammalian spleen indicating and storage and release of red blood cells. Aspects of evolutionary and environmental demands. Experientia 1985;41:159
63. [30] Sørby R, Wein TN, Husby G, Espenes A, Landsverk T. Filter function and immune complex trapping in splenic ellipsoids. J Comp Pathol 2005;132:313
21. [31] Bronte V, Pittet MJ. The spleen in local and systemic regulation of immunity. Immunity. 2013;39(5):806
18. [32] Udroiu I. Storage of blood in the mammalian spleen: an evolutionary perspective. J Mammal Evol 2017;24:243
60. [33] Onkar DP, Govardhan SA. Comparative histology of human and dog spleen. J Morphol Sci 2013;30(1):16
20. [34] Okuda M, Togawa A, Wada H, Nishikawa S. Distinct activities of stromal cells involved in the organogenesis of lymph nodes and Peyer’s patches. J Immunol 2007;179:804
11. [35] Bryant BJ. Lymph node structure. An ontogenetic explanation for divergence in Eutheria, Metatheria and Prototheria. Amer Zool 1975;15:147
53. [36] Ashman RB, Papadimitriou JM. Development of lymphoid tissue in a marsupial Setonix brachyuris (quokka). Acta Anat (Basel) 1975;91 (4):594
611. [37] Willard-Mack CL. Normal structure, function and histology of lymph nodes. Toxocologic Pathol 2006;34:409
24. [38] Allen CD, Okada T, Cyster JG. Germinal-center organization and cellular dynamics. Immunity 2007;27:190
202.

168

SECTION |1 Mammalian immunology

[39] Sixt M, Kanazawa N, Selg M, et al. The conduit system transports soluble antigens from the afferent lymph to resident dendritic cells in the T cell area of the lymph node. Immunity 2005;22:19
29. [40] Harwood NE, Batista FD. The antigen expressway: follicular conduits carry antigen to B cells. Immunity 2009;30:177
9. [41] Roozendaal R, Mempel TR, Pitcher LA, et al. Conduits mediate transport of low molecular weight antigen to lymph node follicles. Immunity 2009;30:264
76. [42] von Andrian UH, Mempel TR. Homing and cellular traffic in lymph nodes. Nat Rev Immunol 2003;3:867
78. [43] Binns RM, Pabst R. Lymphoid tissue structure and lymphocyte trafficking in the pig. Vet Immunol Immunopathol 1994;43:79
87. [44] Vukovic S, Lucic H, Gomercic H, et al. Morphology of the lymph nodes in bottlenose dolphin (Tursiops truncatus) and striped dolphin (Stenella coeruleoalba) from the Adriatic sea. Acta Vet Hung 2005;53:1
11. [45] Thorp BH, Seneque S, Staute K, Kimpton WG. Characterization and distribution of lymphocyte subsets in sheep hemal nodes. Dev Comp Immunol 1991;15:393
400. [46] Sakita K, Fujino M, Koshikawa T, Ohmiya N, et al. The structure and function of the hemolymph node in rats. Nagoya J Med Sci 1997;60:129
37. [47] Galeotti M, Sarli G, Eleni C, Marcato PS. Identification of cell types present in bovine haemolymph nodes and lymph nodes by immunostaining. Vet Immunol Immunopathol 1993;36:319
31. [48] Velinova M, Theilen C, Melot F, et al. New histochemical and ultrastructural observations on normal bovine tonsils. Vet Rec 2001;149:613
17. [49] Pabst R, Gehrke I. Is the bronchus-associated lymphoid tissue (BALT) an integral structure of the lung in normal mammals, including humans? Am J Respir Cell Mol Biol 1990;3:131
5. [50] Heron M, Grutters JC, Ten Dam-Molenkamp KM, Hijdra D, et al. Bronchoalveolar lavace cell pattern from healthy human lung. Clin Exp Immunol 2012;167(3):523
31. [51] Sinkora M, Stepanova K, Butler JE, Francis D, et al. Ileal Peyer’s patches are not necessary for systemic B cell development and maintenance and do not contribute significantly to the overall B cell pool in swine. J Immunol 2011;187:5150
61. [52] Smith HF, Fisher RE, Everett ML, Thomas AD, et al. Comparative anatomy and phylogenetic distribution of the mammalian cecal appendix. J Evol Biol 2009;22:1984
99. [53] Smith HF, Parker W, Kotze SH, Laurin M. Multiple independent appearances of the cecal appendix in mammalian evolution and an investigation of related ecological and anatomical factors. CR Palevol 2013;12:339
54. [54] Smith HF, Parker W, Kotze SH, Laurin M. Morphological evolution of the mammalian cecum and cecal appendix. CR Palevol 2017;16:39
57. [55] Eberl G, Sawa S. Opening the crypt: current facts and hypotheses on the function of cryptopatches. Trends Immunol 2009;32(2):50
5. [56] Pabst O, Herbrand H, Worbs T, Feiedrichsen M, et al. Cryptopatches and isolated lymphoid follicles. Dynamic lymphoid tissues dispensable for the generation of intraepithelial lymphocytes. Eur J Immunol 2005;35:98
107. [57] Chuluunbaatar I, Ichii O, Masum MA, Namba T. Genital organ-associated lymphoid tissues arranged in a ring in the mucosa of cow vaginal vestibule. Res Vet Sci 2022;147
58. [58] Neyt K, Perros F, Corine H, van Kessel G, et al. Tertiary lymphoid organs in infection and autoimmunity. Trends Immunol 2012;33:297
305.

Section 2

Mammalian orders

This page intentionally left blank

Chapter 12

The monotremes: echidnas and platypus

Short-beaked echidna, Tachyglossus aculeatus

As described in Chapter 1 the ancestral amniotes separated into the Synapsids and Sauropsids about 310 mya. One branch of the synapsids evolved into the Therapsids and eventually developed both homeothermy and lactation. Over time, the egg-bearing Prototheria diverged from the live-bearing Theria while the Theria in turn subsequently diverged into two subclasses, the placental mammals
the Eutheria, and the Marsupials - the Metatheria. The divergence of the Prototheria, from the main mammalian line probably occurred sometime around 200 mya [1
3]. These early oviparous Prototherians eventually gave rise to the monotremes. The term Monotremata refers to the fact that they have a single external opening, a cloaca, for both their digestive and reproductive systems. The currently extant monotreme genera, the echidnas, and the platypus probably diverged from each other around 55 mya [2]. The two living genera of the Monotremata are the semi-aquatic platypus (Ornithorhynchidae) and the terrestrial echidnas (Tachyglossidae). The single species of platypus has a restricted distribution in Eastern and Southern Australia. The three species of echidna are distributed in two genera found across Australia and New Guinea. The platypus relies on a diet of invertebrates such as insect larvae, mollusks, and crustaceans whereas echidnas are strictly insectivorous. Both the platypus and echidnas have had their genomes sequenced in detail [2].

12.1

Reproduction

The echidnas are solitary animals that meet only to mate. The species that has been most intensively investigated is the short-beaked echidna (Tachyglossus aculeatus). After mating, a single echidna embryo develops within the maternal uterus and layers of the shell are progressively added (Fig. 12.1). After a gestation period of 15
21 days, the female lays a single, 13
17 mm diameter, leathery shelled, egg into a temporary pouch (incubatorium) made by folding over a flap of her lower abdominal skin. The mother echidna incubates the egg in her burrow keeping it both warm (32 C) and moist by holding it between her curled tail and abdomen [4]. After 10
11 days of incubation, the egg hatches releasing a blind and hairless baby (a puggle), that attaches itself to a mammary patch (areola) inside the pouch. The pouch progressively deepens as the puggle grows. The subsequent lactation lasts for about 160
210 days depending on location. The echidna puggle nurses continuously for 8
12 weeks until it completes development, and its spines begin to grow. Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00002-2 © 2023 Elsevier Inc. All rights reserved.

171

172

SECTION | 2 Mammalian orders

15-21 days

11 days Leaves pouch 80-100 days

Egg laid

160-210 days

Pouch/ Lactation Conception

Hatch

21 days

ECHIDNA

10 days Egg laid

90-120 days

Pouch/ Lactation Conception

Hatch

PLATYPUS

FIGURE 12.1 The timing of gestation, egg incubation, and lactation in the two monotreme species.

Even after leaving the pouch, it continues to suckle for the next six months. During this time, the young do not leave the burrow (and so excrete into the burrow environment). Thus the burrow is heavily contaminated with many environmental bacteria. The platypus (Ornithorhynchus anatinus) behaves somewhat differently. Following mating, pregnant females construct a special nesting burrow where, after a gestation period of 21 days, they eventually lay two small, (17 mm), leathery-shelled eggs. The eggs are incubated for 10
11 days as the mother curls herself around them so that her tail touches her bill and incubates them with the warmth of her body
30 C
32 C. (The platypus does not have a pouch). The newly hatched young suckle from specialized fur-covered nipples and remain in the burrow for three to four months before becoming independent. The milk is so nutritious that hatchling weight may increase twenty-fold during the first 14 weeks of life. (Reflecting the incredibly small size of the newly hatched platypus). The platypus is relatively long-lived
20 years in the wild and 23 years in captivity [4].

12.1.1 Lactation As discussed in Chapter 3, the mammary gland most probably evolved from ancestral apocrine-like glands associated with hair follicles. This association persists in both monotremes and marsupials [5]. It has been suggested that the early function of the protomammary gland was to keep the permeable monotreme eggs moist since their porous leathery shells would have permitted evaporation. However, their eggshells would also have provided considerable protection against microbial invasion. The mammary patch, essentially an areola without a nipple, would have secreted an aqueous /lipid emulsion that served as an effective egg-shell moisturizer. The mammary gland eventually developed a nutritional role as well. Echidna milk is a rich mixture of sugars, lipids, and proteins that are both nutritional and antimicrobial. The major carbohydrates in both echidna and platypus milk are acidic oligosaccharides with O-acetylated sialic acid residues. It has been suggested that this O-acetylation may protect the oligosaccharides from degradation by skin bacteria [6]. Monotreme genomes contain most of the milk genes that have been identified in therian mammals [2]. The platypus has however lost one of the vitellogenin genes that, in oviparous species, encode nutrient-rich proteins stored in the egg yolk. This pseudogenization appears to have happened about 30
70 mya. However, a second vitellogenin gene appears to be functional in the monotremes. (Chickens have three of these genes while viviparous mammals have none.) Monotremes are egg-laying but the amount of yolk present in their eggs is minimal. The nutrients lost with the vitellogenins have been replaced by nutritionally rich caseins in the milk. Most mammals have three casein genes while the platypus genome contains a cluster of three to five casein genes [2]. This casein thus provides the suckling platypus with essential nutrition, including especially calcium [7]. Immunoglobulins have not been detected in monotreme milk. However, their milk contains a unique monotreme lactation protein (MLP) that is constitutively expressed in milk cells throughout lactation. MLP is a 49 kDa N-linked amphipathic α-helical glycoprotein. It is found in the milk of both the platypus and the short-beaked echidna. Their milk also contains many mammary cells that express MLP in large amounts [8]. However, while MLP does not affect E coli, Pseudomonas aeruginosa, Staph epidermidis, or Salmonella enterica, it has significant antimicrobial properties against Staph aureus and the commensal Enterococcus faecalis [9].

The monotremes: echidnas and platypus Chapter | 12

173

12.1.2 Venom A unique feature of the male platypus is that it has venomous spurs on the inner side of each hind leg ankle that are connected to venom glands over the thighs. The venom can kill animals such as dogs and cause intense pain in humans. The venom is a complex mixture with at least 19 different components. These include beta-defensin-like peptides, nerve growth factors, and C-type natriuretic peptides. These toxins, especially the defensins, have evolved from genes that originally had antimicrobial functions and played a role in innate immunity [10]. In contrast, the related antimicrobial peptides, the alpha-defensins are absent from monotremes and first appeared in marsupials [11].

12.2

Hematology

The erythrocytes of the platypus possess no unusual features and are similar to those in other monotremes, marsupials, and eutherians. As in other wild mammals, their white cell parameters change rapidly after capture. This is largely due to a major drop in blood lymphocyte numbers [12] (Table 12.1). One unusual feature of platypus neutrophils however is the occurrence of Do¨hle bodies within the segmented neutrophils. These are pale basophilic inclusions in the peripheral cytoplasm. Their significance is unknown, but they may be remnants of the rough endoplasmic reticulum. Echidna neutrophils generally have 3
7 lobes, which is more than other mammals. In the absence of infection, their numbers may range from 10% to less than 50% of the blood leukocytes [13]. Monocytes typically have a U-shaped nucleus with a basophilic granular cytoplasm. In addition, the nuclei of monotreme eosinophils and basophils are unsegmented so that they have an ovoid or minimally segmented nucleus [14]. They also contain many rod-shaped eosinophilic granules. Basophils, when present, contain few cytoplasmic granules [15].

12.3

Innate immunity

With the exception of the modified defensins employed in platypus venom, the monotreme innate immune system is similar to that in other mammals as befits a system that originated hundreds of millions of years prior to the origin of the mammals. Though generally similar to eutherian innate immunity, monotremes have some interesting differences. Both the platypus and opossum genomes contain expansions of the cathelicidin gene family. Among the eutherians, primates and rodents have a single cathelicidin gene whereas ruminants have numerous such genes [16]. The platypus has eight cathelicidin genes. The monotreme cathelicidins may serve to protect their developing young especially when vulnerable within the burrow. The platypus genome also contains an expansion of genes encoding the macrophage differentiation antigen CD163. This is a hemoglobin scavenger receptor. Two groups of type I interferon genes have been identified in the genome of the short-beaked echidna. They appear to be homologous to eutherian IFN-α and IFN-β. They possess three IFN-α genes, considerably fewer than the ten plus genes present in marsupials and eutherians confirming that the initial gene duplication probably occurred prior to the divergence of monotremes and eutherians [17]. The echidna also has a single functional IFN-β gene [18].

TABLE 12.1 The blood leukocyte counts in the two major Monotreme species and their percent composition. Short-beaked echidna [13]

Platypus [12]

Total WBCs x103/μl

16.7
16.8

25
30

Neutrophils %

54
57

25

Lymphocytes %

40
44

70

Monocytes %

2
3

2.1

Eosinophils %

1

1.7

Basophils %

0

,1

174

12.4

SECTION | 2 Mammalian orders

Lymphoid tissues

Echidnas and the platypus possess all the lymphoid organs associated with the mammalian immune system.

12.4.1 Thymus In the echidna, the thymus arises from the third pharyngeal pouch and is located in the anterior mediastinum under the heart and great vessels adjacent to the thyroid and parathyroids. In the platypus, it is also thoracic but extends back towards the pericardium. As with other mammals, the thymus is a thin lobulated organ. It has a connective tissue capsule and is divided into many lobules. It is similar histologically to the eutherian thymus [19]. The cortex consists of densely packed lymphocytes with macrophages and reticular epithelial cells. The medulla contains numerous Hassall’s corpuscles with thymocytes and obvious epithelial cells. Most of the medullary cells are CD3-positive T cells. There are scattered plasma cells found in the cortex and medulla [20].

12.4.2 Spleen The platypus has a relatively large, bilobed (V-shaped) or trilobed (Y-shaped) spleen that is enclosed in a thick connective tissue capsule that contains very little smooth muscle [20]. As a result, the platypus spleen has the highest relative mass among mammals amounting to about 1% of its body weight [21]. It appears to be a significant site of erythropoiesis in this species. The white pulp is prominent with periarteriolar sheaths and lymphoid follicles suggesting it is primarily a defensive spleen. CD31 lymphocytes (T cells) are located around the follicles and in the periarteriolar lymphoid sheaths. The sheath is surrounded by an intermediate zone that consists of an inner domain of lymphoreticular tissue and an outer domain of venous capillaries. It is separated from the white pulp by an arterial net. The intermediate circulation is closed but the arterioles from the white pulp join the venules in the red pulp. Overall, the well-developed white pulp resembles other mammalian spleens [21].

12.4.3 Lymphoid nodules In contrast to the conventional structure of their thymus and spleen, monotremes lack conventional lymph nodes. Instead of typical mammalian lymph nodes, they have structures consisting of a cluster of lymphoid nodules, each containing a germinal center suspended by its blood vessels within the lumen of a lymphatic plexus. (Fig. 12.2) Each nodule is bathed in the lymph. CD3 positive cells are found in clusters and as individual cells throughout the follicles. Plasma cells are inconsistently present [20]. Each nodule represents a single lymphoid follicle that is comparable to a cortical follicle in a eutherian lymph node [22]. In the echidna, the capsule of the nodule is formed by the wall of the lymph vessel [23]. There is also a network of smaller lymph channels running through it. In the platypus, the lymph nodules are scattered within loose connective tissue located in the same strategic sites as other mammals,
cervical, pharyngeal, mediastinal, mesenteric, and pelvic regions. They each consist of single primary or secondary follicles supported by a network of reticular fibers. The active germinal centers, usually one per nodule, have a densely packed outer mantle with larger dividing cells in the center. Macrophages are abundant within these lymphoid clusters so that any passing antigen is readily captured and processed.

Echidna

Eutherian mammal

FIGURE 12.2 The structure of Monotreme lymph nodules. Each nodule is strategically placed on a lymphatic vessel. The structure of a typical eutherian lymph node is shown in comparison. Both structures permit lymphatic fluid to flow past B cell populations so that any antigens can be captured and responded to.

The monotremes: echidnas and platypus Chapter | 12

175

12.4.4 Gut-associated lymphoid tissues Platypus has tonsils that consist of submucosal follicles and intraepithelial lymphocytes. They are not visible macroscopically. Echidna also possesses tonsils associated with their submandibular salivary glands. The densely packed cells within are CD3-positive. Likewise, their submucosal Peyer’s Patches are only detectable by histology. As in other mammals, they consist of secondary follicles together with clusters of lymphocytes located within the lamina propria. Similar cecal lymphoid tissues are also present in the submucosa. There are also lymphoid aggregates found within the lungs and associated with the bronchi. These are also not obvious macroscopically. Similar tissues have been reported in the echidna. Appendices have been described in the large bowel of both Echidnas and platypus. The appendix in the platypus is 2.5 cm long and 3 mm in diameter. The structure in the echidna is smaller being about 1 cm in length and 3
4 mm in diameter. These appendix-like structures may represent primordial appendices that may have immunological functions but have not yet evolved into fermentation chambers. Antigen may also be taken up by cells within the appendix, the Peyer’s patches, and Hassall’s corpuscles.

12.4.5 Monotreme major histocompatibility complex The major histocompatibility complex (MHC) has been characterized in both the echidna and platypus. There is a separate monotreme MHC class I clade and there are no orthologous relationships between the MHC class 1 genes of the platypus and echidna [24]. In both species, the MHC genes are located on two different chromosomes. However, the classical class I and class II genes are located in a single cluster in each genome [2]. The monotreme MHCs are located on their sex chromosomes. The platypus chromosomal karyotype consists of 21 pairs of autosomes and five pairs of sex chromosomes per diploid cell. The sex chromosomes come in different sizes and gene contents. Thus males possess five different Y chromosomes (Y1
Y5) as well as five different X chromosomes (X1
X5). In contrast, females possess two copies of each of the five X chromosomes. In effect, five pairs of X1X1
X5X5. X1 is the largest of these while X4 is the smallest. During the prophase of meiosis, the sex chromosomes join together to form a chain in an alternating XY pattern. This chain then segregates into XXXXX and YYYYY -bearing sperm [25]. The echidna karyotype also contains 27 autosomes with ten sex chromosomes in females and nine sex chromosomes in males. Unlike eutherian mammals, the monotreme MHCs are not contiguous since they are found on the pseudoautosomal regions of two pairs of sex chromosomes [2]. The MHC core region containing both classes I and class II genes is located on platypus and echidna X3 and Y3 (Fig. 12.3). A similar class I/II complex is present in the marsupial opossums and suggests that this may be the ancestral organization pattern. In addition, echidna X4 and Y4 and platypus Y4 and X5 each contain a cluster of genes that show synteny with the human distal class III region. An intron-containing class I pseudogene is located on platypus Y4/X5 at a genomic location equivalent to the human HLA-B, -C regions also suggesting ancestral synteny. The complete platypus MHC contains seven genes. These include two class I genes, two class II genes, and three framework genes [26]. One of the class II genes has 97% sequence identity to the eutherian class II β chain and thus is PLATYPUS SEX CHROMOSOMES X1

X2

X3

X4 X5

Class I/II Class III

Y1

Y2

Y3

Y4 Y5

FIGURE 12.3 The platypus has ten sex chromosomes. The major histocompatibility complex loci are found on two of these. Note that class I and II loci are adjacent whereas class III loci are on a separate sex chromosome.

176

SECTION | 2 Mammalian orders

Chromosome X5 PLATYPUS

III

Chromosome X3 I-1 I-2 DZB DZA II

Chromosome X4

FIGURE 12.4 The platypus and echidna major histocompatibility complex are each located on two sex chromosomes. Class I and II are adjacent. Class III is separate.

I

Chromosome X3

ECHIDNA

III

II

I

considered a novel DZB allele. The other class II gene is orthologous with the DRA gene in the tammar wallaby (Notamacropus eugenii) and in Eutheria. The platypus class I loci are named class I-1 and class I-2 (Fig. 12.4). It appears that the MHC system of monotremes is both complex as well as continuing to evolve [26].

12.4.5.1 Major histocompatibility complex class I The MHC class 1 genes of the three mammalian classes, eutherians, marsupials, and monotremes are derived from three separate MHC class 1 lineages as a result of two rounds of duplications and deletions. The first round probably occurred prior to the monotreme-therian split. The second before the eutherian-marsupial split. Thus this process is consistent with the birth and death model of MHC evolution. The pseudogenes created by this duplication often contain intron/ exon structures identical to the gene from which they were derived. In addition to the pseudogenes created by duplication, others were probably created by retrotransposition. In this case, the pseudogenes lack introns. These are known as processed pseudogenes and make up the majority of known pseudogenes. Clearly monotremes have many processed pseudogenes. Thus the platypus and echidna class I genes have formed their species-specific clades, unlike the class II loci [24]. They have clearly evolved at very different rates. A unique family class I related genes called UT has been identified in both marsupials and monotremes. They are present but have not been mapped in the platypus. These have been lost from the eutherian lineage [27]. The UT gene products while highly divergent, fold in such a way that they can form the MHC class I α-chain structure. They appear to be non-classical molecules, but their function is unclear. In the opossum, they are mainly expressed on the skin and thymus [28].

12.4.5.2 Major histocompatibility complex class II Four MHC class II sequences have been isolated from the echidna and two from the platypus. Because they are nonorthologous with the other mammals, the two platypus genes have been classified as DZA and DZB. Studies on disparate platypus populations across Australia have shown high DZB polymorphism with 57 DZB β1 alleles identified in 70 individuals [29]. There is clear evidence of positive selection occurring within genes encoding the DZB peptide binding region. There is an ongoing debate regarding the orthology of marsupial and eutherian DRB genes [30]. At least two of the echidna β chain loci are transcribed. Sequence analysis suggests that the monotreme gene cluster is a sister group to all mammalian β chains. Around 200 mya, a gene duplication event occurred that gave rise to two gene clusters, the DP/DQ lineage, and the DO/DR/DA/ DB lineage. Marsupials do not possess a DP/DQ cluster suggesting that it was lost in marsupials before their radiation. The platypus DZB sequence is also present among the echidna sequences.

12.4.6 Natural killer cell receptors As described in Chapter 10, eutherian mammals use two functionally similar families of receptors on their NK cells; the KIR receptors encoded within the leukocyte receptor complex and the KLR (Ly49) receptors encoded within the natural killer complex to identify virus-infected and stressed cells. The platypus genome contains 213 natural killer C-type lectin (KLR) domain genes within its NK complex but their functions are unclear [31
33]. These platypus KLR receptor-like genes are found in at least two separate genomic regions. Thus 34 KLR genes are found on chromosome 7, two are located on a small autosome, while the remainder are currently unanchored [33]. About 18 of these are LRC-like exons although the orthology of most remains unclear. Despite their numbers, no clear orthologs of the NKC

The monotremes: echidnas and platypus Chapter | 12

177

complex families Ly49 or CD94/NKG2, have been identified in monotremes [32]. More recent studies have also identified four LRC genes within the platypus genome [34].

12.5

B cells and immunoglobulins

In a simple study designed to compare the antibody responses of the echidna, platypus, and rabbit, Wronski et al. followed their responses to two intraperitoneal injections of sheep red blood cells [35]. They found that the echidna and platypus produced lower antibody titers than the rabbit in their primary response. They also found that the secondary response in the monotremes was very weak. This was attributed to the fact that both species failed to undergo the IgM to IgG switch so characteristic of eutherians. They also investigated the heat stability of the antibodies produced and determined that the echidna and platypus antibodies were more susceptible to heat inactivation than those in the rabbit. Similar low antibody responses have been reported in the echidna in response to Salmonella adelaide antigens and bovine serum albumin [35]. All the major structural changes that gave rise to the immunoglobulin classes of modern eutherian mammals evolved well before the separation of the monotremes from the marsupials and placental mammals and probably soon after the split from reptile lineages B310 mya. Monotreme antibodies consist of the usual eutherian immunoglobulin classes with some significant exceptions (Fig. 12.5). For example, in the platypus, they are encoded by eight heavy chain constant region genes arranged in order -M-D-O-G2-G1-A1-E-A2- that spans a region of B200 kb and encodes six distinct isotypes. These isotypes include a single IgM, IgD, and IgE, two IgG subclasses, and two IgA subclasses. Between the D and G2 constant region genes is an additional heavy chain gene (IGHO) that encodes a unique monotreme isotype called immunoglobulin O (IgO). While similar to avian IgY, IgO is more closely related to eutherian IgG. In contrast, platypus IgD is more closely related to its homologs in reptiles and fish than to eutherian IgD. For example, it has five constant region domains rather than the three-domain IgD of eutherians. Thus all heavy chain isotypes found in eutherian mammals are also present in the platypus despite their divergence B200 mya.

12.5.1 Immunoglobulin M IgM is structurally conserved in all the tetrapods. Short-beaked echidna (T. aculeatus) polymeric IgM has a molecular weight of 950 kDa, with a heavy chain of 69 kDa and carbohydrate content of 6.4%. Each unit has four constant region domains with highly conserved cysteine and tryptophan residues. It shows approximately 47% amino acid identity to the possum, and dog Cμ and approximately 30% amino acid identity to Cμ from the chicken and turtle [36]. Platypus IgM shows 87% sequence identity with the echidna, T. aculeatus. Its variable genes belong to clan III. This differs from the echidna that uses IGHV genes from all three heavy chain clans [37].

FIGURE 12.5 The organization of platypus immunoglobulin genes. LC, Light chain constant gene; Dψ, IgD domain pseudogene.

IGH D ?

> 44

11 M

D

O

G2

IGK 16 V

G1

A1

E

A2

IGL 11

8-10 J

V

C

M

D

G

O

V

D

E J

A LC

JC

JC

JC

178

SECTION | 2 Mammalian orders

12.5.2 Immunoglobulin D IgD is an immunoglobulin whose function is unclear. An ortholog has been discovered in the catfish (Ictalurus punctatus) and some reptiles otherwise it has generally been considered to be present only in rodents and primates among the eutherians. An IGHD gene has also been identified in pigs, cows, sheep, giant pandas, and dogs. Analysis of this IGHD gene suggests that it is phylogenetically related to IgW found in lungfish and cartilaginous fish and is thus evolutionarily ancient. The platypus IGHD gene is very different from the IGHD gene found in eutherian mammals but more similar to the δ gene of the non-mammalian vertebrates, especially reptiles [38]. The platypus IGHD gene encodes a heavy chain containing ten CH domains without a hinge region and thus resembles the IGHD gene in reptiles, amphibians, and fish. Sequence-wise, the platypus IGHD gene closely resembles the leopard gecko (Eublepharis macularius) delta chain gene. The eutherian δCH1, δCH2, and δCH3 exons share sequence homology with platypus δCH1, δCH6, and δCH7.

12.5.3 Immunoglobulin O The IGHO (Ornithorhynchus) constant gene encodes a heavy chain with four novel CH domains and a hinge attached to the N-terminus of CH2 resulting in a structure that is different from the other known heavy chains. It appears to be structurally intermediate between the υ chain of avian IgY and the γ chain of IgG. It is a possible evolutionary precursor to IgG [39]. When first identified and sequenced it was classified as an IgY [38].

12.5.4 Immunoglobulin A Two highly divergent IgA isotypes have been identified in the platypus [40]. Sequence analysis indicates that these isotypes form a distinct branch that is separate from the eutherian IgAs. They do however have a hinge region and three constant domains. They may have arisen from gene duplication possibly preceding the metatherian/eutherian split. They may also have arisen as a result of recombination between an IGHM and IGHY gene [38]. One of these isotypes, IgA1 is expressed in the platypus but not in the spleen of the echidna although the gene is apparently present in the echidna.

12.5.5 Immunoglobulin G Echidna IgG and IgE have amino acid sequences that separates them from their placental counterparts [41]. Nevertheless, they still retain the basic immunoglobulin structure. Echidna IgG has a molecular weight of 150 kDa, a heavy chain of 50 kDa, and contains 2% carbohydrate. Light chains have a molecular weight of 22.5 kDa [42]. Two IgG isotypes have been identified in both platypus and the echidna. These two isotypes share fewer sequences than do human and mouse IgG suggesting that they diverged much earlier, - well before the placental radiation. Both of the IgG heavy chains, like other mammals, have three constant domains. The second domain in the IgG ancestral to platypus IgG1 and G2 was probably deleted prior to the splitting-off of the monotremes. The differences in the two IgG subclasses are reflected in the position of the disulfide bridge connecting the heavy and light chains. Thus in IgG1 the bridge is located in the N terminus of the CH1 domain whereas in IgG2 it is located at the C-terminus. This latter position is unique to platypus IgG2. The two platypus IgG isotypes also differ with respect to their number of N-linked glycosylation sites. IgG2 has four potential sites while IgG1 has only one. The single site on IgG1 is conserved among the other Ig isotypes as well as opossum IgG.

12.5.6 Immunoglobulin E IgE from Tachyglossus aculeatus has been cloned. Its overall structure including the presence of four constant domains, the position of its N-glycosylation sites, and overall charge distribution have been conserved. There is increased homology at its putative receptor-binding site reflecting the importance of Fc binding for its functions. Only minor changes in the overall structure of IgE have occurred since the separation of the monotremes from the other mammalian lineages. These changes include small insertions and deletions and alterations in the positions of disulfide bridges and glycosylation sites. However, due to the sheer number of single point mutations, the divergence among the nucleotide and amino acid sequences is substantial [41]. The amount of mRNA for IgE is relatively high in the platypus spleen. Only a sixfold difference in mRNA levels exists between free-living platypus IgG1 and IgE. However, it must be remembered that mRNA levels may not correlate well with the levels of their protein products Box 12.1.

The monotremes: echidnas and platypus Chapter | 12

179

BOX 12.1 Immunoglobulin Y. IgY is the major immunoglobulin class found in amphibians, reptiles, and birds. It appears to be the ancestral class from which mammalian IgG and IgE eventually evolved and thus played a key role in their evolution. It has the same domain architecture as IgM and IgE but it lacks a hinge region. However, the closely related platypus IgO does possess a hinge as do the other mammalian immunoglobulin classes. Platypus IgO was initially classified as IgY when first discovered. Its gene is in the expected location for IgY, and it has four constant domains as well as a hinge region. It appears that IgM and IgA evolved into IgY. The precise order of this is debatable but it is likely that IgM first gave rise to IgA and the IgA, in turn, gave rise to IgY. The crystal structure of IgY shows a close relationship with mammalian IgG and IgE. Its membrane and cytoplasmic exons also show close similarities to those of IgG and IgE. IgY in birds can bind to three specific Fc receptors. These receptors are structurally related to the mammalian Fc receptors FcαR1, FcRn, and FcγR, for IgA and IgG supporting the close relationship between IgY and the mammalian immunoglobulins. Zhang X, Calvert RA, Sutton BJ, Dore´ K. IgY: a key isotype in antibody evolution. Biol Rev 2017; 92:2144
2156

12.5.7 Light chains Monotremes use both kappa and lambda light chains however the λ chains account for more than 90% of the light chain transcripts in the spleen [43]. There are single κ and λ light chain loci in both the echidna and platypus. They contain 11 IGLV genes. There are probably at least three tandem Jλ-Cλ pairs located downstream from the IGLV gene cluster [42]. The platypus has only two IGLV subfamilies but as in the heavy chain V regions, their CDR3 segments are unusually long and diverse. The kappa chain locus contains a single IGKC gene in two allelic forms, and 8
10 IGKJ gene segments as well as 16 IGKV genes [43]. There appear to be at least four platypus and nine echidna IGKV subfamilies. These IGKV subfamilies are highly divergent within each species with some chains sharing as little as 57% nucleotide identity [44]. There are at least 29 IGKV gene segments in the platypus, as well as 5
6 IGKJ gene segments [45].

12.5.8 V region genes As discussed in Chapter 9, mammals collectively possess a very large number of IGHV genes that cluster into three major “clans” (I, II, and III). Comparative studies have shown that these three clans have probably existed for more than 400 million years. Fish IGHV genes are most closely related to mammalian clan III although they also possess two additional clans not found in mammals. The IGHV genes of birds (chickens), monotremes, marsupials, and some eutherians (rabbits, and pigs) also belong to clan III. This has led to the suggestion that clan III is the most ancient of the mammalian clans. There are at least 44 IGHV gene segments in the platypus and 11 IGHJ segments [38]. Some studies have suggested that the two monotreme species use different methods of generating V region diversity [36]. Thus the short-beaked echidna appears to use seven different IGHV region subfamilies from all three V region clans and suggests that it relies on using a large numberer of germline V genes to create its diversified VJ repertoire. In contrast to the Echidna, all 44 platypus IGHV genes belong to a single branch within mammalian clan III. They compensate for this by encoding significant V region sequence variation in their CDR3, the region where VJD recombination occurs. They have long and highly diversified D gene segments as well as 5
7 IGHJ gene segments. These, together with N-region addition generate a CDR3 region that ranges from 9 to 19 amino acids in length (mean 13.4). This is longer than the CDR3 in most eutherians [46]. This region also contains intrachain cysteine bridges, a feature also observed in the CR3 regions of camels and cattle (Chapters 14 and 17). These bridges appear to be required to stabilize the long CDR3 regions [47].

12.5.9 Immunoglobulin receptors Mammals possess four major types of IgG Fc receptors in addition to one high-affinity receptor for IgE and one each for IgA and IgM. All these receptors appear to have originated from a common ancestor by repeated gene duplications. It is likely that as the number of immunoglobulin isotypes increased during evolution, so too did the number of their receptors. Genomic studies have indicated that the polymeric-Ig receptors (pIgR), FcR-like receptors (FCRL), and the common FcR γ chain receptors first appeared in teleosts. Fc receptors for IgM and IgA/M receptors first appear in the platypus. The platypus does not appear to have IgG and IgE receptors. These latter receptors first appeared in

180

SECTION | 2 Mammalian orders

marsupials and eutherians while the receptors for IgA are only present in eutherian mammals. It appears that these receptors have two origins, classical IgG and IgE receptors evolved from Fc-Receptor-like (FCRL) precursors whereas IgM and IgA/M receptors probably arose from a pIgR precursor [48]. Eight FCRL proteins have been identified on the B cells of monotremes. Some bind immunoglobulins while others bind MHC molecules. They are distantly related to the NK cell receptors such as the natural killer Ig-like receptors (KIRs), the leukocyte Ig-like receptors (LILRS), and the leukocyte-associated Ig-like receptors (LAIRS) (Chapter 10). Two genes homologous to those encoding the high-affinity IgG receptor, FcγR1 have been identified on platypus chromosome 7. These two genes are FcγRIa and FcγRIαL together with two genes homologous to FCRL3L and the pIgR. No IgA receptor genes have been found in the platypus genome but FcμR and FcαμR genes are present.

12.6

T cells and cell-mediated immunity

Monotremes have a full complement of T cells each covered with antigen receptors (TCR). In addition to the eutherian alpha/beta and gamma/delta chains, non-eutherians also possess antigen receptors with a mu chain. Thus they have five TCR peptide chains available for use. The T cell receptors that recognize antigenic peptides carried within the groove of the MHC molecules fall conveniently into two pairs in eutherians: α and β chains and δ and γ chains. While the monotreme receptors originate from the same ancestral roots as the eutherian TCRs they have diverged over time (Fig. 12.6). The platypus genome contains about 200 V genes belonging to the five TCR loci. There are thirty-six such genes in TRBV, eight in TRMV, 89 in TRAV, 31 in TRDV, and 36 in TRGV. This is the largest number of TRGV genes reported in any mammal. Thus the platypus can generate many diverse T cell antigen receptors.

12.6.1 TRA and TRB genes The TRAV and TRBV genes diversified in a very different manner in monotremes and marsupials when compared to Eutherians [49]. In mammals, all TRAV genes are derived from five ancestral genes while their TRBV genes originate from four ancestors. The TRA locus has also been consistently associated with the TRD locus for a very long time. Monotreme and marsupial TRAV and TRBV gene sequences have been compared to those of primates. Both monotremes and marsupials have sequences related to four of the five ancestral primate TRAV genes (Vα I, II, III, and V). In the case of the platypus, these ancestral genes diversified into platypus- unique clades. The platypus TRAV locus contains 89 V genes. In the TRBV locus of monotremes, there are only seven functional V genes (plus one pseudogene). This is similar to the situation in reptiles. The echidna single TRA constant region shares about 37% amino acid sequence identity with eutherian TRAC genes. The two TRB constant region genes share about 63% nucleotide identity [50].

PLATYPUS

TRA/D

5’

TRG

TRM

31

2

7

53

VA

VD

DD

JD CD VD

JA

4

5

36

TRB

3’

89

36

7 5

V

D

J

C

4

5

CA

FIGURE 12.6 The organization of platypus T cell receptor genes. The structure of the TCRμ locus is derived from Wang X, et al. Platypus TCRμ provides insight into the origins and evolution of a uniquely Mammalian TCR locus. J Immunol 2011; 187: 5246
5254.

The monotremes: echidnas and platypus Chapter | 12

181

12.6.2 TRG The platypus has 36 TRGV genes that are classified into three subfamilies. They also have at least seven TRGJ genes and three TRGC genes. TRGV genes appear to have evolved from four ancestral genes. One is unique to eutherians the other three (Vγ I, II, and IV) are also found in monotremes and marsupials [49]. All mammalian TRDV genes originated from five ancestral genes of which three, Vδ III, IV, and V share a common origin with the TRVA genes. The platypus has three distinct V gene subfamilies encoding its expressed TRG chains, and these belong to a clade absent from eutherians and birds [51]. Each subfamily contains five V gene segments, and their overall divergence is limited, consistent with a recent emergence. The TRG locus also contains three C-region genes. These three TRGC chains have major variations in the length of their connecting peptide regions [52].

12.6.3 TRD The platypus TRD locus contains 31 V genes. One of these is a highly divergent V gene that is almost indistinguishable from an immunoglobulin heavy chain VHδ gene and is closely related to the V genes used in the TRMV locus. (This same gene is also present in the Florida manatee. Chapter 25) It appears that there may have been ancient and recurring translocations of gene segments between the IgG and TRDC genes as well as translocations of TRDC genes out of the TRA/D locus early in mammalian evolution to create the TRM locus [51].

12.6.4 TRM A population of T cell antigen receptors is found exclusively in non-eutherian mammals. These use a unique μ chain. Phylogenetic analysis indicates that the TRM locus is probably derived from the duplication and translocation of a TRD ancestor. As a result, monotremes and marsupials can generate T cells with γ/μ receptors. These are heterodimers with the γ chain containing two γ domains (Vγ 2 Cγ) while the μ chain consists of three domains as a result of utilizing two V domain genes (Vμ-Vμj-Cμ) [53]. (Fig. 12.7) These receptors have restricted diversity in their Vγ and Vμj domains but a highly diverse unpaired Vμ domain [54]. As in marsupials, they are expressed as a peptide chain containing double V domains. These V domains more closely resemble immunoglobulin V domains than do conventional TRVs [18]. The TRM locus contains two clusters of TRMV genes containing nine and six genes respectively. The TRMV genes are flanked by RSS containing a 23 bp spacer while the TRMJ genes have an RSS with a 12 bp spacer. Its genomic organization appears to closely resemble the ancestral form of these genes. Downstream of the TRMV1 cluster is a cluster of at least 35 TRMD genes [18]. Platypus TRM genes are not linked to those that encode the conventional TCR chains. The diversity in the double Vμ peptide chain is generated differently than in conventional TCR chains [53]. The N -terminal V domain is encoded by the somatically recombined V, D, and J genes. This generation of clonal diversity occurs within the thymus. These are related to the IGHV clade and may have resulted from a duplication of these immunoglobulin heavy chain V regions. Thus the N terminal V domain (V1) of the μ chain is encoded by the somatic recombination of V, D, and J segments within developing thymocytes. The second C-proximal V2 domain is encoded by an exon in which the VD, and J segments are already joined and as a result, are relatively invariant. The V1 domains are longer and more diverse since P chain

VP1 VP2 CP

J chain VJ CJ

FIGURE 12.7 The structure of platypus T cell γ/μ receptor. Note that the Vμ1 domain is unpaired. It is unclear what effect this might have on major histocompatibility complex or antigen binding.

182

SECTION | 2 Mammalian orders

they may also incorporate multiple D gene segments. This is the only known example of germline-joined VDJ genes being used in a TCR. The V genes in the platypus TRM locus belong to the mammalian clan III, the immunoglobulin IGHV family. Marsupials, however, do not fall within the three traditional IGHV clans [18]. However, the TRMV genes are orthologous in platypus and marsupials. Because there are no other V chain three-domain structures in these mammals, TRMV2 is believed to pair with the TRGV domain leaving the N-terminal TRMV1 domain unpaired Box 12.1. The TRM locus is also organized in tandem clusters which is not usual for TCR genes. The double V cluster feature resembles the immunoglobulin V genes rather than conventional TCR. As a result, it also requires two rounds of somatic DNA recombination to assemble it. While present in monotremes and marsupials TRM has been lost from the placental mammals. TRM has not been identified in the available amphibian, reptile, or avian genomes either. Since marsupials and monotremes diverged more than 190 mya, TRM clearly preceded that and possibly represents an ancestral mammalian T cell receptor lineage [18].

References [1] Messer M, Weiss AS, Shaw DC, Westerman M. Evolution of the monotremes: phylogenetic relationship to marsupials and eutherians, and estimation of divergence dates based on a-lactalbumin amino acid sequences. J Mammal Evol 1998;5(1):95
105. [2] Zhou Y, Shearwin-Whyatt L, Li J, Song Z, et al. Platypus and echidna genomes reveal mammalian biology and evolution. Nature 2021;592:756
62. [3] Van Rheede T, Bastiaans T, Boone DN, Hedges SB, et al. The platypus in its place: Nuclear genes and indels confirm the sister group relation of monotremes and therians. Mol Biol Evol 2005;23(3):587
97. [4] Griffiths M. The Platypus. Sci Am 1988;258(5):84
91. [5] Oftedal OT. The mammary gland and its origin during synapsid evolution. J Mamm Gland Biol Neopl 2002;7(3):225
52. [6] Urashima T, Inamori H, Fukuda K, Saito T, et al. 4-O-acetyl-sialic acid (Neu4,5Ac2) in acidic milk oligosaccharides of the platypus (Ornithorhynchus anatinus) and its evolutionary significance. Glycobiology 2015;25(6):683
97. [7] Brawand D, Wahli W, Kaessmann H. Loss of egg yolk genes in mammals and the origin of lactation and placentation. PLoS Biol 2008;6(3): e63. Available from: https://doi.org/10.1371/journal.pbio.oo60063. [8] Enjapoori AK, Grant TR, Nicol SC, Lefevre CM, et al. Monotreme lactation protein is highly expressed in monotreme milk and provides antimicrobial protection. Genome Biol Evol 2014;6(10):2754
73. [9] Stannard HJ, Miller RD, Old JM. Marsupial and monotreme milk
a review of its nutrient and immune properties. Peer J 2020;8:e9335. Available from: https://doi.org/10.7717/peerj.9335. [10] Warren WC, Hillier LW, Marshall Graves JA, Birney E, et al. Genome analysis of the platypus reveals unique signatures of evolution. Nature 2008;453(7192):175
83. Available from: https://doi.org/10.1038/nature06936. [11] Lynn DJ, Bradley DG. Discovery of α-defensins in basal mammals. Dev Comp Immunol 2007;31:963
7. [12] Whittington RJ, Grant TR. Hematological changes in the platypus (Ornithorhyncus anatinus) following capture. J Wildl Dis 1995;31 (3):386
90. [13] Hawkey CM. Comparative mammalian hematology. Cellular components and blood coagulation of captive wild animals. London: Heinemann Medical Books; 1975. [14] Canfield PJ, Whittington RJ. Morphological observations on the erythrocytes, leukocytes and platelets of free-living platypuses Ornithorhynchus anatinus (Shaw) (monotremata: Ornithothynchidae). Aust J Zool 1983;31(4):421
32. [15] Bolliger A, Backhouse TC. Blood studies on the Echidna Tachyglossus aculeatus. Aust J Zool 1983;31:421
32. [16] Zanetti M. Cathelicidins, multifunctional peptides of the innate immunity. J Leukoc Biol 2004;75(1):39
48. [17] Harrison GA, McNicol KA, Deane EM. Type I interferon genes from the egg-laying mammal, Tachyglossus aculeatus (Short-beaked echidna). Immunol Cell Biol 2004;82:112
18. [18] Wang X, Parra ZE, Miller RD. Platypus TCRμ provides insight into the origins and evolution of a uniquely mammalian TCR locus. J Immunol 2011;187:5246
54. [19] Haynes JI. The marsupial and monotreme thymus revisited. J Zool Lond 2001;253:167
73. [20] Connolly JH, Canfield PJ, McClure SJ, Whittingtton RJ. Histological and immunological investigation of lymphoid tissue in the platypus (Ornithorhynchus anatinus). J Anat 1999;195:161
71. [21] Udroiu I, Sgura A. The phylogeny of the spleen. Quart Rev Biol 2017;92(4):411
43. [22] Diener E, Ealey EHM, Legge JS. Phylogenetic studies on the immune response. III. Autoradiographic studies on the lymphoid system of the Australian echidna Tachyglossus aculaetus. Immunology. 1967;13:339
47. [23] Basir MA. Note on the histology of lymph nodes in Echidna. J Anat 1941;75(Pt 2):267
8. [24] Miska KB, Harrison GA, Hellman L, Miller RD. The major histocompatibility complex in monotremes: an analysis of the evolution of MHC class I genes across all three mammalian subclasses. Immunogenetics 2002;54:381
93. [25] Grutzner F, Rens W, Tsend-Ayush E, El-Mogharbel N, et al. In the platypus a meiotic chain of ten sex chromosomes shares genes with the bird Z and mammal X chromosomes. Nature 2004;432:913
17.

The monotremes: echidnas and platypus Chapter | 12

183

[26] Dohm JC, Tsend-Ayush E, Reinhardt R, Grutzner F, Himmelbauer H. Disruption and pseudoautosomal localization of the major histocompatibility complex in monotremes. Genome Biol 2007;8(8):R175. Available from: https://doi.org/10.1186/gb-2007-8-8-r175. [27] Papenfuss AT, Feng Z-P, Krasnec K, Deakin JE, et al. Marsupials and monotremes possess a novel family of MHC class I genes that is lost from the eutherian lineage. BMC Genomics 2015;16(1):535. Available from: https://doi.org/10.1186/s12864-015-1745-4. [28] Krasnek KV, Papenfuss AT, Miller RD. The UT family of MHC class I loci unique to non-eutherian mammals has limited polymorphism and tissue specific patterns of expression in the opossum. BMC Immunol 2016;. Available from: https://doi.org/10.1186/s12865-016-0181-9. [29] Lillie M, Woodward RE, Sanderson CE, Eldridge MDB, Belov K. Diversity at the major histocompatibility complex class II in the platypus, Ornithorhynchus anatinus. J Heredity 2012;103(4):467
78. [30] Belov K, Lam MKP, Hellman L, Colgan DJ. Evolution of the major histocompatibility complex: isolation of class IIb cDNAs from two monotremes, the platypus and the short-beaked echidna. Immunogenetics 2003;55:402
11. [31] O’Brien SJ. The platypus genome unraveled. Cell 2008;133:953
5. [32] Yoder JA, Litman GW. The phylogenetic origins of natural killer receptors and recognition: relationships, possibilities and realities. Immunogenetics 2011;63(3):123
41. [33] Wong ESW, Sanderson CE, Deakin JE, Whittington CM, et al. Identification of natural killer cell clusters in the platypus genome reveals an expansion of C-type lectin genes. Immunogenetics 2009;61:565
79. [34] Guselnikov SV, Taranin AV. Unraveling the LRC evolution in mammals: IGSF1 and A1BG provide the keys. Genome Biol Evol 2019;11 (6):1586
601. [35] Wronski V, Woods GM, Munday BL. Antibody response to sheep red blood cells in platypus and echidna. Comp Biochem Physiol A 2003;136:957
63. [36] Belov K, Hellman L, Cooper DW. Characterization of echidna IgM provides insights into the time of divergence of extant mammals. Dev Comp Immunol 2002;26:831
9. [37] Belov K, Hellman L. Platypus immunoglobulin M and the divergence of the two extant monotreme lineages. Austral Mammal 2003;25:87
94. [38] Gambo´n-Deza F, Sa´nchez-Espinel C, Magada´n-Mompo´ S. The immunoglobulin heavy chain locus in the platypus (Ornithorhynchus anatinus). Mol Immunol 2009;46(13):2515
23. Available from: https://doi.org/10.1016/j.molimm.2009.05.025. [39] Zhao Y, Cui H, Whittington CM, Wei Z, et al. Ornithorhinchus anatinus (Platypus) links the evolution of immunoglobulin genes in Eutherian mammals and nonmammalian tetrapods. J Immunol 2009;183:3285
93. [40] Vernersson M, Aveskogh M, Hellman L. Cloning of IgE from the echidna (Tachyglossus aculeatus) and a comparative analysis of ε chains from all three extant mammalian lineages. Dev Comp Immunol 2004;28:61
75. [41] Vernersson M, Aveskogh M, Munday B, Hellman L. Evidence for an early appearance of modern post-switch immunoglobulin isotypes in mammalian evolution. II; cloning of IgE, IgG1 and IgG2 from a monotreme, the duck billed platypus Ornithorhynchus anatinus. J Immunol 2002;32:2145
55. [42] Belov K, Hellman L. Immunoglobulin genetics of Ornithorhynchus anatinus (platypus) and Tachyglossus aculeatus (short-beaked echidna). Comp Biochem Physiol 2003;136:811
19. [43] Nowak MA, Parra ZE, Hellman L, Miller RD. The complexity of expressed kappa light chains in egg-laying mammals. Immunogenetics 2004;56(8):555
63. [44] Vernersson M, Belov K, Aveskogh M, Hellman L. Cloning and structural analysis of two highly divergent IgA isotypes, IgA1 and IgA2 from the duck-billed platypus Ornithorhynchus anatinus. Mol Immunol 2010;47:785
91. [45] Johansson J, Salazar JN, Aveskogh M, Munday B, et al. High variability in complementarity-determining regions compensates for a low number of V gene families in the λ light chain locus of the platypus. Eur J Immunol 2005;35:3008
19. [46] Miller RD, Hansen VL. Marsupial and monotreme immunoglobulin genetics. Chapter 4 in the Comparative Immunoglobulin Genetics book (2014). [47] Johansson J, Aveskogh M, Munday B, Hellman L. Heavy chain V region diversity in the duck-billed platypus (Ornithorhynchus anatinus): long and highly variable complementarity-determining region 3 compensates for limited germline diversity. J Immunol 2002;168:5155
62. [48] Akula S, Mohammadamin S, Hellman L. Fc receptors for immunoglobulins and their appearance during vertebrate evolution. PLoS One 2014;9 (5):e96903. Available from: https://doi.org/10.1371/joutnal.pone.0096903. [49] Olivieri DN, Gambon-Cerda S, Gambon-Deza F. Evolution of V genes from the TRV loci of mammals. Immunogenetics 2015;67:371
84. [50] Belov MK, Miller DR, Ilijeski A, Hellman L, Harrison GA. Isolation of monotreme T-cell receptor α and β chains. Immunogenetics 2004;56:164
9. [51] Parra ZE, Lillie M, Miller RD. A model for the evolution of the mammalian T-cell receptor α/δ and μ based on evidence from the duckbill platypus. Mol Biol Evol 2012;29(10):3205
14. [52] Parra ZE, Arnold T, Nowak MA, Hellman L, Miller RD. TCR gamma chain diversity in the spleen of the duckbill platypus (Ornithorhynchus anatinus). Dev Comp Immunol 2006;30:699
710. [53] Miller RD. Those other mammals: the immunoglobulins and T cell receptors of marsupials and monotremes. Semin Immunol 2009;22(1):3
9. [54] Morrisey KA, Wegrecki M, Praveena T, Hansen VL, et al. The molecular assembly of the marsupial γδ T cell receptor defines a third T cell lineage. Science 2021;371:1383
8.

This page intentionally left blank

Chapter 13

Marsupials: Opossums to Kangaroos

Common Wallaroo. O. robustus.

Marsupials constitute about 7% of the world’s living mammalian species. The distinctive feature common to most (but not all) marsupials is the possession of a pouch in which their highly altricial young develop. This is a consequence of a very short gestation period that avoids immunological rejection of the fetus. Marsupials are a clade that originated from the last common ancestor of the Metatherians about 180 mya. There are about 330 extant species of marsupial. About 70% of these are found on the Australian continent including Tasmania and New Guinea. The remaining 113 species are natives of South America with some extending north into Central and North America. As a result, the marsupials can be subdivided into those in the Americas—Superorder Ameridelphia and those in Australasia—Superorder Australidelphia (Fig. 13.1). Currently, it is believed that the marsupials probably originated in Gondwana, around 187
166 mya. As the continents drifted, they prospered in the region that became South America. Evidence suggests that they reached Australia from South America by way of Antarctica in a single migration occurring around 65 mya. One theory suggests that a small marsupial opossum from South America, related to the monito del monte (Dromiciops gliroides)—a member of the Microbiotheria, succeeded in reaching Australasia before it separated from Antarctica. Thus all living marsupials have South American ancestors. The American species include the family Didelphidae, the opossums, and the Paucituberculata the shrew opossums. In the absence of eutherian mammals, the marsupials in Australia were able to diversify into the numerous unique species present today; the order Dasyuromorphia including the Tasmanian devil (Sarcophilus harrisii), and the numbat (Myrmecobius fasciatus); the order Notoryctenirphia, the marsupial moles; the order Peramelemorphia including the bandicoots and bilbies; and the order Diprotodontia which includes all the other 148 extant species (including the wombats, koalas (Phascolarctos cinerus), possums, gliders, kangaroos, and wallabies) [1]. (By convention, North American species are called opossums, and Australian species are called possums).

Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00016-2 © 2023 Elsevier Inc. All rights reserved.

185

186

SECTION | 2 Mammalian orders

MARSUPIAL PHYLOGENY 80 70

60 50

AMERICANS

40 30

20 10

0

Million years ago

Opossums

FIGURE 13.1 A phylogeny of the marsupials. As mentioned in the text of this chapter. The Australian species all originated from an ancestral Microbiotherian that probably originated in what is now South America but succeeded in colonizing Australia prior to the final breakup of the southern continents.

Monito del monte

Marsupial moles, Bilbies, Bandicoots

AUSTRALIANS

Dasyurids,Tasmanian devil Possums Wombats, Koalas

Diprotodonts

Macropods Kangaroos Wallabies Gliders

Marsupials are highly diverse with enormous variations in size and life history. They range in size from less than 5 g in the Planigales (Planigale spp) to over 80 kg in the red kangaroo (Macropus rufus). Their lifespan ranges from less than a year in some small male Dasyurids (carnivorous marsupials) to up to 27 years for the large kangaroos. Herbivorous species may be K strategists and produce fewer than one young each year whereas some marsupial insectivores have litters larger than ten and an interbirth interval shorter than three months [2]. The defining feature of female marsupials is a triple vagina. All marsupials have placentas, and many have pouches. All marsupials have other common features such as a very short gestation period (9
42 days) and give birth to highly altricial young that weigh less than 1% of the mass of the mother. As a result of this short gestation period, newborn marsupials essentially lack a functioning adaptive immune system. They must therefore complete their development either within their mother’s pouch or otherwise closely attached to her. Some marsupials such as the Caenolestids (shrew-opossums), some Didelphids (opossums), and most Dasyurids, do not have a fully developed or enclosed pouch. Instead, they use raised skin folds that develop during gestation and have rings of muscle surrounding the teats that contract during lactation to tightly bind the suckling young.

13.1

Reproduction and lactation

Marsupial embryos, like eutherians, develop within the uterus and receive oxygen and nutrients through a placenta. Their pregnancy is however very short, and their placenta is much less invasive than in eutherians. As the fertilized ovum passes down the oviduct it is covered by a glycoprotein mucoid coat and a shell membrane. Thus for much of their development, the embryos are covered by a maternally-secreted protein layer called the “shell membrane” but no shell. As a result, placental invasion of the maternal endometrium does not occur until relatively late in pregnancy. As in eutherians, the marsupial placenta expresses non-classical MHC class I antigens that minimize immune rejection. Morphological changes in the pregnant uterus mirror those seen in eutherians. Transcriptomic studies on the uteruses of pregnant opossums show that an inflammatory reaction develops at the maternal-fetal interface at a time corresponding to the breakdown of the shell membrane [3]. It appears that, at least in the opossum, this inflammatory reaction may be a trigger for the subsequent parturition. (It also suggests that the key to a successful prolonged pregnancy in eutherians is the suppression of this inflammatory response). The median vagina opens into a urogenital sinus and serves as a transient birth canal. Marsupials have very altricial young that, following birth, migrate immediately to the mother’s mammary skin or pouch where they attach tightly to her teats and complete the development process in what can be considered to be an extrauterine gestation (Fig. 13.2). The newborn young are required to crawl from the urogenital opening, (a site within millimeters of the maternal anus), through the abdominal fur, to their mother’s pouch. Newborn marsupials do not have a functioning immune system but are exposed to a diverse array of microorganisms while developing outside the uterus. Protection of the developing young comes from passive immunity through milk, prenatal immunoglobulin transfer, the

Marsupials: Opossums to Kangaroos Chapter | 13

187

TAMMAR WALLABY LACTATION Birth 25 day pregnancy

Permanently Attached to teat

Feeds intermittently

200 days

100 days

IgG transfered

20

40

60

80

38 5 60 Mature cervical thymus white pulp in spleen lymphocytes in thymus and\lymph node 10-15 40 60 first immune Lymph node Mature thoracic responses follicles thymus 45-55 14 cathelicidin expression Hematopoiesis in skin

120 90 Germinal centers in lymph nodes 97 Serum IgG rises

120 spleen germinal centers 120 lymphocyte clusters in intestine

FIGURE 13.2 The development of the immune system within the pouch of the tammar wallaby (Notamacropus eugenii) [8].

presence of multiple antimicrobial compounds secreted by the pouch epithelium, and repeated cleaning as a result of maternal licking. Young marsupials also possess many innate defenses, especially antimicrobial peptides. There does not appear to be any marsupial equivalent of colostrum. Only one marsupial species, the tammar wallaby (Notamacropus eugenii), has been shown to transfer maternal IgG to the fetus via the yolk sac [4]. Marsupial young are however born with an additional outer layer of epidermis consisting of a partially keratinized layer of squamous cells called the periderm [5]. This serves as a protective coating probably serving to reduce desiccation as well as acting as an antibacterial barrier. It is lost about one week after birth in D. virginianus and Dasyurus hallucatus [6,7]. Similar developmental stages are also seen in the common brushtail possum (Trichosurus. vulpecula). Additionally, these joeys are born without the ability to maintain their body temperature. They develop homeothermy and endothermy after 100 days within the warm pouch.

13.1.1 Protection in the pouch During their migration from the urogenital opening and after arrival at the pouch, the young are confronted with enormous microbial challenges. For example, females groom the pouch with their tongue in the period leading up to birth suggesting that oral bacteria are almost certainly present (Fig. 13.3) [5]. The pouch contains its diverse microbiota including both gram-positive and -negative bacteria. This has been extensively studied in the tammar wallaby (N. eugenii), the brushtail possum (T. vulpecula), the koala (P. cinereus), the quokka (Setonix brachyurus), and the Tasmanian devil (S. harrisii). The culture of brushtail possum pouches grew 46 Gram-positive and 20 Gram-negative bacterial species. 16S RNA screening identified 227 separate clones. Actinobacteria were the predominant phylum accounting for over 78% of the microbial diversity. Studies of the Tasmanian devil pouch by 16S RNA screening identified 1907 phylotypes. The devil pouch microbiota was dominated by Firmicutes (36%), and Proteobacteria (34%), with fewer Fusobacteria, Bacteroidetes, and Actinobacteria. In addition, of course, some bacteria can be acquired from the urogenital tract and on the ventral fur through which the neonate must crawl to reach the pouch. The composition of the pouch microbiota also changes as lactation proceeds, the young develop, and waste accumulates. The pouch in addition to being a physical shelter also provides chemical protection. Thus antibacterial peptides such as multiple cathelicidins, dermcidin, eugenin, and cystatin C (a cysteine protease inhibitor) are present in pouch secretions. Tasmanian devil pouch secretions include cathelicidins that can kill Staph aureus and Enterococcus faecalis.

188

SECTION | 2 Mammalian orders

FIGURE 13.3 Some of the antimicrobial factors and other protective factors found in marsupial pouches that presumably assist in controlling potential infections in the newborn.

MILK Immunoglobulins Transferrin Lysozyme Complement Cytokines Tricosurin COMMENSAL MICROBIOTA Lactobacilli

MATERNAL Salivary IgA SECRETED Defensins Eugenin Cystatin C Dermcidin Cathelicidins

Eugenin, a small peptide [pGlu-Gln-Asp-Tyr(SO3)-Val-Phe-Met-His-Pro-Phe-NH2] has been isolated from the pouch of the tammar wallaby during the early lactation period. It can contract smooth muscle, but eugenin also enhances the proliferation of splenocytes by acting through cholecystokinin receptors. It appears therefore to be an immunomodulatory peptide that may also serve to control the pouch microbiota [9]. The mammary transcriptome of koalas reveals many immune system transcripts including immunoglobulins, lysozyme, cytokines, MHC classes I and II, and complement components as well as antimicrobial peptides such as the cathelicidins [10]. All marsupial immunoglobulin heavy chains are present in the mammary transcriptome except for IgE. IgA is most abundant in the late milk samples as is the pIgR (secretory component). Koala milk contains four cathelicidins. Three highly abundant complement components are C2, C3, and C4a (2.4% of peptides). Koala milk also has a relatively high concentration of trichosurin, and alpha-1-B glycoprotein potentially linked to the specialized eucalyptus diet as well as very early lactation protein that may also have an antimicrobial role. (Trichosurin is a highly conserved protein found in the milk of marsupials throughout lactation. Its function is unclear. It may act as a chemical attractant to guide the newborn animal to the pouch and teat. It can also bind phenolic compounds and may also serve to neutralize plant toxins.) [11] The pouch microbiota is controlled largely by the local release of antimicrobial compounds, most notably lysozyme and dermcidin. Lysozyme degrades the cell wall of Gram-positive bacteria. It is the most highly expressed gene in midlactation milk in the Tasmanian devil. Dermcidin is a potent antibacterial protein secreted by sweat glands in eutherians [2]. Diverse other antimicrobial agents such as immunoglobulins, transferrin, and lysozyme, in addition to the cathelicidins, are found in marsupial milk. As pointed out elsewhere, in the short-tailed opossum (Monodelphis domestica) the concentration of lysozyme in milk increases progressively throughout lactation. Cathelicidin genes are expressed in the skin and pouch lining of the Tasmanian devil. These cathelicidins may be selective in their bactericidal activities thus permitting commensal survival. The overall effect of all these antibacterial molecules is to suppress pathogen growth and regulate the relative abundance of the commensal bacterial population. Some lactobacilli may produce lactic acid and so reduce the pouch pH like the eutherian vagina and hence inhibit fungal growth. The tammar wallaby possesses 14 cathelicidin genes that are expressed in the mammary gland in early lactation [12]. Other proteins detected in pouch washes of tammar wallabies and common wombats (Vombatus ursinus) include β-lactoglobulin, α-lactalbumin, hornerin, eugenin, and dermcidin [5].

13.1.2 Lactation The progressive and prolonged extrauterine development of marsupial pouch young ensures that their milk composition must change over time to meet the changing needs of the neonate. In general, early in lactation, their milk is somewhat dilute. It contains relatively low levels of proteins and fats but high levels of carbohydrates. Later in lactation, the composition reverses so that carbohydrates drop while proteins and fats increase. In addition, early in lactation certain proteins may predominate but they may then disappear to be replaced by others. This may happen several times. Thus there are multiple development /lactation stages seen in pouch young marsupials.

Marsupials: Opossums to Kangaroos Chapter | 13

189

13.1.3 Ameridelphia As might be anticipated, the early divergence and radiation of the marsupials on two different continents have ensured that significant differences exist between the American and Australasian marsupials. The American marsupials are typified by the gray short-tailed opossum, M. domestica. Like the other marsupials, this species’ extremely short gestation period (14.5 days) ensures that its altricial pouch young spend a very long-time suckling. This is in contrast to placentals where suckling takes place after the birth of precocious young. When eutherian young suckle their first milk, the colostrum loads their circulation with immunoglobulins. This is a temporary state until the newborn can begin to synthesize its own antibodies. Marsupial pouch young, on the other hand, require ongoing protection during the entire time they are developing. However, the short-tailed opossum lacks a pouch, and as a result, the behavior of its young is different from the Australian marsupials. Some opossum species carry their young in vestigial pouches that are skin folds. In the case of those that have no pouch, the young must firmly attach to teats using their mouths and front paws. These neonatal opossums lack most of the cells required for an adaptive immune response and there is no evidence of any passive transfer of immunoglobulins early in lactation. As pointed out above, marsupial young develop most of their immune system postnatally [13]. In opossums, transcription of the B cell receptor signal transduction molecules CD79a and CD79b, can be detected in embryonic tissue prior to birth. V(D)J gene recombination in B cells begins shortly after birth, but Ig heavy chain transcription is not detected until 24 hours postnatally while transcription of Ig light chains is not detectable for 7 days. IgG synthesis is first detected in the neonatal opossum at about five weeks of age. This delayed onset is due to a lack of expression of the enzyme terminal deoxynucleotidyl transferase in early developing B cells. Studies on opossums and their secretions have also shown that IgA transcripts are abundant in their mammary tissues and together with IgM, increase progressively from birth to weaning. In addition, the polymeric Fc receptor, (pIgR), responsible for transporting IgA across epithelial surfaces increases in expression throughout lactation. Neonatal opossums do not make their own IgA until eight weeks of age just prior to weaning. The opossum mammary gland contains few transcripts for IgG and IgE. Even though opossum milk contains IgG, it appears that although IgA and IgM are locally synthesized within the mammary gland, IgG is more likely to be actively transferred from the maternal circulation [14]. The neonatal immunoglobulin receptor FcRn is also expressed in the mammary glands early in lactation with peak levels being reached about week four, just before neonatal IgG synthesis commences. There is no evidence that a colostral phase as such, occurs in opossums. γ/δ T cells predominate in opossum mammary tissues during lactation [15]. In a study comparing M. domestica with the brushtailed possum T. vulpecula, IgA transcripts have been detected in T. vulpecula as early as postnatal day 18, prior to IgG whereas in M. domestica IgA1 B cells were not detected until week eight in both the spleen and the intestine.

13.1.4 Australidelphia Lactation in macropod marsupials can be divided into four phases. The quokka (S. brachyurus) gives birth to and suckles its developing young for about 300 days. For the first 70 days during Phase 1, the joey is permanently attached to the teat. From 70 to 180 days the joey lives in the pouch and suckles intermittently (Phase 2). From 180 to 200 days the joey makes excursions outside the pouch (Phase 3). After 200 days the joey only returns to the pouch to suckle until lactation ceases (Phase 4). Over the entire lactation period, a joey increases in weight from 0.3 to 1689 or 2024 g for females and males respectively. This is a 5600- and 6700-fold increase during lactation. Eutherians such as rabbits, kittens, calves, and humans increase only four to sixfold reflecting their advanced state of maturity, the capacity of their uterus, and the large size of their neonates [16]. Another macropod, the tammar wallaby (N. eugenii) has a pregnancy of 26 days that is followed by a lactation of 300
350 days. The very small joey attaches to the teat in its mother’s pouch and spends its first 100 days permanently attached (Phase 2). From 100 to 200 days, it is intermittently attached (Phase 3) [17]. The beginning of Phase 3 marks the first exit from the pouch and is thus somewhat similar to the birth of a placental mammal. It is at this time that the young wallaby begins to supplement its diet with plant material. As described above, opossums lack this biphasic immunoglobulin secretion pattern. Over this prolonged lactation period the composition of proteins and nutrients in the mother’s milk changes in a predictable fashion [18]. The tammar wallaby undergoes two periods in which increased immunoglobulin transfer from the mother to the joey occurs. The first is in phase 1, immediately postpartum. Thus up to 48 hours postpartum the milk is characterized by high levels of IgA, IgG, lipid, and protein. The second is when the joey exits the pouch (Phase 3) and is exposed to the external environment and the world of environmental microbes for the first time [19]. (Fig. 13.4) At this time, release from the teat is associated with increases in serum IgG. The developing joey lacks a functioning

190

SECTION | 2 Mammalian orders

_ _ 1 _ 0.5 o 0 _o o -.5 _ -1 _ 1.5

Effect

o o

o

o

FIGURE 13.4 Changes in the expression levels of IgG in the milk of the tammar wallaby throughout lactation. The increased expression occurs immediately after birth and later at the beginning of phase 3 when the young joey is beginning to leave the pouch [19]. From Daly KA, et al. Analysis of the expression of immunoglobulins throughout lactation suggests two periods of immune transfer in the tammar wallaby (Macropus eugenii). Vet Immunol Immunopathol 2007; 120:187
200. With permission.

o

-1.5_

o l 0

l 50

l 100

l 150

l l l 200 250 300 Days after birth

adaptive immune system until well after 100 days in the pouch. Gut closure, the time when the intestinal tract can no longer absorb large macromolecules such as immunoglobulins, does not occur until 170
180 days of age. Thus compared to eutherians the length of the immunoglobulin passive transfer period is much longer in marsupials. In wallaroos (Osphranter robustus), for the first 90
100 days of pouch life, blood levels of IgG are low. Thereafter these levels rise rapidly to reach adult levels by the time that the young leaves the pouch [20]. IgA is also present in tammar wallaby milk, but it does not cross the intestinal epithelium. It persists in the gut where it prevents invasion by pathogens and probably promotes the selective growth of the commensal microbiota.

13.1.4.1 Immune cells Immune cells are present in marsupial milk and may provide another layer of protection [5]. Thus when born, the young joey lacks both circulating lymphocytes and organized lymphoid tissues. As a result, it is dependent on maternal neutrophils, lymphocytes, and macrophages, supplied in the first milk. Neutrophils are present in high numbers in tammar wallaby milk up to 105 days postpartum. On the other hand, neutrophils do not appear in quokka milk until 20 days postpartum. Lymphocytes are present at 12 days postpartum in quokka milk and up to 250 days postpartum in tammar wallaby milk. Macrophages are found in the milk of tammar wallabies after 105 days [5].

13.2

Hematology

The primary hematopoietic organ of newborn marsupials is the liver. As they develop within the pouch the bone marrow gradually assumes the hematopoietic role [21]. The leukocytes of marsupials are, in general, typically mammalian in numbers, functions, and morphology. Some unusual features include small numbers of granules in neutrophils, some elongated eosinophil granules, and many irregular nuclei in monocytes [22]. (Table 13.1)

13.3

Innate immunity

Marsupials diverged from the eutherian mammals about 180 mya (A debatable timing). The complete sequencing of the genome of the opossum, M. domestica has allowed its immunome to be examined in detail. It contains members of all the key immune gene families. There have been substantial duplication or gene conversion events involving leukocyte receptors, NK receptor complexes, immunoglobulins, type I interferons, and defensins [12]. A study of serum bacterial killing activity in five marsupial species (Brushtail possum, Eastern grey kangaroo, Tasmanian devil, ringtail possums and koalas), has shown great variation in the ability of their serum to kill Escherichia coli in vitro. These differences are independent of the animal’s social structure, captivity status, sex, or phylogenetic distance but are associated with diet and body size. For example, the two eucalypt-eating specialists, the koala and the ringtail possums have the lowest serum bacterial killing ability while the brushtail possum and the Eastern grey kangaroo had the greatest [28]. Antimicrobial peptides play a key role in protecting marsupial young during their early vulnerable period of development. The opossum genome contains 12 cathelicidins, 32 beta-defensins, and a single alpha-defensin gene [29]. The 12

TABLE 13.1 The numbers of blood leukocytes of selected Marsupial species. These vary seasonally as well as by sex, nutrition, reproductive, and lactation status. Species

Quokka [23]

Common Wombat [23]

Red kangaroo [24]

Eastern gray kangaroo [25]

Brushtail possum [22]

Monodelphis domestica [26]

Koala [27]

Total WBCs 3 103 /μL

13

8.8

3.8
7.1

3
13

13.4

13.5
16.8

2.8
11.2

Neutrophils (%)

86

10

50
62

50
60

43

23
28

48

Lymphocytes (%)

12

76

23
34

20
34

50

48
52

44

Monocytes (%)

1

13

7

7
7.6

4.6

14
19

2.8

Eosinophils (%)

1

1

2
7

2.6
7.6

1.6

6
10

5.7

Basophils (%)

0

0

4

4

,1

,1

,1

192

SECTION | 2 Mammalian orders

cathelicidin genes are highly diverse with amino acid identities in the mature peptides ranging from 1% to 94%. This presumably reflects positive selection for those molecules critical to the survival of developing young. There are also 14 cathelicidin genes in the tammar wallaby genome. They are expressed in the mammary gland during early lactation and can kill a wide range of potential pathogens [12]. Thirty-seven defensin genes and pseudogenes have been identified in the opossum genome. Many can be assigned to known eutherian defensin gene categories but there is a large number that are opossum-specific. The presence of the single alpha defensin gene in M. domestica suggests that the duplication of beta-defensin genes occurred before the marsupial/eutherian divergence [29,30]. Defensin genes are abundant in the Australidelphians. Thus there are 48 in the Tasmanian devil, 34 in the koala, and 39 in the tammar wallaby. Gene duplication has driven unique and species-specific expansions They are found in three genomic clusters [31]. Koalas appear to lack the toll-like receptors TLR1 and TLR13 [32]. However, TLR13 is present in M domestica and in the Tasmanian devil.

13.3.1 Cytokines The sequencing of the opossum genome has enabled investigators to identify the presence of multiple cytokines in this species [33]. Thus IFN-γ, IL-2, IL-4, IL-6, IL-12, and IL-13 are readily identified despite having only limited sequence identity with their eutherian counterparts. Additional analysis identified 36 chemokine genes including a lineagespecific expansion of the macrophage inflammatory protein family [34]. Other divergent cytokines identified include IL-7, IL-9, IL-31, IL-33, and CSF2. All this suggests that diverse opossum immune responses involving multiple functional T cell subpopulations are governed by cytokines in a manner similar to that in eutherians. Interferons are also produced by Australian marsupials with some interesting differences [8]. For example, the tammar wallaby possesses a large number of type I IFN genes, similar to eutherians. These include 10 to 12 related to IFNA and about four related to IFNB. Some of these are probably pseudogenes. Many of these genes likely encode functional class I IFN subtypes. This is distinctly different from the echidna which has only three such genes (Chapter 12).

13.4

Lymphoid organs

Marsupials possess all the key organs associated with the mammalian lymphoid system.

13.4.1 Thymus When it leaves the uterus, the newborn marsupial has yet to develop any clearly defined lymphoid tissues although has a well-developed jaw to allow it to attach firmly to the teat and strong forelimbs that enable it to climb into the pouch [35]. Presumably, it also has some innate protection, but it largely relies on the maternal defenses operating within the pouch environment. Things start to happen only after the young enter the pouch and attaches to the teat. In general, the thymus develops first followed by the secondary lymphoid tissues. Kangaroos are different since they possess two pairs of thymuses. One pair is located within the thoracic cavity. The other pair is found in the ventral cervical region. Depending on the species, they may be present in one or both locations. Given the diversity of the marsupials, it is unsurprising that there are also significant differences in gross thymic morphology [36]. Thus in the Polyprotodonts such as the Dasyurids and bandicoots there is only a thoracic thymus [37]. Diprotodonts (the macropods and possums) have both thoracic and cervical thymuses. At the other extreme, wombats (Vombatus spp) and koalas have only cervical thymuses. Among the Ameridelphians, the opossums have paired thoracic thymuses only. Irrespective of their location, there appear to be minimal functional and structural differences between the thymuses in these species. Their structure resembles that of eutherians with a distinct cortex and medulla, and obvious Hassall’s corpuscles. The cervical thymus generally develops and matures before the thoracic thymus [38]. Likewise, involution results in the replacement of thymic tissues with fibrous tissue or adipose tissue [39]. In many species, the thymus and parathyroid tissues are intermingled [40]. Neonatal thymectomy in the quokka results in lymphocyte depletion in lymph nodes and spleen [41]. It also results in a significantly decreased lifespan. Depending on the species, lobulation of the thymus occurs within the first 7 days after birth [35]. Hassall’s corpuscles appear slightly earlier in Ameridelphians than in Australidelphians (days 4
5 rather than days 13
21). Thus in opossums, lymphocytes first appear in the thymus at 2 days. They appear in the blood and lymph nodes around 6 days, in the bone marrow at 10
12 days, and in the spleen at 17
22 days after birth [42]. The thymus develops rapidly and

Marsupials: Opossums to Kangaroos Chapter | 13

193

resembles the adult organ by about 32 days. The secondary lymphoid organs develop much more slowly. Thus lymph nodes only develop germinal centers by about 65 days when the young opossums become independent of the mother.

13.4.2 Bone marrow At birth, the largely cartilaginous bones of the neonatal opossum lack a bone marrow. However, it develops rapidly. As a result, hematopoiesis begins to occur in the bone marrow between 7
14 days after birth.

13.4.3 Spleen Splenic hematopoiesis is first seen on days 4
7. The differentiation of splenic tissue into red and white pulp occurs on days 10
12 of age in the Virginia opossum but on days 40
60 in the tammar wallaby, quokka, and the dunnart (Sminthopsis macroura). The spleen is fully functional by the time the young leave the pouch [43]. The marsupial spleen is trilobed (Y shaped) as in monotremes and is enclosed in a thick monolayered capsule with a poorly developed trabecular system (Fig. 13.5). [44]. There is a marginal zone, the venous vessels are nonsinusal, and the intermediate circulation is open. Splenic erythropoiesis occurs in the Ameridelphia such as the opossums but not in the Australidelphia. Interestingly, the relative splenic weight in the Ameridelphia is about twice that in the Australidelphia [44]. CD31 and CD51 cells are present in large numbers in the periarteriolar lymphoid sheaths of koalas, brushtail possums, and ringtail possums. They are also scattered through the white and red pulp. MHC class II positive cells are mainly found in germinal centers and splenic cords [37].

13.4.4 Lymph nodes Lymph nodes have been described in many marsupial species. They tend to be fewer than in eutherians and they are not arranged in clusters [38]. In general, they conform to the usual mammalian pattern with a distinct cortex and medulla, with primary and secondary follicles in the cortex, and an obvious cellular paracortex. CD3- and CD5- positive T cells are densely packed in the interfollicular areas and paracortex. Plasma cells are most numerous in the periphery of the follicles and the medullary cords. MHC class II positive cells are most numerous in the follicular mantles, germinal centers, the non-follicular cortex, and the medullary cords [37] (Fig. 13.6). Mesenteric lymph nodes in eastern gray kangaroos have a similar structure to those in eutherian mammals [45]. They have a distinct cortex and medulla with prominent follicles and germinal centers. T cells are mainly located in the cortex and paracortex. B cells are found in the marginal zones of the follicles. Lymph aggregates appear around lymphatic vessels in the pouchless opossum and the quokka at 5 days of age but do not differentiate into cortex and medulla until 10
20 days. They appear in the northern brown bandicoot (Isoodon macrourus) and the stripe faced dunnart (S. macroura) around day 30 [35]. Germinal centers generally appear between days 60 and 100. In both the quokka and the opossum, the cervical lymph nodes develop before more caudal ones [42]. There are some differences in the details of the anatomical location and lymphatic drainage of nodes when compared to those in eutherians [46].

FIGURE 13.5 A section of the spleen of a Virginia opossum (Didelphis virginiana). It shows significant hematopoiesis. Courtesy of Dr Brian Porter.

194

SECTION | 2 Mammalian orders

FIGURE 13.6 lymph node from a Virginia opossum (D. virginiana). It appears to be highly active with many lymphoid follicles. probably because this was a heavily parasitized animal. Courtesy of Dr Brian Porter.

13.4.5 Gut-associated lymphoid tissues Oropharyngeal tonsils have been described in the koala, brushtail possum, and ringtail possum. They contain many lymphoid follicles oriented towards the overlying epithelium which is infiltrated with lymphocytes. Koala tonsils have deep crypts unlike those of possums. By contrast, the tonsils of the kowari (Dasyuroides byrnie), a small carnivorous marsupial, have been reported to consist of a single follicle overlaid by infiltrated epithelium [37]. Peyer’s patches may be difficult to identify grossly since, as in monotremes, they do not cause obvious protrusions above the gut epithelium. These submucosal follicles have been described in many different marsupial species. In Australidelphia, they consist of multiple closely packed lymphoid follicles with well-developed germinal centers and mantle zones orientated towards the epithelium. They have dense interfollicular accumulations of lymphocytes. In contrast, Ameridelphians such as the opossums lack well-defined domes, caps, or mantle zones. T cells are found in the peripheral regions of the germinal centers in secondary follicles whereas B cells are abundant in the primary follicles [45]. Peyer’s patches first appear on day 42 in the quokka and around day 100 in the tammar wallaby. Cecocolic lymphoid patches have been described in the koala as well as brushtail and ringtail possums. They have a similar structure to the intestinal Peyer’s patches. In koalas, they are covered by epithelial M cells. The distribution of T and B cells in these mucosal lymphoid tissues is similar to that seen in eutherians [37]. Two marsupials, the koala and the greater glider (Petauroides volans) have long tapering ceca. These appear superficially similar to the rabbit appendix. However, in the koala, it is not sharply demarcated from the cecum and cannot be considered a true appendix. The greater glider in contrast has an abrupt stricture delineating the presence of an appendix. A long tapering cecum may have a similar function to an appendix [47].

13.5

The marsupial MHC

13.5.1 Opossum Analysis of the first completely sequenced marsupial genome, the gray short-tailed opossum (M. domestica) has demonstrated that, as in other mammals, the marsupial MHC is dense and complex [48]. It also shares many features with nonmammals. The class I genes have also been amplified within the class II region resulting in a unique class I/II region [49]. The MHC is the most gene-dense and polymorphic region of the mammalian genome. The MHC gene products serve as antigen-presenting molecules and as a result, its genes are associated with resistance to infectious disease, immunologic diseases such as autoimmunity, and reproductive success. MHC genes are grouped into three classes, I, II, and III. These three classes are typically arranged along the chromosome in the order I -III- II. The eutherian MHC is large and dense. For example, the human MHC contains 264 genes and pseudogenes over a length of 3.6 mb. Another feature of the eutherian MHC is the presence of many framework genes. The number and order of these framework genes are conserved between species. In between them, the class I genes have independently expanded and diversified. The M. domestica MHC is similar in size and gene content to eutherian MHCs. It is located on chromosome 2q. However, it is organized in a manner more similar to that seen in amphibians and birds. The main difference between the MHC of opossums and eutherians is the location of the class I genes. (Fig. 13.7) Thus the opossum has a class I/II region that contains interspersed class I and class II genes. It contains eleven putative class I and ten class II genes. The

Marsupials: Opossums to Kangaroos Chapter | 13

195

class II loci also contain the non-classical DMA, and DMB genes Three marsupial specific class II gene families are also present, including DA, DB, and a family designated DC [49]. The opossum class II loci are not separated from class I by the class III region. This suggests that the class I and II genes may also have been adjacent in the mammalian ancestor (as in monotremes). In addition, the opossum possesses a framework region that lacks the interspersed class I loci. (In eutherians the framework regions also contain class I genes). Opossums have a class III region that is very similar in gene order and content to eutherian class III. It has two extended flanking regions that correspond to the extended eutherian class I and class II regions. Two tightly linked classical class I-like genes (UB and UC) have been translocated outside the MHC although they remain syntenic on chromosome 2. The opossum MHC contains a homologous cluster of framework genes next to the class III region. They are arranged in the same order as the eutherian framework genes. An additional family of unique nonclassical MHC class I genes the UT family has been identified that is restricted to monotremes and marsupials. This family designated ModoUT1 to 17 (In the monotreme M. domestica), is located on opossum chromosome 1 and is unlinked to the main MHC locus. They encode proteins that vary in size between five to eight exons and show limited polymorphism and minimal evidence of positive selection. The UT products while highly divergent, fold in such a way that they can form the MHC class I α-chain structure. These UT proteins are expressed on specific tissues such as the thymus and skin. For example, UT8 is only expressed on immature α/β T cells in the thymus of M. domestica but is absent from mature T cells. The UT gene family has also been identified in the tammar wallaby, brushtail possum, and Tasmanian devil. They are present but have not been mapped in the platypus but appear to have been lost in eutherians [50]. Their function is unknown [50,51]. There may be a functional connection between the expression of these UT molecules and TCRμ.

13.5.2 Australidelphia The opossums and macropods last shared a common ancestor B80 mya. As a result, the structure of the N. eugenii MHC is very different [52]. Thus while class I, II, and III genes are invariably closely linked in eutherians, they are unlinked in the wallaby. (Fig. 13.7) The MHC class II and III genes as well as the antigen processing gene TAP2 and the MHC framework genes are located together on the wallaby chromosome 2q. However, the classical class I genes have been completely translocated out of the MHC and are distributed in ten locations across six different chromosomes. Seven of them appear to encode functional antigen-presenting molecules. This does not appear to be typical of all marsupials and could be a recent development in the wallaby [53]. The rest of the wallaby MHC on chromosome 2q covers 4.5 mb and contains 129 genes. It has also undergone extensive rearrangement. Its class II genes have expanded and are now separated into two clusters by the class III OPOSSUM Class I/II region II

I

III

II

I

WALLABY

II

I

Class I Extended class I

II

III

II

Class II

Class III

Extended class II

FIGURE 13.7 A comparison of the organization of the opossum and tammar wallaby MHC. Note that in the opossum, the Class I and class II genes are not separated by the class III region. Note that in the wallaby MHC, most class I genes are not found in the MHC but on other chromosomes.

196

SECTION | 2 Mammalian orders

region. The class IIa region contains the DBA, DAA, and DBB genes. The class IIb region contains DAB genes as well as additional DBB and DBA genes. Wallabies may have 6
8 DAB loci of which at least five are expressed [53]. The MHC of the Tasmanian Devil is located on chromosome 4q. It encodes six unique β chain sequences from at least three loci that belong to the marsupial Class II DA family. In this species, there are at least 25 MHC types and 53 unique class I sequences that represent seven class I loci, two of which are non-classical [54]. These genes show little heterogeneity indicating recent divergence. However, there are significant variations in the number of their expressed alleles. Thus there are two to seven sequence variants expressed in each individual. They can be divided into two groups based on these sequences. Twenty percent of individuals have a restricted MHC repertoire and express only group I or group II sequences. Thus individual devils can have 0
6 group I sequences and 0
4 group II sequences [54]. The restricted MHC repertoire in some Tasmanian Devil populations may account for their susceptibility to infection with the transmissible tumor cells that cause Devil Facial Tumor Disease. (Box 13.1) The koala also has both class I and class II genes, that encode both alpha and beta chains [32]. Koalas from Northern populations appear to have a greater MHC diversity than those in Southern populations [55]. The class I sequences have about 80% similarity to the red-necked wallaby (Macropus rufogriseus) [56]. Some of these MHC variants in the koala are linked to susceptibility or resistance to Chlamydial disease.

13.5.3 The natural killer complex The M. domestica natural killer complex (NKC) is located on chromosome 8 and contains nine C-type lectin receptor genes. Their sequence identity ranges from 4% to 26%. At least seven of these genes appear to predate the divergence of marsupials and eutherians. There is a close relationship between the opossum MHC and the NKC. Within the opossum MHC, there are two genes called MIC and OSCAR. Opossum MIC is a distant homolog of the human polymorphic class I genes MICA and MICB. The MIC genes are class I related genes that encode the ligands for NKG2D a C-type lectin NK cell receptor. NKG2D receptors are homodimers expressed on NK cells, macrophages, and T cells. This MIC ortholog is an exception to the usual class I paralog situation. The opossum NKG2D has 60% amino acid identity with human and 65% identity with the dog NKG2 [29]. The opossum MIC is basal to a clade that includes human MICA/B and mouse MILL1/2. MILL1/2 genes however are found in the leukocyte receptor complex (LRC) of rodents. This suggests that there once existed an ancestral genomic region that probably contained both MHC and NKC genes in both the KIR and the C-type lectin forms. The function of the MILL gene products is currently unknown. They are β2-microglobulin-associated glycoproteins encoded outside the MHC [57]. The marsupial homolog of CD1 has also been found in the opossum genome but it is located on a different arm of chromosome 2p and could be a pseudogene [58]. The osteoclast-associated receptor (OSCAR), appears to participate in class II antigen processing in dendritic cells. The OSCAR gene is also found within the MHC in the opossum. (It is located within the LRC in humans and rodents). Thus the opossum MHC region points to the likely existence of a supercomplex of genes containing MHC class I and II, in addition to C-type and Ig-type NK receptor genes [49].

13.5.4 The leukocyte receptor complex The LRC of the M. domestica complex contains 233 exons encoding Ig-like domains on chromosome 4. There are two additional exons on chromosome 2 while 55 are unmapped. 193 of these exons appear to be functional. They include two belonging to CEACAM1 (CD66), five siglec genes, as well as over 100 immunoglobulin domains similar to KIR and LILR [29]. They are predicted to encode 45 different receptors. The Tasmanian devil has 140 Ig-like LRC domains. Of these, 126 appeared to be functional and are predicted to encode at least 24 receptors [59].

13.6

B cells and immunoglobulins

Comparative studies on the immune responses of marsupials and eutherians have been limited. It does appear, however, that the marsupials mount a weaker antibody response to some antigens than do eutherian laboratory mammals. Thus the opossum (M. domestica) responds to a single intramuscular injection of sheep red blood cells like other mammals with antibody titers remaining high for as long as 37 weeks [60]. However, the secondary response is weaker than the primary response and declines between 11 and 15 weeks. Responses to repeated boosting remain relatively low for many weeks. This suggests, that like the monotremes, immunologic memory has yet to fully evolve in marsupials.

Marsupials: Opossums to Kangaroos Chapter | 13

197

The marsupial opossum (Didelphis) also resembles more primitive vertebrates in that it responds better to particulate antigens, such as bacteria than to soluble foreign proteins.

13.6.1 Opossum The availability of the M. domestica genome has permitted the production of annotated maps of its IGH, IGL, and IGK loci [15,48]. This has confirmed that the opossum predominantly uses a single IGHV subfamily and clan. They also have only a single IgG subclass.

13.6.2 Heavy chains The M. domestica IGH locus is found on chromosome 1. The 3’ end contains the IGHC genes and is centromeric while the 5’end containing the IGHV genes is telomeric. The IGH locus spans 1418 kb The IgH constant genes in the opossum encode four single immunoglobulin subclasses, IgM, IgG, IgE, and IgA. (Fig. 13.8) There is an IGHM pseudogene located upstream of the functional IGHM gene. Both IGHM genes also have three IGHJ gene segments located upstream of each that do not appear to be functional. No cDNAs encoding a heavy chain with homology to IgD have been found in any marsupial species [15]. At the site downstream of IGHM where an IGHD gene would be expected, there is a cluster of inserted retro-elements consisting of both endogenous retrovirus sequences and LINE elements. These may have replaced the IGHD gene [61]. Opossum IgM, IgG, and IgE, each have two transmembrane exons while IgA has only one. Their extracellular domain structure is typical of all mammalian immunoglobulins. IgM and IgE have three CH exons and no evidence of a hinge. IgG and IgA have three CH exons. IgG has a single hinge whereas the IgA hinge is an extension of the CH1 exon. Twenty-five IGHV gene segments have been identified in the opossum. Twenty-three of these belong to subfamily 1 and 18 are fully functional. A single-member belongs to the IGHV2.1 subfamily. The remaining five are pseudogenes with premature stop codons. The IGHV1 family forms a monophyletic clade. Sister IGHV genes have been reported from the tammar wallaby, bandicoot, and brushtail possum. An IGHV3.1 gene segment has been identified that appears to have recombined with a D segment in the germline. The IGHV genes of the opossum, all belong to clan III. Nine D gene segments (eight functional) have been identified as well as six IGHJ segments (four functional). The D gene segments are of variable length and the shorter segments appear to be preferentially used. Likewise, of the four J segments, only the two that are immediately upstream of the functional IGHM segment are used. FIGURE 13.8 The organization of the immunoglobulin heavy and light chain genesin Monodelphis domestica. LC- Light chain constant gene. Ψ- pseudogene

OPOSSUM IGH 5’

19 V

D

7

3

5

V D VV J

M D

J

IGL

64

M

D

G

V

D

M

2

122

IGK

3’

3

E J

A LC

G

E

A

198

SECTION | 2 Mammalian orders

13.6.3 Light chains The opossum IGK locus is about three kb in length and is found on chromosome 1. There is a single IGKC gene at the 3’ end and four functional IGKJ segments but upstream of these are 122 IGKV gene segments of which 89 are functional. These IGKV gene segments form two large clusters separated by an 800 kb gap. The two clusters appear to have arisen from a duplication event occurring before the marsupial/eutherian divergence [15]. These IGKV segments are classified into seven subfamilies. The IGL locus is located on chromosome 3. It is almost four kb in length. The M. domestica IGL locus contains 64 IGLV gene segments divided into four Vλ subfamilies together with eight IGLJ and eight IGLC genes arranged in a series of Jλ-Cλ pairs. Of the 64 IGLV gene segments, 54 belong to the IGLV1 subfamily and all but six are functional. Despite the complexity of the κ light chain loci, opossum B cells transcribe λ to κ light chains in a ratio of 2:1 [62]. To determine if M. domestica had a typical variable gene diversity for marsupials the IGHV genes of T. vulpecula have also been investigated [63]. As in Monodelphis, T. vulpecula also uses a restricted IGHV pool and a diverse IGLV pool. It is suggested that a pattern has emerged with IGLV expanding to compensate for the reduced diversity of IGHV.

13.6.4 Fc receptors Almost all the classical eutherian Fc receptor families together with the Fc receptor-like receptors are found in marsupials. For example, M. domestica possesses all the receptors (IgM, pIgR, IgG, and IgE). An exception is the IgG receptor, FcγR1 which appears to be missing [64]. Marsupials have an IgE receptor but do not have an FcαR receptor. This receptor is only found in eutherian mammals [64].

13.7

T cells and cell-mediated immunity

T cell functions and the structure of their antigen receptors are relatively highly conserved between marsupials and eutherians. However, newborn M. domestica cannot generate a T-dependent antibody response before seven days of age. The ability to mount a cell-mediated immune response such as allograft rejection also develops postnatally. Thus neonatal opossums and macropods tolerate skin allografts until almost two weeks of age. Graft rejection coincides with the appearance of functional T cells. In tammar wallabies, lymphocytes expressing the B cell marker CD79b (Ig-β) can be found in the intestinal lymphoid tissues of pouch young as early as day seven but CD31 T cells are not detected in the thymus until day 12.

13.7.1 The T cell antigen receptors 13.7.1.1 Opossum 13.7.1.1.1

TRA/D genes

As in eutherian mammals, the opossum TCR delta chain genes are embedded within the alpha gene locus (Fig. 13.9). As a result, their gene order is—29 TRAV, one TRDV, 24 TRAV, 5 TRDV, 2 TRDD, 6 TRDJ, TRDC1, 1 TRDV (reverse orientation), 53 TRAJ, and one TRAC [62]. Of the TRAV genes, 56 are functional. Of the six TRDV genes, four are functional. The TRAC and TRDC genes contain three exons; these encode the antigen-binding domain, the connecting peptide, and the transmembrane/cytoplasmic region of the receptor peptide chains [65]. 13.7.1.1.2 TRB genes The opossum TRB locus has been mapped to chromosome 8q and it spans 400 kb which is smaller than human or mouse. The opossum has 35 TRBV genes, four TRBD genes (three functional), 18 functional TRBJ genes, and four functional TRBC genes [66]. The TRBV genes are very diverse and divided into six subfamilies. The rest of the locus consists of four -D-J-C- cassettes. The four TRBC genes are very similar to each other, and each encodes four exons. Cassettes 2 and 3 appear to be derived from a recent tandem duplication event. 13.7.1.1.3 TRG genes The opossum TRG locus has been mapped to chromosome 6q. It is relatively small, occupying only 90 kb. (Human TRG is 150 kb while the mouse TRG is 205 kb in length). The opossum has nine TRGV genes of which all are functional, seven TRGJ genes of which five are functional; and a single functional TRGC gene [67]. The TRGV gene

Marsupials: Opossums to Kangaroos Chapter | 13

TRA/D

5’

44 AV

DV

35

TRB

21

4

2

6

A/DV

DV

DD

DJ

5

TRG

TRM

9

2

AJ

AC

C2

C3

C4

7

3

2 C1

DV

5

4

4 C1

DC

FIGURE 13.9 The organization of the T cell antigen receptor genes in Monodelphis domestica.

3'

53

199

3

2 C4

C3

C2

3 C5

C6

C7

C8

mu cluster

V

D

J

C

segments are arranged differently in the opossum than in humans and mice with a translocon organization. Thus the V gene segments are clustered upstream of the clustered TRGJ genes that are located upstream of the single TRGC gene. The single TRGC gene contains three exons, each encoding a single domain. 13.7.1.1.4 TRM genes M. domestica, like the platypus and wallaby, produces a unique TCR chain expressed early in development before conventional TCRs, which may protect during the first few days of life before their immune system is fully functional. This receptor chain is called TCRμ. (μ or M for marsupial) [68,69]. The TRM locus is located on chromosome 3q of M. domestica. (TRA/D is on chromosome 1p; TRB is on chromosome 8q, and TRG is on chromosome 6q). Homologs to TRM have not been detected in any eutherians. The locus is organized into six tandem cassettes covering 614 kb. Each cassette contains one V, two or three-D, and one J gene, along with constant region genes. The exons of the TRMC encode three domains of the μ chain, transmembrane, connecting peptide, and cytoplasmic. TRM appears to be a hybrid locus generated by recombination between ancestral IGH and TCR genes. The TRMC genes are related to TRDC while the TRMV genes are related to immunoglobulin V genes. The TRM locus is organized in tandem cassettes where each cassette contains two classes of gene segments. One class consists of a single nonrearranged gene segment, while the other class consists of a TRMV-gene segment already joined to TRMD and TRMJ genes in the germline DNA. The opossum possesses six complete cassettes as well as two incomplete cassettes that lack the TRMV and TRMD genes [67]. There are therefore two TRM isoforms produced, TRMVJ and TRMV. TRMVJ uses the pre-joined V(D) J segments. TRMV in contrast requires V(D)J recombination and contains the double V configuration. In addition to the six functional TRMV gene segments, there is also a single TRMV orphan gene that appears to be nonfunctional since it lacks a leader sequence. During postnatal development of the opossum, it is possible to compare the timing of VDJ recombination and the appearance of TCRμ transcripts relative to the conventional TCR chains [67]. TRMV is detectable as early as 24 hours after birth when the opossum thymus is still undifferentiated epithelium. TRMVJ is not detectable until day 13 when the thymus is structurally complete. The other TCRs, α, β, and δ, are also detectable on day 1 of age. TRGV transcripts

200

SECTION | 2 Mammalian orders

are not detectable until day 8. Thus unlike eutherians, α/β T cell development precedes γ/δ/μ T cell development in the opossum [67].

13.7.1.2 Other marsupials In the three marsupial species whose genomes have been sequenced, the TRAV clade sequences are intermediate between those in monotremes and the eutherians. In marsupials, there are more TRBV genes than in monotremes. Thus there are 22 in the tammar wallaby, 36 in Monodelphis (of which 27 are functional), and 33 in the Tasmanian devil. The TRD genes have been identified and sequenced in three species of marsupial, M. domestica, the tammar wallaby, and the northern brown bandicoot, (I. macrourus). The TRDC genes in all three species are highly divergent and not orthologous to each other [70]. Thus unlike eutherian mammals, the three marsupials investigated have multiple, highly divergent, TRDC loci. One subfamily of these TRDC genes is closely related to the TRDC sequences in eutherians. A second subfamily forms a unique marsupialspecific clade. A unique expression pattern of TRDV and TRDC gene segments is also seen in the bandicoot and the wallaby. The eutherian-related sequences are only found in association with each other in the cDNAs from both species. A similar pattern is also seen in the close association between the marsupial TRDV and TRDC genes.

13.7.2 Other T cell receptors Cytotoxic T cells recognize antigenic peptides presented by MHC class I molecules. This recognition is facilitated by the presence of CD8 expressed on the T cell surface. It serves as a linker for the MHC molecule and serves to lock the T cell and its target in a tight embrace. The CD8 gene locus has been characterized in both M. domestica and N. eugenii. The CD8 alpha and beta chains are about 63% identical in their alpha chain and 57% identical in their beta-chain amino acid sequences [71]. They are 36%
45% identical to their eutherian counterparts and retain all the major structural features of these proteins (Box 13.1). BOX 13.1 Devil facial tumor disease. The large carnivorous marsupial, the Tasmanian devil (Sarcophilus harrissii), is on the brink of extinction as a result of devil facial tumor disease, a transmissible cancer. This disease first appeared in 1996 and has spread across Tasmania. It has reduced some devil populations by as much as 90%. Tumor cells are transmitted when devils bite each other around the face, a common behavior. Animals die within 3
6 months of acquiring cancer since they are unable to mount an immune response to the foreign cells. The tumor cells grow and form a large mass that is eventually lethal. Almost every devil “infected” with these tumor cells dies of cancer. Although devils have a functioning immune system, their limited MHC diversity prevents them from recognizing the tumor cells as foreign. (The tumor cells originated from Schwann cells from a female devil in the early-1990s but are continuing to evolve.) A second, genetically distinct facial tumor lineage (DFT2) has also been recognized in devils. Facial tumor cells do not express surface MHC class I due to down-regulation of their β2-microglobulin and TAP genes. This downregulation is a result of epigenetic deacetylation of histones. Thus there is no histocompatibility barrier to tumor growth. Although devils have functioning NK cells these cannot kill the tumor cells for unknown reasons. MHC expression can be restored by exposing facial tumor cells to recombinant devil IFN-γ and subsequent activation of the MHC class II transactivator, a critical transcription factor. Blood mononuclear cells activated by mitogens in vitro can also kill devil tumor cells. Encouraging results have also been obtained by vaccinating animals with killed adjuvanted tumor cells. It also appears that resistance to this cancer is emerging in some wild populations. Brown, GK., Tovar, C., Cooray, AA., Kreiss, A., Darby, J., Murphy, JM., Corcoran, LM., Bettiol, SS., Lyons, AB. & Woods, GM. Mitogen-activated Tasmanian devil blood mononuclear cells kill devil facial tumor disease cells. Immunol Cell Biol, 94, 673
9., 2016 Pye, RJ., Pemberton, D., Tovar, C., Tubio, JM., Dun, K. A., Fox, S., Darby, J., Hayes, D., Knowles, GW., Kreiss, A., Siddle, HV., Swift, K., Lyons, AB., Murchison, EP. & Woods, GM. A second transmissible cancer in Tasmanian devils. Proc Natl Acad Sci U S A, 113, 374
9, 2016a. Pye, RJ., Woods, GM. & Kreiss, A. Devil Facial Tumor Disease. Vet Pathol, 53, 726
36, 2016b. Siddle HV, Kreiss A, Eldridge MD, et al.: Transmission of a fatal clonal tumor by biting occurs due to depleted MHC diversity in a threatened carnivorous marsupial, Proc Natl Acad Sci USA, 104:16221
16226, 2007. Siddle HV, Kreiss A, Tovar C, Yuen CK, Cheng Y, et al. Reversible epigenetic down-regulation of MHC molecules by devil facial tumor disease illustrates immune escape by contagious cancer. Proc Natl Acad Sci. 110:5103
5108, 2013.

Marsupials: Opossums to Kangaroos Chapter | 13

201

References [1] Eldridge MDB, Beck RMD, Croft DA, Travouillon KJ, Fox BJ. An emerging consensus in the evolution, phylogeny and systematics of marsupials and their fossil relatives (Metatheria). J Mammal 2019;100(3):802
37. [2] Cheng Y, Belov K. Antimicrobial protection of marsupial pouch young. Front Microbiol 2017;8:354. Available from: https://doi.org/10.3389/ fmicb.2017.00354. [3] Griffith OW, Chavan AR, Protopapas S, Maziarz J, et al. Embryo implantation evolved from an ancestral inflammatory attachment reaction. Proc Natl Acad Sci U S A 2017;114(32):E6566
75. Available from: https://doi.org/10.1073/pnas.1701129114. [4] Deane EM, Cooper DW. Immunology of pouch young marsupials. I. Levels of immunoglobulin, transferrin and albumin in the blood and milk of euros and wallaroos (Hill kangaroos: Macropus robustus, Marsupiala). Dev Comp Immunol 1984;8:863
76. [5] Edwards MJ, Hinds LA, Deane EM, Deakin JE. A review of complementary mechanisms which protect the developing marsupial pouch young. Dev Comp Immunol 2012;37:213
20. [6] Krause WJ, Cutts JH, Leeson CR. Postnatal development of the epidermis in a marsupial Didelphis virginiana. J Anat 1978;125(1):85
99. [7] Pralomkarn T, Nelson J, Gemmell RT. Postnatal development of the skin of the marsupial native cat Dasyurus hallucatus. J Morphol 1990;205:233
42. [8] Harrison GA, Young LJ, Watson CM, Miska KD, et al. A survey of type I interferons from a marsupial and monotreme: implications for the evolution of the interferon type I gene family in mammals. Cytokine 2003;21:105
19. [9] Baudinette RV, Boontheung P, Musgrave IF, Wabnitz PA, et al. An immunomodulator used to protect young in the pouch of the tammar wallaby Macropus eugenii. FEBS J 2005;272:433
43. [10] Morris KM, O’Meally D, Zaw T, Song X, et al. Characterization of the immune compounds in koala milk using a combined transcriptomic and proteomic approach. Sci Rep 2016;6:35011. Available from: https://doi.org/10.1038/srep35011. [11] Watson RP, Demmer J, Baker EN, Arcus VL. Three-dimensional structure and ligand binding properties of trichosurin, a metatherian lipocalin from the milk whey of the common brushtail possum Trichosurus vulpecula. Biochem J 2007;408:29
38. [12] Wang J, Wong ESW, Whitley JG, Li J, et al. Ancient antimicrobial peptides kill antibiotic resistant pathogens: Australian mammals provide new options. PLoS One 2011;6(8):e24030. Available from: https://doi.org/10.1371/journal.pone/0024030. [13] Wang X, Sharp AR, Miller RD. Early postnatal B cell ontogeny and antibody repertoire maturation in the opossum Monodelphis domestica. PLoS One 2012;7(9). Available from: https://doi.org/10.1371/journal.pone.0045931e45931. [14] Fehrenkamp BD, Morissey KA, Miller RD. Opossum IgG is from maternal circulation and timing of transfer correlates with neonatal immune development. Reprod Fertil Dev 2019;31(7):1246
51. [15] Fehrenkamp BD, Miller RD. γδ T cells are the predominant T cell type in opossum mammaries during lactation. Dev Comp Immunol 2019;95:96
100. [16] McClellan HL, Miller SJ, Hartmann PE. Evolution of lactation: nutrition v protection with special reference to five mammalian species. Nutr Res Rev 2008;21:97
116. [17] Pharo EA. Marsupial milk: a fluid source of nutrition and immune factors for the developing pouch young. Repro Fert Devel 2019;31: 1252
65. [18] Joss JL, Molloy MP, Hinds L, Deane E. A longitudinal study of the protein components of marsupial milk from birth to weaning in the tammar wallaby (Macropus eugenii). Dev Comp Immunol 2008;33(2):152
61. Available from: https://doi.org/10.1016/j.dci.2008.08.002. [19] Daly KA, Digby M, Lefevre C, Mailer S, et al. Analysis of the expression of immunoglobulins throughout lactation suggests two periods of immune transfer in the tammar wallaby (Macropus eugenii). Vet Immunol Immunopathol 2007;120:187
200. [20] Deane EM, Cooper DW, Renfree MB. Immunoglobulin G levels in fetal and newborn tammar wallabies (Macropus eugenii). Reprod Fertil Dev 1990;2:369
75. [21] Belov K, Miller RD, Old JM, Young LJ. Marsupial immunology bounding ahead. Aust J Zool 2013;61:24
40. [22] Barbour RA. The leukocytes and platelets of a marsupial, Trichosurus vulpecula. A comparative morphological metrical and cytochemical study. Arch Histol Jap 1972;34:311
60. [23] Hawkey CM. Comparative mammalian hematology. Cellular components and blood coagulation of captive wild animals. London: Heinemann Medical Books; 1975. [24] Wilson GR, Hoskins L. Haematology and blood chemistry of the red kangaroo Megaleia rufa in captivity. Aust Vet J 1975;51:146
9. [25] Green- Barber JM, Ong OTW, Kanuri A, Stannard HJ, Old JM. Blood constituents of free ranging eastern grey kangaroos (Macropus giganticus). Aust Mamm 2017;40(2):136
45. Available from: https://doi.org/10.1071/AM17002. [26] Evans KD, Hewett TA, Clayton CJ, Krubitzer LA, Griffey SM. Normal organ weights, serum chemistry, Hematology, and cecal and nasopharyngeal bacterial cultures in the gray short-tailed possum (Monodelphis domestica). J Am Ass Lab Anim Sci 2010;49(4):401
6. [27] Canfield PM, O’Neill ME, Smith EF. Haematological and biochemical reference values for the koala (Phascolarctos cinereus). Aust Vet J 1989;66(10):324
6. [28] Odewahn R, Wright BR, Czirjak GA, Higgins DP. Differences in constitutive innate immunity between divergent Australian marsupials. Developmental and Comparative Immunology 2022;132. Available from: https://doi.org/10.1016/j.dci.2022.104399. [29] Belov K, Sanderson CE, Deakin JE, Wong ESW, et al. Characterization of the opossum immune genome provides insights into the evolution of the mammalian immune system. Genome Res 2007;17:982
91. Available from: https://doi.org/10.1101/gr.6121807. [30] Lynn DJ, Bradley DG. Discovery of α-defensins in basal mammals. Dev Comp Immunol 2007;31:963
7.

202

SECTION | 2 Mammalian orders

[31] Jones EA, Cheng Y, O’Meally D, Belov K. Characterization of the antimicrobial peptide family defensins in the Tasmanian devil (Sarcophilus harrisii), koala (Phascolarctos cinereus), and tammar wallaby (Macropus eugenii). Immunogenetics 2017;69:133
43. [32] Abts KC, Ivy JA, DeWoody A. Immunomics of the koala (Phascolarctos cinereus). Immunogenetics 2015;67(5-6):305
21. Available from: https://doi.org/10.1007/s00251-015-0833-6. [33] Wong ESW, Young LJ, Papenfuss AT, Belov K. In silico identification of opossum cytokine genes suggests the complexity of the marsupial immune system rivals that of eutherian mammals. BMC Immunol Res 2006;2:4. Available from: https://doi.org/10.1186/1745-7580-2-4. [34] Wong ESW, Papenfuss AT, Belov K. Genomic identification of chemokines and cytokines from the opossum. J Interferon Cytokine Res 2011;31(3):317
30. [35] Borthwick CR, Young LJ, Old JM. The development of the immune tissues in marsupial pouch young. J Morphol 2014;275:822
39. [36] Yadav M. The presence of the cervical and thoracic thymus lobes in marsupials. Aust J Zool 1973;21(3):285
301. [37] Canfield PJ, Hemsley S. The roles of histology and immunohistology in the investigation of marsupial disease and normal lymphoid tissue. Dev Comp Immunol 2000;24:455
71. [38] Cisternas P, Armati PJ. The immunolymphatic system. In: Amati PJ, Dickman CR, Hume ID, editors. Marsupials. Cambridge University Press; 2006. [39] Peel E, Belov K. Immune-endocrine interactions in marsupials and monotremes. Gen Comp Endocrinol 2017;244:178
85. [40] Haynes JI. The marsupial and monotreme thymus revisited. J Zool Lond 2001;253:167
73. [41] Ashman R, Keast D, Stanley NF, Waring H. The immunological responses of marsupials. Amer Zool 1975;15:155
66. [42] Ashman RB, Papadimitriou JM. Development of lymphoid tissue in a marsupial Setonix brachyuris (quokka). Acta Anat (Basel) 1975;91 (4):594
611. [43] Cisternas PA, Armati PJ. Development of the thymus, spleen, lymph nodes and liver in the marsupial Isoodon macrourus (Northern brown bandicoot. Peramelidae). Anat Embryol 1999;200:433
43. [44] Udroiu I, Sgura A. The phylogeny of the spleen. Quart Rev Biol 2017;92(4):411
43. [45] Old JM, Deane EM. Histology and immunohistochemistry of the gut-associated lymphoid tissue of the eastern grey kangaroo, Macropus giganteus. J Anat 2001;199:657
62. [46] Hanger JJ, Heath TJ. Topography of the major superficial lymph nodes and their efferent lymph pathways in the koala (Phascolarctos cinereus). J Anat 1991;177:67
73. [47] Smith HF, Fisher RE, Everett ML, Thomas AD, et al. Comparative anatomy and phylogenetic distribution of the mammalian cecal appendix. J Evol Biol 2009;22:1984
99. [48] Wong ESW, Papenfuss AT, Belov K. Immunome database for marsupials and monotremes. BMC Immunol 2011;12:48
53. [49] Belov K, Deakin JE, Papenfuss AT, Baker ML, et al. Reconstructing an ancestral mammalian immune supercomplex from a marsupial major histocompatibility complex. PLoS Biol 2006. Available from: https://doi.org/10.1371/journal.pbio.0040046. [50] Papenfuss AT, Feng Z-P, Krasnec K, Deakin JE, et al. Marsupials and monotremes possess a novel family of MHC class I genes that is lost from the eutherian lineage. BMC Genomics 2015;16(1):535. Available from: https://doi.org/10.1186/s12864-015-1745-4. [51] Krasnek KV, Papenfuss AT, Miller RD. The UT family of MHC class I loci unique to non-eutherian mammals has limited polymorphism and tissue specific patterns of expression in the opossum. BMC Immunol 2016;17:43. Available from: https://doi.org/10.1186/s12865-016-0181-9. [52] Siddle HV, Deakin JE, Coggill P, Whilming LG, et al. The tammar wallaby major histocompatibility complex shows evidence of past genomic instability. BMC Genomics 2011;12:421. Available from: https://doi.org/10.1186/1471-2164/12/421. [53] Deakin JE, Siddle HV, Cross JG, Belov K, et al. Class I genes have split from the MHC in the tammar wallaby. Cytogenet Genome Res 2007;116(3):205
11. [54] Siddle HV, Sanderson C, Belov K. Characterization of major histocompatibility complex class I and class II genes from the Tasmanian devil (Sarcophilus harrisii). Immunogenetics 2007;59:753
60. [55] Madden D, Whaite A, Jones E, Belov K, et al. Koala immunology and infectious diseases: How much can the koala bear? Dev Comp Immunol 2018;82:177
85. [56] Houlden BA, Greville WD, Sherwin WB. Evolution of class I gene loci in marsupials: Characterization of sequences from koala (Phascolarctos cinereus). Mol Biol Evol 1996;13(8):1119
27. [57] Kajikawa M, Baba T, Tomaru U, Watanabe Y, et al. MHC class I-like molecules are b2-microglobulin-associated, GPI anchored glycoproteins that do not require TAP for cell surface expression. J Immunol 2006;177:3108
15. [58] Baker ML, Miller RD. Evolution of mammalian CD1: marsupial CD1 is not orthologous to the eutherian isoforms and is a pseudogene in the opossum Monodelphis domestica. Immunology 2007;121:113
21. [59] Guselnikov SV, Taranin AV. Unraveling the LRC evolution in mammals: IGSF1 and A1BG provide the keys. Genome Biol Evol 2019;11 (6):1586
601. [60] Croix DA, Samples NK, Vandeberg JL, Stone WH. Immune response of a marsupial (Monodelphis domestica) to sheep red blood cells. Dev Comp Immunol 1989;13:73
8. [61] Miller RD, Hansen VL. Marsupial and monotreme immunoglobulin genetics. In: Kaushik AK, Passman Y, editors. Chapter 4 Comparative immunoglobulin genetics. Toronto, ON: Apple Academic Press; 2014. [62] Schraven AL, Hansen VL, Morrissey KA, Stannard HJ, et al. Developmental and comparative immunology single-cell transcriptome analysis of the B-cell repertoire reveals the usage of immunoglobulins in the gray, short-tailed opossum (Monodelphis domestica). Dev Comp Immunol 2021;123:104141.

Marsupials: Opossums to Kangaroos Chapter | 13

203

[63] Baker ML, Belov K, Miller RD. Unusually similar patterns of antibody V segment diversity in distantly related marsupials. J Immunol 2005;174:5665
71. [64] Akula S, Mohammadamin S, Hellman L. Fc receptors for immunoglobulins and their appearance during vertebrate evolution. PLoS One 2014;9 (5). Available from: https://doi.org/10.1371/joutnal.pone.0096903. [65] Olivieri DN, Gambon-Cerda S, Gambon-Deza F. Evolution of V genes from the TRV loci of mammals. Immunogenetics 2015;67:371
84. [66] Parra ZE, Baker ML, Hathaway J, Lopez AM, et al. Comparative genomic analysis and evolution of the T cell receptor loci in the opossum (Monodelphis domestica). BMC Genet 2008. Available from: https://doi.org/10.1186/1471-2164-9-111. [67] Parra ZE, Baker ML, Lopez AM, Trujillo J, et al. TCRμ recombination and transcription relative to the conventional TCR during postnatal development in opossums. J Immunol 2009;182:154
63. [68] Parra ZE, Baker ML, Schwartz RS, Deakin JE, et al. A unique T cell receptor discovered in marsupials. Proc Natl Acad Sci U S A 2007;104 (23):9776
81. [69] Morrissey KA, Wegrecki M, Praveena T, Hansen VL, et al. The molecular assembly of the marsupial γμ T cell receptor defines a third T cell lineage. Science 2012;371:1383
8. [70] Baker ML, Osterman AK, Brumburgh S. Divergent T-cell receptor delta chains from marsupials. Immunogenetics 2005;57:665
73. [71] Duncan LG, Nair SV, Deane EM. The marsupial CD8 gene locus: molecular cloning and expression analysis of the alpha and beta sequences in the gray short-tailed opossum (Monodelphis domestica) and the tammar wallaby (Macropus eugenii). Vet Immunol Immunopathol 2009;129:14
27.

This page intentionally left blank

Chapter 14

Tylopoda: Camels and llamas

Vicunas; Vicugna vicugna

The order Artiodactyla includes the even-toed ungulates that bear their weight on two of their five toes. The aquatic cetaceans have evolved from the even-toed ungulates and taxonomists have therefore combined the two in the superorder, Cetartiodactyla. It is probable that the first artiodactyls first emerged about 75 mya. The order currently contains more than 200 species classified into about 10 families. (Fig. 14.1).

70

60

50

Million years ago 40 30 20 10

FIGURE 14.1 The evolution and phylogeny of the artiodactyls. As in much phylogeny the divergence times are a matter of dispute among paleontologists.

0 TYLOPODA Camels SUIDAE

70

Pigs

35 TAYASSUIDAE Peccaries CETACEA Whales, dolphins

55

HIPPOPOTAMIDAE Hippos TRAGULIDAE CERVIDAE 30-40

Chevrotains

Deer

MOSCHIDAE Musk deer BOVIDAE

Cattle, sheep, goats

ANTILOCAPRIDAE Pronghorns GIRAFFIDAE

Giraffes

Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00022-8 © 2023 Elsevier Inc. All rights reserved.

205

206

SECTION | 2 Mammalian orders

Camelids of the suborder Tylopoda were probably the first to split from the basal artiodactyl clade. Their last common ancestor with the other artiodactyls probably existed around 70
65 mya during the Paleocene epoch. Since then, they have undergone large-scale adaptive radiation. They consist of two distinct tribes. One consists of the Old World Camelinae including the dromedary camel (Camelus dromedarius), and the wild and domesticated Bactrian camels (Camelus ferus and Camelus bactrianus). The other tribe is the New world Laminae that contain four extant species, the llama (Lama glama), guanaco (Lama guanicoe), alpaca (Vicugna pacos), and vicun˜a (Vicugna vicugna). These can all interbreed and produce fertile hybrids. The Tylopoda have several characteristic features that differentiate them from the other artiodactyl suborders. They have unusual two-toed feet with toenails and soft footpads. They are strict herbivores and are adapted to feed on forage consisting of very fibrous and thorny plants. As a result, they have a three-chambered stomach that is very different from that in the other suborders. It enables them to retain ingested material for a very long time and so effectively digest their diet of lignocellulose. They have a split upper lip that helps grasp forage, and they have elliptical red blood cells that may help them resist dehydration. They also differ in placental structure. Camelids originated in present-day North America. For eons, they were restricted to that landmass. They diversified and thrived until about 5 mya when some members eventually migrated to Asia by way of the Behring land bridge while others took advantage of the newly formed Isthmus of Panama to migrate into South America. The camelids that remained in North America became extinct during the last ice age. As a result, the number of extant species is small. Nevertheless, they are well adapted to survive in extreme conditions—hot deserts in Asia and Africa and cold dry mountain deserts in South America. Camels were probably first domesticated about 4000 years ago in the Arabian peninsula and Turkmenistan. No naturally wild dromedary populations remain, and the wild Bactrian camel population is probably less than 1000 individuals. Recently, mitochondrial and genomic sequencing has confirmed that these wild Bactrian camels are a separate species from the domesticated Bactrian camel. They are therefore classified as C. ferus [1]. C. ferus lives in desert and semidesert areas, mostly in China. Dromedary camels are socially and economically important animals, especially in the desert regions of Africa and Asia. They are used for meat, hides, and milk as well as beasts of burden and racing. They are however the natural hosts of the Middle-East Coronavirus that can cause lethal respiratory disease in humans.

14.1

Reproduction and lactation

Camels have an epitheliochorial placenta that does not permit the transfer of immunoglobulins from mother to young. It resembles that of the horse in that it is a diffuse type. (Interestingly, the fetus only develops in the left horn of the uterus.) [2] As a result, both newborn camels and the crias of llamas and alpacas are agammaglobulinemic and rely on the intake of colostral immunoglobulins for protection at birth. IgM, IgG, and IgA have been identified in camel colostrum. IgG is the most abundant of these immunoglobulins. Camel milk contains the classical four chain antibody (IgG1) as well as the two heavy-chain-only antibodies (IgG2 and IgG3) that are unique to camels. Immunoglobulin synthesis begins early in the young calf and serum IgG levels rise about two months after birth. Camel lactation typically lasts for 8
9 months, and the volume of milk produced ranges from 800 to 3600 L [3]. The milk immunoglobulin content is similar in both C. bactrianus and C. dromedarius [4]. It tends to vary seasonally with the highest immunoglobulin values in the winter. The IgG concentrations in mammary secretions decrease from 13,200 to 475 mg/dL through the first 7 days postpartum [4]. These levels are slightly higher than those found in cow’s milk. Immunoglobulin concentrations in llamas, alpacas, and their crias have also been measured. Llama and alpaca milk contain an average of 19,262 mg/dL IgG at parturition and 633 mg/dL 6 days later [5].

14.2

Hematology

The published total leukocyte counts for the dromedary range from 6.5 to 19.6 cells 3 103/μL blood [6]. (Table 14.1). Similar counts have been reported in the Bactrian camel. Lymphocytes account for about 30%
45% of these cells. Lymphocyte numbers tend to be higher in younger animals [11]. Cellular phenotyping has identified three subsets of blood monocytes in camels. About 80% are CD14hi, CD163hi, and major histocompatibility complex (MHC)IILo. A second subset is MHCIIhi and the third subset is CD14lo, CD163lo, and MHC IIhi. Camel monocytes also express CD26, the receptor for the MERS coronavirus [12].

Tylopoda: Camels and llamas Chapter | 14

207

TABLE 14.1 The numbers and composition of blood leukocytes of selected camel species. Dromedary [9]

Bactrian [9]

Alpaca [10]

Llama [10]

Total WBCs 3 103/μl

6.5
19.6

8.6
24.4

9.8
17.5

4.8
16.4

Neutrophils. (%)

50
77

67

79

66

Lymphocytes. (%)

30
45

27

18

22

Monocytes. (%)

2
8

3

2

3

Eosinophils (%)

0
6

2.5

1

3

Basophils (%)

0
2

0.5

0

0.5

14.3

Innate immunity

14.3.1 Leukocytes Camel neutrophils have been reported to be smaller than those in other species [6]. One unique structural finding is that they possess a perinuclear membrane. They also contain both primary and secondary granules, but tertiary granules have not been reported. Camels possess eosinophils. As in other mammals, these eosinophils likely play a key role in resistance to parasites. Ultrastructurally, camel eosinophils have some unique features. While the basic structure of their granules with an electron-dense crystalloid core surrounded by a homogeneous matrix is maintained, they are very polymorphic. They vary in shape and size and are often segmented. The granules may demonstrate lamellae and even multiple crystalloid cores within a single granule. When activated, camel eosinophils may develop hypersegmented nuclei [6].

14.3.1.1 Pattern recognition receptors Dromedaries possess all ten toll-like receptors. They have a similar structure and function to those observed in other artiodactyls [13].

14.3.2 Complement Camels have demonstrable levels of hemolytic complement activity in their bloodstream [6]. Calves also appear to acquire hemolytic complement through maternal colostrum. Maximum levels of classical pathway components appear to be reached between 1 and 5 years of age and decline thereafter. Males appear to possess higher levels than females. Likewise, alternate pathway complement activity peaks between 1 and 3 years of age. Low levels of conglutinins have been found in camels [6].

14.3.3 Cytokines The Th1 cytokines, IL-2, IL-12, and IFN-γ, have been identified in the Bactrian camel as have the Th2 cytokines, IL-4, IL-10, and IL-13 [11]. Camels produce a similar diversity of interferons as do cattle. Thus 11 IFN-α subtypes and 1 IFN-ε have been reported in the dromedary [14]. Likewise, multiple llama cytokines have been sequenced and characterized including IL-2, IL-4, IL-10, IL-12p35, IL-12p40, IL-13, and interferon-γ. Llama IL-2 has 85%
90% sequence homology with bovine and pig IL-2 [15,16].

14.4

Lymphoid organs

14.4.1 Thymus As in other mammals, the dromedary thymus is located in the anterior mediastinum. It may extend into the neck where it is located between the trachea and the left external jugular vein [3]. It consists of two lobes covered by a thin connective tissue capsule. Histologically it is similar to the thymus in other artiodactyls.

208

SECTION | 2 Mammalian orders

14.4.2 Spleen The dromedary spleen is crescent-shaped and is covered with a thick capsule consisting of an outer layer of connective tissue and an inner layer of smooth muscle cells. Vascular and avascular trabeculae extend into the spleen parenchyma from the capsule. There are unique subcapsular and peritrabecular sinuses around the primary and secondary trabeculae. The central artery that emerges from the periarteriolar lymphoid sheath branches into up to four penicillary arteries that extend as sheathed arterioles. A wide marginal zone surrounds the white pulp and contains sheathed arteries but no marginal sinuses. The camel spleen is clearly of the blood storage type. It has both closed and open circulation. The venous return is also unique since the blood flow from the venous sinuses goes from the peritrabecular sinuses to the subcapsular sinuses to the splenic vein [17].

14.4.3 Lymph nodes Dromedary lymph nodes are generally small and occur in clusters. They tend to be covered by a two-layered capsule. The outer layer of the capsule consists of connective tissue while the inner layer consists of smooth muscle. Trabeculae may extend into the nodes from the capsule [3]. Most dromedary lymph nodes appear to be of conventional structure with no obvious unique features.

14.4.4 Hemal nodes Hemal nodes are found in ruminants, some rodents, and camels. They have the same general structure as all mammals. (Fig. 17.3). They have been divided into two structural types [18]. Hemal nodes contain blood-filled sinuses only and are restricted to ruminants. Hemolymph modes, in contrast, contain both blood and lymph sinuses. The parotid, mandibular and lateral retropharyngeal lymph nodes of the dromedary appear to be a type of hemolymph node. They contain lymphatic nodules, dense nodular lymphoid tissue, and diffuse lymphoid tissue. They lack a medulla and cortex. Networks of sinuses are present in the diffuse lymphoid tissue and there are numerous erythrocytes within and around this network. The nodal sinuses are contiguous with the septal blood vessels. Thus they resemble the hemolymph nodes of other mammals but uniquely, are located in sites that are typical of ordinary lymph nodes [18]. Dromedaries also possess multiple hemolymph nodes that differ from typical ruminant hemal nodes. They are generally spherical or kidney-shaped with one or two hili and a capsule and trabeculae containing connective tissue and smooth muscle cells. The parenchyma consists of a cortex and a medulla. The cortex is formed from lymphoid follicles and diffuse interfollicular lymphocytes. The medulla consists of lymphoid cords separated by medullary sinuses. The interfollicular lymphocytes and those in the medullary cord are T cells (CD31). The lymphoid follicles contain B cells (CD221). MHC class II/DR is expressed by most parenchymal cells. They also contain subcapsular, peritrabecular, and medullary blood sinuses as well as afferent and efferent lymphatics and lymphatic sinuses [19].

14.4.5 Mucosal lymphoid tissues Dromedaries have five groups of tonsils in their oropharyngeal region. Their palatine tonsils are located on the lateral walls of the oropharynx There is a deep elongated depression, the tonsillar sinus, just behind the fold of the soft palate diverticulum. The tonsils appear as round macroscopic nodules protruding into the lumen. The tonsillar crypt is lined with stratified squamous epithelium. The tonsils are formed by solitary and aggregated lymphoid nodules in which the follicles are separated by interfollicular tissues. The interfollicular tissues are also infiltrated with lymphocytes. There are many CD81 T cells at the reticular epithelium but fewer in the mucosal epithelium. Many of the cells in the mucosal epithelium appear to be γ/δ T cells [3]. The lingual tonsil is located at the root of the tongue where it consists of spherical macroscopic lymphoid nodules covered by keratinized stratified squamous epithelium [20]. The velar tonsil consists of lymphoid nodules with associated crypts in the soft palate. The tubal tonsil is located on the lateral walls of the nasopharynx and the paraepiglottic tonsil is located at both sides of the base of the epiglottis [20]. Dromedaries possess significant amounts of bronchus-associated lymphoid tissue. These consist of lymphoid nodules that may vary from a few clustered lymphocytes to relatively large structures containing a germinal center. They may be large in young camels, depending upon exposure to inhaled antigens, but they involute with advancing age [21].

Tylopoda: Camels and llamas Chapter | 14

209

FIGURE 14.2 The gastric lymphoid tissue of a camel. (A): A view of the mucosa illustrating the approximately triangular shape of the aggregated lymphoid nodule area (ALNA) along the ventral wall of the isthmus from the origin of the proximal enlargement and along the curvature verntriculi minor. (a): glandular sac area of the second compartment of the stomach: (b): Isthmus. (c): ALNA; and (d) non-ALNA of the cardiac glandular region of the third compartment. (B). Histological characteristics of mucosal folds of ALNA from the abomasum of Bactrian camels of different ages. From Zhang W-D et al. Vet Immunol Immunopathol 2012 147:147
153. With permission.

The camels, although they are forestomach fermenters and regurgitate their digesta in the same manner as ruminants, have a distinctly different digestive tract. Their forestomach is divided into three compartments. Unlike the ruminants, all three compartments have glandular regions and none have papillae. The third compartment of the Bactrian camel, called, for convenience, the abomasum, contains a large, aggregated area of lymphoid nodules (ALNA) in the cardiac region (Fig. 14.2). This structure has not been reported in any other species including the dromedary [22]. The ALNA is confined to a long, triangular region along the ventral wall of the isthmus from the origin of the proximal enlargement along the ventral curvature. The mucosal folds in this area are much thicker than in the rest of the mucosa. The folds contain as many as 26 lymphoid follicles. The follicles are distributed on both sides of the mucosal folds and are visible as round structures. They are covered by a follicle-associated epithelium. Germinal centers are obvious within the follicles. The ALNA enlarges prior to puberty around 3
5 years of age but regresses thereafter so that by 17
20 years the structure is severely atrophied. IgA1 B cells are diffusely scattered through the lamina propria of these aggregated lymphoid nodules.

14.4.5.1 Bactrian camel The mucosal-associated lymphoid tissues in the form of Peyer’s patches or isolated follicles and lamina propria lymphocytes are present along the entire intestinal submucosa and lamina propria of the Bactrian camel. The overall shape of these patches is highly variable. Some are nodular or cystic, a structural type unique to camels. The number of Peyer’s patches tends to increase with age to reach a peak around 5 years before declining. In the large intestine, they are mainly located at the ileocecal junction, in the first third of the colon, and the beginning of the cecum [23].

14.4.5.2 Dromedary The ileal PP in the dromedary are dark rose-colored, cup-shaped elevated structures arranged in three irregular rows [24]. Their numbers range from 25 to 27 in the anterior ileum and 31
38 in the posterior ileum [3]. The latter are located in the terminal 20 cm near the ileocecal junction. The central row is, as in other species, antimesenteric. Each patch is formed by several elongated dome regions. There is no evidence of any long continuous ileal patch as seen in other artiodactyls. The domes are formed by one to three secondary lymphoid follicles in the submucosa covered by a typical dome-associated epithelium that contains both columnar epithelial cells and M cells but not goblet cells. This epithelium is infiltrated by lymphocytes. The lymphoid follicles contain obvious germinal centers. There are high endothelial venules in the interfollicular region. Their numbers decline in adults. (IPP disappear at 15 months of age in sheep and goats, in cattle over 2 years of age, and in pigs by 4). Peyer’s patches are not present in the duodenum or jejunum [11]. IgA1 B cells are diffusely scattered through the lamina propria.

14.5

The major histocompatibility complex

The MHC of the camels has an overall organization similar to that of other artiodactyls. The MHC is located on the long arm of chromosome 20 in both Old- and New-world camels and organized in the conventional Class II-Class IIIClass I order. The Old-world camels show relatively limited diversity in both their MHC class I and class II genes. This

210

SECTION | 2 Mammalian orders

is likely linked to a loss of wild populations, years of selective domestication, and loss of habitat. The nucleotide sequences in some subregions are more similar to the pig rather than to cattle [25]. Thus for example, they have a similarly organized class II subregion. Camels lack functional DY genes, and their MIC genes in the class I region are also organized differently. In general, Old-world camels have significantly lower diversity in their class I and class II genes than do New-world camels, despite their different geographic origins [26]. The MHC loci are highly conserved among the camelids so that the MHC class I of V. pacos is almost identical to that in the Old-world camels [27].

14.5.1 Major histocompatibility complex class I While the overall structure of the camel MHC class I region is well conserved, camelid class I genes have relatively little diversity (Fig. 14.3) [28]. For example, in selected MHC class I genes such as B-67 and BL3
7 there is unexpectedly low variability when compared to other mammals. In the classical B-67 locus there is only one synonymous substitution, and it is shared between dromedaries and Bactrian camels. BL3
7 is predicted to be orthologous to HLA-E but it only shares 48% nucleotide homology with humans. Camel β2-microglobulin is encoded by a gene on chromosome 6. Contrary to the situation in other mammals, the class I-related MR1 and MICA loci are more polymorphic than the genes in the classical class I locus. Their products are important ligands for NK cell receptors. Thus the MR1 protein is an antigen-presenting molecule encoded by a gene located outside the MHC (Chapter 10). It presents metabolites of bacterial vitamin B2 and plays an important role in regulating the microbiota since it recognizes microbial antigens within the intestine [28]. MR1 is highly polymorphic in camels with 170 allelic sites. These are partially shared between Bactrians and dromedaries. It is very similar to cattle MR1 showing 79.9% amino acid homology with Bos taurus and 80.2% homology with Sus scrofa. Similarly, the MICA gene encoding the class I related stress signaling molecule recognized by NKG2 on NK cells and T cells contains 40 SNPs. It is closely related to MICA in other camelids as well as in humans (58.6 nucleotide identity with H. sapiens) [28].

14.5.2 Major histocompatibility complex class II There is very little polymorphism among the DRA, DRB, and DQB genes in the three old-world camel species. Thus among the DRA genes, there is only one synonymous and one nonsynonymous SNP shared between all three species [29]. DRB has five polymorphisms, DQA in the Bactrian camel has 11 SNPs of which four are synonymous, and it shares nine with C. ferus. DQB is the most polymorphic gene with 21 SNPs as well as a 12 bp functional insertion not found in other mammals. There is no DYA locus in the camel MHC. However, it does contain a nonfunctional DYA fragment as do pigs. While the overall structure of the class II region in Old- and New-world camelids is very similar, Alpacas possess MHC-B and -E genes that are absent in the Old-world camels while Bactrian camels possess an HLA-A-like, and an MHC-G gene that is absent in the other species [27].

14.5.3 Major histocompatibility complex class III The camel class III region is well conserved. It has an organization similar to that seen in pigs and cattle [25]. Like others, it contains the genes encoding TNF-α and the Ly6 gene family. There are three SNPs within the TNF-α gene that are shared between the Old-world camels.

14.5.4 The Natural Killer receptor complexes Natural killer cells have been described in both Old- and New-world camels. The organization of the NK cell receptor genes in camels differs significantly from that of cattle but closely resembles those found in pigs [30]. As in other mammals, two gene complexes encode NK cell receptors
the natural killer complex (NKC) and the leukocyte receptor complex (LRC). The LRC contains genes encoding receptors with Ig-like protein domains while the NKC encodes receptors with C-type lectin domains. TAP1 DQB DQA DRB DRA

II

MICA BL3-7 A126 B67

III

Ib

Ia

FIGURE 14.3 The overall structure of the dromedary major histocompatibility complex.

Tylopoda: Camels and llamas Chapter | 14

211

14.5.4.1 Leukocyte receptor complex The camel LRC is located on chromosome 9. It is about 0.7 mb in length and contains fifteen genes. These consist of six LILR genes arranged in two clusters, two KIR pseudogenes, and a single NCR1 gene. It also contains a gene encoding a unique Ig-like receptor comprising two Ig-like domains and a long cytoplasmic tail. This appears to function as an inhibitory receptor. A similar protein has been described in pigs. There is a high sequence similarity between the alpaca (V. pacos) and the Old-world camel species in these genes. Their level of polymorphism is low [30]. In addition to NRC1, the NCR2 and NCR3 genes are located outside the LRC on chromosome 20. Of the three, only NCR1 and NCR2 appear to be functional. NCR3 appears to be a pseudogene.

14.5.4.2 Natural killer complex The camel NKC is located on chromosome 34. It encompasses 0.9 mb and contains 26 genes encoding proteins with Ctype lectin-like domains. One cluster encodes a minimal set of seven functional KLR genes (KLRA, -C1, C2, -D, -E, -I, and -K) as well as two pseudogenes and thus resembles the organization in the pig. Camel KLRC is unique in that it is the only gene family with two members that deliver opposing signals. Thus KLRC1 codes for an inhibitory receptor while KLRC2 encodes an activating receptor. The other cluster of genes codes for members of three other receptor families, KLRB, -F, and -G. These consist of both activating and inhibitory receptors [30].

14.6

B cells and Immunoglobulins

14.6.1 Old-world camels Camels employ the usual four major immunoglobulin classes, IgM, IgG, IgE, and IgA (Fig. 14.4) [31,32]. However, while the common four chain, Y-shaped structure is ubiquitous throughout the class Mammalia, camels are different. Normally, each immunoglobulin molecule consists of two heavy and two light chains. Each heavy chain consists of a variable domain linked to a chain of three or four constant domains. The light chains consist of one variable and one constant domain. Normally, the paired variable domains from a light and a heavy chain then interact to form an antigen-binding site. These domains are variable since, during B cell development, the light chain genes undergo sequential rearrangement by recombining variable (V) and joining (J) gene segments in the kappa (IGK) and lambda (IGL) light chains. In the heavy chain genes (IGH) the complete V region gene is formed by combining one V with diversity (D) and J gene segments to form complementarity determining region 3 (CDR3). The recombined V gene segments are then joined to the constant domain genes to encode a complete heavy chain.

14.6.1.1 IGHM Camel IgM is of conventional structure
A pentameric star-shaped molecule formed by five Y-shaped immunoglobulin units. The IgM heavy chains (μ) of the Bactrian camel have an overall amino acid sequence similarity of 64% to humans, 60% to mice, 97% to dromedary, and 95% to alpaca μ chains. Alignment analysis showed that the location of 5’ IGH

50VH /42VHH V

6

D

J

3’ M

D

G2

G1

G3

E

53

IGK

IGL

6

33 10 KO, 10 CO

M

D

G

V

D

E J

A LC

A

FIGURE 14.4 The arrangement of the dromedary immunoglobulin genes. Note that there are two IgG1 isotypes, three IgG2 isotypes (a, b, and c) and possibly two IgG3 isotypes in llamas. Camels have only two IgG1 isotypes and two IgG2 isotypes (a and c).

212

SECTION | 2 Mammalian orders

the cysteines that are responsible for inter- and intra-chain bonding in IgM are conserved between the different camel species. The cysteine residues in the CH3 domain that are responsible for the disulfide bonds that link the IgM units to form the pentameric structure are also conserved [33].

14.6.1.2 IGHG Nine IGHG constant region genes have been detected in the dromedary genome. Five are known to be functional and encode heavy chains. Four other IGHG genes have been detected but no products have been detected so they are assumed to be pseudogenes. Members of the camel family from both the Old and New Worlds make three IgG subclasses: IgG1, IgG2, and IgG3. As discussed below, IgG1 is of conventional structure and contains two heavy and two light chains. However, neither IgG2 nor IgG3 contain light chains and so rely exclusively on their paired heavy chains for antigen binding.

14.6.1.3 IgG1 Camel IgG1 is of a conventional four-chain structure. It consists of two identical heavy chains paired with two identical light chains and connected by inter-chain disulfide bonds. Its heavy chain genes contain four constant exons, (V, CH1, CH2, CH3), and these chains have a molecular weight of 170 kDa. Camel conventional IgG1 accounts for up to 25% of their circulating IgG. There are two IgG1 allotypes, IgG1a and IgG1b [33]. Different hinge regions are used for the IgG1a (19 amino acids) and 1b (12 amino acids) allotypes.

14.6.1.4 IGHE Camel IgE also appears to be a conventional four-chain immunoglobulin. The Bactrian camel IGHE gene shares 98.1% nucleotide identity with the corresponding regions in the alpaca, and 55.2% with the human IGHE gene [31].

14.6.1.5 IGHA Camel IgA is unremarkable. The bactrian camel IGHA gene sequence is homologous to human and alpaca IgA with 62.8% and 91.3% nucleotide homology, respectively [31].

14.6.1.6 IGHV All camel immunoglobulin heavy chains have an N-terminal V domain. This domain consists of four relatively conserved framework segments interspersed with three CDRs. CDR 1 and 2 are encoded by the germline IGHV gene alone while CDR3 is encoded by the V-(D)-J rearranged gene segments. Camel immunoglobulins have distinct differences in their CDR lengths with CDR1 being 5
7 residues and CDR2 being 16
17 residues in length. CDR3, in contrast, is highly variable since they employ two distinctly different types of IGHV genes. These encode a conventional VH and an unconventional VHH domain [33]. The average CDR3 length in the dromedary VHH and VH domains are 17
18 and 11
13 amino acids respectively [34,35]. The average CDR3 length of the llama VHH domain is 15 amino acids. These differences will of course, directly influence the size of the antigen-binding site. Camel IgM V genes belong to clan III. Likewise, 60% of their IgG1 variable region sequences align with the human IGHV3 family (clan III) [33]. The remaining V gene sequences align with Clan II.

14.6.2 Light chains Camel light chains differ greatly in size. While lambda chains average about 33 kDa, at least 10 different-sized IGLC gene sequences have been identified. About five of these C genes are probably a result of allotypic variation rather than the products of multiple genes. All camel kappa chains share the same C region indicating the presence of a single functional IGKC gene. Though dromedary genomes show the presence of 53 IGKV and 33 IGLV sequences, relatively few are used in any single individual [33]. They show great positional variability, especially in framework region 2. All these variations appear to be due to single amino acid substitutions suggesting that somatic mutation rather than recombination is the main mechanism of their variation. The IGLV genes belong to subfamilies IGLV2 and IGLV3. These camel lambda V regions cluster with the clan I, human IGLV1, and mouse IGLV2 sequences [33]. The camel IGKV regions cluster with clan II, human IGKV2, and mouse IGKV1 sequences

Tylopoda: Camels and llamas Chapter | 14

14.7

213

Heavy-chain-only antibodies

The dromedary has three heavy-chain-only IgG subclasses (IgG2a, IgG2c, and IgG3). They are constructed from paired 100-kDa heavy-chain disulfide-linked dimers unconnected to light chains (Fig. 14.5). These heavy-chain-only antibodies (HCAbs) are produced as a result of a mutation in the camel CH1 splicing site where light chains normally attach to the heavy chains. HCAbs are significantly smaller than conventional immunoglobulins being 90 kDa compared to about 150 kDa [34]. The three camel HCAbs share the IGH locus with conventional IgG1. Three constant genes, IGHG2a, IGHG2c, and IGHG3 encode their heavy chains. All three lack the first heavy chain constant exon, and so encode only one hinge and two constant domains (Fig. 14.5). As a result, they cannot bind to light chains [36]. Normally, the CH1 domain contains a cysteine that can form a disulfide bond that connects the heavy chain to a light chain. This does not happen in this these antibodies because of a point mutation on the donor splicing site from G to A at the first constant exon/intron boundary. This disrupts the consensus splicing sequence and may be the cause of the loss of CH1 [37]. As a result, the heavy chain variable domain is joined directly to the hinge region in the HCAb. Thus they have VHH, H, CH2, and CH3 domains. The IGHG2a, IGHG2c, and IGHG3 genes encode hinge regions of 35, 15, and 12 amino acids, respectively.

14.7.1 V domain structure HCAbs constant region genes are linked to IGHV region genes encoded in germline sequences and that undergo conventional V(D)J recombination. The conventional IGHG1 links to a “normal” VH domain whereas the heavy-chain-only immunoglobulins link to specialized VHH domains that are fully capable of binding antigen by themselves. Each VHH gene contains all the essential components of the VH region such as a promoter region and the recombination signal sequences [34]. The IGHV region sequences encode both VH and VHH domains. There are a total of 33 VHH and 39 VH unique sequences encoded by 42 and 50 different genes respectively in the dromedary. The VHH domain and the conventional VH domains share considerable sequence identity, but VHH domains possess some unique features. In a conventional IGH locus, VH domains normally link to CH1 domains. VHH domains cannot do so because of the defect in the CH1 donor splice site of HCAbs. In these cases, the VHH domains link directly to the hinge regions. In some molecules, the VH genes may contribute to the HCAb V gene diversity. In these cases, the IGHV domain attaches to the hinge using a short junction without an additional cysteine. The IGHV4 subfamily genes may also contribute to VH and VHH to generate both conventional IgG and HCAbs.

14.7.2 VHH gene segments The 33 VHH gene segments fall into seven subfamilies of which five are expressed. The produced immunoglobulins preferentially use the V gene segments located closest to the D-gene cluster. An analysis of the VHH gene segments in C. dromedarius has shown that they belong to the IGHV3 subfamily (clan III) [38]. The HCAb repertoire is also largely generated by somatic hypermutation in addition to recombination as well as by a large number of N-region additions and deletions. Both VH and VHH likely make use of some common D gene segments. Both also appear to have identical 3’ sequences downstream of the DH-JH segments. In normal VH domains, the amino acids in all three CDRs but especially CDR3 interact with the antigen. As a result, CDR3 plays a dominant role in determining antibody affinity and specificity. The CDR1 and CDR2 sequences of the VHH domain are very similar to those in the VH of conventional camel IgG1. However, their CDR3 region is much longer and more variable. VHH domains also contain three CDR hypervariable regions. In conventional VH domains, amino acids at positions 42, 49, 50, and 52 (IMGT numbering) usually form a hydrophobic surface that associates with the light chain variable domain to form the antigen-binding groove. In VHH domains, these amino acids are replaced by more hydrophilic amino VH

VHH VHH 35 amino acids

IgG1

IgG2

12 amino acids

IgG3

FIGURE 14.5 The structural organization of the dromedary conventional immunoglobulin, IgG1 and two different heavy chain immunoglobulins, IgG2 and IgG3. The two IgG2 isotypes differ in the size of their hinge region. The red arrow denotes the antigenbinding site.

214

SECTION | 2 Mammalian orders

L

VHH

L

D J

H

CDR1 CDR2 FR1

FR2

CH2

CH3

CDR3 FR3

FR4

S-S FIGURE 14.6 The structure of a heavy chain from dromedary IgG2 or IgG3. Note that the CDR3 of the VHH region is cross-linked to the CDR 1 region by an intrachain disulfide bond. L. Leader sequence.

acids. As a result, in addition to failing to bind light chains, these changes enhance the solubility of HCAbs. The long hydrophilic CDR3 in the VHH domain greatly increases the shape and sequence diversity of the antigen-binding site. (Interestingly, these differences are greater in C. bactrianus than in C. dromedarius or the llama family). The CDR3 region of the VHH domain has an unusual structure that may enable it to bind unusual epitopes [37]. Thus in both the Old- and New-world camels the long rearranged CDR3 of the VHH domain contains a pair of extra cysteines that form an intrachain disulfide bond. This links to either position 50 in C. dromedarius or position 55 in L. glama. This intrachain bond stabilizes the VHH domain and ensures that the CDR3 sequence remains fixed in an optimal configuration [26]. (Fig. 14.6).

14.7.3 Heavy Chain-only Antibody functions Camel HCAbs have different biological properties than conventional immunoglobulins. They can still interact with Fc receptors on macrophages and opsonize pathogens and form immune complexes. They can also neutralize viruses as shown by plaque reduction assays. On the other hand, camel HCAbs appear to be poor agglutinins probably because their antigen-binding sites are located so close together. IgG2 does not appear to activate complement by the classical pathway however, IgG3 from a dromedary immunized with sheep red cells can trigger complement-mediated lysis [34]. HCAbs have shorter half-lives than conventional immunoglobulins as a result of their much smaller size. Both HCAbs (IgG2 and IgG3) have a half-life of about 14 days compared to IgG1 which has a half-life of 20
25 days Despite lacking light chains, camel HCAbs can still bind to many antigens with high affinity. It has been noted that these antibodies bind especially effectively to the substrate pockets of enzymes. The antigen-binding site on these VHH domains is convex. This enables them to fit snugly into the concave active site on an enzyme. The CDR 1 and CDR2 regions are larger and have greater structural variability in VHH than classical VH domains. Functionally, HCAbs have greater access to cryptic epitopes because of the compact nature of their paratopes. As a result, VHH domains are very effective at neutralizing the active sites of enzymic antigens which are often difficult for conventional VH-VL antigen binding sites to reach. The unique properties of these HCAbs have made them of special interest to the biotechnology industry. These "nanoantibodies" are under investigation for many therapeutic uses as well as being a valuable research tool.

14.8

New-world camels

The New-world camel IGH locus is located on chromosome 4. It encodes IgM, IgA, and IgE in addition to five IgG subclasses. These are IgG1a, IgG1b, IgG2b, IgG2c, and IgG3 (Fig. 14.4). IgG1 is of a conventional four-chain structure. IgG2 and IgG3 are heavy-chain only antibodies. The llama possesses two IgG1 allotypes (a and b) and three IgG2 allotypes (a, b, and c). Alpaca IgG3 consists of two variants that may be isotypes [39]. As in the Old-world camels, the IGH locus contains both VH and VHH gene segments followed by a unique DH-JH cluster and multiple C region genes. Alpaca B cells pass through an initial IgM- expressing stage and subsequently undergo a class switch similar to that in conventional antibodies [40].

14.8.1 Constant domains The llama HCAbs, IgG2a, and IgG3 are almost identical to dromedary IgG2 and IgG3. As in the camels, the CH1 domain is missing and as a result, their heavy chain is of reduced length and cannot link to light chains [41]. The donor

Tylopoda: Camels and llamas Chapter | 14

215

splice site flanking the CH1 exon is mutated in genes encoding the short heavy chain so that the resulting CH1 domain cannot splice to the hinge region. In llamas, the predominant IgG2 isotypes are G2a and G2c. Their HCAbs consist of long-hinge IgG2b and shorthinge IgG2c. Analysis of B cells shows that IgG2b1 and IgG2c1 B cells are present in similar proportions in the repertoire. They also have two subsets of IgG2b and IgG2c antibodies that entirely lack a hinge domain and so employ direct VHH-CH2 splicing. These hingeless antibodies in llamas may result from somatic hypermutation or errors in transcript processing. In addition to the changes that prevent light chain binding, there are also changes in the two other intrachain disulfide loops that affect the conformation of the VHH antigen-binding site. The llama IGH locus contains six functional germline V gene segments. One encodes conventional VH segments whereas the other five encode VHH segments. Published camel VHH sequences are classified into four subfamilies [8]. Three of these subfamilies are present in llamas but absent from dromedaries. Further analysis of these subfamilies indicates the very long CDR3 with the intrachain disulfide bond is present in only two of the subfamilies. The CDR3 length in IgG2b and IgG2c is modestly longer (2
3 residues) than in IgG1 (less than in camels .5 residues) largely as a result of N-region addition. The VHH subfamilies also differ from each other and conventional VH domains in the structure of their CDR1 and CDR2 as well as in CDR3 length and sequence variability. All the llama VHH clones belong to clan III based on the structure of their framework regions [42]. IgG2b and G2c use a more restricted set of V region genes than IgG1. IgG2b and G2c also have increased somatic mutation rates in both their framework and CDR regions compared to IgG1 [43]. The high sequence similarity between the IGHG constant regions in both Old- and New-world camels supports the concept that heavy-chain-only antibodies appeared after the tylopoda diverged from the other artiodactyls. The camel heavy-chain antibodies IgG2 and IgG3 make up about 50%
75% of all the serum IgG. However, in the llama, they only make up about 30%. On a molar basis, they make up a higher proportion of antibody molecules (62%
82% in camels and 43% in llamas). All the γ chain variants are found in similar relative concentrations in alpaca serum, colostrum, and milk [39]. However, following the passive transfer, the concentration of the HCAbs in newborn cria serum drops faster than the concentration of regular IgG1 - unsurprising considering its shorter half-life.

14.9

T cells and cell-mediated immunity

Camels belong to the “γ/δ-high” group of mammals. Up to 35% of the circulating T cells in young camels use γ/δ chains for their antigen receptors (Fig. 8.9). As in other artiodactyls, young camels have a higher percentage of γ/δ T cells than adults. Adult blood lymphocytes in dromedaries consist on average, of about 26% B cells, 25% CD41 T cells, and 7.4% γ/δ T cells [11]. Expression of the integrin CD11a differentiates naive and effector CD41 T cells [11]. Many camel blood lymphocytes also express WC1 (Chapter 15) [3,14]. T cell antigen receptors (TCR) consist of paired heterodimeric peptide chains that are each constructed from an N-terminal V domain and C terminal constant domains. Conventionally the variable domain of the α and γ TCR chains is formed by V-J recombination whereas the β and δ chain genes also employ D gene segments so that their V domains result from V-(D)-J recombination. α/β T cells only recognize a peptide antigen when it is bound to the groove of an MHC class I or class II molecule on an antigen-presenting cell. In addition, and uniquely among mammals, in C. dromedarius, the rearranged variable regions of their TCR γ and δ chains can undergo somatic mutation to increase their antigen-binding diversity. The paired V domains come together to form the antigen-binding site of the TCR. The six CDR regions of the paired V domains of TCTγ/δ collectively form the antigen-binding site. However, in TCR α/β only the paired CDR3s recognize the antigen. The two germline-encoded CDRs (CDR 1 and CDR2) recognize the external conformation of the antigen-presenting MHC. Camel γ/δ T cells behave somewhat differently from α/β T cells. Thus they prefer epithelial and mucosal sites rather than the lymph node paracortex and blood circulation. They can also employ an immunoglobulin-like antigen binding process in addition to an MHC-restricted one. Thus they can bind some free antigens in the absence of an MHC molecule. In effect, they can participate in both innate and adaptive immune responses.

14.9.1 TRA/D As in other mammals, the T cell receptor delta (TRD) locus is embedded within the TRA locus [44]. It is 870 kb in size and contains a limited number of TRA genes (83), compared to sheep (307), or cattle (183) As a result, the TRA locus relies on somatic mutation for the generation of much of its diversity. The entire dromedary TRA/D locus is arranged thus from 5’ to 3’; 83 TRAV genes, two TRAV/TRDV genes, mingled with only 12 TRDV genes. These are followed

216

SECTION | 2 Mammalian orders

DROMEDARY

TRA/D

TRB

5'

*3

88

6

AV/DV

DV

DD DJ DC DV

34

6

7

TRG

*

4

2

V

3'

60

4

AJ

AC

FIGURE 14.7 The structure of the three dromedary T cell receptor loci. The asterisks denote the delta and gamma V genes that may undergo somatic hypermutation and so can generate greatly increased receptor diversity.

6

2

D

J

C

subject to * Genes somatic hypermutation

by six TRDD, four TRDJ, one TRDC and as in other mammals a single TRDV gene in the opposite orientation. The locus is completed with 60 TRAJ segments and a single TRAC gene (Fig. 14.7). Twenty-six of the V genes are arranged in the opposite orientation. The 83 TRAV genes are classified into 33 different subfamilies. Of these genes, 36 appear to be functional with the remainder being either pseudogenes or ORFs. The twelve TRDV genes are assigned to five subfamilies but only TRDV1 has multiple members. Three of the TRDV genes appear to be pseudogenes [44]. The TRDC gene encodes a chain of 156 amino acids and is composed of three translated exons and a single untranslated one. The TRAC gene has a similar structure and encodes a peptide chain of 121 amino acids.

14.9.2 TRB The organization of the TRB locus is similar to that in other artiodactyls with a pool of TRBV genes located at the 5’ end of three tandem D-J-C clusters followed by a single TRBV gene in an inverted orientation at the 3’ end [45]. Thus this pattern was almost certainly established prior to the Tylopoda divergence as a result of a duplication event and unequal crossing over between the ancestral TRBC1 and TRBC2 genes [46]. Compared to T cell receptor gamma (TRG) and TRD, TRB diversity is not affected by somatic hypermutation but relies only on combinatorial and junctional diversity. All three TRBDJ-C clusters are used to generate a functional β chain. The pool of TRBV genes is smaller in camels than in other mammals. Thus there are 30 genes in C. ferus, 33 in C. dromedarius, and C. bactrianus arranged in 26 different subfamilies. This is a low number compared to 134 TRBV genes in bovines, 68 in humans, and 77 in rabbits but about the same as the 43 in the pig. Although the gene content of the TRB locus is similar among the camels, their functional receptor repertoire differs insofar as genes that are functional in one species may be pseudogenes in another. The TRBV genes in the camels form a monophyletic group with other mammals including humans, dogs, sheep, and pigs, supporting their ancient origin.

14.9.3 TRG The dromedary has a single TRG locus organized into three V-J-J-C cassettes. The TRG locus spans about 105 kb. It consists of six functional TRGV genes, three functional TRGJ genes, and three TRGC genes [45,47].

14.9.4 Somatic hypermutation Somatic hypermutation has not been detected in the T cell receptor genes of any other mammalian species. The absence of somatic hypermutation in α/β T cells in most mammals is attributed to the requirement that their CDR1 and CDR2 regions must be able to bind to MHC molecules for antigen recognition purposes. Thus the diversity of their α/β TCRs is generated solely by combinatorial and junctional diversity [11]. Dromedaries have a limited number of TRDV and TRGV genes. However, the binding repertoire of their γ/δ T cell receptors is significantly expanded by somatic hypermutation. Somatic hypermutation acts on both the γ and δ chains of the dromedary TCR [44]. In general, the process appears to be similar to the process in B cell antigen receptors.

Tylopoda: Camels and llamas Chapter | 14

217

Analysis of the TRD mutations indicates that they result from sequential point mutations. They are not the result of gene conversion. The mutations tend to occur in the bone marrow and spleen. They result from the combined action of AID, uracil-DNA glycosylase, and mismatch repair pathways. Dromedary TRG and TRD evolution is favored by productive rearrangements in rearranged TRG V-J and TRD V-D-J genes generating a large, diversified T cell repertoire [7]. The mutation profiles of the TRGV and TRDV genes are similar to those of immunoglobulins in other mammals. Replacement mutations in the framework regions occur early in the lineages and are related to enhanced structural stability. There is no asymmetry in targeting G:C or A:T base pairs. Once completed, clonal expansion of mutated productive rearrangements occurs [44]. γ/δ T cells unlike α/β cells can bind free, soluble antigens as well as with nonclassical MHC molecules and generally have limited diversity. Somatic hypermutation increases this γ and δ chain diversity and also potentially increases their binding affinity for soluble antigens. Alpacas (V. pacos) possess a subset of γ/δ T cells called Vγ9Vδ2T cells whose activation depends upon their receptors binding phosphoantigens. These phosphoantigens mediate interaction with B7-like butyrophilin molecules (BTN-3) in an MHC independent manner [48]. (Box 8.2). This unique T cell subset has only been reported previously in primates and is a major subpopulation in human blood! [49].

References [1] Jirimutu E, Wang Z, Ding GH, Chen GL, et al. Genome sequences of wild and domestic bactrian camels. Nat Commun 2012;3:1202. Available from: https://doi.org/10.1038/ncomms2192. [2] Ghazi SR, Oryan A, Pourmirzaei H. Some aspects of microscopic studies of the placentation in the camel (Camelus dromedarius). Anat Histol Embryol 1994;23:337
42. [3] Al Ramadan SY, Al-Mohammed Salem KT, Alshubaith IH, Al-Ali AM, et al. Selected aspects of camel immune system and immune responses. Open J Vet Med 2021;11:177
211. [4] Konuspayeva G, Faye B, Loiseau G, Levieux D. Lactoferrin and immunoglobulin contents in camel’s milk (Camelus bactrianus, Camelus dromedarius, and hybrids) from Kazakhstan. J Dairy Sci 2007;90:38
46. [5] Bravo PW, Garnica J, Fowler ME. Immunoglobulin G concentrations in periparturient llamas, alpacas and their crias. Small Rumin Res 1997;26:145
9. [6] Johnson EH, Al-Habsi KR, Al-Busaidi RM. A review of observations made on select parameters of the camel immune system. Ag Mar Sci 2013;18:1
6. [7] Ciccarese S, Vaccarelli G, Lefranc M-P, Tasco G, et al. Characterization of the somatic hypermutation in the Camelus dromedarius T cell receptor gamma (TRG) and delta (TRD) variable domains. Dev Comp Immunol 2014;46(2):300
13. Available from: https://doi.org/10.1016/j. dci.2014.05.001. [8] Harmsen MM, Ruuls RC, Nijman IJ, Niewold TA, et al. Llama heavy-chain V regions consist of at least four distinct subfamilies revealing novel sequence features. Mol Immunol 2000;37:579
90. [9] Hawkey CM. Comparative mammalian hematology. Cellular components and blood coagulation of captive wild animals. London: Heinemann Medical Books; 1975. [10] Foster A, Bidewell C, Barnett J, Sayers R. Hematology and biochemistry of alpacas and llamas. Practice 2009;31:276
81. [11] Hussen J, Schuberth H-J. Recent advances in camel immunology. Front Immunol 2020;11:614150. Available from: https://doi.org/10.3389/ fimmu.2020.614150. [12] Hussen J, Shawaf T, Al-herz AI, Alturaifi HR, et al. Expression patterns of cell adhesion molecules on CD41 T cells and WC11 T cells in the peripheral blood of dromedary camels. Pak Vet J 2018;38(3):231
6. [13] Choudhary J, Purva M, Milind M, Gahlot K, et al. Phylogenetic and protein domain architecture analysis of toll-like receptor 4
10 genes in camel (Camelus dromedarius). Int J Curr Microbiol Appl Sci 2021;10(1):2887
95. Available from: https://doi.org/10.20546/jcmas.2021.1001.335. [14] Premraj A, Aleyas AG, Nautiyal B, Rasool TJ. Camelid type I interferons: Identification of functional characterization of interferon alpha from the dromedary camel (Camelus dromedarius). Mol Immunol 2020;119:132
43. [15] Odbileg R, Lee S-I, Yoshida R, Chang K-S, et al. Cloning and sequence analysis of llama cytokines related to cell-mediated immunity. Vet Immunol Immunopathol 2004;102(1
2):93
102. [16] Odbileg R, Lee S-1, Ohashi K, Onuma M. Cloning and sequence analysis of llama (Lama glama) Th2 (IL-4, IL-10, and IL-13) cytokines. Vet Immunol Immunopathol 2005;104(3
4):145
53. [17] Zidan M, Kassem A, Dougbag A, Ghazzawi El, et al. The spleen of the one humped camel (Camelus dromedarius) has a unique histological structure. J Anat 2000;196:425
32. [18] Abdel-Magied EM, Taha AA, al-Qarawi Am Elfaki MG. The parotid, mandibular and lateral retropharyngeal lymph nodes of the camel (Camelus dromedarius). Anat Histol Embryol 2001;30(4):199
203. [19] Zidan M, Pabst R. Histological, histochemical and immunohistochemical study of the haemal nodes of the dromedary camel. Anat Histol Embryol 2004;33(5):284
9. [20] Al-Ramadan SY, Alluwaimi AM. The lymphoid tissue in the palatine tonsils of the dromedary camel (Camelus dromedarius). J Camel Pract Res 2018;25(1):65
73.

218

SECTION | 2 Mammalian orders

[21] Elhussieny O, Zidan M. Temporospatial characterization of the bronchus associated lymphoid tissue (BALT) of the one humped camel (Camelus dromedarius). Trop Anim Hlth Prodn 2021;53(2):265. Available from: https://doi.org/10.1007/s11250-021-02694-3. [22] Xu X-H, Wang W-H, Gao Q, Qi S-S, et al. The anatomical characteristics of the aggregated lymphoid nodule area in the stomach of Bactrian camels (Camelus bactrianus). Vet J 2010;362
5. [23] Zhaxi Y, Wang W, Zhang W, Gao Q, et al. Morphological observation of the mucosa-associated lymphoid tissue in the large intestine of Bactrian camels (Camelus bactrianus). Anat Rec 2014;297:1292
301. [24] Zidan M, Pabst R. Unique microanatomy of ileal Peyer’s patches of the one humped camel (Camelus dromedarius) is not age-dependent. Anat Rec 2008;291:1023
8. [25] Plasil M, Wijkmark S, Elbers JP, Oppelt, et al. The major histocompatibility complex of old-world camels
a synopsis. Cells 2019;8 (10):1200. Available from: https://doi.org/10.3390/cells8101200. [26] Ciccarese S, Burger PA, Ciani E, Castelli V, et al. The camel adaptive immune receptors repertoire as a singular example of structural and functional genomics. Front Genet 2019;10:997. Available from: https://doi.org/10.3389/fgene.2019.00997. [27] Richardson MF, Munyard K, Croft LJ, Allnutt TR, et al. Chromosome-level alpaca reference genome VicPac3.1 improves genomic insight into the biology of New World camelids. Front Genet 2019;10:586. Available from: https://doi.org/10.3389/fgene.2019.00586. [28] Plasil M, Wijkmark S, Elbers JP, Oppelt J, et al. The major histocompatibility complex of Old- World camelids: Class I and class I-related genes. HLA 2019;93:203
15. [29] Plasil M, Mohandesan E, Fitak RR, Musilova P, et al. The major histocompatibility complex in old world camelids and low polymorphism of its class II genes. BMC Genomics 2016;17:167. Available from: https://doi.org/10.1186/s12864-016-2500-1. [30] Futas J, Oppelt J, Jelinek A, Elbers JP, et al. Natural killer cell receptor genes in camels: another mammalian model. Front Genet 2019;10:620. Available from: https://doi.org/10.3389/fgene.2019.00620. [31] Liang Z, Wang T, Sun Y, Yanf W, et al. A comprehensive analysis of immunoglobulin heavy chain genes in the Bactrian camel. Front Ag Sci Eng 2015;2(3):249
59. [32] Ming L, Wang Z, Yi L, Batmunkh M, et al. Chromosome level assembly of wild bactrian camel genome reveals organization of immune gene loci. Mol Ecol Resour 2020;20:770
80. [33] Griffin LM, Snowden JR, Lawson ADD, Wernery U, et al. Analysis of heavy and light chain sequences of conventional camelid antibodies from Camelus dromedarius and Camelus bactrianus species. J Immunol Methods 2014;405:35
46. [34] Conrath KE, Wernery U, Muyldermans S, Nguyen VK. Emergence and evolution of functional heavy chain antibodies in Camelidae. Dev Comp Immunol 2003;27:87
103. [35] Li X, Duan X, Yang K, Zhang W, et al. Comparative analysis of immune repertoires between Bactrian camel’s conventional and heavy-chain antibodies. PLoS One 2016;11(9):e0161801. Available from: https://doi.org/10.1371/journal.pone.0161801. [36] Brooks CL, Rossotti MA, Henry KA. Immunological functions and evolutionary emergence of heavy chain antibodies. Trends Immunol 2018;39(12):956
60. [37] Vadnais ML, Criscitiello MF, Smider VV. In: Vaugn T, Osbourn J, Jallal B, editors. Antibodies from other species in protein therapeutics. Wiley-Verlag; 2017. [38] Nguyen VK, Hamers R, Wyns L, Muyldermans S. Camel heavy-chain antibodies diverse germline VHH and specific mechanisms enlarge the antigen-binding repertoire. EMBO J 2000;19(5):921
30. [39] Daley -Bauer LP, Purdy SR, Smith MC, Gagliardo LE, et al. Contributions of conventional and heavy chain IgG to immunity in fetal, neonatal and adult alpacas. Clin Vaccine Immunol 2010;17(12):2007
15. [40] Achour I, Cavelier P, Tichit M, Bouchier C, et al. Tetrameric and homodimeric camelid IgGs originate from the same IGH locus. J Immunol 2008;181:2001
9. [41] Woolven BP, Frenken LGJ, van der Logt P, Nicholls PJ. The structure of the llama heavy chain constant genes reveals a mechanism for heavychain antibody formation. Immunogenetics 1999;50:98
101. [42] Vu KB, Ghahroudi MA, Wyns L, Muyldermans S. Comparison of llama VH sequences from conventional and heavy chain antibodies. Mol Immunol 1997;34(16
17):1121
31. [43] Henry KA, van Faassen H, Harcus D, Marcil A, et al. Llama peripheral B cell populations producing conventional and heavy chain-only IgG subtypes are phenotypically indistinguishable but immunologically distinct. Immunogenetics 2019;71:307
20. [44] Massari S, Linguiti G, Giannico F, D’Addabbo P, et al. The genomic organization of the TRA/TRD locus validates the peculiar characteristics of dromedary δ-chain expression. Genes 2021;12(4):544. Available from: https://doi.org/10.3390/genes12040544. [45] Antonacci R, Bellini M, Linguiti G, Ciccarese S, Massari S. Comparative analysis of the TRB locus in the Camelus genus. Front Genet 2019;10:482. Available from: https://doi.org/10.3389/fgene.2019.00482. [46] Antonacci R, Bellini M, Pala A, Minececcia M, et al. The occurrence of three D-J-C clusters within the dromedary TRB locus highlights a share evolution in Tylopoda, Ruminantia and Suina. Dev Comp Immunol 2017;76:105
19. Available from: https://doi.org/10.1016/j.dci.2017.05.021. [47] Antonacci R, Linguiti G, Burger PA, Castelli V, et al. Comprehensive genomic analysis of the dromedary T cell receptor gamma (TRG) locus and identification of a functional TRGC5 cassette. Dev Comp Immunol 2020;106:103614. Available from: https://doi.org/10.1016/j. dci.2020.103614. [48] Kulkarni SS, Falzarano D. Unique aspects of adaptive immunity in camelids and their applications. Mol Immunol 2021;134:102
8. [49] Fichtner AS, Karunakaran MM, Gu S, Boughter CT, et al. Alpaca (Vicugna pacos), the first nonprimate species with a phosphoantigen-reactive Vγ9Vδ2 T cell subset. Proc Natl Acad Sci U S A 2020;117(12):6697
707.

Chapter 15

Suiformes: Pigs and Peccaries

Indian wild boar, Sus scrofa cristatus.

The order Artiodactyla first emerged about 70 million years ago (mya). The suborder, Suiformes diverged from the Ruminantia about 55 mya. Its members are even-toed ungulates with a characteristic single stomach and tooth morphology. The earliest fossil members of the suborder Suiformes date from the Oligocene (34
24 mya) in Asia. They are among the earliest branches of the Artiodactyla. It has proved difficult to determine which of the two families, the Suiformes, or the Tylopoda branched first. The divergences occurred over a relatively short period of time, so it has proved difficult to determine their exact branching sequence. Seventeen extant species are currently recognized in this suborder and classified into 4
8 genera. A major split has occurred between the family of New-world peccaries (Tayassuidae) and the Old-world pigs (Suidae). Their last common ancestor likely lived around 34
39 mya in the late Eocene/early Oligocene based on molecular estimates [1] (Fig. 14.1). Peccaries differ from Suids in many ways, including possession of a three-chambered stomach, three hoofs on each hind leg, as well as differences in dentition [2]. For many years the hippopotamus was considered a member of the Suiformes but recent studies have indicated that they are members of the same artiodactyl clade that includes the whales and porpoises. As expected, the vast majority of studies on the immunology of the Suiformes have focused on the domestic pig [3]. European and Chinese wild boar (Sus scrofa) were independently domesticated about 9000
11,000 years ago in Eastern Anatolia and around the same time in the Yellow River region in China. However, it is also apparent that European and Chinese wild boar are distinct subspecies that probably diverged around B1 mya [4]. Thus European and Chinese pigs are genetically very different. As a result of multiple domestication events, there are at least 16 recognized subspecies of S. scrofa found throughout Europe, Asia, and North Africa [2]. Domestic pigs have been transported throughout the world and some hybridization has inevitably occurred. These genetic differences are most readily seen in the complexity of their immunoglobulin-heavy chain loci. All Suidae are native to the Old World. It is interesting to note that the estimated divergence dates for rodents and primates (Euarchontoglires) date from about 75 mya, around the same time as the cetartiodactyls began to diversify. However, the short generation time of rodents such as the mouse means that the mouse has diverged to a much greater extent while the pig genomic sequences have remained more closely related to the longer-lived primates such as humans. Thus the great majority of human genes that were lost in the mouse, have been retained in the pig. As a result, the overall genetic similarity to humans is much greater in the pig than in the mouse.

Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00025-3 © 2023 Elsevier Inc. All rights reserved.

219

220

15.1

SECTION | 2 Mammalian orders

Reproduction and lactation

Pigs have a diffuse epitheliochorial placenta that while complexly folded, does not destroy endometrial tissue [5]. The endometrium and trophectoderm are very thin so that oxygen and nutrients do not have to diffuse far. The gestation period of the domestic sow is about 115 days [6]. B cells appear in the fetal yolk sac on day 20, in the fetal liver by day 30, and in the bone marrow by day 45. The first SWC31 leukocytes can be found in the yolk sac and liver on day 17. The thymus appears by 40 days post-conception and is colonized by two waves of T cell progenitors beginning on day 38. γ/δ T cells appear in the thymus and peripheral blood about 10 days later. γ/δ T cells are present in the bloodstream by day 55, but their numbers grow rapidly so that they predominate late in gestation. The intestinal lymphoid tissues are devoid of T cells at birth. CD41 T cells appear in the intestine at 2 weeks of age, and CD81 T cells appear at 4 weeks. Their immigration and proliferation appears to be driven by the intestinal microbiota. Natural killer (NK) cell activity does not develop until several weeks after birth, although cells with an NK cell phenotype have been identified at 45 days’ gestation in spleen and umbilical blood.

15.1.1 Cell-mediated immunity and colostrum Porcine colostrum contains large numbers of lymphocytes, but milk does not. Sow colostrum contains between 1 3 105 and 1 3 106 lymphocytes/mL. Of these, 70%
80% are T cells. Colostral lymphocytes may survive up to 36 hours in the intestine of newborn piglets. They can pass between the intestinal epithelial cells, enter the lymphatic vessels and reach the cortex of the mesenteric lymph nodes. Within two hours after receiving colostrum that contained radiolabeled cells, maternal lymphocytes appeared in the bloodstream of piglets [7]. As a result, cell-mediated immunity is transferred to the newborn. Piglets that have received these colostral cells show enhanced responses to mitogens compared with control animals. The mechanisms of this protective effect are unclear. Transcriptome analysis of colostral T cells in sows however has indicated that they are more activated than are peripheral blood T cells. Ingestion of maternal colostral leukocytes immediately after birth stimulates the development of the neonatal immune system [8]. Some of these passively transferred cells may also be NK cells [9].

15.1.2 Antibody-mediated immunity B cells are the first lymphocytes to appear in fetal peripheral blood. IgM1 B cells can be found in the fetal liver at 40 days, the spleen by day 50, and in the bone marrow by day 60 post-conception. The number of circulating B cells rises significantly between 70- and 80-days of gestation. Fetal piglets can produce antibodies to parvoviruses at 58 days and can reject allografts at about the same time. Their blood lymphocytes can respond to mitogens between 48 and 54 days. The response to antigens in the fetus is predominantly of the IgM class. As noted later, B cells can be found in the thymus of newborn pigs. The molecular development of the antibody repertoire has also been followed in the developing piglet. Thus VDJ rearrangement is first seen in the fetal liver on day 30. However, the fetal piglet does not initially use all its IGHV or IGHD genes. Likewise, N-region addition does not occur before day 40, suggesting that the onset of terminal deoxynucleotidyltransferase activity occurs after that time. IgM, IgA, and IgG transcripts are present by 50 days in all major lymphoid organs. Piglets are thus born with relatively limited B cell diversity. B cell numbers increase for the first 4 weeks after birth, but their antigen-binding repertoire does not begin to expand until 4
6 weeks of age. The production of IgA is controlled by exposure to the intestinal microbiota [10]. A limited amount of B cell class switching occurs in the developing fetus so that, as a result, newborn piglets already possess some intestinal IgA. Thus piglets at birth are exposed to new antigens from the growing microbiota as well as from colostrum and milk. It is these antigens that initiate IgA production and diversification. At weaning, the animal is exposed to new dietary antigens and must also develop oral tolerance. Lambda light chain rearrangements can be detected in the developing yolk sac of piglets between 20- and 50-days’ gestation, well before kappa chain rearrangements. Junctional diversity within VDJ regions is limited at all stages of development. B cell lymphogenesis and gene rearrangements continue for at least 5 weeks postpartum. All of these features appear to be somewhat unique to the pig. ’As in the other artiodactyls, the pig placenta does not permit the transfer of immunoglobulins to developing piglets. As a result, newborn piglets are born with agammaglobulinemia. They receive a concentrated solution of immunoglobulins when they suckle maternal colostrum. This maternal antibody is transported intact across the intestinal wall. In other mammals, the IgG accumulates in the colostrum as a result of active transport from the maternal bloodstream to

Suiformes: Pigs and Peccaries Chapter | 15

221

the mammary gland mediated by the Fc receptor, FcRn. FcRn does not however appear to be the responsible receptor in pigs since FcRn-"knockout" pigs still accumulate IgG in their colostrum [11].

15.2

Hematology

15.2.1 Blood leukocytes Total white cell counts in domestic pig blood average 14.7 6 4.5 x103/μl although these numbers may differ significantly between farms. Of these, neutrophils average about 53% (31
81), lymphocytes average 38% (16
70), and 4% (0
7) are monocytes. Eosinophils are usually fewer than 6% (0
22) (Fig. 15.1). As in other mammals, basophils are rare, constituting less than 1% of the total blood leukocytes. In most mammals, the percentage of lymphocytes in the blood tends to be higher in young animals. In pigs, the opposite is the case. Lactating sows tend to have higher white cell numbers than pregnant sows. Overall, white cell numbers decrease with age [12].

15.3

Innate immunity

15.3.1 Pattern recognition receptors The pattern recognition receptors in pigs closely resemble those of primates such as humans [13]. Pigs have ten functional toll-like receptors as do humans. Sequencing studies show significant homology with their human counterparts. Thus porcine and human TLR1, 3, and 9 have about 80%; TLR7 has 85% and TLR8 has 73% homology. They share similar functional domains. TLR2 is expressed in porcine thymus, spleen, Peyer’s patches (PPs), mesenteric lymph nodes, and palatine tonsils. It is expressed on M cells and macrophages but not on lymphocytes. There is extensive polymorphism among porcine TLRs [14]. Thus 21 SNPs have been identified in TLR1, 27 in TLR2, 33 in TLR6, and 33 in TLR10. Studies on these polymorphisms indicate that there are significant differences between wild boars and domestic pigs as well as between pigs of different breeds. This genetic variability may significantly influence an individual animal’s ability to detect and respond to an invading pathogen. Other pattern recognition receptors such as NOD1 and NOD2 have been cloned and characterized in the pig. Likewise, pigs possess and express RIG-1, MDA5, and LGP2. These appear to be functionally similar to their human counterparts. Several porcine C-type lectin receptors have also been characterized including CD69, CD205, CD207 (Langerin), DC-SIGN, and Dectin-1. Pig lungs are rich in pulmonary surfactant proteins such as the C-type lectins, SPA and SP-D. Porcine SP-D appears to have activity against influenza viruses [13]. Porcine mannose-binding lectins (MBLs) have been cloned and sequenced and shown to have 65% identity with human MBL. MBL concentrations are highly heritable in some pig breeds. Pigs have a functional complement system that appears very similar to the system in humans. Pattern recognition receptors act through multiprotein inflammasomes to drive inflammatory cytokine production. There are several different inflammasome pathways and not all are active in all mammalian species. In the case of the pig, inflammatory mediators such as alum crystals will activate inflammasomes in LPS-primed cells through the NLRP3 pathway [15].

LEUKOCYTES 14.7 x 103/ml

ADULT LYMPHOCYTES

Neutrophils

53%

TDE NK

TJG

Lymphocytes

38% Monocytes

4% 2%

Eosinophils

B

FIGURE 15.1 The white cell counts and proportions in pig blood as well as the composition of the blood lymphocytes.

222

SECTION | 2 Mammalian orders

The NLRC inflammasome plays a critical role in the recognition of flagellin and other parts of the type III secretion system in bacteria. It is thus an important component of the innate immune system and in the response to bacterial pathogens. Two key components of the NLRC4 inflammasome are CARD-containing 4 (NLRC4) and NLR apoptosis inhibitory protein. Sequencing these genes in multiple domestic pigs and wild boars has revealed a loss of intermediate exons in both genes and a lack of valid open reading frames. They are therefore pseudogenes. This lack of an effective pathway for recognizing bacterial proteins associated with TLR5 suggests that the other polymorphic porcine TLRs such as TLR2, TLR4, and TLR5 may also have different roles to play [16].

15.3.2 Acute-phase proteins The acute-phase response is a rapid reaction to tissue injury that develops within 12 hours and is characterized by a rapid rise in certain serum proteins. Pigs possess several such acute-phase proteins. These include C-reactive protein, haptoglobin, major acute-phase protein (MAP), and serum amyloid A (SAA). Other negative acute-phase proteins decline in sick animals. These include albumin, transferrin, and transthyretin [17]. MAP is a 120 kDa plasma glycoprotein that acts as a significant protease inhibitor in pigs and a moderate one in cattle. Porcine MAP shares homology with inter-α-trypsin inhibitor heavy chain 4, a moderate acute phase protein in dogs. SAA, a protein of 15 kDa, is also an acute-phase protein in pigs. Other porcine acute-phase proteins include apolipoprotein A-1, sialic acid, and ceruloplasmin. Pigs with infected tail lesions and carcass abscesses due to biting have elevated levels of CRP, SAA, and haptoglobin compared with control pigs. In experimental Actinobacillus pleuropneumoniae infections, CRP and SAA are significantly increased. Salivary haptoglobin and chromogranin A may also be used to measure stress in pigs however SAA is not elevated in stressed pigs [18].

15.3.3 Antimicrobial peptides Multiple host defense peptides have been identified in pigs [19]. They are active against a diverse spectrum of organisms both in vitro and in vivo. Among the gene families that have expanded in the porcine immunome are the cathelicidins. Thus pigs have ten cathelicidin genes compared to one in humans and mice. This expansion may be artiodactyl specific since cattle also possess ten cathelicidin genes. Thirty-four beta-defensin genes are present in the porcine genome. This is similar to the human with 39 genes but very different from cattle with 106. This may be a ruminantspecific feature. Porcine β-defensin 1 is absent from the upper respiratory tract of newborn piglets perhaps explaining their susceptibility to pneumonia at that age. No alpha-defensins have been detected in pigs. There are also significant breed differences in the expression of defensins in different pig tissues [20]. Thus their levels are much higher in Meishan pigs than in Duroc x Yorkshire x Landrace crossbreeds. CD163 is a hemoglobin scavenger receptor expressed on the surface of pig macrophages. It is expressed at high levels in activated macrophages such as those responding to inflammation. By binding free hemoglobin, CD163 reduces the oxidative and proinflammatory effects of hemoglobin and hence has an anti-inflammatory effect. During inflammation, it is released into the bloodstream and hence is also an acute-phase protein. CD163 is a member of the scavenger receptor, cysteine-rich protein family and as such is closely related to another prominent pig protein, SWC1 which plays a key role in the regulation of γ/δ T cells [21].

15.3.4 Cytokines Studies on the porcine immunome reveal significant expansions in some gene families encoding components of the innate immune system. For example, pigs possess 39 type 1 interferon genes (compared to 19 in humans and 51 in cattle), in addition to 16 IFN pseudogenes [22]. The expansion has resulted in the production of novel subtypes of IFN-δ and IFN-Ω. Interferon lambda-2 is missing from pigs. Duplication of the IL-1β gene (IL-1β and IL-1βL) in the pig appears to be unique among mammals. Both predicted proteins are 267 amino acids in length but are only 86% identical. Pigs also have a duplication of the OAS1 gene. This encodes the interferon-induced, key antiviral effector molecule, 20 -50 -oligoadenylate synthase. Porcine IL-4 is not a stimulatory factor for pig B cells [23]. While a functional IL-37 gene is present in cattle, sheep, horses, and dogs, it is a pseudogene in pigs and mice [22]. In addition to the other interleukins, pigs possess an interleukin-related gene, IL-15L of unknown function. IL-15L is also present in cattle but absent in mice and humans [24].

Suiformes: Pigs and Peccaries Chapter | 15

15.4

223

Lymphoid organs

15.4.1 Thymus The porcine thymus is relatively large and can extend from the thorax to the larynx or even the submandibular space. As in other mammals, it is derived from the pharyngeal endoderm and the ectoderm of the third pharyngeal cleft. The thymus grows backward into the thorax as it develops; however, two lobes remain in the cervical region on each side of the trachea. It appears to be functionally similar to the thymus in other artiodactyls. The venules in the corticomedullary area have a high endothelium. The developing thymus is colonized by waves of T cell progenitors that rearrange their antigen receptors and undergo both positive and negative selection. A unique feature of the porcine thymus is that it is also a site of some B cell development [25]. The immature lymphocytes that migrate to the thymus remain pluripotent and retain the ability to develop into either B cells or T cells. At some stage however a career choice must be made, and cells become irrevocably committed to a specific cell phenotype. Thus there is a population of mature B cells and plasma cells found within the thymic medulla. They only account for about 1% of all thymocytes. There is also a population of developing B cells. Many of these are IgA positive. Interestingly, there is also a population of committed T cells that can still rearrange immunoglobulin genes but do not transcribe them. All three cell populations have rearranged immunoglobulin heavy chain VDJ genes [25].

15.4.2 Spleen The pig spleen is tightly attached to the stomach by short gastric blood vessels. Arteries entering the spleen pass through muscular trabeculae before entering the white pulp and branching into arterioles. Immediately on leaving the trabeculae, each arteriole is surrounded by a periarteriolar lymphoid sheath. The arterioles eventually leave this sheath and branch into penicillary arterioles. These penicillary arterioles are surrounded by ellipsoids (periarteriolar macrophage sheaths). The arterioles then open, either directly or indirectly, into venous sinuses that drain into the splenic venules. Ellipsoids are relatively large and prominent in pigs, mink, dogs, and cats; they are small and indistinct in horses and cattle; and are absent in laboratory animals such as mice, rats, guinea pigs, and rabbits. In species that lack ellipsoids, particles are trapped primarily in the marginal zone of the white pulp.

15.4.3 Lymph nodes Domestic pigs and related swine (warthogs), hippopotamuses, rhinoceroses, and some dolphins have structurally unique lymph nodes reflecting different patterns of lymphocyte recirculation (Fig. 11.6 and Fig. 15.2). Their lymph nodes consist of several lymphoid “nodules” oriented so that the cortex of each nodule is located toward the center of the node, whereas the medulla is at the periphery. Each nodule is served by a single afferent lymphatic that enters the central cortex as a lymph sinus. Thus afferent lymph is carried deep into the node. The afferent lymph contains 35
200/μl lymphocytes as well as a few neutrophils. A cortex surrounds the lymph sinus. Outside this region are a paracortex and a medulla. This medulla may be shared by adjacent nodules. Lymph flows from the cortex at the center of the node to the

FIGURE 15.2 A section of a pig lymph node. Note how the germinal centers are located in the interior of the node. Original magnification 3 12. (From a specimen provided by Dr. Brian Porter.) From Tizard, Vet Immunol 10e.

224

SECTION | 2 Mammalian orders

Bloodstream

Bloodstream

Efferent vein

Pig

High endothelial venules

Thoracic duct

Other mammals

Efferent lymphatics

Lymph node paracortex

High endothelial venules

FIGURE 15.3 Comparison between the major route of circulation of T cells in the pig and that in other mammals. Note that pig lymphocytes are largely confined to the bloodstream. (From Tizard, Vet Immunol 10e).

Lymph node paracortex

medulla at the periphery through specialized perifollicular sinuses to reach the B cell follicles before leaving through efferent vessels in the subcapsular area. The medulla has very few sinuses but consists of a dense mass of cells that is relatively impermeable to cells suspended in the lymph. As a result, few lymphocytes migrate through the medulla. T cells in pigs enter the lymph node conventionally through high endothelial venules (HEVs). However, they do not leave the lymph node through the lymphatics but migrate directly back to the bloodstream through the HEVs of the paracortex. B cells in contrast leave the efferent lymphatics to eventually return to the bloodstream via the major lymphatic ducts (Fig. 15.3) [26].

15.4.4 Mucosa-associated lymphoid tissues 15.4.4.1 Tonsils The circle of lymphoid tissues in the pharynx,
Waldeyer’s Ring, is well developed in pigs [27]. Pigs possess five tonsils; lingual, paraepiglottic, and tubal, as well as the palatine and pharyngeal tonsils [28]. They have a single pharyngeal tonsil on the median roof of the nasopharynx, two palatine tonsils on each side, a well-developed lingual tonsil at the root of the tongue, and tubal tonsils. The tubal tonsils form a patch with an accumulation of lymphoid tissue around the Eustachian tube. Epithelial crypts extend into these tonsils. The tonsils are rich in B cells with moderate numbers of α/β T cells and low numbers of γ/δ T cells (Fig. 11.8).

15.4.4.2 Bronchus-associated lymphoid tissues There are no bronchus-associated lymphoid tissues detectable in germ-free pigs. In conventional pigs, their size varies between individuals, but it is present in all conventionally reared adult pigs [27]. These bronchial lymphoid tissues consist of single lymphoid follicles that do not bulge into the airways. They are primarily located around the bronchioles. There do not appear to be any M cells on their epithelial surfaces. Pigs also possess numerous pulmonary intravascular macrophages (Fig. 15.4).

15.4.4.3 Gastric lymphoid tissues Pigs, like camels, have a cluster of lymphoid nodules within the submucosa and lamina propria of the lesser curvature of their gastric cardia and cardiac fundic diverticulum. These nodules develop in fetal pigs and are present at birth. The cluster may enlarge and fill with a diffuse lymphoid infiltrate, follicles, and germinal centers in pigs in a microbe-rich environment [29]. These nodules resemble the lymphoglandular complexes found in the pig colon [27].

15.4.4.4 Peyer’s patches PPs are found in the walls of both the ileum and jejunum of pigs. Isolated lymphoid follicles are also found in the walls of the large intestine. Lymphoid infiltrates are found in the jejunal lamina propria as early as 50 days gestation. The number of jejunal PPs in individual pigs ranges from 20 to 30 with an average diameter of 3.8 cm (Fig. 11.2) [30]. These jejunal PPs persist for the life of the animal. They are of conventional structure and contain pear-shaped follicles separated by extensive interfollicular tissue. These consist of mainly B cells with up to 30% T cells. The pig has a single, large ileal PP. It is a long, ribbon-like patch in the terminal ileum extending forward from the ileocecal junction for 1
3 m and occupying 15% of the length of the small intestine [30]. This ileal patch consists of densely packed lymphoid follicles, each separated by a connective tissue layer, and contains only B cells. This large patch regresses within the first year of life, but it does not appear to be a primary lymphoid organ since it is not required for B cell development. It appears to be a secondary organ that plays a role in the initial immune responses to the intestinal microbiota.

Suiformes: Pigs and Peccaries Chapter | 15

225

FIGURE 15.4 An intravascular macrophage (M) from the lung of a 7-day-old pig. The cell has numerous pseudopods, electron-dense siderosomes, phagosomes, and lipid droplets. It is closely attached to the thick portion of the air-blood tissue barrier that contains fibroblasts (F) and a pericyte (P) between basal laminae of the capillary endothelium (E) and the alveolar epithelium. At sites of close adherence, intercellular junctions with subplasmalemmal densities are seen (arrow). Bar 5 2 μm. Original magnification 3 8000. With permission From Winkler GC, Cheville NF. Postnatal colonization of porcine lung capillaries by intravascular macrophages: an ultrastructural morphometric analysis, Microvasc Res 1987;33:224
32.

The development of the porcine ileal PP begins prenatally between embryonic days 76 and 91 when submucosal lymphoid clusters begin to form. However, isolated mucosal lymphoid follicles continue to develop postnatally as well. Thus they develop central, marginal, and subepithelial zones, and contain large numbers of proliferating and apoptotic B cells. IgM1 cells predominate in the central zone [31]. Interestingly, these zones are not present in the developing mesenteric lymph nodes. The epithelium covering these patches contains multiple M cells, characterized, as their name implies, by microfolds. The PPs grow rapidly from about 10 days prior before to several weeks after birth. They do not develop in germ-free animals, so this growth is presumably driven by the presence of the intestinal microbiota. Interestingly, the ileal PP grow by increasing the size and number of their follicles. In contrast, the jejunal PP grow by increasing the size of their follicles only. While the IPP and JPP look identical at birth, their morphology changes. Thus the IPP follicles become ovoid while their interfollicular areas and domes shrink. This is mainly associated with an increase in B cells rather than T cells. Very few lymphocytes migrate into the IPP, but immigration is high in the JPP and in that region of the IPP located close to the ileocecal junction. These intestinal lymphoid tissues are covered by a specialized lymphoid epithelium. In the IPP these are mainly enteroabsorbtive cells with only a few M cells. The JPP surfaces have relatively more M cells. In the pig large intestine, there are two major types of lymphoid tissue. First, there are large submucosal multifollicular lymphoid clusters that develop prenatally. These are anatomically similar to jejunal Peyer’s patches. Second, there are mucosal isolated lymphoid follicles that develop after birth [75]. An accumulation of individual lymphoid follicles develops at the ileocecal junction. Individual follicles are present in all parts of the large intestine. In many cases, these develop into lymphoglandular complexes. There are ten irregular-shaped, large lymphoid clusters in the first 20 cm of the spiral colon [30].

15.4.4.5 Lymphoglandular complexes Lymphoglandular complexes are present in the wall of the large intestine and cecum in pigs. (Chapter 11).

15.4.5 Dendritic cells Dendritic cells play important roles in both innate and adaptive immunity insofar as they recognize, capture, and present foreign antigens to the adaptive immune system. Pigs have heterogeneous populations of these cells. They have both conventional cDCs and plasmacytic pDCs. Pig cDCs are CD172a1, CD11R11, CD11/2, and CD80/861/2, whereas their pDCs are CD172a1, CD41, CD11/2, and CD80/861/2. Both types secrete IL-10 and IL-12. Porcine DCs also express FcγRII and FcγRIII on their surface and thus can be activated by immune complexes. They express TLRs and are responsive to stimulation by bacterial LPS and CpG DNA. The functions of pig DCs have been characterized based

226

SECTION | 2 Mammalian orders

on their transcriptome. Thus pig blood pDCs are the major source of TNF-α, IL-12p40, and IFN-α in addition to some complement components. Pig cDCs are most efficient at antigen presentation and T cell stimulation. Neonatal pig dendritic cells are also activated by ligands of TLR4 or TLR9 [32]. TLR7 and TLR9 are restricted to pDCs and not expressed on cDCs as in mice Pig pDCs are also unique in that they express functional NKp46 receptors. In other mammals, these are considered to be strictly NK cell markers. However, pig DCs do not express perforins so they are not cytotoxic. They may be activating receptors [33]. It is also of interest to note that pig pDCs produce IFN-α in response to several common viruses, including transmissible gastroenteritis, pseudorabies, and swine flu, but not porcine reproductive and respiratory syndrome virus (PRRSV). This virus impairs antigen presentation by cDCs and enhances their IL-10 production. It is not surprising therefore that PRRSV causes persistent infection and stimulates only a weak immune response.

15.5

Major histocompatibility complex

The major histocompatibility complex (MHC) of the pig (SLA) spans about 2.4 mb on chromosome 7 [34]. At least 150 genes within the MHC have been characterized [35]. It differs from other mammals in that its class II region is separated from the class I and class III regions by the presence of the centromere [36]. Thus the class II region is located on the q arm while the telomeric class I and the centromeric class III regions are located on the p arm of the chromosome. This is an unusual example of the loss of the tight linkage that is usually present between MHC classes I, II, and III [37] (Fig. 15.5).

15.5.1 Major histocompatibility complex class Ia Because of the complex nature of the evolution of MHC class I molecules, there are no orthologous relationships between the class I genes of humans and pigs [38]. Pigs possess seven classical class I genes: SLA-1, SLA-2, SLA-3, SLA-4, SLA-5, SLA-9, and SLA-12. Of the proteins encoded by these loci, SLA-1, SLA-2, and SLA-3 are functional and play a key role in antigen presentation to cytotoxic T cells. SLA-4 and SLA-9 are pseudogenes. SLA-5 does not appear to be expressed. There is an additional class I pseudogene, SLA-11 found adjacent to the main class I loci. The classical genes are arranged in a single contiguous region in the order SLA-1, -5, -9, -3, -2, -4 from the p terminus of the chromosome. Three nonclassical class Ib genes SLA-6, -7, and -8 are also clustered. The SLA class I genes are found in two distinct clusters. One cluster consisting of five classical class I genes and two MIC-like genes is located in the beta block. The other cluster consisting of the seven classical class I genes is located in the kappa block. The number of functional class I genes varies between one and four depending upon the specific haplotype of the animal. For example, haplotype Hp-62.0 contains one SLA-1, two SLA-9, and one SLA-12 locus whereas Hp-28.0 contains two SLA-1, three SLA-5, and two SLA-9 loci [39]. Like other mammalian class I genes they each contain eight exons. These code for a protein chain with a leader peptide, three constant extracellular domains, a transmembrane domain, and a cytoplasmic tail. The mature proteins each contain about 316 amino acids [40]. Class Ia molecules are expressed on most nucleated cells but with great quantitative variations. In pigs, class I molecules have been detected on lymphocytes, platelets, granulocytes, hepatocytes, kidney cells, and sperm. They are not usually expressed on red cells, gametes, neurons, or trophoblast cells. Some cells, such as myocardium and skeletal muscle, may express very few class Ia molecules. Interestingly, MHC class I molecules are also expressed on pig brain FIGURE 15.5 The organization of the pig major histocompatibility complex.

DP

DM

DO

DQ

DR

BA

A B

B

B B A

B B B A

SLA-2 SLA-3 SLA-1

centromere II

III

Ib

Ia

Suiformes: Pigs and Peccaries Chapter | 15

227

cells. This expression pattern is very different from humans and mice where brain cells express only minimal amounts of MHC class I or none at all. High-resolution typing of SLA class I proteins has detected 52 alleles including 18 novel alleles and nine SLA-1 duplication haplotypes [41]. There are significant differences in the distribution of these alleles among different pig breeds. Normally the two ends of the MHC class I peptide-binding groove are closed. This is in contrast to the MHC class II binding groove where the ends are open and long antigenic peptides can extend out of the groove at either end. An interesting feature of pig MHC class I peptide binding is the fact that they can also recognize long antigenic peptides whose N-terminus can extend out beyond the MHC binding groove. This novel form of antigen presentation has also been observed in some human MHC class I molecules [42].

15.5.2 Class II The pig is unique in that its SLA class II region is separated from the class III and class I regions by the centromere
a heterochromatic dark staining structure. It is located between duplicated butyrophilin genes [43]. The SLA class II genes encode multiple alleles of pig DRA, DQA, DRB, and DQB. Thus they make at least one functional SLA-DR and one SLA-DQ heterodimeric product. Pigs have no functional genes encoding either DPA or DPB. The class II DR α-chain contains 229 amino acids in two extracellular domains, a transmembrane domain, and a cytoplasmic domain. It is of interest to note that pig DRA alleles more closely resemble human DRA than they do pig DQA. The length of the DRα chain is identical in pigs and humans. This suggests that the class II α chain precursor gene was duplicated into DRA and DQA before the separation of the primate and artiodactyl lineages. The DQα chain has the same general structure as the DRα chains. Likewise, the class II β chains are similar, with each consisting of two extracellular domains, an intermediate domain, a transmembrane domain, and a cytoplasmic domain. Unlike the class II α chains, the β chains show great structural variability. Among these class II genes, they are either monomorphic like the DRA gene or oligomorphic like the DQA genes. In contrast, the DRA and DQB genes show extensive polymorphism. The SLA class III region is linked to the class I region. It spans about 700 kb of DNA and contains at least 33 characterized genes. It has some unique features. Thus in the pig, the region is known as the RCCX module that contains the CYP21A2, TNXA, and C4B loci and is single whereas in humans and mice its copy number depends upon their haplotype [35].

15.5.3 The natural killer receptor complex In contrast to other mammals in which NK cells are characteristically large, granular lymphocytes, in pigs, these appear to be small to medium-sized cells that largely lack granules. Porcine NK cells are found in the spleen and peripheral blood, but very few are found in lymph nodes or thymus. Pig NK cells are CD21, CD42, CD52, NCR11, CD251, CD8α1, and CD161 but subsets exist. Likewise, in contrast to other species where NKp46 is considered to be a pan-NK cell marker, many porcine NK cells fail to express it at all. On the other hand, they do express NKp44 even in the absence of additional stimulation [44]. This expression can be enhanced still further by exposure to IL-2 or IL-15. (NKp44 is expressed exclusively on activated NK cells in humans). Porcine NK cells are effectively cytolytic against pseudorabies-, swine fever-, and coronavirus-infected cells [45]. Their activities are enhanced by interferon-γ, and IL-2. IL-2, IL-12, and IL-18 act synergistically to promote pig NK cell expression of perforin and IFN-γ. Only single genes encoding KIR and Ly49 have been reported in pigs. The Ly49 (KLRA1) gene has a mutation in the codon for a highly conserved cysteine residue and may therefore be a pseudogene. It has two possible orthologs. Despite this apparent lack of NK cell receptor diversity, it does not appear to affect their numbers. NK cell numbers vary greatly among individual pigs ranging from 1% to 24% of circulating lymphocytes, and this may have a direct effect on disease resistance [13]. Thus it remains uncertain how pig NK cells recognize their targets

15.6

B cells and immunoglobulins

Pig B cells are mainly derived from stem cells located in the bone marrow. The bone marrow is especially productive during the second half of gestation, but it remains functional throughout life [46]. A second major source of B cells is the developing thymus. The ileal PPs are not a significant source of B cells in the pig unlike the situation in other

228

SECTION | 2 Mammalian orders

artiodactyls. Unlike the great combinatorial diversity seen in humans, the early pre-immune antibody repertoire in pigs is relatively limited with only 224 variants accounting for almost 100% of the immunoglobulin variation. Most of this diversity is accounted for by heavy chain V gene rearrangements. The presence of the intestinal microbiota significantly promotes the development and diversity of pig B cells, whose numbers within the gastrointestinal tract, especially the ileal PP, increase greatly during the first 2 weeks of age, although receptor diversity may not increase until 4
6 weeks of age. Isolator-raised, germ-free pigs have serum immunoglobulin levels 20- to 100-fold less than conventional pigs. Conventional pigs exhibit much greater diversity in their mucosal IgM and IgA V genes than do germ-free pigs. The cells that appear after microbial colonization include classswitched IgM1 and IgA1 B cells as well as CD21CD212 memory B cells and CD41CD82 α/β Th cells [47].

15.6.1 Immunoglobulin heavy chains IgG is the predominant serum immunoglobulin, accounting for about 85% of the total serum antibodies. IgM accounts for about 12% and dimeric IgA for about 3% of serum immunoglobulins.

15.6.1.1 IGHG The pig IGH region is about 190 kb in size. The first genomic studies performed on the European breeds such as Landrace and Duroc identified single IGHM, IGHD, IGHE, and IGHA genes in addition to six IGHG subclass genes (Fig. 15.6) [48,49]. Eleven pig IGHG gene sequences have since been described [50]. These encode the six IgG subclasses named IgG1 thru IgG6. There are two allelic forms of each of these subclasses except for IgG3. The IGHG genes form a cluster located between IGHD and IGHE. They encode the six subclasses as well as their alleles. They are arranged in order: 50 -G3-G5a-G5b-G6a-G6b-G4a-G4b-G2b-G2a-G1a-G1b-30 These sequences were initially derived from the genome of a single Landrace pig with a haplotype designated Lan1. Subsequent genomic studies have shown that other breeds of pigs have different numbers of IGHG genes. It is believed that all of these IgG genes were present in the ancestral wild boars, but some have subsequently been lost following domestication. For example, Landrace pigs may have only four IgG heavy chain genes. They appear to have two IGHG5 and two IGHG6 genes but no genes encoding IgG2 or IgG4. Other pig breeds may lack the genes encoding IgG4 or IgG6. Only IGHG1, and IGHG3, are present in all pig breeds. Presumably, these two IgG genes are ancestral. IGHG3 appears to be the most ancient and is found at the upstream (50 ) end of the cluster. The initial expansion of the IGH 5’

30

5

3'

5 M

IGL

34

IGK

60

D

G3 G5a G5b G5c

G1 G2a G2b G2c G4

5

M

D

G

E

A

V

D

J

LC

E

A

FIGURE 15.6 The organization of the porcine immunoglobulin loci, Depending upon their breed, pigs may have four to seven different IgG heavy chains.

Suiformes: Pigs and Peccaries Chapter | 15

229

IGHG gene numbers presumably resulted from repeated duplications. There are also structural differences within their switch regions that may be relevant. The differences between these pig IgG subclasses and their alleles may be quite minor. For example, IgG2a and IgG2b differ by only three amino acids. IgG3 has an extended hinge, is structurally unique, and appears to be the most evolutionarily conserved of the porcine IgGs. IgG5b differs most from its allele and its CH1 domain shares sequence homology with the CH1 of IgG3 [50]. In many cases, exon shuffling has led to the sharing of immunoglobulin domains. Thus IgG1, IgG2, and IgG4 all appear to use a CH1 domain derived from the same ancestor. The hinge used by IgG1b is the same as that used by IgG4b while the IgG6b hinge is shared with IgG5a. As pointed out at the beginning of this chapter, European and Chinese domestic pigs are descended from disjunct populations of wild boar and are thus genetically very different. This is most readily seen in their immunoglobulin heavy chain gene loci. Thus one of the remarkable features of the porcine immune system is that the number of immunoglobulin heavy chain genes varies among breeds. When indigenous Chinese breeds such as the Erhualian, Luchuan, and Xiang are compared to the European Duroc, it has been found that the Erhualian breed has seven genes, the Duroc, Landrace, and Luchuan have six (not identical) and the Xiang pigs have five genes. All the pigs had the four basic heavy chain genes IGHM, IGHD, IGHA, and IGHE. They differ however in their IGHG genes. Collectively, these can be divided into nine subclasses, IGHG1, IGHG2a, IGHG2b, IGHG2c, IGHG3, IGHG4, IGHG5a IGHG5b, and IGHG5c [2]. It is important to note at this stage that in all the cases studied to date, mammalian IgG genes appear to have diversified into different subclasses after speciation. Thus their effector functions such as receptor binding or complement activation cannot be extrapolated to the “same name” subclasses in other species [50]. In the case of the pig, IgG3 is most likely to activate complement and bind to FcγRs. All of them except IgG6a and IgG5 should bind well to FcγRs and FcRn [50].

15.6.1.2 Other classes One IgM allotype has been reported in the pig. The exons encoding the CH1 domains of IGHM and IGHD are very similar to each other. An IgD gene has been identified in pigs. Its first constant domain may be coded by either an IGHD gene or an IGHM gene. It encodes multiple hinge regions. Thus pig IgD heavy chain transcripts may contain either -VDJ-CH1μ-CH2δ-CH3δ
or -VDJ-CH1δ-CH2δ-CH3δ 2 . These two genes, however, show almost 99% identity, so the biological consequences are probably not great. The exon encoding the hinge region between CH1 and CH2 in IGHD is also duplicated, a feature that is also found in other artiodactyls. The second hinge region gene however lacks a critical nucleotide and is, therefore, a pseudogene. While IgD transcripts are common in fetal piglets, IgD has not been identified and may not be expressed in adult pigs. Pigs have a single IGHA gene with two alleles. IgAb differs from IgAa by a 12-nucleotide deletion in the hinge region owing to a mutation in its splice acceptor site. The consequences of this are unclear. Pig IgE has been identified. The IGHE gene structure is similar to other mammalian species with four functional exons encoding its constant domains.

15.6.1.3 Variable domains Pigs have about 30 IGHV gene segments (20 functional), five D genes (two functional), and five germ-line IGHJ gene segments of which only one is functional [51]. The neonatal piglet has very little immunoglobulin diversity at birth. Early in fetal life, the pig uses only four of the IGHV segments, and their early antibody repertoire, therefore, consists of only eight to ten combinations. Later in fetal life, this restricted repertoire is compensated for by early TdT activity and extensive, in a frame, N-region addition leading to significant junctional diversity. VH use is independent of gene position, but four IGHV genes (VHA, VHB, VHC, and VHE), account for 80% of the pre-immune repertoire. When combined with VHF, VHY, and VHZ, these seven genes account for .95% of the pre-immune repertoire [49]. They must use extensive somatic hypermutation to develop a fully functional and expansive repertoire. Pigs do not appear to use gene conversion to a significant extent. All the known pig IGHV genes belong to a single ancestral family, IGHV4 [52,53]. Pigs however can also employ hybrid VH genes in which they use the CDR1 of one VH gene and the CDR2 of another [54].

230

SECTION | 2 Mammalian orders

15.6.1.4 Immunoglobulin light chains Unlike cattle, sheep, and horses that predominantly use lambda light chains, pigs produce kappa and lambda light chains in approximately equal amounts. The pig IGK and IGL chain loci are located on chromosomes 3 and 14 respectively. The pig lambda locus spans about 229 kb. It contains at least 23 IGLV genes (10 functional) and 4 IGLC genes arranged in four sets of tandem cassettes [52]. The constant-proximal cluster contains IGLV3 family members while the C-distal cluster contains IGLV8 and IGLV5 family members. The IGLV8 group appears to have expanded relatively recently perhaps in response to specific porcine pathogens. The IGLV genes can be subdivided into seven families, VL1, VL2, VL3, VL5, VL7, VL8, and a poorly defined group III [55]. IGLV expression is almost exclusively limited to the use of IGLV3 and IGLV8. The C-distal IGLV cluster also contains three IGLV1 pseudogenes that are orthologous to the IGLV genes expressed in cattle. Ten of these IGLV genes and two IGLJ genes appear to be functional and belong to either the IGLV3 or IGLV8 families. Eight appear to possess stop codons, frameshifts, or both and one is missing a V-exon. There is mutational variation along the entire length of the IGLV genes in both framework and CDRs. There is also evidence that non-allelic homologous recombination occurs in three recombination hotspots [56]. The expression of IGLV genes is highly variable between individual pigs. Thus at least one IGLV gene in an individual may be functional and highly expressed but may be completely absent from others [57]. The sequence from which the light chain transcript was obtained has been designated IGLV3
6. These IGLV3
6 transcripts account for about 20% of all IGL transcripts and are highly utilized. The pig kappa locus contains at least 60 IGKV genes spanning 89 kb, five IGKJ genes (two pseudogenes), and one IGKC gene. (It is a quarter of the size of the human kappa locus) [58]. Despite this, about ten IGKV and one IGLJ segments encode 70%
80% of the pre-immune light chain repertoire [58]. These genes can be divided into four subfamilies. However, three of these subfamilies representing 90% of the expressed IGKV genes belong to a family with 87% sequence similarity to human IGKV2. Studies on domestic pig populations have indicated that allelic variation is high in IGKV genes [58].

15.7

B cell receptor development

In humans and mice, when B cell antigen receptors are generated, the first chain to be expressed on the cell surface is the heavy chain. This chain is capable of generating much more junctional and combinatorial diversity than the light chain and as a result, is the major contributor to antigen binding. This heavy chain is linked to signal transduction molecules, and a surrogate light chain is provided so that the pre-B cell can respond in a limited way to antigens. As a result, a small clone of B cells expressing only the heavy chain develops. Signaling through this pre-receptor triggers limited proliferation. Once heavy chain expression is complete, this is followed by the rearrangement and expression of one of the B cell’s IGK genes. If this fails to produce a functional light chain, the cell switches to the other IGK allele for a second attempt. If this does not work, the Cκ segments are deleted and the B cell is obliged to use one of the IGL alleles, and if this fails, the second IGL allele represents the last resort. If all these efforts fail to produce a functional light chain, the B cell cannot make a functional immunoglobulin. It will undergo apoptosis without participating in an immune response. Thus κ1 B cells are produced first while λ1 B cells are produced later. Pigs do it differently! In pig B cells, the order of B cell receptor assembly is reversed [59]. Thus in this species, the first step in B cell development is light chain gene rearrangement. Immature B cells first express a light chain in the absence of IgH and surrogate light chains are not required. One κ light chain rearranges first but only on one chromosome. The B cells then rapidly replace it with rearranged λ light chains [60]. Thus κ2λ1 B cells are produced before IgH rearrangement occurs. Only after the λ genes are rearranged does heavy chain rearrangement begin to occur. So, for a period, the developing B cells express only λ chains on their surface. The IGH rearrangement is similar to the process in mice and complete immunoglobulins form only after the rearranged heavy chain is proven to be functional [61]. The proportion of λ1 B cells increases during ontogeny. The κ/λ ratio at birth is 2:1 but this increases to 1:1 by 5 weeks and 1:5 by 20 weeks. Interestingly, this change in the κ/λ ratio does not occur in developing germ-free pigs where it can reach 4:1 by 8 weeks of age. Most κ1 B cells are rearranged later using germline IGLV genes. This results in many κ1 B cells also carrying λ light chain transcripts [60,62] (Fig. 15.7). Lambda chain rearrangements can be detected in the developing yolk sac of piglets between 20- and 50-days gestation well before kappa rearrangements. Junctional diversity within VDJ regions is limited at all stages of development. B cell lymphogenesis and gene rearrangements continue for at least 5 weeks postpartum. All of these features appear to be somewhat unique to the pig.

Suiformes: Pigs and Peccaries Chapter | 15

PIGS

N genes rearranged first

IGH genes rearranged Surrogate light chains

Mature N+ cells carry O transcripts O+ cells generated

FIGURE 15.7 The order in which porcine B cells generate complete immunoglobulin molecules. This is the reverse of the “conventional process” seen in primates and rodents.

HUMANS and MICE

IGL genes rearranged

231

IGL genes expressed

IGH genes expressed

IGH genes rearranged

IGL genes rearranged

IGH genes expressed

IGL genes expressed

Complete B cell receptor

Complete B cell receptor

Pattern recognition receptor Leptospira Borrelia

WC1.1 Mainly antiviral - IFNJ

Pre-B cell receptor formed Use all N genes first

Regulatory functions JG T cells monocytes granulocytes

Growth arrest in JGT cells

FIGURE 15.8 The workshop cluster 1 proteins, are transmembrane glycoproteins belonging to the scavenger receptor cysteine rich family. They are expressed in camels, pigs, cattle and sheep. They are formed by multiple extracellular repeats of the scavenger receptor cysteine-rich class B domain.

CD163 Hemoglobin scavenger receptor - anti-inflammatory

WC 1.2 Regulatory - TGFE IL-10

15.8

T cells and cell-mediated immunity

Among the many unique features of pigs compared to other highly studied mammals is that they produce high numbers of extrathymic double-positive CD41CD81 T cells. Pigs are also a γ/δ T cell-high species [63].

15.8.1 Workshop cluster1 cells Pig leukocytes express several unique surface proteins that do not carry a CD designation. These are designated as “Workshop clusters” (WC). The most important of these is WC1. Swine WC1 (SWC1) is a type 1 transmembrane glycoprotein made up of six extracellular scavenger receptor cysteine-rich domains (Fig. 15.8). SWC1 is homologous to CD52 in other species. It is expressed exclusively on γ/δ T cells. These WC11 γ/δ T cells play an important defensive role in pigs and are the predominant γ/δ T cell population in this species [64]. The WC1 family is polygenic in pigs as in cattle and small ruminants but only eleven WC1 genes have been identified in the pig [65]. The WC1 family consists of lineage-restricted pattern recognition molecules that can directly bind to pathogenderived molecules. They also serve as co-receptors that work in association with γ/δ T cell receptors to bind and respond to specific pathogens. In ruminants, not all γ/δ T cells express WC1. (Bovine uterine γ/δ T cells are WC1- negative while circulating T cells are WC1-positive). It should be noted that orthologs of WC1 are found in the platypus, as well as other artiodactyls such as camels, but not in either humans or mice. On the other hand, WC1 is closely

232

SECTION | 2 Mammalian orders

related to the hemoglobin scavenger receptor CD163. This is expressed in humans, rats, and pigs. CD5 and CD6 also belong to the WC1 family. Other pig WC1 family receptors have also been described. For example, SWC2 is homologous to CD27 and is expressed on T and NK cells. SWC3 is expressed on monocytes and macrophages. SWC9 is expressed only by mature pig macrophages.

15.8.2 T cell receptors 15.8.2.1 α/β T cells Pig α/β T cells can be subdivided into three populations based on their expression of CD4 and CD8. These are CD8α/ β1CD42, CD82CD41 and CD8α/α1CD411 [66]. The CD8α/α1 population does not develop in the thymus but is generated as a result of peripheral activation. These peripheral double-positive T cells are considered to be effector or memory T cells. Other mammals possess CD8α/α 1 T cells but in contrast to them, the pig α/α cells are not downregulated after activation. Some of the CD41 T cells that express CD8α/α homodimers are cytotoxic. Pig T cells also constitutively express MHC class II molecules which is also a unique feature [66]. Other important T cell subsets that have been characterized in pigs are NKT cells, follicular helper T cells (Tfh), and Th17 cells. NKT cells can be identified by their expression of CD31NKp461. The cells can be identified by their expression of transcription factor- B cell lymphoma 6 (Bcl6). ProinflammatoryTh17 cells can be characterized by their use of the transcription factor RORγt and, of course, by their production of IL-17 [67].

15.8.2.2 γ/δ T cells T cells expressing γ/δ receptors can be detected in the developing piglet thymus by 40 days gestation. Most of these are CD4-negative. γ/δ T cells, in contrast to α/β T cells, are readily activated and as a result, serve as the first responders to infections or inflammatory diseases. In young pigs, their numbers in the blood rise to reach a maximum of around 19
25 weeks. Up to 66% of their blood T cells may be γ/δ positive at this time, but this eventually drops to 25%
50% in adults. In the spleen 18%
25% of the T cells are γ/δ1. In the thymus, the figure is 5%
31%. Lower proportions are found in the other secondary lymphoid organs. γ/δ T cells are present in especially large numbers in body surface tissues such as the dermis and the intestinal epithelium. Some have differing levels of TCR γ/δ expression [68]. Up to 60% of γ/δ T cells in pig blood are double-positive (CD41, CD81). The rest are predominantly doublenegative (CD42, CD82). Pigs also have two subpopulations of circulating γ/δ T cells based on their expression of the adhesion molecule CD2. Thus CD2 and CD21 may reflect two different cell lineages. Otherwise, CD22 cells may simply be naı¨ve [67]. CD22 cells have not been identified in other species. They exhibit two distinct γ-chain clonotypes. Some pig γ/δ T cells can function as antigen-presenting cells using MHC class II molecules. As in humans and mice, the porcine γ/δ TCR repertoire is diverse in young piglets but becomes oligoclonal and compartmentalized in 2
5year-old pigs at mucosal sites [69]. A subset of porcine γ/δ T cells may express the activating NK cell receptor NKG2D. Many circulating γ/δ T cells in the pig are also WC11. These subpopulations of γ/δ T cells are functionally diverse. Some may produce IFN-γ alone, TNF-α alone, or both IFN-γ and TNF-α [70]. Other subpopulations can produce IL-17. Pig γ/δ T cells show increased expression of TLR2, 5, 9, 10, and NOD2 following salmonella infection. Pigs possess three unique γ/δ TCR clonotypes. They share a common delta chain but use different gamma chains. One γ chain of 38 kDa is expressed on blood lymphocytes; the second γ chain of 37 kDa is distributed between blood and tissue T cells, while the third γ chain of 46 kDa is expressed on CD21 T cells within lymphoid tissues [68]. These γ chains probably originate from different cassettes within the TRG locus. (Trans-cassette gene-splicing occurs in the pig in addition to the more normal V-J recombination within cassettes.) Pigs also have a high TCR-δ diversity with B 39 unique clonotypes and share one specific clonotype with the CD2 cells (Vδ1DδxJδ4).

15.8.2.3 TCR genes Several different whole-genome assemblies of the pig genome have been generated [66]. Given the genetic heterogeneity of the domestic pig populations, it is unsurprising that these assemblies differ between individuals. This was described previously for the IGHG genes, but it is also a feature of the pig T cell receptor gene loci (Fig. 15.9).

Suiformes: Pigs and Peccaries Chapter | 15

TRA/D

TRB

5'

48 VA 43

28 6 VD DD 7

5

CA

FIGURE 15.9 The organization of the porcine T cell antigen receptor genes. The TRG locus may contain differing numbers of V-J-C cassettes depending upon their haplotype.

6 C3

2

3'

50 JA

7 C1

TRG

4 JD CD

233

C2

2 C1

V

C3

C2

D

J

C

15.8.2.4 TRA/D As in other species, the pig TRD locus is embedded within the TRA locus. The TRDJ, TRDD, and TRDC genes are positioned between the V and J genes of TRA. The pig TRD locus has a more complex structure than humans or mice. The existence of 28 TRDV, 6 TRDD, 4 TRDJ, and 1 TRDC gene have been reported Thus while pigs have at least six TRDD genes, humans have three and mice only two. The pig TRDD genes may be concatenated into a chain thus generating up to four D segments in series. Thus given the large numbers of γ/δ T cells in this species the delta chain can have a very diverse repertoire. Porcine α/β T cells express 48 TRAV and 50 TRAJ sequences. The TRA region is highly conserved in comparison to humans and mice. It contains 33 functional TRAV genes that can be divided into 20 subfamilies. These all have human functional counterparts.

15.8.2.5 TRB The pig TRB locus occupies 407 kb on chromosome 10. It is arranged in reverse orientation. It consists of 43 TRBV genes, three TRBD genes, 20 TRBJ genes, and three TRBC genes. These are arranged in three DJC cassettes. Note that sheep also have three cassettes, while rabbits, dogs, cats, and primates only have two [22]. Some V-D- J combinations are nonrandomly expressed and there is a bias towards the use of several specific V-J combinations [64]. For example, four of these combinations account for .80% of all TRBV usage in thymocytes and peripheral T cells [66].

15.8.2.6 TRG The pig TRG locus contains three or four TRGC genes each arranged with its V and J genes in the form of two or three cassettes. A typical arrangement would be, first a cassette (TRGC1) containing five V genes and two J genes plus the C gene followed by three -V-J-C- cassettes. Multiple frameshifts have occurred in TRGV1 and TRGV5 resulting in their pseudogenization. It appears that TRGC4 is only present in some haplotypes [65]. The genes in the pig TRGC1 cassette align with those in the bovine TRGC5 cassette. The other small pig TRGC cassettes do not align with any bovine TRG cassettes. Pig γ/δ CD22 T cells preferentially use the TRGC1 cassette [68].

15.8.3 Natural killer T cells Pigs, unlike other Artiodactyls, have an established NKT cell
CD1d system [71]. These NKT cells are thymusderived T cells that share a phenotype with NK cells. NKT cells are a subset of innate-like CD3ε1 T cells found in some mammals. Unlike conventional α/β T cells that bind peptide antigens presented by classical MHC molecules, NKT cells can respond to self and endogenous glycolipid antigens presented by nonpolymorphic CD1d proteins

234

SECTION | 2 Mammalian orders

expressed on some antigen-presenting cells and hematopoietic cells. This results in the rapid production of multiple effector cytokines such as IFN-γ and the mediation of a range of innate responses. The NKT cells interact with the CD1d-bound antigens through a conserved invariant TCR. Pig CD1- reactive T cells predominantly express an invariant Vα-Jα rearrangement that is highly homologous to its human counterparts. The co-expressed β chain also uses a semivariant V gene. The invariant α chain interacts with the antigen through its CDR1α and CD3Rα domains. The β chain in contrast makes stabilizing contacts with CD1d and modulates the receptor binding affinity. Pigs possess a full repertoire of CD1 molecules (CD1a, CD1b, CD1c, CD1d, and CD1e) that may activate NKT cells. (Mice have only CD1, humans have six, while horses have 13 functional CD1 genes) [72,73]. Pig NKT cells are CD8α and CD44 positive while CD11b and NKp46 are negative [73]. They account for up to 1% of pig T cells [74].

References [1] Frantz L, Meijard E, Gongora J, Haile J, et al. The evolution of suidae. Ann Rev Anim Biosci 2016;. Available from: https://doi.org/10.1146/ annurev-animal-021815-111155. [2] Ruvinsky A, Rothschild MF. Systematics and evolution of the pig. In: Rothschild MF, Ruvinsky A, editors. The genetics of the Pig. CAB International; 1998, p. 1
16 [3] Pabst R. The pig as a model for immunology research. Cell Tissue Res 2020;380:287
304. [4] Zhang M, Li Z, Li J, Huang T, et al. Revisiting the pig IGHC gene locus in different breeds uncovers nine distinct IGHG genes. J Immunol 2020;205:2137
45. [5] Furukawa S, Kuroda Y, Sugiyama A. A comparison of the histologic structure of the placenta in experimental animals. J Toxicol Pathol 2014;27:11
18. [6] Sinkora M, Butler JE. The ontogeny of the porcine immune system. Dev Comp Immunol 2009;33:273
83. [7] Tuboly S, Berna´th S, Gla´vits R, Medveczky I. Intestinal absorption of colostral lymphoid cells in newborn piglets. Vet Immunol Immunopathol 1988;20(1):75
85. [8] Bandrick M, Ariza-Nieto C, Baidoo SK, Molitor TW. Colostral antibody-mediated and cell-mediated immunity contributes to innate and antigen-specific immunity in piglets. Dev Comp Immunol 2014;43:114
20. [9] Hlavova K, Stepanova H, Faldyna M. The phenotype and activation status of T and NK cells in porcine colostrum suggest these are central/ effector memory cells. Vet J 2014;202:477
82. [10] Levast B, Berri M, Wilson HL, Meurens F, Salmon H. Development of gut immunoglobulin A production in piglet in response to innate and environmental factors. Dev Comp Immunol 2014;44:235
44. [11] Ke C, Ma Y, Pan D, Wan Z, et al. FcRn is not the receptor mediating the transfer of serum IgG to colostrum in pigs. Immunology 2021;163:448
59. [12] Jeˇzek J, Staric J, Nemec M, Plut J, et al. The influence of age, farm, and physiological status on pig hematological profiles. J Swine Health Prod 2018;26(2):72
8. [13] Mair KH, Sedlak C, Ka¨ser T, Pasternak A, et al. The porcine innate immune system: an update. Dev Comp Immunol 2014;45:321
43. [14] Nova´k K. Functional polymorphisms in Toll-like receptor genes for innate immunity in farm animals. Vet Immunol Immunopathol 2014;157:1
11. [15] Vrentas CE, Schaut RG, Boggiatto PM, Olsen SC, et al. Inflammasomes in livestock and wildlife: insights into the intersection of pathogens and natural host species. Vet Immunol Immunopathol 2018;201:49
56. [16] Sakuma C, Toki D, Shinkai H, Takenouchi T, et al. Pigs lack functional NLRC4 and NAIP genes. Immunogenetics 2017;69:125
30. [17] Heegaard PMH, Stockmarr A, Pin˜iero M, Carpintero R, et al. Optimal combinations of acute phase proteins for detecting infectious diseases in pigs. BMC Vet Res 2011;42:50. [18] Huang Y, Liu Z, Liu W, Yin C, et al. Salivary haptoglobin and chromogranin A as noninvasive markers during restraint stress in pigs. Res Vet Sci 2017;114:27
30. [19] Sang Y, Blecha F. Porcine host defense peptides: expanding repertoire and functions. Dev Comp Immunol 2009;33:334
43. [20] Chen J, Qi S, Guo R, Yu B, et al. Different messenger RNA expression for the antimicrobial peptides, beta-defensins between Meishan and crossbred pigs. Mol Biol Rep 2010;37:1633
9. [21] Herzig CTA, Warers RW, Baldwin CL, Telfer JC. Evolution of the CD163 family and its relationship to the bovine gamma delta T-cell coreceptor WC1. BMC Evol Biol 2010;10:181. [22] Dawson HD, Loveland JE, Pascal G, Gilbert JGR, et al. Structural and functional annotation of the porcine immunome. BMC Genomics 2013;14:332
48. [23] Murtaugh MP, Johnson CR, Xiao Z, et al. Species specialization in cytokine biology: is interleukin-4 central to the Th1-Th2 paradigm in swine? Dev Comp Immunol 2009;33:344
52. [24] Dijkstra JM, Takizawa F, Fischer U, Friedrich M, et al. Identification of a gene for an ancient cytokine, interleukin 15-like, in mammals; interleukins 2 and 15 co-evolved with this third family member, all sharing motifs of IL-15Ralpha. Immunogenetics 2014;66:93
103. [25] Sinkorova J, Stepanova K, Butler JE, Sinkora M. T cells in swine completely rearrange immunoglobulin heavy chain genes. Dev Comp Immunol 2019;99:103396. Available from: https://doi.org/10.1016/j.dci.2019.103396.

Suiformes: Pigs and Peccaries Chapter | 15

235

[26] Binns RM, Pabst R. Lymphoid tissue structure and lymphocyte trafficking in the pig. Vet Immunol Immunopathol 1994;43:79
87. [27] Liebler-Tenorio EM, Pabst R. MALT structure and function in farm animals. Vet Res 2006;37:257
80. [28] Casteleyn C, Breugelmans S, Simoens P, Van den Broeck W. The tonsils revisited: review of the anatomical localization and histologic characterization of the tonsils of domestic and laboratory animals. J Immunol Res 2011;2011:472460. Available from: https://doi.org/10.1155/2011/ 472-460. [29] Driessen A, Van Ginneken C, Creemers J, Lambrichts I, et al. Histological and immunohistochemical study of the lymphoid tissue in the normal stomach of the gnotobiotic pig. Virchow’s Arch 2002;441:589
98. [30] Kararli TT. Comparison of the gastrointestinal anatomy, physiology, and biochemistry of humans and commonly used laboratory animals. Biopharm Drug Dispos 1995;16:351
80. [31] Fujikawa M, Ito S, Suzuku S, Fuchimoto D, et al. Organogenesis of ileal Peyer’s patches is initiated prenatally and accelerated postnatally with comprehensive proliferation of B cells in pigs. Front Immunol 2020;11:604674. Available from: https://doi.org/10.3389/fimmu.2020.604674. [32] Vreman S, Auray G, Savelkoul HFJ, Rebel A, et al. Neonatal porcine blood derived dendritic cell subsets show activation after TLR2 or TLR9 stimulation. Dev Comp Immunol 2018;84:361
70. [33] Mair KH, Stadler M, Saalmueller A, Gerner W. Porcine plasmacytoid dendritic cells are unique in their expression of a functional NKp46 receptor. Front Immunol 2022;13:822258. Available from: https://doi.org/10.3389/fimmu.2022.822258. [34] Kulski JK, Shiina T, Anzai T, Kohara S, Inoko H. Comparative genomic analysis of the MHC: the evolution of class I duplication blocks, diversity and complexity from shark to man. Immunol Rev 2002;190:95
122. [35] Renard C, Hart E, Sehra H, Beasley H, et al. The genomic sequence and analysis of the swine major histocompatibility complex. Genomics 2006;88:96
110. [36] Chardon P, Renard C, Vaiman M. The major histocompatibility complex in swine. Immunol Rev 1999;167:179
92. [37] Lunney JK, Ho C-S, Wysocki M, Smith DM. Molecular genetics of the swine major histocompatibility complex, the SLA complex. Dev Comp Immunol 2009;33:362
74. [38] Velten F, Rogel-Gaillard C, Renard C, et al. A first map of the porcine major histocompatibility complex class I region. Tissue Antigens 1998;51:183
94. [39] Tanaka-Matsuda M, Ando A, Rogel-Gaillard C, Chardon P, Genishi H. Difference in number of loci of swine leukocyte antigen classical class I genes among haplotypes. Genomics 2009;93:261
73. [40] Ho CS, Lunney JK, Franzo-Romain MH, et al. Molecular characterization of swine leucocyte antigen class I genes in outbred pig populations. Anim Genet 2009;4:468
78. [41] Le MT, Choi H, Lee H, Le VCQ, et al. SLA-1 genetic diversity in pigs: extensive analysuis of copy number variation, heterozygosity, expression, and breed specificity. Sci Rep 2020;10:743. Available from: https://doi.org/10.1038/s41598-020-57712-5. [42] Wei X, Wang S, Wang S, Xie X, et al. Structure and peptidomes of swine MHC class I with long peptides reveal the cross-species characteristics of the novel N-terminal extension presentation mode. J Immunol 2022;208:480
91. [43] Ho CS, Lunney JK, Lee JH, et al. Molecular characterization of swine leucocyte antigen class II genes in outbred pig populations. Anim Genet 2010;41:428
32. [44] Mair KH, Crossman AJ, Wagner B, Babasyan S, et al. The natural cytotoxicity receptor NKp44 (NCR2, CD336) is expressed on the majority of porcine NK cells Ex Vivo without stimulation. Front Immunol 2022;13:767530. Available from: https://doi.org/10.3389/fimmu.2022.767530. [45] Gerner W, Ka¨ser T, Saalmu¨ller A. Porcine T lymphocytes and NK cells
An update. Dev Comp Immunol 2009;33:310
20. [46] Sinkora M, Sinkorova J. B cell lymphogenesis in swine is located in the bone marrow. J Immunol 2014;193:5023
32. [47] Potockova H, Sinkorova J, Karova K, Sinkora M. The distribution of lymphoid cells in the small intestine of germ-free and conventional piglets. Dev Comp Immunol 2015;51:99
107. [48] Eguchi-Ogawa T, Wertz N, Sun X-Z, Puimi F, et al. Antibody repertoire development in fetal and neonatal piglets. XI. The relationship of variable heavy chain gene usage and the genomic organization of the variable heavy chain locus. J Immunol 2010;184:3734
42. [49] Eguchi-Ogawa T, Toki D, Wertz N, Butler JE, Uenishi H. Structure of the genomic sequence comprising the immunoglobulin heavy constant (IGHC) genes in Sus scrofa. Mol Immunol 2012;52:97
107. [50] Butler JE, Wertz N, Deschacht N, Kacskovics I. Porcine IgG: structure, genetics, and evolution. Immunogenetics 2009;61:209
30. [51] Butler JE, Wertz N. The immunoglobulin genes of domestic swine. In: Kaushik AK, Passman Y, editors. Comparative Immunoglobulin genetics. Toronto, ON: Apple Academic Press; 2014. [52] Dawson HD, Sang Y, Lunney JK. Porcine cytokines, chemokines and growth factors:2019 update. Res Vet Sci 2020;131:266
300. [53] Vadnais ML, Criscitiello MF, Smider VV. Antibodies in other species. In: Vaughan T, Osbourn J, Jallal B, editors. Protein therapeutics. Verlag: Ie, Wiley; 2017. [54] Butler JE, Weber P, Wertz N. Antibody repertoire development in fetal and neonatal piglets. XIII. Hybrid VH genes and the preimmune repertoire revisited. J Immunol 2006;177:5459
70. [55] Schwartz JC, Lefranc M-P, Murtaugh MP. Organization complexity and allelic diversity of the porcine (Sus scrofa domestica) immunoglobulin lambda locus. Immunogenetics 2012;64(5):399
407. [56] Guo X, Schwartz JC, Murtaugh MP. Genomic variation in the porcine immunoglobulin lambda region. Immunogenetics 2016;68(4):285
93. [57] Schwartz JC, Murtaugh MP. Characterization of a polymorphic IGLV gene in pigs (Sus scrofa). Immunogenetics 2014;66:507
11. [58] Butler JE, Wertz N, Sun J, Wang H, et al. Antibody repertoire development in fetal and neonatal pigs. VII. Characterization of the preimmune κ light chain repertoire. J Immunol 2004;173:6794
805.

236

SECTION | 2 Mammalian orders

[59] Sun X, Wertz N, Lager K, Sinkora M, et al. Antibody repertoire development in fetal and neonatal piglets. XXII. λ rearrangement precedes κ rearrangement during B-cell lymphogenesis in swine. Immunology 2012;137:149
58. [60] Sinkora M, Stepanova K, Sinkorova J. Immunoglobulin light chain κ precedes λ rearrangement in swine but a majority of λ1 B cells are generated earlier. Dev Comp Immunol 2020;111:103751. [61] Sinkora M, Stepanova K, Butler JE, Sinkora Jr. M, et al. Comparative aspects of immunoglobulin gene rearrangement arrays in different species. Front Immunol 2022;13:823145. Available from: https://doi.org/10.3389/fimmu.2022.823145. [62] Sinkora M, Stepanova K, Sinkorova J. Consequences of the different order of immunoglobulin gene rearrangements in swine. Dev Comp Immunol 2022;126:104196. Available from: https://doi.org/10.1016/j.dci.2021.104196. [63] Charerntantanakul W, Roth JA. Biology of porcine T lymphocytes. Anim Health Res Rev 2007;8:1
16. [64] Pillai MR, Lefevre EA, Carr VB, Charleston B, O’Grady P. Workshop cluster 1, a γδ T cell specific receptor is phosphorylated and down regulated by activation induced Src family kinase activity. Mol Immunol 2007;44:1691
703. [65] Le Page L, Gillespie A, Schwartz JC, Prawits L-M, et al. Subpopulations of swine γδ T cells defined by TCRγ and WC1 gene expression. Dev Comp Immunol 2021;125:104214. Available from: https://doi.org/10.1016/j.dci.2021.104214. [66] Sinkora M, Butler JE. Progress in the use of swine in developmental immunology of B and T lymphocytes. Dev Comp Immunol 2016;58:1
17. [67] Ka¨ser T. Swine as biomedical model for T-cell research
Success and potential for transmittable and non-transmittable human disease. Mol Immunol 2021;135:95
115. [68] Le Page L, Baldwin CL, Telfer JC. γδ T cells in artiodactyls: focus on swine. Dev Comp Immunol 2022;128:104334. Available from: https:// doi.org/10.1016/j.dci.2021.104334. [69] Holtmeier W, Ka¨ller J, Geisel W, et al. Development and compartmentalization of the porcine TCR delta repertoire at mucosal and extraintestinal sites: the pig as a model for analyzing the effects of age and microbial factors. J Immunol 2002;169:1993
2002. [70] Sedlak C, Patzl M, Saalmu¨ller A, Gerner W. IL-12 and IL-18 induce interferon-γ production and de novo CD2 expression in porcine γδ T cells. Dev Comp Immunol 2014;47:115
22. [71] Van Beeck FAL, Reinink P, Hermsen R, Zajonc D, et al. Functional CD1d and/or NKT invariant chain transcript in horse, pig, African elephant and Guinea pig, but not in ruminants. Mol Immunol 2009;46:1424
31. [72] Yang G, Artiaga BL, Lewis ST, Driver JP. Characterizing porcine invariant natural killer T cells: a comparative study with NK cells and T cells. Dev Comp Immunol 2017;76:343
51. [73] Yang G, Artiaga BL, Lomelino CL, Jayaprakash AD, et al. Next generation sequencing of the pig αβ TCR repertoire identifies the porcine invariant NKT cell receptor. J Immunol 2019;202:1981
91. [74] Wu M, Jiang Q, Nazmi A, Yin J, Yang G. Swine unconventional T cells. Dev Comp Immunol 2022;128:104330. Available from: https://doi. org/10.1016/j.dci.2021.104330. [75] Jørgensen PB, Eriksen LL, Fenton TM, Bailey M, et al. The porcine large intestine contains developmentally distinct submucosal lymphoid clusters and mucosal isolated lymphoid follicles. Dev Comp Immunol. 2022. Available from: https://doi.org/10.1016/j.dci.2022.104375.

Chapter 16

The cetaceans: whales and dolphins

Common dolphin. Delphinus delphis. Courtesy R. Tizard.

The divergence of the whales and dolphins, the cetaceans, from the early Artiodactyls appears to have occurred around 60 mya in what is now the Indian subcontinent. Fossils found in northwest Pakistan and northern India reveal the progressive changes from a land mammal to a marine one. This process of transforming land into marine mammals was slow and took at least 15 million years. The first cetacean fossils appear around 53.5 mya. (The first artiodactyl fossils are dated around 55 mya). This evolutionary process probably began during the early Eocene period as carnivorous mammals moved into the shallow warm waters of the Tethys sea. They developed skeletal changes such as a thick bony wall around the middle ear, a large powerful tail, their nasal orifice moved backward, and their eyes moved to the sides of their head. They remained air breathers and like all mammals continued to be viviparous and fed their newborns with milk. They still retain many skeletal features reflecting their ancestry including vestigial hind limb bones. Molecular phylogeny studies and analysis of mitochondrial genomes suggest not only that the whales are closely related to the artiodactyls but that hippopotamuses are their closest relatives on land [1,2]. Thus the cetaceans and hippos form a monophyletic group deeply embedded within the Artiodactyla. Camels and pigs are considered basal to this order. The early cetaceans are believed to have lived in freshwater habitats as wading quadrupeds before adapting to a fully aquatic lifestyle. The whales and hippos probably shared a common semiaquatic ancestor about 60 mya. Mitochondrial genome studies suggest that they diverged B54 mya (Fig. 16.1) [2]. During their move from land to sea, the early cetaceans would have been required to have made significant adaptations. Not only would they have needed structural changes such as the conversion of limbs into flippers and flukes, the loss of hair, the development of a blubber layer, and the development of echolocation skills. They also had to make physiological changes that permitted deep and prolonged diving. In addition, they would have had to adapt to changes in the number and diversity of the parasites and microbial pathogens they encountered. It has been assumed that there were fewer pathogenic microbes and infectious diseases in their new aquatic environment and as a result, there would have been less pressure to adapt and maintain polymorphism in immune genes, especially in the major histocompatibility complex (MHC). The newly aquatic mammals, however, would have carried old pathogens with them and encountered new pathogens that would have had to be defeated [3]. Recent studies have suggested that while different, the selective pressures exerted by pathogens would have been no less [4,5]. Recent mass die-offs of marine mammals have been attributed to viral infections, especially morbilliviruses
in many cases derived from terrestrial sources. Moreover, differential pressures must have been placed on their antigen recognition molecules such as their TLRs, their immunoglobulins, and their MHC. Their immune functions would Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00008-3 © 2023 Elsevier Inc. All rights reserved.

237

238

SECTION | 2 Mammalian orders

60

50

40

30

20

10

0 Million years ago Orcas Dolphins Toothed whales Baleen whales Hippopotamuses Ruminants Pigs Camels

FIGURE 16.1 The phylogeny of the cetaceans relative to the other artiodactyls. The hippopotamus is a member of the cetacea that diverged at least 55 mya.

have also had to adapt to a completely new set of challenges. It is clear that there is reduced genetic diversity in marine mammals, but these are equally likely to have been the result of population bottlenecks due to recent whale hunting and random drift in small populations. The MHC has provided some evidence for the timing of these divergences [5]. Thus using the phylogenies derived from the MHC class II region, specifically the DQB and DRB loci, and postulating sequential gene duplication, it has been possible to investigate two dolphin species, the bottlenose dolphin (Turciops truncatus) and the Indo-Pacific bottlenose dolphin (T. aduncus) and compare these to the artiodactyls such as the pigs and ruminants. On this basis, the estimated divergence time of the cetaceans was B60 mya while the divergence time of the two dolphin species was about 24 mya [5]. Sometime around 35 million years ago in the early Oligocene, a second radiation occurred, and the early cetaceans diverged into the Odontoceti, the toothed whales, including the porpoises and dolphins, and the Mysticeti, the baleen whales [6]. The Odontocetes have lost their olfactory sense and as a result, use sophisticated echolocation to find their prey. The Mysticeti have retained their olfactory senses but have lost their teeth in favor of baleen. They are characterized by extremely large body sizes.

16.1

Reproduction and lactation

Modern cetaceans must give birth underwater and nurse their offspring underwater. The cetacean placenta is an epitheliochorial diffuse tissue that fills the entire uterus. The body of the developing fetus usually occupies one uterine horn while its tail may extend into the other. These structures appear to be similar in toothed dolphins and baleen whales [7,8]. Cetaceans have two very different lactation patterns. Thus in baleen whales, lactation lasts 5
7 months. Their milk is full of fat (30%
50%) and energy-rich. In toothed whales, lactation may last 1
3 years, and their milk is lower in fat (10%
30%) [9]. Given a long-standing prohibition of the killing of mothers with calves, there is no information on the levels of immunoglobulins in whale colostrum [9]. The level of immunoglobulins in cetacean colostrum has yet to be measured. Traces of immunoglobulins have been detected in the prepartum secretions of bowhead whales [9]. However, given the use of an epitheliochorial placenta and its effects on maternal immunoglobulin transfer in other artiodactyls, it is assumed that there is a minimal transplacental transfer of immunoglobulins and that newborn cetaceans require immunoglobulins from maternal colostrum to provide passive immunity over the first weeks of life.

16.2

Hematology

In general, cetacean leukocyte counts are similar to those in humans with a preponderance of neutrophils [10]. Thus total counts range from 5.5 3 103 cells/μL in Dall’s porpoise. (Phocoenoides dalli) to 11.5 3 103 cells/μL in pacific pilot whales (Globicephala macrorhynchus). The proportion of neutrophils ranges from 43% in Pacific white-sided dolphins (Lagenorhynchus obliquidens) to 81% in orcas (Orcinus orca). In captive dolphins, eosinophil counts may be

The cetaceans: whales and dolphins Chapter | 16

239

high in the absence of detectable parasites. Thus in a population of apparently healthy Atlantic bottlenose dolphins (T. truncatus) the total leukocyte count averaged 9.65 6 1.79 x 103 cells/μL. Of these, lymphocytes accounted for 1.84 6 0.74 x 103 cells/μL, while segmented eosinophils accounted for 3.33 6 1.26 x 103 cells/μL [11]. Other reports suggest that the number of leukocytes is lower (6.6 3 103/μL) if hand-counted! In captive dolphins, the relative proportions are 60.3% neutrophils, 21.7% lymphocytes, 2.3% monocytes, 14.5% eosinophils and ,1% basophils. On the other hand, free-ranging dolphins had 39.5% neutrophils, 17% lymphocytes, 2.7% monocytes, 39.6 eosinophils and 0.9% basophils [12].

16.3

Innate immunity

Innate immunity is an ancient and highly conserved defensive system. As a result, few significant differences are seen between the mechanisms of inflammation in different mammalian species. For example, most TLRs appear to be well conserved suggesting little change in function. There is however evidence of significant adaptive evolution of TLR4 during the transition of cetaceans from a terrestrial to an aquatic existence [13]. These changes have primarily occurred in the lipopolysaccharide-binding site suggesting that they reflect adaptation to the new pathogens encountered in the marine environment.

16.3.1 Neutrophils Polymorphonuclear neutrophils in dolphin blood function like those in other artiodactyls. Upon appropriate stimulation, they can extrude neutrophil extracellular traps (NETS). For example, they release NETS upon exposure to tachyzoites of Neospora caninum. As in other mammals the NETS contain histones, elastase, myeloperoxidase, and pentraxins [14]. As described above, cetaceans tend to have elevated eosinophil counts compared to other mammals [15]. These have been reported to range from two to thirty percent of blood leukocytes. While this has generally been attributed to high parasite burdens, there is little evidence that these mammals carry unusually high numbers of parasites. Eosinophils may serve other functions in these mammals [15].

16.3.2 Cytokines IL-1α, IL-1β, and IL-1RA have been cloned and sequenced from bottlenose dolphins. Their active sites are conserved while the rest of the molecules have not. The major positive acute-phase proteins in the cetaceans have also been investigated in bottlenose dolphins. They include C-reactive protein, serum amyloid A, haptoglobin, and fibrinogen. Serum albumin is a negative acute-phase protein in many dolphins and whales [16]. It is interesting to note that toothed whales including dolphins and orcas lack the key antiviral proteins Mx1 and Mx2 [17] as a result of a loss of myxovirus (MX) genes. However, baleen whales possess the MX genes. These are interferon-induced enzymes that help defend mammals against many viral infections. As a result, the toothed whales appear to be relatively vulnerable to influenza and other RNA viruses such as dolphin morbillivirus.

16.4

Lymphoid organs

16.4.1 Thymus As in other mammals, the thymus is located in the anterior superior mediastinum, behind the sternum, and close to the thyroid gland. It is enclosed in a connective tissue capsule. In some dolphins, the parathyroid glands or even lymph nodes may be enclosed within the thymic parenchyma. On histology, it has a darker staining cortex and a lighter staining medulla with extensive aggregates of T cells [18]. A reticular network of epithelial cells is present throughout each lobule. Hassall’s corpuscles are scattered throughout the medulla. They are more prominent in younger animals but can be found in old animals. Age-related thymic atrophy appears to be a slow process in harbor porpoises (Phocoena phocoena), but in other cetaceans, an apparently normal thymus may be found in healthy adults. The thymus of the bottlenose dolphin persists into adult life although its cortex progressively thins. The network of epithelial cells condenses and forms cysts as the numbers of lymphocytes decrease [19]. The cysts increase in number and size with increasing body size. They initially are small and irregularly shaped, but over time they enlarge and become spherical. The cysts are lined by squamous epithelium and filled with a colloid-like material. The shrunken thymus is progressively replaced by adipose tissue (Fig. 16.2).

240

SECTION | 2 Mammalian orders

FIGURE 16.2 Cyst formation in the thymus of harbor porpoises, Phocoena phocoena. (A). Multiple thymic microcysts (*) in an involuted thymus. L, lymphocytes; H&E stain. Bar 5 100 μm. (B). Multiple thymic macrocysts (up to 2 cm in diameter) containing gelatinous material: Formalin fixation; bar 5 1 cm. From Beineke A, et al. Vet Immunol Immunopathol 2010;133: 81
94. With permission.

16.4.2 Spleen Cetacea have relatively small spleens (0.2% of body weight) when compared to land mammals so they have limited blood storage capability. Like other lymphoid organs in bottlenose dolphins, they reach maximum size at the onset of puberty and subsequently shrink. Even so, they share the same muscular capsule as the terrestrial Artiodactyla. As in terrestrial mammals, the splenic tissue consists of white pulp distributed evenly through the red pulp. The white pulp consists of lymphoid nodules at the end of arterioles. The periarteriolar lymphoid sheaths are prominent. Germinal centers are inconsistently present. The splenic capsule has an outer fibrous and inner muscular layer. The spleen is involved in hematopoiesis in bottlenose dolphins and harbor porpoises. Beluga (Delphinapterus leucas) splenic white pulp consists predominantly of periarteriolar lymphoid sheaths and follicles are rare or absent in adults. This suggests that they live in deep clean marine areas where microbial challenges are minimal [15,20]. Marginal zones are difficult to identify in older belugas [15]. The red pulp appears to have two layers, an intermediate zone, and a perivenous layer. The perivenous layer is narrow and consists of venules and an intravascular reticular tissue rich in myeloid cells. This corresponds to the red pulp of a conventional mammalian spleen. The arteriovenous connection is closed, and no ellipsoids are present around the terminal arterioles. The Odontocete spleen has two separate venous drainage routes, the hilar and capsular systems [21]. Multiple accessory spleens may be present in whales and dolphins (Fig. 16.3). Thus a survey of 14 species of 63 individual Odontocetes and Mysticetes stranded on the northern coasts of Brazil, found that accessory spleens were present in 38 animals [22]. The number ranged from one to 14 spleens per animal. They were either firmly fixed to the main spleen or the large curvature of the first stomach or both. Their presence was not related to age or sex, but they were more prevalent in larger, deeper diving species. Their germinal centers were progressively reduced in numbers with increasing age [22]. Accessory spleens have been found in 21% of common dolphins (Delphinus delphis) and 18% of striped dolphins (Stenella coeruleoalba) [20].

16.4.3 Lymph nodes There is great variability in lymph node definition in marine mammals. Some are clearly defined single nodes, others, such as the mesenteric nodes, are aggregated to form a single large irregular mass. The lymph nodes of the bottlenose dolphin occur in four well-defined groups based on their location. Thus some are somatic and found in the cervical and pelvic regions. Others are strategically located close to the lung marginal zone and associated with the bronchi in the hilus and diaphragm and collect the lymphatic drainage from the lung. Lymph nodes are also associated with the aortic arch and with the thymus and thyroid. Dolphins have visceral lymph nodes of which the mesenteric is the largest and most obvious since they form a large conglomerate nodular lymphoid mass (the pseudopancreas) [15,20]. The visceral lymph nodes in the bottlenose dolphin

The cetaceans: whales and dolphins Chapter | 16

241

FIGURE 16.3 The spleen and seven accessory spleens from a harbor porpoise, P. phocoena; bar 5 1 cm. From Beineke A, et al. Vet Immunol Immunopathol 2010;133:81
94. With permission.

have smooth muscle fibers within their capsule and trabeculae as well as a network of muscle fascicles within the organ, suggesting that it may have a contractile function and assist in maintaining lymph circulation. In marine mammals, lymph node structure is highly variable. All the lymph nodes of bottlenose dolphins are of conventional structure with a peripheral cortex and a central medulla. In striped dolphins (Stenella coerulealba), on the other hand, the mesenteric lymph nodes are of conventional structure, whereas the mediastinal lymph nodes have an inverted structure with a central cell dense “cortex” and a peripheral less dense “medulla” [23]. Beluga (D. leucas) lymph nodes, while of conventional structure, possess significant amounts of smooth muscle in their lymph mode medullas in close proximity to their sinuses [15]. While prominent in young belugas, follicles are small or absent in older animals.

16.4.4 Mucosal associated lymphoid tissues The digestive and respiratory tracts of cetaceans are completely separate. As a result, they require “tonsils” both in their pharynx and in their larynx (Fig. 16.4).

16.4.4.1 The complex laryngeal gland The upper respiratory tract requires its immune defenses. When cetaceans inhale, air enters through the blowhole and is directed directly to the larynx. Thus a complex lymphoepithelial “gland” has been described in the larynx of several cetaceans including bottlenose dolphins. It is located in the rostroventral mucosa of the larynx overlying the cricoid cartilage [24]. It is a well-defined raised trabeculated area covered by a pseudostratified columnar epithelium that branches into the underlying submucosal layer. As a result, there are multiple epithelial-lined folds and crypts surrounded by aggregations of lymphocytes. The lymphocytes often infiltrate the epithelium. In addition, mucus glands are associated with this lymphoid epithelium. This mucus presumably traps particles on inhalation. Histologically the laryngeal gland closely resembles the palatine or oropharyngeal tonsils. Age-related involution is much less obvious in this organ than in the other mucosal lymphoid tissues. In their oral cavity, the pharynx has moved anteriorly, and dolphins have both dorsal and ventral oropharyngeal tonsils. The paired oropharyngeal tonsils are lymphoepithelial organs with multiple branching crypts and secretory glands. The dorsal pair are found in the mucosa of the palate. They vary in size and are larger in young dolphins. They have a typical tonsillar structure with 2
6 obvious crypt openings per tonsil [20]. The second pair of tonsils are much smaller and are found just anterior to the airway and larynx. They contain many mucus glands. Scattered small tonsil-like structures may also be present in the ventral pharyngeal mucosa [20]. Like other artiodactyls, many cetacean species have retained a multichambered stomach. The forestomach can serve as a reservoir for water ingested with their prey so that the water can be expelled once their prey is secured. The intestine of the bottlenose dolphin is long and slender and lacks a cecum and appendix [20]. The mucosa-associated lymphoid tissue is most prominent in the distal gastrointestinal tract. In contrast to the beluga, large confluent lymphoid structures are present in the intestine of the bottlenose dolphin. The bottlenose dolphin colonic mucosal lymphoid tissue undergoes nearly complete involution with puberty and increased age. Cetaceans such as the bottlenose dolphins, orcas, minke (Balaenoptera acutorostrata), and sperm whales (Physetus macrocephalus) appear to have lost the gastrointestinal host defense gene NOX1. This is possibly related to a decreased gut microbiome diversity in these marine carnivores. NOX1 is a transmembrane NADP oxidase that generates reactive

242

SECTION | 2 Mammalian orders

Blow hole

Esophagus Oropharyngeal tonsils Trachea

Laryngeal gland

FIGURE 16.4 The location of the tonsillar tissues in the bottlenose dolphin, Turciops truncatus.

oxygen species in the colonic epithelium. It is activated in response to microbial invasion thus killing pathogens and triggering the wound healing process [25].

16.4.5 Anal tonsils A specialized mucosal lymphoepithelial organ is found in the anal canal of several whale species including the sperm whale, gray whale (Eschrichtius robustus), bottlenose, and Ganges river dolphins (Platanista gangetica) [26]. A cluster of circumferential tonsil-like lymphoid aggregations is located where the rectal mucosa meets the stratified squamous epithelium of the anus [27]. They extend for a variable distance anteriorly. On the surface, these appear as irregular longitudinal folds and nodules. In the gray whale, distinct glandular openings are apparent. The anal tonsils are 30
40 cm long and about 10 cm wide. In the sperm whale, they are obvious as bumps surrounding the anal opening. The tonsils are covered in the stratified squamous epithelium but there is a central branching crypt that is invaginated into the lymphoid tissue mass where the crypts branch and divide the lymphoid masses into nodules. The invaginated epithelium gets progressively thinner as it goes deeper into the tonsil and disappears in the mid-portion of the crypt. The crypts of two or three tonsils may join to open onto the surface through a common opening. Lymphoid nodules without crypts may be present next to the tonsils. On histology, the tonsils consist of a mass of lymphatic nodules containing germinal centers or follicles. Their size and complexity vary between species [27]. They appear to be most active in young animals [26]. Lymphocyte loss and cystic dilatation of the anal tonsil begin after puberty as the anal tonsil degenerates with age. The anal tonsils are believed to capture foreign antigens that enter the body by rectal water influx during diving. This would be a problem at depths where the water pressure is significant and whales normally have to return to the surface to defecate.

16.5

The major histocompatibility complex

A feature unique to cetaceans is their global range distribution. This is much greater than that seen in typical terrestrial mammals. As a result, it is unreasonable to assume that such widespread species are genetically homogeneous. Likewise, there are clear differences in MHC profiles between deep water species compared to coastal or riverine species. Presumably reflecting differences in microbial exposure between the two habitats [5]. The initial studies on cetacean MHCs indicated that they showed limited diversity in whales. Subsequent studies have however indicated that there is great variability in MHC diversity between species. Given that diverse pathogens place significant selective pressure on immune system genes then it can be detected in widely distributed species such as the cetaceans. Thus the effects of longitudinal span, migration, and the occupation of different biomes have been studied for their influences on the cetacean MHC DQB locus [28]. Using a database of 15 species and 121 sequences it is clear that there is greater MHC diversity in resident tropical species when compared to temperate or migratory species. Oceanic cetaceans also show much greater MHC diversity. This cannot all be explained by pathogenic pressure and other processes must also be involved, perhaps including the adoption of different lifestyle strategies. Early studies showed low levels of MHC genetic diversity in fin whales (B. physalis) and sei whales (B. borealis) [29]. Sequence analysis of beluga MHC class II genes also showed low but measurable polymorphism. Subsequent studies

The cetaceans: whales and dolphins Chapter | 16

243

have demonstrated great variations in MHC polymorphism between toothed and baleen whale populations. Thus analysis of 42 bottlenose dolphins detected a gene encoding a highly polymorphic DQB receptor with six alleles and 21 SNPs over a length of 172 bp. In contrast, beluga whales had only five alleles and eleven SNPS over 172 bp in 233 animals. Even humpback whales have 23 DQB alleles expressed in 30 individuals. This suggests that even these very large K strategists sustain a significant mutation rate within their MHC. The data also argues against the idea that a marine environment has fewer pathogens and exercises lower selection pressure than a terrestrial environment [29]. Further studies on different cetacean species have shown a spectrum of differences in MHC diversity. For example, two highly endangered species, the Indo-Pacific humpback dolphin (Sousa chinensis) and the North Atlantic Right Whale (Eubalaena glacialis) show very limited diversity [30
32]. Likewise, whales that were subjected to extreme hunting pressure such as sei and fin whales also have significantly limited polymorphism [33]. Not all populations of baleen whales however lack MHC diversity. Thus a study on the DQB locus of 80 blue whales (B. musculus) from the Gulf of California identified 22 functional allotypes. Up to five functional DQB alleles per individual were found. Thus it appeared that the level of MHC variation in this population was comparable to that found in terrestrial mammals and that they too were subjected to pathogen-induced selective pressures [34]. On the other hand, river dolphins are remarkably diverse [35]. Thus the Chinese river dolphin (Lipotes vexillifer) class II DQB locus has been investigated [35]. Out of 18 individuals that were accidentally captured or stranded, 48 variable sites were determined, and 43 alleles were identified. Further analysis showed significant evidence of positive selection. It is postulated that the increased polymorphism in these species may be an adaptation to a freshwater environment with higher potential microbial contamination and pathogen burden. A similar example of a high level of MHC class II polymorphism is also seen in another freshwater species, the Amazon River Dolphin (Inia geofrensis) [36].

16.5.1 The MHC Class I region MHC class I antigens are expressed widely on nucleated cells where their role is to present endogenous antigens, including viral antigens, to cytotoxic T cells (Fig. 16.5). MHC class I molecules are expressed on all lymphocytes. Cetaceans have seven MHC class I genes as in T. truncatus but only three in orcas, sperm whales, North Atlantic right whales, and blue whales [32]. Presumably, this low number reflects a shift to a mid-ocean marine environment with a lower pathogen selection pressure on MHC diversity.

16.5.2 The MHC Class II region Among the dolphin class II genes, nine loci have been identified. They include the classical clusters DQA, DQB, DRA, DRB, and DPB, as well as the nonclassical clusters DMA, DMB, DOA, and DOB. (DPA is completely missing while DPB has only a few traces in some but not all species). Both types of MHC genes encode alpha-beta protein heterodimers. Each MHC cluster contains different numbers of alpha and beta genes, and this number varies between species. Most of these genes are functional but others have accumulated mutations and thus become nonfunctional pseudogenes. Sequences from eight of these dolphin genes include intact coding regions and are presumably functional. Their overall pattern conforms to the usual mammalian pattern although their conformation, orientation, and the number of alpha and beta genes vary between species. The genomic organization of the class II region in cetaceans resembles the bovine since they all share the cetartiodactyl inversion that separates the class II genes into two subregions. Class IIa contains the DR and DQ genes while class IIb contains the non-classic genes and a DRB pseudogene [37]. The function of MHC class II molecules is to present processed exogenous antigens including extracellular bacteria and viruses to CD41 helper T cells and so trigger an adaptive immune response. MHC II antigens are expressed on T cells of beluga whales, harbor porpoises, and bottlenose dolphins. The second exon of MHC class II genes, especially DQA and DQB, encodes the antigen-binding domain and thus has a very high degree of polymorphism [4]. There is also evidence of duplication in the DQA promoter region, and this too may reflect a response to a pathogen threat. The nonclassical MIC proteins act as molecular chaperones that promote peptide binding to the classical class II molecules. DR DO TAP DM

IIb

DQ DR

IIa

III

I

FIGURE 16.5 The organization of the major histocompatibility complex of the bottlenose dolphin, Turciops truncatus.

244

SECTION | 2 Mammalian orders

16.5.3 DR loci Cetaceans have a single functional DRA gene. Cetaceans and cattle share a one codon deletion in the first exon of their DRA genes. Cetaceans have up to three functional DRB loci in class IIa and a presumed DRB pseudogene located next to DOB in the class IIb region. The multilocus DRB (B1
B4) contains two to four putative gene copies depending on the species. It is believed to have duplicated during the evolution of cetaceans. Diversity is driven by polymorphism at one or two specific loci. Twenty-three DRB sequences have been identified in dolphins. The cetacean DRB pseudogene is located in the same position as DSB in cattle. The position and orientation of the class IIb pseudogenes are compatible with that of DRB in the non-inverted class II region of other mammals.

16.5.4 DQ loci Most cetaceans have only a single copy of both the DQA and DQB genes. No DQ loci have been found in the minke whale genome. There is a single DQB locus in toothed whales, but duplicate DQB genes have been reported in baleen whales and the Yangtse finless porpoise (Neophocaena phocaenoides). This species also shows locus reduction in DRB, DQA, DQB, and a loss of DY [38]. Cetacean DQA is orthologous to BoLA-DQA2

16.5.5 Other MHC class II loci Like cattle, whales have no DP sequences. The only trace of their presence is a DPB pseudogene found in the minke whale class IIb region [37]. Likewise, no DY loci have been found in any cetacean. Interestingly they all possess a DMA gene except for the sperm whale. This species has also lost one of the two DRB genes and its DQ genes are inverted. There is one DQB and two DRB loci in T. truncatus. In their products, the antigen-binding groove shows multiple nonsynonymous mutations suggesting that they have been under significant positive selective pressure. The two nonclassical clusters, DO and DM have only a single copy except for a missing first exon of DOB in the bottlenose dolphin. Yang et al. used cDNA sequences to estimate MHC class II divergence times in cetaceans and artiodactyls. Surprisingly they found that they formed two clades. One clade consisted of whales and dolphins. The other clade included artiodactyls such as the pig, hippo, and ruminants. It is assumed that artiodactyl DQB and DRB are paralogous rather than orthologous and compatible with the birth and death model of MHC evolution [5].

16.5.6 The natural killer receptor complex A single study has been conducted showing that belugas (D. leucas) possess functional NK cells, but their receptor system has yet to be described [39].

16.6

B cells and immunoglobulins

The B cells in the beluga account for about 3%
12% of the blood lymphocytes. In bottlenose dolphins, they constitute 10%
15%. Mean orca serum IgG concentration ranges from 15.04 6 3.97 g/L for animals aged 0
5 years and 26.65 6 9.8 g/L for animals aged .10 years [40]. This figure is somewhat higher than the level of IgG in bottlenose dolphins (3.2- . 11.49 g/L). However free-ranging dolphins have higher levels than captive dolphins. This possibly reflects a greater parasite burden [41,42].

16.6.1 IGH genes The immunoglobulin heavy chain genes of the cetaceans cluster with the terrestrial artiodactyls and they possess an antibody repertoire similar to other land mammals. Thus all cetaceans examined have at least one gene for each of the immunoglobulin classes found in land mammals. Likewise, some have significant duplication of their IGHG genes (Fig. 16.6).

16.6.2 IgM The bottlenose dolphin has a single copy of the IGHM gene. Its sequence is very similar to the IGHM genes in the cow, sheep, and pig as expected.

The cetaceans: whales and dolphins Chapter | 16

IGH

5’

47

3’

?D

2

M

IGK

4

D

G1

G2

E

A1

245

FIGURE 16.6 The immunoglobulin genes of the bottlenose dolphin, Turciops truncatus.

A2

?J

37

IGL

3J? 3C

M

D

G

E

A

V

D

J

LC

16.6.3 IgA Bottlenose dolphins possess IgA molecules that are structurally similar to those found in other mammals. Thus they have three constant chain domains and a hinge region located between CH1 and CH2. Its sequence is very similar to that of the terrestrial artiodactyls. The dolphin genome appears to contain multiple IGHA sequences suggesting the presence of multiple pseudogenes and possibly more than one IgA isotype [43].

16.6.4 IgD All cetacean IGH loci contain three IGHD exons. As in Artiodactyla, the predicted CH1 domain of IgD is nearly identical to the CH1 of IgM while the CH2 and CH3 domains are IgD-specific. No IGHD gene has yet been identified in the sperm whale.

16.6.5 IgG There are usually two to four IGHG constant genes present in the cetacean IGH locus [44]. Two IGHG genes are present in most cetacean species examined (T. truncatus, O. orca, L. vexillifer, P. catodon), but four are present in the minke whale (B. acutorostrata). In general, the cetaceans retain the typical IGH gene order namely: 50 -M-D-G-E-A-30 . An exception is found in the minke whale in which there is a single IGHG gene in the standard position and three additional IGHG genes downstream of IGHA. In the sperm whale (P. macrocephalus) there are also additional IGHG exons present, but these appear to be pseudogenes. Cetacean IgG heavy chains each consist of three constant domains plus a hinge segment and their genes closely resemble the IGHG in the other Artiodactyla. They also tend to cluster within a species rather than between species, reflecting their evolution by gene duplication. However, there is some species clustering between bottlenose dolphins and orca
species that likely diverged within the last three million years [44]. There is evidence of positive selection in the IGHM and IGHA but not in the IGHG constant region genes. Further studies of IgG subclasses from whales have indicated that orcas (O. orca), fin whales (B. physalus), and minke whales (B. acutorostrata) possess both IgG1 and IgG2 [45]. There are minor differences in the sizes of these IgG molecules between the baleen whales and the orca. There are also differences in their sensitivity to digestion by papain or pepsin. Analysis of the T. turciops IGHG genes shows that they possess two isotypes, IgG1 and IgG2 [46]. Their sequences closely resemble those in other artiodactyl species. Indeed, the phylogenetic tree of the artiodactyl IgG genes is identical to that of the species themselves [46]. The hinge region in both is encoded by a single exon. The hinge region of the membrane-bound forms of these IgGs is also similar to terrestrial mammals. Likewise, they have a conserved antigen receptor transmembrane (CART) motif in their transmembrane domains. The CART motif is an amino acid sequence found in the transmembrane domain of the antigen receptor. It is required for signaling to the B cell [47]. Investigation of the dolphin transmembrane domain has indicated that it differs from other species in that a highly

246

SECTION | 2 Mammalian orders

conserved Ser residue is replaced by a Gly in the CART motif. The effect of this on B cell signal transduction is unclear [46]. The membrane-bound forms of IgG are closely linked to the paired CD79 signal transduction molecules that together form the B cell antigen receptor. These molecules interact through the CART motif.

16.6.6 IGV genes There are great variations in the numbers of IGHV genes among the cetaceans. For example, at the extremes, the bottlenose dolphin has 47 IGHV genes while the white-beaked dolphin (Lagenorhynchos obliquidens) has only five. Orcas have 26, but minke whales have only eight. In contrast to the pig and cow that only use IGHV genes from a single family, the dolphin uses genes from two [48]. Analysis of the dolphin heavy chain V framework sequences indicates that most belong to clan III, none belong to clan I, and some may belong to clan II. However there are also many IGHV sequences that cannot be assigned to any of the three clans. Further studies indicate that these unassigned sequences containing many prolines and cysteines are generally aligned with the long unassigned CDR3 sequences in the bovine [49]. These form the stalk and knob structures in some bovine variable domains (Chapter 17). Some CDR3 regions of the dolphin V domains are also unusually long. They may contain up to 21 amino acids. These also contain multiple Cys residues that presumably stabilize the CDR by forming intrachain disulfide bonds. This is a feature also seen in camelid VHH domains [48] (Chapter 14).

16.6.7 Light chains All cetacean species examined have a single constant gene encoding the kappa light chain, IGKC but the number of constant genes in the lambda chain locus varies from one to three [50]. There are also great variations in the light chain V genes. Thus IGKV numbers are relatively low and range from seven in the minke whale to one in the narwhal (Monodon monoceros). Conversely, the number of IGLV genes is very much higher ranging from ten in the Vaquita (Phocoena sinus) to 37 in the bottlenose dolphin [50]. Three IGKV clans and five IGLV clans have been described.

16.7

T cells and cell-mediated immunity

16.7.1 Pressure adaptation Cetaceans have the ability to dive deeply and are therefore subjected to greatly increased water pressure. Several immune functions mediated by macrophages and lymphocytes are affected by these changes in pressure including both antigen-processing and immunoglobulin production. Cell cycle progression is also inhibited by increased pressures. High pressure has the potential to cause protein denaturation [51]. For example, the increased pressure associated with diving in belugas reduces phagocytosis by granulocytes. Increased pressure also has the potential to alter lymphocyte functionality. Thus it has been shown to have an effect on peripheral blood mononuclear cells derived from belugas [52]. These studies have demonstrated that an increased pressure of up to 1000 G on the beluga cells resulted in increased IL-2 production possibly as a result of T cell activation. (Human cells showed no alteration in IL-2 production when subjected to these same pressures). However, pressure effects on mononuclear cell proliferation differed depending upon whether the belugas were captive or free living. (It decreased for aquarium animals but increased for the freeliving belugas.) These differences are likely a result of stress effects or the fact that the free-living belugas had more recent experience in deep diving [52]. All the cetaceans that have been investigated to date, appear to be “γ/δ-high” species. For example, 31% of beluga T cells express γ/δ antigen receptors [53]. In addition, 30% of their blood T cells are CD41. Cetaceans possess the four key T cell antigen receptor peptide chains, α, β, δ, and γ encoded by TRA, TRB, TRD, and TRG gene loci respectively (Fig. 16.7).

16.7.1.1 TRA/D The T. truncatus genome assembly has confirmed that their TRD locus is embedded within the TRA locus. As in other cetaceans, its conserved structure resembles that in artiodactyls. Gene segments belonging to the TRDV1 subfamily are distributed throughout the TRAV gene region. Relative to primates, cetaceans show a reduction in their numbers of TRAV genes. Primates have 35 while cetaceans have 15. T. truncatus has 16 TRAV and five TRDV genes. However, it also has 61 TRAJ genes [54]. The numbers of these genes in cattle are much higher and more similar to the diverse primate clades. Thus the loss of TRAV genes appears to be cetacean-specific. There is a bias in V-J gene usage in T. truncatus. The number of TRA/DV genes in other cetacean species ranges from 18 in the sperm whale to seven in the orca.

The cetaceans: whales and dolphins Chapter | 16

TRA/D

5'

3 5 2 DV DD DJ DC DV

16 AV

TRB

22

TRG

2

6

61 AJ

247

3'

AC

7

7

3

V

D

J

C

FIGURE 16.7 The T cell antigen receptor genes of the bottlenose dolphin, Turciops truncatus. The arrows denote gene orientation.

16.7.1.2 TRB The number of TRBV genes ranges from twelve in the minke whale to seven in the long-finned pilot whale (Globicephala melas) to 23 in T. truncatus. These TRBV genes are grouped into 20 distinct subfamilies [55]. The organization of the TRB locus is similar to that seen in other artiodactyls with three tandem D-J-C clusters located at its 30 end. It is however unusually short [55]. Thus its length of 276 kb is less than human TRB (620 kb) and much smaller than the other artiodactyls (goat 558 kb, pig 407 kb, dromedary 302 kb). It completely lacks duplications and has many deletions that significantly reduce the number of available V genes. Each of the dolphin 20 TRBV subfamilies forms a monophyletic group suggesting that these subfamilies developed prior to the cetacean divergence. TRBV genes also possess an evolutionary pattern similar to the TRAV genes described above. Thus while primates have 25 subfamilies, cetaceans have only 15. Two subfamilies within clade II are not found in primates and one is not found in the bovine. As with TRAV, this loss of V gene diversity is not seen in the bovine indicating that it occurred after these species diverged. One gene, TRBV30 was expressed in 66% of cDNA clones from different dolphins indicating preferential usage.

16.7.1.3 TRG As in other artiodactyls, there are relatively few TRGV genes in cetaceans. The dolphin TRG locus is the simplest and smallest among the mammalian TRG loci identified to date [54]. It spans only 48 kb. It consists of a single V-J-C cassette containing two TRGV genes belonging to two subfamilies, three TRGJ genes, and a single TRGC gene. The two V genes and three J genes are used in every possible combination. The single TRGC gene has a small exon 2 similar to that in the sheep. In effect, the dolphin TRG locus closely resembles the structure of the TRGC5 cassette, the ancestral unit that gave rise to the repeated iterations in the ruminant TRG locus (Chapter 17) [55]. In other mammals, these V genes belong to three main clades I, II, and III. However, while primates and bovines express all three, cetaceans lack clade III [50].

References [1] Gatesy J, Geisler JH, Chang J, Buell C, et al. A phylogenetic blueprint for a modern whale. Mol Phylo Evol 2013;66:479
506. [2] Ursing BM, Arnason U. Analysis of mitochondrial genomes strongly support a hippopotamus-whale clade. Proc Roy Soc Lond 1998; 265:2251
5. [3] Di Guardo G, Mazzariol S. Cetacean morbilliviruses: a land-to-sea journey and back? Virol Sin 2019;34:240
2. [4] Paga´n HJT, Ferrer T, O’Corry-Crowe G. Positive selection in coding regions and motif duplication in regulatory regions of bottlenose dolphin MHC class II genes. PLoS One 2018;. Available from: https://doi.org/10.1371/journal.pone.0203450. [5] Yang W-C, Hu J-M, Chou L-S. Phylogenetic analysis of MHC class II genes in bottlenose dolphins and their terrestrial relatives reveal pathogen-driven directional selection. Zoo Stud 2010;49(1):132
51. [6] Hassanin A, Delsuc F, Ropiquet A, Hammer C, et al. Pattern and timing of diversification of Cetartiodactyla (Mammalia Laurasiatheria) as revealed by a comprehensive analysis of mitochondrial genomes. C R Biol 2012;335:32
50.

248

SECTION | 2 Mammalian orders

[7] De Silva VMF, Carter AM, Ambrosio CE, Carvalho AF, et al. Placentation of dolphins from the Amazon River Basin. The Boto Inia geoffrensis, and the Tucuxi Sotalia fluviatilis. Reprod Biol Endocrinol 2007. Available from: https://doi.org/10.1186/1477-7827-5-26. [8] Kitayama C, Sasaki M, Ishikawa H, Mogoe T, et al. Structure and functions of the placenta in common minke (Balaenoptera acutorostrata), Bryde’s (B. brydei) and sei (B. borealis) whales. J Reprod Dev 2015;61:415
21. [9] Oftedal OT. Lactation in whales and dolphins: evidence of divergence between baleen- and toothed-species. J Mammary Gland Biol Neoplasia 1997;2(3):205
30. [10] Hawkey CM. Comparative mammalian hematology. Cellular components and blood coagulation of captive wild animals. London: Heinemann Medical Books; 1975. [11] Bossart GD, Romano TA, Peden-Adams MM, Schaefer AM, et al. Comparative innate and adaptive immune responses in Atlantic bottlenose dolphins (Turciops tunicatus) with viral, bacterial and fungal infections. Front Immunol 2019. Available from: https://doi.org/10.3389/ fimmu.2019.01125. [12] Nouri-Shirazi M, Bible BF, Zeng M, Tamjidi S, Bossart GD. Phenotyping and comparing the immune cell populations of free-ranging Atlantic bottlenose dolphins (Tursiops truncatus) and dolphins under human care. BMC Vet Res 2017. Available from: https://doi.org/10.1186/s12917017-0998-3. [13] Shen T, Xu S, Wang X, Yu W, et al. Adaptive evolution and functional constraint of TLR4 during the secondary aquatic adaptation and diversification of cetaceans. BMC Evol Biol 2012. Available from: https://doi.org/10.1186/1471-2148-12-39. [14] Villagra-Blanco R, Silva LMR, Aguilella-Segura A, Arcenillas-Herna´ndez J, et al. Bottlenose dolphins (Tursiops truncatus) do also cast neutrophil extracellular traps against the apicomplexan parasite Neospora caninum. Int J Parasit Para Wildl 2017;6:287
94. [15] Romano TA, Felten SY, Olschowka JA, Felten DL. A microscopic investigation of the lymphoid organs of the Beluga, Delphinapterus leucas. J Morphol 1993;215:261
87. [16] Gelain ME, Bonsemblante F. Acute phase proteins in marine mammals: state of art, Perspectives and challenges. Front Immunol 2019. Available from: https://doi.org/10.3389/fimmu.2019.01220. [17] Braun BA, Marcovitz A, Camp JG, Jia R, Bejerano G. Mx1 and Mx2 key antiviral proteins are surprisingly lost in toothed whales. Proc Natl Acad Sci U S A 2015;. Available from: https://doi.org/10.1073/pnas.1501844112. [18] Silva FMO, Guimara˜es JP, Rodrigues MN, Vergara-Parente JE, et al. Microscopic study of the thymus of Guiana dolphin and humpback whale. Microscopy: Adv Sci Res Educ (Mendez-Vilas Ed) 2014. Available from: https://doi.org/10.13140/2.1.2513.9205. [19] Cowan DF. Involution and cystic transformation of the thymus in the bottlenose dolphin Tursiops truncatus. Vet Pathol 1994;31:648
53. [20] Cowan DF, Smith TL. Morphology of the lymphoid organs of the bottlenose dolphin, Tursiops truncatus. J Anat 1999;194:505
17. [21] Nakamine H, Nagata S, Yonezawa M, Tanaka Y. The whale (Odontoceti) spleen: a type of primitive mammalian spleen. Kaibogaku Zasshi 1992;67(2):69
82. [22] Silva FMO, Carvalho VL, Guimara˜es JP, et al. Accessory spleens in cetaceans and its relevance as a secondary lymphoid organ. Zoomorphology 2014;133(3):343
50. [23] Vukovik S, Lucic H, Gomercic H, Goercic MD, et al. Morphology of the lymph nodes in bottlenose dolphin (Turciops truncatus) and striped dolphin (Stenella coeruleoalba) from the Adriatic sea. Acta Vet Hungarica 2005;53(1):1
11. [24] Smith TL, Turnbull BS, Cowan DF. Morphology of the complex laryngeal gland in the Atlantic bottlenose dolphin, Tursiops truncatus. Anat Rec 1999;254:98
106. [25] Hecker N, Sharma V, Hiller M. Convergent gene losses illuminate metabolic and physiological changes in herbivores and carnivores. Proc Natl Acad Sci U S A 2019;116(8):3036
41. [26] Cowan DF, Smith TL. Morphology of complex lymphoepithelial organs of the anal canal (“anal tonsil”) in the Bottlenose dolphin, Tursiops truncatus. J Morphol 1995;223:263
8. [27] Yamasaki F, Komatsu S, Kamiya T. A comparative morphology of anal tonsils in Platanistidae. Sci Rep Whales Res Inst 1977;29:95
100. [28] Andre´ani M, Freitas L, Ramos EKS, Nery MF. Longitudinal diversity gradient and cetaceans from the perspective of MHC genes. Immunogenetics 2020;72:393
8. [29] Yang W-C, Chou L-S, Hu J-M. Sequence analysis of MHC class II genes in Cetaceans. Shanghai: InTech; 2012. ISBN: 978-953-51-0589-3. [30] Zhang X, Lin W, Zhou R, Gui D, et al. Low major histocompatibility complex class II variation in the endangered Indo-Pacific humpback dolphin (Sousa chinensis): inferences about the role of balancing selection. J Heredity 2016. Available from: https://doi.org/10.1093/hered/esv138. [31] Zhang Z, Sun X, Chen M, Ll L, et al. Genomic organization and phylogeny of MHC class II loci in cetaceans. J Heredity 2019;332
9. Available from: https://doi.org/10.1093/hered/esz005. [32] Gillett RM, Murray BW, White BN. Characterization of class I
and class II-like major histocompatibility complex loci in pedigrees of North Atlantic right whales. J Heredity 2014;105(2):188
202. [33] Trowsdale J, Groves V, Arnason A. Limited MHC polymorphism in whales. Immunogenetics 1989;29:19
24. [34] Moreno-Santillan DD, Lacey EA, Gendron D, Ortega J. Genetic variation at exon 2 of the MHC class II DQB locus in blue whale (Balaenoptera musculus) from the Gulf of California. PLoS One 2015. Available from: https://doi.org/10.1371/journal.pone.0141296. [35] Yang G, Yan J, Zhou K, Wei F. Sequence variation and gene duplication at MHC DQB loci of Baiji (Lipotes vexillifer), a Chinese river dolphin. J Heredity 2005;96(4):310
17. [36] Martı´nez-Agu¨ero M, Flores-Ramı´rez S, Ruiz-Garcı´a M. Amazon River dolphin polymorphism and population differentiation of MHC class II peptides. In: Ruiz-Garcia M, Shostell J, editors. Biology, evolution, and conservation of river dolphins. New York: Nova Science Publishers; 2010.

The cetaceans: whales and dolphins Chapter | 16

249

[37] Sa´ ALA, Breaux B, Buriamqui T, Deiss TC, et al. The marine mammal class II major histocompatibility complex organization. Front Immunol 2019. Available from: https://doi.org/10.3389/fimmu.2019.00696. [38] Ruan R, Ruan J, Wan X-L, Zheng Y, et al. Organization and characteristics of the Yangtze finless porpoise (Neophocaena asiaorientalis). Sci Rep 2016. Available from: https://doi.org/10.1038/srep22471. [39] De Guise S, Ross PS, Osterhaus ADME, Martineau D, et al. Immune functions in beluga whales (Delphinapterus leucas): evaluation of natural killer cell activity. Vet Immunol Immunopathol 1997;58:345
54. [40] Taylor BC, Brotheridge RM, Jessup DA, Stott JL. Measurement of serum immunoglobulin concentration in killer whales and sea otters by radial immunodiffusion. Vet Immunol Immunopathol 2002;89:187
95. [41] Le-Bert CR, Nollens H, Venn-Watson S, Green LG. Baseline circulating immunoglobulin G levels in managed and free ranging bottlenose dolphins (Tursiops truncatus). Dev Comp Immunol 2009;33(4):449
55. [42] Ruiz CL, Nollens HH, Venn-Watson S, Green LG, et al. Baseline circulating immunoglobulin G levels in managed collection and free-ranging bottlenose dolphins (Tursiops truncatus). Dev Comp Immunol 2009;33:449
55. [43] Mancia A, Romano TA, Gefroh H, Chapman RW, et al. Characterization of the immunoglobulin A heavy chain gene of the Atlantic bottlenose dolphin (Tursiops truncatus). Vet Immunol Immunopathol 2007;118:304
9. [44] Li L, Rong X, Li G, Wang Y, et al. Genomic organization and adaptive evolution of IGHG genes in marine mammals. Mol Immunol 2018;99:75
81. [45] Andre´sdo´ttir V, Magnado´ttir B, Andre´sson OS, Pe´tursson G. Subclasses of IgG from whales. Dev Comp Immunol 1987;11:801
6. [46] Mancia A, Romano TA, Gefroh H. Chapman Rw, et al. The immunoglobulin G heavy chain (IGHG) genes of the Atlantic bottlenose dolphin, Turciops truncatus. Comp Biochem Phys B 2006;144:38
46. [47] Campbell KC, Backstro¨m T, Tiefenhaler G, Palmer E. CART: a conserved antigen receptor transmembrane motif. Sem Immunol 1994;6:393
410. [48] Lundqvist ML, Kohkberg KE, Gefroh HA, Arnaud P. Cloning of the IgM heavy chain of the bottlenose dolphin (Tursiops truncatus) and initial analysis of VH gene usage. Dev Comp Immunol 2002;26:551
62. [49] Deiss TC, Vadnais M, Wang F, Chen PL, et al. Immunogenetic factors driving formation of ultralong VH CDR3 in Bos taurus antibodies. Cell Mol Immunol 2019;16:53
64. [50] Gambo´n-Deza F. Immunoglobulins, MHC, and T cell receptor genes in Cetaceans. bioRxiv. https://doi.org/10.1101/2020.10.24.353342. [51] Hallenbeck JM, Anderson JC. Pathogenesis of the decompression disorders. In: Bennet PB, Elliot DH, editors. The physiology and medicine of diving. San Pedro Publishing Co; 1982. p. 435
60. [52] Thompson LA, Romano TA. Pressure induced changes in adaptive immune function in Belugas (Delphinapterus leucas); Implications for dive physiology and health. Front Physiol 2016. Available from: https://doi.org/10.3389/fphys.2016.00442. [53] Beineke A, Siebert U, Wohlsein, Baumga¨rtner W. Immunology of whales and dolphins. Vet Immunol Immunopathol 2010;133:81
94. [54] Linguiti G, Antonacci R, Tasco G, Grande F, et al. Genomic and expression analyses of Tursiops truncatus T cell receptor gamma (TRG) and alpha/delta (TRA/TRD) loci reveal a similar basic public γδ repertoire in dolphin and human. BMC Genomics 2016. Available from: https://doi. org/10.1186/s12864-016-2841-9. [55] Linguiti G, Kossida S, Pierri CL, Jabado-Michaloud J, et al. The T cell receptor (TRB) locus in Tursiops truncatus: from sequence to structure of the alpha/beta heterodimer in the human/dolphin comparison. Genes 2021. Available from: https://doi.org/10.3390/genes12040571.

This page intentionally left blank

Chapter 17

Ruminants: cattle, sheep, and goats

Hereford Cattle. Bos taurus taurus.

The emergence of the artiodactyls about 65 mya was eventually followed by their diversification into multiple suborders. The Tylopods probably diverged first. This was followed by the divergence of the Cetaceans and the Suiformes. It is estimated that the Ruminantia diverged in the late Eocene 30
40 mya [1]. The main branch of the Ruminantia, the Pecora, however, developed a multichambered stomach that enabled them to vastly increase their digestive capabilities and exploit new food supplies, especially cellulose-based plant material. At present, there are six extant families in the suborder Ruminantia. The Tragulidae
the mouse deer; Antilocapridae
the pronghorn; Giraffidae
the giraffes and okapi; Moschidae
the musk deer, and the two largest families, the Cervidae
the true deer and the Bovidae (Fig. 17.1). The Bovidae are the largest extant artiodactyl family. They contain at least 143 species including antelopes, buffalo, wildebeest, hartebeest, sheep, goats, bison, yak, and domestic cattle. The bovids characteristically possess nonbranching horns with a keratinous sheath growing out of their frontal bones. During the late Miocene epoch, 11.6
5.3 mya, there was a progressive change in the world’s climate. Instead of the previous hot, moist conditions, there was a trend towards decreased rainfall. This was a result of global cooling and the decreased ability of cold air to hold moisture. As a result, forests declined, while open woodlands and grasslands expanded. Between 8 and 5 mya, there was a major expansion of these grasslands and the herbivores evolved with them. The grasses adapted too. The grasses that thrived could withstand drought and assimilate CO2 more efficiently than their predecessors. (CO2 levels dropped during the late Miocene.) Grasses that preferentially used a C4 photosynthesis pathway rather than the previous C3 pathway could thrive in a climate with lower levels of CO2 and higher levels of sunlight [2]. This expansion of grasslands was associated with the coevolution and expansion of some grazing herbivores. Not all grazers could eat and digest the newly abundant, tough, silica-rich grasses and many large herbivores went extinct. However, horses prospered, and so did ruminants [3]. The ancestral carnivores also adapted to these changes and followed the herbivores onto the grasslands. The herbivores in response evolved greater speed and agility. Eating tough herbage requires tough teeth such as molars that can withstand constant grinding; it requires the presence of an intestinal microbiota that produces cellulases that can break down the plant fibers. It requires the evolution Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00012-5 © 2023 Elsevier Inc. All rights reserved.

251

252

SECTION | 2 Mammalian orders

30

25

20

15

10

5

0 Million years ago Pronghorn Giraffe Deer Buffalo Cattle Kudu Gazelles Wildebeeste Sheep Goats

FIGURE 17.1 The phylogeny of the Ruminants.

of large fermentation chambers such as the rumen or cecum where microbial digestion can take place and the cellulose is broken down. The products of this fermentation, the volatile fatty acids, provide much of the animal’s calorific needs. In addition, bacterial proteins provide much of their protein needs. Rumens are much more effective at this than ceca since they release these nutrients before their absorption in the small intestine. As a result of this increased efficiency, ruminants have been more successful than hindgut fermenters such as the perissodactyls. Their most recent expansion occurred around 2.0
1.8 mya. A time that coincides with the emergence of humans [4]. However, they also suffered major declines about 100,000
50,000 years ago likely as a result of human activities [1]. There are at least eight species included in the genus Bos. In addition to domestic cattle, B. taurus, and its two subspecies B. taurus taurus and B. taurus indicus, these include, Banteng, Gaur, Gayal, Yak, Wisent (European bison), American bison, and Kouprey as well as the extinct Aurochs, B. primigenius. The two subspecies of cattle were domesticated from wild aurochs populations on at least two different occasions during the Neolithic period. About 10,000 years ago B. t. taurus was domesticated in the Middle East. About 1500 years later B. t. indicus was domesticated in the Indus river valley. Subsequent selection for desirable traits has resulted in the development of hundreds of distinct breeds.

17.1

Reproduction and lactation

Ruminants have cotyledonary placentas. Thus instead of having a single large area of contact between the fetal membranes and the uterus, these mammals have, in effect numerous small placentomes. The placental-maternal interface is epitheliochorial. As a result, the healthy ruminant placenta forms an impenetrable barrier to the transfer of maternal immunoglobulins. Consequently, newborn ruminants are agammaglobulinemic. Although the gestation period of the cow is 280 days, the fetal calf thymus is recognizable by 40 days postconception. The fetal bone marrow and spleen appear at 55 days. Lymph nodes are found at 60 days, but Peyer’s patches do not appear until 175 days. Blood lymphocytes are seen in fetal calves by day 45, IgM1 B cells by day 59, and IgG1 B cells by day 135. The time of appearance of serum antibodies depends on the sensitivity of the techniques used. It is therefore no accident that the earliest detectable immune responses are those directed against viruses, using highly sensitive virus neutralization tests. Fetal calves have been reported to respond to rotavirus at 73 days, parvovirus at 93 days, and parainfluenza 3 virus at 120 days. Fetal blood lymphocytes can respond to mitogens between 75 and 80 days, but this ability is temporarily lost near the time of birth as a result of maternal steroid production. T cell subpopulations are present in calves at levels comparable to adults, but B cell numbers increase significantly during the first 6 months after birth. Calves acquire innate and IgM-mediated immune competence within the first week of life [5].

Ruminants: cattle, sheep, and goats Chapter | 17

253

17.1.1 Secretion and composition of colostrum and milk Cattle do not transfer immunoglobulins to their young in utero. Colostrum contains the accumulated secretions of the mammary gland over the last few weeks of pregnancy together with proteins actively transferred from the bloodstream under the influence of estrogens and progesterone. Neonatal immunoglobulin receptors (FcRn) are expressed in mammary gland ductal and acinar cells and mediate the transfer of IgG from serum into colostrum [6]. The predominant immunoglobulin in the colostrum of ruminants is IgG1, which may account for 65%
90% of its total antibody content. IgG2 is present but in very much smaller quantities. IgA and IgM are usually minor but significant components, 2 7%
10% each. All of the IgG, most of the IgM, and about half of the IgA in bovine colostrum are actively transferred from the bloodstream. In milk, in contrast, only 30% of the IgG and 10% of the IgA are so derived; the rest is produced by plasma cells within the mammary gland stroma. Colostrum also contains secretory component both in the free form and bound to IgA. Colostrum contains many cytokines. For example, bovine colostrum contains significant amounts of IL-1β, IL-6, TNF-α, and IFN-γ. These cytokines may promote the development of the immune system in the young calf [7].

17.1.2 Absorption of colostrum Calves ingest colostrum when they first suckle immediately after birth. Naturally suckled calves ingest an average of two liters of colostrum, although some individuals may ingest as much as six liters. In neonatal calves, protease activity in the digestive tract is low and is further reduced by trypsin inhibitors in colostrum. Therefore, colostral immunoglobulins are not degraded but reach the small intestine intact. These colostral immunoglobulins are subsequently endocytosed by enterocytes and bind to FcRn within their endosomes. Once bound to endosomal FcRn, the molecules are transported across the enterocytes and transferred to the lacteals and possibly the intestinal capillaries [7]. Eventually, the absorbed immunoglobulin reaches the bloodstream, and newborn calves obtain a massive transfusion of maternal immunoglobulins. Young calves have large amounts of free secretory component in their intestine. Colostral IgA and, to a lesser extent, IgM can bind this secretory component, which may then prevent their absorption. The duration of intestinal permeability also varies among immunoglobulin classes. In cattle, permeability is highest immediately after birth and declines after about six hours because of the replacement of FcRn-bearing enterocytes by cells that do not express this receptor. As a rule, absorption of all immunoglobulin classes drops to a very low level after about 24 hours. Feeding colostrum tends to hasten this closure, whereas a delay in feeding results in a slight delay in closure (up to 33 hours). The presence of the mother may be associated with increased immunoglobulin absorption. Thus calves fed measured amounts of colostrum in the presence of their mother will absorb more immunoglobulins than calves fed the same amount in her absence. In laboratory studies in which measured amounts of colostrum are fed to calves, there is a great variation (25%
35%) in the amount of immunoglobulins absorbed [5]. Unsuckled calves normally have very low levels of immunoglobulins in their serum. The successful intake of colostral immunoglobulins immediately supplies them with serum IgG at a level approaching that found in adults. Peak serum immunoglobulin levels are normally reached between 12 and 24 hours after birth. After absorption ceases, these passively acquired antibodies decline through normal metabolic processes. The rate of decline differs among immunoglobulin classes, and the time taken to decline to non-protective levels depends on their initial concentration. In calves, the serum half-life of colostral-derived IgG1, G2, and G4 is about 21 days while IgG3 has a half-life of 7 days [8]. As intestinal absorption is taking place, simultaneous proteinuria may also occur. This is due to intestinal absorption of very small proteins such as β-lactoglobulin that can be excreted in the urine. In addition, the glomeruli of newborn mammals are permeable so that the urine of neonatal ruminants contains intact immunoglobulin molecules. This proteinuria ceases with the termination of intestinal absorption. The secretions of the mammary gland gradually change from colostrum to milk. Ruminant milk is rich in IgG1 and IgA (Fig. 3.4). For the first few weeks in life, while protease activity is low, these immunoglobulins can be found throughout the intestine and in the feces of young mammals. As the digestive ability of the intestine increases, eventually only secretory IgA molecules remain intact. The IgG transferred through a cow’s colostrum represents the results of her history of antigen exposure, B cell responses, and somatic mutation. This maternal IgG in effect represents the immunological experiences of the mother. Maternal antibodies act on the immune system of the newborn during a critical imprinting period and exert a lifelong influence on the calf’s immune development [9]. This influence may be stronger than some genetic predispositions!

254

SECTION | 2 Mammalian orders

Thus maternal antibodies can enhance the newborn immune responses to some antigens and suppress their responses to others. They may also influence Th1/Th2 polarization and the subsequent development of immunologic diseases. Memory, CD41, CD81, and γ/δ T cells can be detected in calves before the development of active immunity [10]. Thus calves can mount a T cell response even in the presence of maternal antibodies.

17.1.3 Cell-mediated immunity and colostrum Bovine colostrum is full of lymphocytes, but milk is not. Bovine colostrum contains between 3 3 104 and 1 3 105 lymphocytes/mL, about half of which are T cells. Colostral lymphocytes may survive up to 36 hours in the intestine of newborn calves, and some may penetrate the intestinal epithelium and reach the lacteal ducts or the mesenteric lymph nodes. Cell-containing and cell-free colostrum have been compared for their ability to protect calves against enteropathic E. coli. The calves receiving colostral cells excreted significantly fewer bacteria than those receiving cell-free colostrum. The concentration of IgA- and IgM-specific antibodies against E. coli in the serum of neonatal calves was higher in those that received colostral cells than in those that did not. Those calves that received colostral cells also had greater responses to the mitogen concanavalin A and foreign antigens such as sheep erythrocytes [11]. The mechanisms of this protective effect are unclear [12]. The CD81 T cells in bovine colostrum can produce large quantities of IFN-γ, that may also influence the early development of Th1 responses in neonatal calves [13]. Thus ingestion of maternal colostral cells appears to promote the development of activated lymphocytes. The monocytes from calves that received colostral cells are more capable of processing and presenting antigens than those that did not [14]. Transfer of a cell-mediated immune response by bovine colostral lymphocytes has been demonstrated. Pregnant cows were vaccinated against the bovine virus diarrhea virus (BVDV). Blood lymphocytes from calves that received cell-free colostrum from these cows were unresponsive to BVDV antigens. In contrast, lymphocytes from calves that had received colostrum containing live cells showed enhanced responses to BVDV antigens at one and two days after colostral ingestion. The lymphocytes of calves that received whole colostrum showed enhanced mitogenic responses to maternal and unrelated leukocytes after 24 hours. They also responded to the nonspecific stimulant staphylococcal enterotoxin B. In contrast, the lymphocytes of calves that received acellular colostrum did not [15]. Ingestion of maternal leukocytes immediately after birth stimulates the development of the neonatal immune system.

17.2

Hematology

Cattle, sheep, and goats, generally have blood leukocyte counts in the region of 4
12 3 103 cells/μL (Table 17.1). It is important to note however that these numbers are influenced by age, breed, and pregnancy as well as by infections. The percentage of neutrophils ranges from 13% to 40%: lymphocytes however range from 55% to 80%. Thus lymphocytes far outnumber neutrophils in adult animals, but this may not be the case in very young animals. Circulating bovine natural killer (NK) cells are relatively large and contain cytoplasmic granules. Monocyte percentages range from 3% to

TABLE 17.1 Ruminant white blood cell counts. Bovine

Sheep

Goat

Total leukocytes 3 10 /μL

4
12

4
12

4
13

Neutrophils (%)

28

41

37

Lymphocytes (%)

62

68

64

Monocytes (%)

5

4

3

Eosinophils (%)

1.5

6

4

Basophils (%)

,1

,1

,1

3

Data from Radostits OM, et al. Veterinary medicine. 9th ed. London: WB Saunders; 2000.

Ruminants: cattle, sheep, and goats Chapter | 17

255

11% and eosinophils up to 7%. However, it must be emphasized that eosinophil numbers are very much influenced by the animal’s parasite burden. Recent studies comparing bovine hematology reference intervals over 50 years since 1957 have noted a significant increase in average blood neutrophil counts from 2.4 3 103/μL in 1957 to 4.0 3 103/μL in 2001. There are many possible reasons for this ranging from background infection levels to changes in nutrition, husbandry, or genetics. Conversely, lymphocyte, monocyte, and eosinophil counts have decreased significantly over that time [16].

17.3

Innate immunity

The key cellular components of the innate immune system in cattle are the same as in other artiodactyls. Thus phagocytic cells, neutrophils and macrophages, dendritic cells, and innate lymphoid cells of all three types, as well as NK cells, are present. As described later, cattle are a γ/δ T cell-high species with up to 60% of T cells in calves being γ/δ1. Most of these γ/δ1 T cells also express the WC1 antigen on their surface [17]. It is clear however that despite being developmentally precocious, neonatal calves are deficient in some aspects of innate immunity at birth [18]. They have minimal production of cytokines such as IL-7, IL-23R, and CXCR3 and increased production of the suppressive cytokines, TGF-β1 and TGF-β2. They do not express any of the TH2 polarizing cytokines at birth. Phagocytosis is inefficient, and they have low expression of some complement components such as C1q and C2.

17.3.1 Toll-like receptors Innate immune responses are triggered through diverse pattern recognition receptors, the most important of which are the toll-like receptors (TLR). As in the other artiodactyls, bovids express ten different TLR genes. They have the usual overlapping PAMP recognition profiles, and it is clear that their primary role is to detect bacterial and viral invaders. They are all type I transmembrane proteins with a large extracellular domain. Cattle possess a typical TLR intron/exon structure [19]. Thus bovine TLR1, 2, 3, 4, 6, and 9 contain 5, 2, 5, 3, 4, and 2 exons respectively. TLR5, 7, 8, and 10 consist of single exons only. There is significant variability among cattle TLR genes with the frequency of polymorphism ranging from one SNP per 300 bases in TLR1 to one SNP per 32 bases in TLR3 and overall, averaging about one SNP per 100 bases. Thus they confer a broad diversity on the bovine pathogen recognition process [19]. Both B. t. taurus and B. t. indicus share haplotypes at all the TLR loci [20]. Some bovine TLR polymorphisms are associated with resistance to mastitis and Johne’s disease. The ten bovine TLR genes are found on seven different chromosomes. TLR10, -1, and -6 form a cluster on bovine chromosome 6. TLR7 and -8 are closely associated on the X chromosome together with a TLR pseudogene [19]. The evolution of TLRs generally corresponds to the phylogeny of their hosts, However, some TLR genes, most notably TLR2, show evidence of positive selection during recent ruminant evolution. The natural ligands of TLR10 remain unclear since it is not present in laboratory rodents, however bacterial flagellin is a prime suspect. Cattle also possess the cytosolic PAMP recognition receptors such as RIG-1 and MDA5 and they activate sentinel cells through the NF-κB, IRF-3, and IRF-τ pathways via the mitochondrial antiviral signaling adaptor (MAVS). These in turn activate the inflammasomes that generate the inflammatory cytokines. Cattle are known to possess the NLRP3 inflammasome although it differs in some respects from that found in humans and mice [21]. There are also breed-related differences in the NLRP inflammasome genes that may influence resistance to gastrointestinal parasites in cattle, sheep, and goats. Once activated, their effector cells produce the usual inflammatory cytokines, TNF-α, IL-1β, and IL-6. These attract other cells to the area and result in the cellular and vascular changes of inflammation. They also generate a diverse array of chemokines and eicosanoids. The complete bovine genome has been sequenced and has been found to contain unusually large numbers of genes associated with innate immunity [22]. For example, cattle have ten cathelicidin genes, compared with only one in humans or mice. They have 106 beta-defensin genes, compared with 30
50 in humans and mice [23]. They also have many more type 1 interferon genes than other species. It has been suggested that this duplication and divergence of the genes involved in innate immunity may be a consequence of the huge load of microorganisms present in the rumen and the resulting increased need to prevent microbial invasion. Alternatively, these diverse antimicrobial peptides may be essential since living within dense herds increases the risks of infectious disease transmission between individuals and thus requires a more effective immune system.

256

SECTION | 2 Mammalian orders

Additionally, cattle show substantial differences from other mammals in the genes related to lactation. Many of these lactation-associated genes, such as those for serum amyloid A, β2-microglobulin, and cathelicidins, are also related to innate immunity. Finally, the cattle genome contains ten lysozyme genes, largely expressed in the abomasum and gastrointestinal tract. It is speculated that they may play an important role in killing bacteria entering the intestine from the rumen.

17.3.2 Defensins Beta-defensins are encoded by the ancestral defensin genes from which α-defensins are also derived. α-defensins are not present in cattle. Four gene clusters encode the beta-defensins in bovines, dogs, and rodents [24]. In cattle, clusters of β-defensin genes are located on chromosome 13 as well as on chromosome 27. This has resulted in a great increase in defensin diversity. Consequently, cattle possess at least 57 β-defensin genes; the most found in any mammal so far. (There are 29 in pigs, 38 in dogs, 48 in mice, 33 in chimpanzees, and 48 in humans). These defensins play a major role in combating bacterial infections and possibly in regulating the gut microbiota. The defensins are mainly expressed in the respiratory tract, the mammary gland, small intestine, colon, and reproductive organs.

17.3.3 Complement Cattle have a complete but conventional complement system. However, one unique feature is their production of conglutinin and immunoconglutinins (IK). IKs are physiologic autoantibodies directed against neoepitopes on complement components C2b, C4b, and especially C3b. The neoepitopes that stimulate IK formation are generated by conformational changes when these complement components are activated. The level of IKs in serum reflects the amount of complement activation. This, in turn, is a measure of the antigenic stimulation to which an animal is subjected. IK levels are thus nonspecific indicators of the prevalence of infectious disease within an animal population. Their physiological role is unclear, but they probably enhance complement-mediated opsonization.

17.3.4 Conglutinin Conglutinin is a member of the subfamily of C-type (Ca11-dependent) lectins called collectins. It was first detected in the serum of cattle by Ehrlich and Sachs in 1902 [25]. It is found in cattle and a few other mammals [26]. Conglutinin plays an important role in antimicrobial resistance. For example, it can bind, and agglutinate (conglutinate) complement-coated particles and it can trigger phagocytosis. It can bind the surface polysaccharides on invading bacteria and parasites and so serve both as a complement activator and an opsonin. It has both antibacterial and antiviral properties. In addition to conglutinin, other bovine collectins include mannose-binding protein, surfactant proteins A and D, and bovine collectin 43 (CL-43). Conglutinin is found not only in adult bovine serum but also in colostrum, milk, and fetal calf serum. Conglutinin is an oligomeric protein that can range in size from 34 to 630 kDa. The reduced conglutinin monomer is 40
44 kDa in size. It is a heat-stable beta-globulin consisting of four trimeric subunits. The trimers are stabilized by interchain disulfide bonds. On electron microscopy the conglutinin molecule is X-shaped. Conglutinin is closely related to surfactant protein D (SP-D) and probably arose by duplication of the original SP-D gene. It has a 78% sequence identity with bovine SP-D. Since conglutinin (or a close relative) is found, not only in the Bovidae but also in goats, sheep, camels, pigs, and diverse antelopes, the gene duplication probably occurred early in artiodactyl evolution [27]. Conglutinin is found not only in serum, but it is synthesized by neutrophils and macrophages as well as by hepatocytes. Conglutinin binds to the alpha chain of iC3b in a calcium-dependent manner. It is this binding and subsequent agglutination of complement-coated particles that first drew attention to it. Similar collectins have been identified in horse and human serum [27].

17.3.5 Cytokines 17.3.5.1 Interferons Type I interferons are produced in response to viral infections. The bovine type I IFN locus has undergone significant rearrangement and expansion compared to mice and humans. Its genes are located in two sub-loci separated by .700 kb found on the short arm of chromosome 9 [28]. The IFNW family for example has undergone major expansion resulting in 24 functional genes and eight pseudogenes. There are 13 IFNA genes (the same number as humans). The

Ruminants: cattle, sheep, and goats Chapter | 17

257

IFNB family is also present in multiple copies in bovines. The IFNT family contains three genes and two pseudogenes [28]. IFNK and IFNE are present as single genes. As a result, there are many different type I interferons produced by ruminants including IFN-α, IFN-β, IFN-ω, IFN-κ, IFN-ε, IFN-τ, and IFN-κ. Cattle make all/most of these. IFN-ε for example, is expressed constitutively in the female reproductive tract and may serve a defensive function there. It is present in other artiodactyls including sheep and pigs [29]. Interferon τ is a recently evolved type 1 interferon encoded by the IFNT gene and has only been detected in ruminants [30]. It is secreted by trophoblast and endometrial cells during days 10
24 of pregnancy in the sheep. It plays a key role as a signal for the maternal recognition of pregnancy by maintaining progesterone production in the corpus luteum at that time and thus preventing a return to ovarian cyclicity. IFN-τ has many effects on the endometrium ensuring that the uterus is receptive to the fetus and has anti-inflammatory effects that aid in preventing maternal rejection of her semi-allogeneic fetus. It acts on cultured trophoblast cells to regulate their production of IL-6 and IL-8 [31]. Multiple IFN-τ polymorphisms, and variants exist. Although not detected in humans and mice, they do respond to its effects. IFN-τ binds to the same receptor as IFN-α (IFNAR) and induces intracellular signaling through the JAK/STAT pathway. As a result, it promotes the production of the typical interferon-stimulated genes that encode antiviral proteins as well as the regulatory cytokines IL-4, IL-6, and IL10. Despite this, it is not, however, virally inducible [32]. Interferon-chi is encoded by a newly identified multigene family (IFNX) of four IFN-χ members, one of which appears to be a pseudogene. It is restricted to cattle [29]. Two of the subtypes are functional genes whose products have antiviral and antiproliferative activities. They bind to type I-IFN receptors, induce production of IRF7, and signal through the JAK-STAT pathway. They also appear to be involved in the positive feedback of interferon production.

17.3.5.2 Chemokines Cattle possess fewer chemokines than humans. However, they also have some chemokines with no obvious human homologs [33]. All chemokines appear to have evolved from a single ancestral protein that appeared about 650 mya, long before the emergence of the mammals. They have since undergone numerous duplication events in a lineagespecific manner. Many chemokine genes are clustered together. These clustered chemokines often bind to the same target cell receptor. Interestingly non-clustered chemokine genes tend to have homeostatic roles with more specific receptors and tend to be more conserved. Cattle have two chemokines that are not found in humans. One resembles human CCL23. The second is called regakine-1. Regakine-1 is a CC chemokine found in bovine serum that acts with CXCL8 and C5a to attract neutrophils and enhance inflammation. It has a homolog in sheep. Chemokines can be detected in many inflammatory diseases of cattle, including bacterial pneumonia, mastitis, arthritis, and endotoxemia. Impaired neutrophil migration is associated with certain specific CXCR2 genotypes, and this may result in increased susceptibility to mastitis in some cattle [34].

17.4

Lymphoid organs

17.4.1 Thymus The thymus in the calf is relatively large. It occupies most of the anterior mediastinal space reaching back to the pericardium. Most of it is however found in the cervical region (Fig. 17.2) [35]. It consists of two lobes that extend along the trachea and esophagus from the thoracic inlet to the thyroid. These lobes are large at the base of the neck but taper anteriorly.

17.4.2 Spleen The bovine sinusal spleen has a bilayered capsule with smooth muscle-rich trabeculae [36]. The red pulp contains an extensive network of smooth muscle cells interspersed within the cords (Fig. 11.4).

17.4.3 Lymph nodes Ruminant lymph nodes conform to the conventional mammalian pattern with a peripheral cortex and central medulla separated by a T cell-rich paracortex.

258

SECTION | 2 Mammalian orders

FIGURE 17.2 The location of the thymus (yellow) in the calf. From Sisson S, Grossman JD. The anatomy of the domestic animals. 4th ed. Philadelphia, PA: WB Saunders Co; 1953. With Permission.

17.4.4 Hemal nodes Hemal nodes (technically hemolymph nodes) are structures similar to lymph nodes found in association with some blood vessels of cattle, deer, and related mammals. Because of their dark red or black color and their size (mm to peasized), they are very obvious in necropsy. They are found in adipose and connective tissue in the neck, around the kidneys, and in large blood vessels. Their numbers and prominence vary greatly between individual animals. Their function is unclear and somewhat controversial [37]. The general arrangement of their lymphoid tissue is similar to that in conventional lymph nodes (Fig. 17.3). However, they differ from conventional lymph nodes in that their sinuses contain numerous red cells, and in some respects, they resemble small spleens. They have a cortex containing germinal centers and B cells. T cells predominate at the medulla in association with lymphatic sinuses. They are surrounded by bloodfilled sinuses. These cells differ, however, from those found in conventional lymph nodes by containing more γ/δ1, WC11 T cells, and fewer CD81 T cells [38]. Intravenously injected carbon particles are trapped in the sinusoids of hemal nodes, suggesting that they may combine features of both the spleen and lymph nodes. As a result, they can both trap antigens and trigger cellular and humoral immune responses.

17.4.5 Mucosal associated lymphoid tissues Organized lymphoid tissue is found on all the bovine mucosal surfaces. Thus it is found in the lacrimal and salivary glands, in the nasopharyngeal tonsils, in the gut-associated lymphoid tissue along the intestine, and in the lungs in the form of bronchus-associated lymphoid tissues. As in other species these tissues generally consist of aggregated lymphoid follicles covered by epithelium The epithelial surface contains the specialized antigen capturing cells known as M cells. Sensitized T and B cells can migrate between these mucosal lymphoid organs. For example, they can move from the intestine to the mammary gland in pregnant cows so that antibody responses to intestinal invaders result in the migration of B cells to the mammary gland where they differentiate into plasma cells and result in the appearance of IgA antibodies in milk. The process of rumination that brings a bolus of semi-digested fermented food to the oral cavity may require some additional defensive measures in the oropharynx. Cattle possess five distinct tonsils [39]. These are the lingual, palatine, pharyngeal, tubal, and palatine tonsils. Lymphocytes are scattered through the submucosa and the connective tissue cores of the lingual papillae [40]. In cattle, a well-developed and visible lingual tonsil is located at the root of the tongue. Multiple parallel crypts are visible, and these are surrounded by aggregations of secondary lymphoid follicles. These lymphocytes are primarily CD31 T cells. Ruminants also possess large ovoid palatine tonsils 3
4 cm in length [41]. These palatine tonsils are not visible on the surface of the mucosa. The crypts and associated mucus glands open into a central cavity, a common tonsillar sinus which then has a short passage to the surface. The lingual and palatine

Ruminants: cattle, sheep, and goats Chapter | 17

259

FIGURE 17.3 Histological section of a bovine temporal hemal node. It is enclosed in a thick capsule of connective tissue and smooth muscle (1). It contains large numbers of lymphoid follicles (2). Between these are scattered lymphocytes (3). The sinuses of the node are filled with red blood cells (4). At the center of the node are arteries and veins (5). From Castelyn CR et al. Morphological and immunological characteristics of the bovine temporal lymph node and hemal node. Vet Immunol Immunopathol 2008;126:339
50. With Permission.

tonsils form follicles with crypts. The pharyngeal and tubal tonsils however lack crypts. The tubal tonsils are visible as a cluster of irregular bumps around the opening of the Eustachian tubes. Bronchus-associated lymphoid tissue is not present in neonatal calves, but aggregates of lymphocytes accumulate progressively over the first 18 months of age [41]. This developing lymphoid tissue is located under the bronchial epithelium and in the submucosa as well as in the bronchiolar adventitia. Multiple Peyer’s patches are present in the bovine jejunum. There may be as many as 76 patches present in lateterm fetuses. There is also a single very large, elongated patch in the ileum (IPP) that may reach 9 meters in length in newborn calves. This long IPP undergoes age-related involution and by 2 years of age, only a few small follicles remain. In ruminants, these PPs are considered to be primary lymphoid organs since they are the major sites of neonatal B cell diversity generation (See Chapter 11). Cattle also have a single patch of lymphoid tissue in the proximal colon located near the ileocecal opening [41]. The presence of an enormous rumen full of microbes and their antigens has the potential to place a significant burden on the ruminant immune system. However, the rumen is lined by stratified squamous epithelium and rarely leaks. If leaks do occur as a result of ruminal acidosis or traumatic reticulitis, then both local and systemic inflammation result. However, the immune system also plays an important role in regulating the ruminal microbiota [42]. Thus saliva is rich in IgA and when swallowed this IgA can readily bind to both pathogenic bacteria as well as commensals. When these IgA-coated bacteria are analyzed, it is found that the composition and abundance of the major bacterial orders,
Bacteroidetes, Actinobacteria, Fibrobacter, Tenericutes, and candidate phyla TM7 are clustered tightly with those coated by salivary IgA. They closely resemble the whole rumen microbiota. Thus the binding of salivary IgA appears to have a direct influence on the composition of the ruminal bacterial population [42].

260

SECTION | 2 Mammalian orders

17.4.6 The genital lymphoid ring Lymphoid tissues consisting of organized lymphoid follicles, as well as diffuse lymphocytic infiltrates, are found in the lamina propria of the vaginal vestibule of cows where they form a ring of lymphoid tissue [43]. These lymphoid follicles may contain germinal centers. They also contain both lymphocytes and macrophages and high endothelial venules (Fig. 17.4). The lymphocytes are predominantly B cells. The epithelial covering of these follicles is infiltrated with lymphocytes as well. It is suggested that this genital lymphoid ring serves a similar function to Waldeyer’s ring in the pharynx.

17.5

The major histocompatibility complex

The bovine major histocompatibility complex (MHC) has been mapped to autosome 23 and is collectively termed the BoLA (Bovine leukocyte antigen) system (Fig. 17.5) [44]. The MHC class 1 molecules are transmembrane heterodimeric glycoproteins with a 45 kDa alpha chain that associates with a β2-microglobulin molecule of 12 kDa. They are expressed on all nucleated cells. MHC class II molecules are also heterodimers with an alpha chain of 33 kDa linked to a beta chain of 28 kDa. In general, class II MHC molecules are only expressed on professional antigen-presenting cells in this species. MHC class III molecules, as in other mammals, are a diverse mixture of molecules connected in some way with adaptive or innate immune functions.

17.5.1 The MHC class I region The BoLA class I region ranges from 770 to 1650 kb in length. Its genes are clustered in two groups, A and B but only the A cluster appears to be functional. Unlike other mammals that have a fixed number of classical class I genes, ruminants have variable numbers of them. In cattle, there are six classical class I genes [45]. Between one and three of these classical genes are expressed on any individual haplotype. As in other mammals, class I genes are much more divergent than class II genes. As a result, there are no orthologous relationships even between related ruminant species. There are also many gene fragments and pseudogenes located in the class I region. Currently 96 class I alleles have been FIGURE 17.4 The histologic structure of the gonadal-associated lymphoid tissues in the bovine vaginal vestibule. (A) An isolated lymphoid follicle. (B) A follicle is associated with scattered lymphocytes between it and the vaginal epithelium. (C) A follicle in close approximation to the epithelium and the epithelium is densely infiltrated with lymphocytes. (D) A double follicle in close association with the vaginal mucosa. From Chuluunbaatar I, et al. Genital organ-associated lymphoid tissues arranged in a ring in the mucosa of cow vaginal vestibules. Res Vet Sci 2022;145:147
58. With Permission.

Ruminants: cattle, sheep, and goats Chapter | 17

DI

DY

DN

DM

B

A B

A

A B

DO 15 cM DQ

B

B A

IIa

FIGURE 17.5 The overall structure of the bovine major histocompatibility complex. Note how the class II region is divided by an inserted sequence 15 cM in length.

DR

B B B A

A IIb

261

III

B I

identified and 29 different haplotypes characterized [46]. All these proteins are functional and capable of presenting processed antigenic peptides and all are polymorphic to some degree. They are not all consistently expressed. Indeed, no cattle haplotype has been found to express more than three class I genes [47]. At least 13 configurations have been identified to date. There are likely many more in less common cattle breeds. Certain MHC class I genes appear to be preferred when regulating cell-mediated immune responses. Their positions within the class I locus may not be fixed, and their relative positions may change in different haplotypes [45]. Some animals may even possess two copies of the same class I gene. Polymorphism in the expressed MHC class I molecules is usually restricted to the paired α1 and α2 domains that together form the peptide-binding groove. Endogenous peptides 8
10 amino acids long bind to the groove and are presented to CD81 T cells. Cattle also possess four nonclassical MHC class I genes (NC1-NC4) whose products play a role in regulating NK cell function. NC1 appears to be present on all haplotypes. They are relatively nonpolymorphic receptors with the number of alleles ranging from one to eight [48].

17.5.2 The MHC class II region The ruminant MHC class II region is unique in that it is split into two subregions (IIa and IIb) from DOB in the class IIb region to DQB in the class IIa region. They are separated by a spacer sequence of 15 cM. This splitting of the class II region resulted from a chromosomal inversion. The class IIb region is located close to the centromere while the class I region is telomeric to the class IIa genes. The class IIa region contains two gene clusters DR and DQ that are tightly linked in most haplotypes. A DP region is totally absent. The DQA and DQB genes are duplicated. The BoLa-DRA gene product shows limited polymorphism with only four alleles and three SNPs that are located in its second exon forming the antigen-binding site. Conversely, the DRB genes are highly polymorphic [46]. Again, this polymorphism occurs in the second exon that encodes the antigen-binding domain Cattle have three DRB loci of which only one (DRB3) is functional. DRB1 is a pseudogene, DRB2 may not be expressed. However, more than 130 different alleles from exon 2 of DRB3 have been identified [49]. The DQ cluster contains five DQA and five DQB genes [46]. There are 81 alleles in BoLA-DQB located across at least five loci. Like other such genes, their numbers vary among different haplotypes. There are also at least 61 DQA alleles. Fifty-six different class IIa haplotypes have been identified to date. This is significantly more polymorphism than in other mammals.

17.5.3 The MHC Class IIb region The class II region contains both an extended class II region as well as a classical class II region. The classical region contains the DSB, DYA, and DYB genes. DYA and DYB encode proteins expressed by a subpopulation of dendritic cells and are believed to be involved in antigen processing [46]. DYA does not appear to have homologs in other species and may be bovine specific. DYA and DIB form a closely linked pair. Their function is unknown [49]. DIB is also tightly linked to DOB, and both show limited polymorphism. The function of DSB is also unknown. In the extended region are multiple nonclassical genes encoding proteins involved in antigen processing including the LMP complex (low molecular mass polypeptide), LMP2, and LMP7, and the TAP genes, TAP 2.1, TAP1, and TAP2.

262

SECTION | 2 Mammalian orders

17.5.4 The MHC class III region The bovine class III region contains a heterogeneous mixture of genes encoding proteins in some way linked to immunity. These include complement factors BF and C4, steroid 21-hydroxylase (CYP21), heat shock protein 70 (HSP70), and tumor necrosis factors α and β (TNFA and TNFB).

17.5.5 The natural killer receptor complex Cattle have a unique NK cell receptor system in that they use genes from an expanded leukocyte receptor complex (LRC) in addition to an expanded NKC complex [50]. The KIR and KLR gene families are located on different chromosomes while their potential MHC class I ligands are located on a third chromosome. As a result, bovine NK cells can express different combinations of different cell surface receptors that are encoded by genes from either the LRC or the NKC. The two complexes encode proteins that are structurally unrelated [51]. Thus overall, bovine NK cells are highly heterogeneous, and cattle possess a multigenic array of activating and inhibitory receptors that recognize MHC class I molecules [45,52].

17.5.6 Leukocyte receptor complex The bovine LRC complex is located on chromosome 18 and contains a large KIR cluster flanked by LILR and FCAR genes. Killer immunoglobulin-like receptors (KIRs) are the primary receptors for MHC class I antigens on bovine NK cells The usual arrangement in mammals is that the LRC usually has no more than a single functional KIR gene. A notable exception to this rule occurs in domestic cattle (Fig. 17.6). As in the higher primates, cattle possess an expanded KIR locus, (although they use a different gene lineage). Of 18 bovine KIR genes, eight are functional while ten have been inactivated by point mutations. The members of the cattle KIR family have evolved through successive duplications of a block containing ancestral KIR3DL and KIR3DX genes that predated eutherian mammals. As a result, cattle KIR consists of two clades [52]. One clade represented by a single protein is most closely related to human KIRs. The other clade that accounts for most bovine KIRs is closely related to the nonfunctional human KIR3DX1. This human gene is located within the LILR gene cluster. Twelve bovine genes belong to the KIR3DX lineage while six belong to the KIR3DL lineage. Selective inhibition of KIR3DL and other activating receptor genes leaves just one functional inhibitory receptor KIR3DL, one activating KIR3DX, and six inhibitory KIR3DX. Thus functional KIR diversity evolved from an ancestral KIR3DX in cattle and from KIR3DL in simian primates. The bovine KIR gene family that has expanded is related to a single-family found in humans, KIR3DX1. These results are consistent with a model where duplication of the ancestral KIR gene occurred B135 mya, before the radiation of the placental mammals. Successive duplications of KIR3DL gave rise to the primate NK cell family. On Lymphocyte Receptor Complex

8 LILR

LAIR

NCR1

13 KIR

Human

Bovine

Natural Killer Complex CD69

KLR

Ly49\

Human CD69 Bovine Functional genes only

KLR

KLRA (Ly49)

FIGURE 17.6 A comparison of the human and bovine NK cell receptor complex and the leukocyte receptor complex.

Ruminants: cattle, sheep, and goats Chapter | 17

263

the other hand, successive duplications of the ancestral KIR3DX created the diversity of bovine functional KIRs. Cattle also have at least 16 LILR genes within the LRC complex. Of their products, seven are likely inhibitory, four are activating and five are soluble.

17.5.7 Natural killer complex The bovine NKC is divided between chromosomes one and five [53]. The key genes within the LRC encode the CD94/ NKG2 heterodimeric receptors. Most of these NKG2A genes can generate multiple alternatively spliced forms. With respect to the CD94/NKG2 receptor complex, cattle possess four CD94 (KLRD) sequences, ten NKG2A (KLRC1) sequences, and three NKG2C (KLRC2) sequences [54]. Some of the NKG2A sequences show minor allelic variations. They all have two ITIMs in their cytoplasmic tails. In contrast, NKG2C appears to be monomorphic. Their specific ligands are not currently known but presumably, they are nonpolymorphic MHC molecules [54]. In cattle and goats, a second KLRC locus has evolved separately. It is flanked by KLRA and KLRJ genes and a novel KLRH-like gene has evolved an activating tail [55]. There has been no expansion of the bovine Ly49 repertoire, the receptor system employed by rodents and horses. The single bovine Ly49 (KLRA) gene does, however, encode three alleles, so cattle are the only species known to express both polymorphic Ly49 and KIR receptors [56].

17.5.8 Natural killer receptor ligands The MHC class I chain-related (MIC) genes are located in the bovine class I region. Their products are expressed by cells in response to stress. Four MIC genes have been identified in cattle compared to only two, MICA and MICB in humans [45]. Their products act as ligands for several different receptors such as the activating receptor NKG2D. Some of these MIC genes are polymorphic and their presence varies with haplotype, although the MIC4 gene is consistently present in all haplotypes. The presence of the other three is variable. MIC4 is an NKG2D ligand and may play a similar role to MICA and MICB.

17.5.9 Natural killer cell functions Bovine NK cells are large granular NKp461, CD3 lymphocytes. Their proportion among bovine blood lymphocytes ranges from 2% to 3.5% in healthy animals [57]. Their numbers also tend to be higher in young calves and they are rapidly mobilized by infections. NK cells can be found in the secondary lymphoid organs as well as in the bone marrow [58]. They are found in the highest concentrations in the spleen, lymph nodes, and peripheral blood [59]. Different subsets of these NK cells are found within different tissues. For example, CD21 cells predominate in the blood whereas CD2 cells predominate in lymph nodes and afferent lymph vessels [60]. NK cells can leave lymph nodes by returning directly to blood vessels rather than emigrating through the lymphatics. Cattle NK cells can be activated by IL-2, IL-12, IL-15, IFN-α, and IFN-γ. Activation by IL-2 enables them to express CD25 and CD8 and lyse tumor cell lines. CD21 NK cells appear to be activated and respond more rapidly to IL-2 in culture although their cytotoxic activities are the same [51]. If activated by IL-12 and IL-15, NK cells express increased amounts of granulysin, IFN-γ, and perforin, and can kill both human tumor cells and bacille Calmette-Gue´rin (BCG)-infected macrophages. Cattle also produce NK-lysin and uniquely, have four functional NK-lysin genes (As opposed to one in other mammals) [61,62]. Three of these genes (NK1, NK2A, and NK2B) are predominantly expressed in the Peyer’s Patches while one (NK2C) is expressed exclusively in the lungs. Three NK subpopulations have been described. For example, most bovine NK cells express both CD2 and NCR1, but subpopulations may be CD2 or NCR negative. Other bovine NK cell surface molecules include CD16, perforins, CD5, CD94, WC1, MHC class II, and asialo-GM1. Cattle also have some circulating NCR1-positive T cells. Bovine NK cells can readily kill bovine target cells using their cytotoxic molecules, perforin, granulysin, and NK lysin [57]. They can kill human cancer cell targets as well as bovine cells infected with parainfluenza-3, bovine leukemia virus, or bovine herpesvirus type 1 [61]. They generate resistance to mycobacteria by preventing its replication within macrophages [63]. They play a role in resistance to the protozoan parasite Neospora caninum by producing IFNγ which kills infected cells.

264

SECTION | 2 Mammalian orders

17.5.10 Dendritic cells Bovine dendritic cells can be divided into four subsets. Conventional DC1 cells (cDC1s); conventional DC2 cells (cDC2s), plasmacytoid DCs (pDCs), and monocyte-derived DCs, each with its own characteristic phenotype. All bovine DCs express MHC II, CD80, CD86, and CD40. Cattle also possess two dendritic cell subpopulations that differ in their ability to stimulate CD41 and CD81T cells. One population synthesizes more IL-12, whereas the other population produces more IL-1 and IL-10. These may well represent cDC1 and cDC2 subpopulations respectively. Cattle pDCs produce large amounts of type I interferons [64]. Evidence indicates that DYA and DIB class II MHC genes are transcribed in bovine dendritic cells to form a distinct subpopulation. These MHC-positive DCs are found in the lymph nodes, lungs, and thymus [65]. The transcription of these MHC products by thymic epithelial cells is probably associated with antigen presentation by this DC subset.

17.6

B cells and immunoglobulins

Cattle and other ruminants possess a full range of immunoglobulins. Most but not all, are of conventional four-chain structure. However, cattle are the only mammals known to possess two functional IgM heavy chain loci [66]. Bovine immunoglobulins have the same functions in other species although the role of IgD remains unclear as does the specific function of IgG3.

17.6.1 Immunoglobulin heavy chains The cattle IGH locus contains eight immunoglobulin heavy chain genes,
IGHM1, IGHM2, IGHD, IGHG1, IGHG2, IGHG3, IGHA, and IGHE (Fig. 17.7). Most of these heavy chain genes are located on chromosome 21. There are two additional functional IGHM loci and an IGHJ locus located on chromosome 11. The genes on chromosome 21 are organized in a continuous 678 kb genomic sequence.

17.6.2 IGHM Both IgM genes, IGHM1 and IGHM2, can be expressed independently or sequentially by class switching [66]. The IGHM2 gene is expressed dominantly in most antibody-producing cells. As in other species, membrane-bound IgM functions as a B cell antigen receptor and is the first isotype expressed during B cell ontogeny. It is required for B cell survival and the subsequent expression of other IGH classes. It is unclear however how the two functional IgM loci on different chromosomes interact. Cattle also differ from humans by having the majority of their circulating B cells expressing IgM rather than IgG. FIGURE 17.7 The overall structure of the bovine immunoglobulin heavy and light chain loci. Ψ 5 pseudogene.

23 6

42

5' IGH

3' M1

(D

V

Dn ) 3 Jn M2

D

G3

G1

G2

E

A

63

IGL

IGK

22

3

M

D

G

E

A

V

D

J

LC

Ruminants: cattle, sheep, and goats Chapter | 17

265

17.6.3 IGHD The bovine IGHD gene appears to have been duplicated from the IGHM gene over 300 mya [67,68]. The persistence of IgD over such a long period suggests that it confers some survival advantage [68]. Cow, sheep, and pig IgD H chains all contain three constant domains and a hinge region (Fig. 9.7). They share about 45% homology with their human ortholog. A short Sδ switch sequence also resulted from the second duplication and controls μ 2 δ class switch recombination in cattle. The ruminant Cμ1 exon and its upstream region appear to have been duplicated before the separation of the cow and sheep lineages about 20 mya. The new genes were inserted upstream of the IGHD hinge exon and subsequently modified by gene conversion.

17.6.4 IGHG Cattle have three IGHG genes corresponding to their three subclasses: IgG1, IgG2, and IgG3. Two heavy chain allotypes (a and b) have been identified in each of the three IgG classes. IgG1 constitutes about 50% of serum IgG and is the predominant immunoglobulin in cows’ milk rather than IgA. Bovine IgG1 expression is positively regulated by IL4 and IgG2 expression by IFN-γ. IgG2 levels are highly heritable, and their concentrations vary greatly among cattle. Cattle possess a unique Fc receptor on their macrophages and neutrophils that binds only IgG2. Since bovine IgG2 has a very small hinge region, this receptor might represent a specific adaptation to the structure of this immunoglobulin. IgG3 differs from IgG1 and IgG2 in the structure of its hinge region. The hinge is encoded by two exons, one of which encodes a 22 amino acid extension of its CH1 exon. It also differs as a result of an additional glycosylation site in its CH3 domain. Like the other IgG subclasses, it has two alleles that differ by a six amino acid substitution in the coding region [69]. Its specific function remains unclear.

17.6.5 IGHV To date, B36 functional IGHV genes (42 total), 23 IGHD, and four functional IGHJ genes have been identified in cattle [66,70]. All the functional IGHV genes belong to the IGHV1 family since the members of the IGHV2 and IGHV3 families found in cattle are all pseudogenes. In general, bovine V region diversity is primarily mediated by somatic hypermutation and junctional diversity. Cattle likely employ recombination for their light chains and a combination of recombination and gene conversion for their heavy chains. Initial diversification occurs in the bone marrow and other lymphoid organs followed by somatic mutation in ileal Peyer’s patches.

17.6.6 Ultralong VH CR3 A unique feature of some bovine IgG molecules is the presence of a subset (,10%), that possesses an unusually long third complementary determining region (CDR3) in their variable domains and thus has significantly different antigenbinding properties. Human and mouse IgG heavy chain variable domains have CDR3 regions that normally average about 10
13 amino acids in length. Cattle are different. Because they have two unusually long V and D gene segments, a subset of bovine IgG molecules possess V domains with CDR3s that may contain from 45 to over 70 amino acids [71]. The 23 gene segments in the D cluster can be divided into nine subfamilies, D1 to D9 [59]. Except for D9 which contains only 14 nucleotides, all the other D gene segments contain more than 30 nucleotides. By far the longest is D8 which contains 149 nucleotides. The extreme length of CDR3 in some bovine antibodies is therefore mainly due to their use of D8. This segment encodes four cysteines that form intrachain bonds with each other, together with repeated glycine, serine, and tyrosine residues. As a result, the mixture of mini-domains generated by these cysteine residues can fold to form a tightly linked knob structure [72] (Fig. 17.8). In addition to D8, these ultralong CDR3 regions may also use an unusually long IGHV gene (IGHV1
7). An eightnucleotide duplication at the 30 end of IGHV1
7 results in its increased length. This duplication generates a β-strand that contributes to the stalk structure of the ultralong CDR3 (Fig. 17.9) [72]. (The CDR1 and CDR2 regions of this V gene segment show limited sequence variability and are of normal length.) These joined V and D segments together can therefore encode an unusually long CDR3 that forms an unusual stalk and knob structure in the V domain [66]. These long heavy chain CDR3s fold into a long beta-stranded stalk supporting a disulfide-bonded “knob” located far

266

SECTION | 2 Mammalian orders

Knob and ball structures

Hinge region

FIGURE 17.8 The structure of a bovine long-chain immunoglobulin showing the stalk and knob structure generated by the use of unusually long V and D gene segments.

FIGURE 17.9 A ribbon diagram showing the structure of a bovine Fab (Left) as compared to a “normal” Fab showing the superlong β-stranded stalk and knob protruding from the VH immunoglobulin domain. From Wang et al. Reshaping antibody diversity. Cell 2013;153:1379
93. With permission.

from the rest of the molecule. This protruding knob and stalk can still bind antigens effectively. These ultralong CDRs may also show significant structural diversity as a result of somatic mutation and combinations of their somatically generated disulfide bonds [73]. The knob can fold into multiple mini-domains and can bind diverse antigens. The benefits of this unusual structure are unclear, but it almost certainly enables the CDR3 region of these molecules to bind otherwise inaccessible antigenic determinants on some viruses. These ultralong CDRs may compensate for their limited V region diversity.

17.6.7 Light chains The bovine IGL locus contains 63 IGLV genes, of which 17 are functional, organized in three subclusters 50 to four J-C genes [74]. (This number is much lower than in humans or mice). The IGLV genes are organized into eight subfamilies.

Ruminants: cattle, sheep, and goats Chapter | 17

267

The predominantly expressed IGLV1 genes are found in two 50 clusters, whereas the rarely expressed IGLV2 and IGLV3 genes are proximal to the J-C genes. Cattle have more than one IGLJ gene, but only one is expressed. Many of the pseudogenes are fused to IGLJ in the germline. Cattle also have four IGLC genes. Two of these (IGLC2 and IGLC3) are functional, whereas the other two (IGLC1 and IGLC4) are pseudogenes. IGLC3 is preferentially expressed [75]. About 95% of bovine antibodies use lambda light chains. Allotype B1 is found on light chains of some cattle but is relatively uncommon [66]. There are 22 variable kappa chain genes in cattle organized into four subfamilies [74]. The kappa chain locus is compact (280 kb) and simple, reflecting the preferred use of lambda chains.

17.6.8 Receptor assembly In most mammals, excluding pigs, when B cell antigen receptors are generated, the first chain to be assembled is the heavy chain. This chain is capable of generating much more junctional and combinatorial diversity than the light chain and is the major contributor to antigen binding. This heavy chain is linked to signal transduction molecules, and a surrogate partner chain is provided so that the pre-B cell can respond in a limited way to antigens. As a result, a small clone of B cells expressing only the heavy chain is formed. Signaling through this pre-receptor triggers limited proliferation. This is followed by the assembly of a partner chain. In B cells this is the light chain. The partner chain is the one that uses only V and J genes and thus contributes much less diversity to the antigen receptor although it “fine-tunes” its antigen-binding specificity. Three bovine surrogate light chain genes have been identified. These surrogate chains are temporarily expressed during B cell development. They encode three polypeptides VPREB1, IGLL1, and VPREB3. The genes for these surrogate chains are located close to the λ chain locus but in the opposite orientation [74]. The surrogate chains are required to stabilize the heavy chains during rearrangement and before light chain joining [74].

17.7

T cells and cell-mediated immunity

17.7.1 T cell antigen receptors T cells expressing γ/δ receptors may comprise up to 60% of the circulating T cell population in young calves. They decrease steadily with age, but their numbers remain relatively high until adulthood. Thus 8%
18% of adult bovine peripheral blood T cells are γ/δ T cells. These antigen-binding heterodimers are paired with the complex signaling structure classified as CD3. While superficially similar, the α/β and γ/δ antigen receptors differ in their structure, glycosylation, plasma membrane organization, and signaling motifs. Although many γ/δ T cells can bind antigens specifically based upon the shapes of their combined V regions and binding groove, there is a subpopulation of γ/δ T cells that depend upon WC1 binding in a manner similar to the role of CD4 and CD8 molecules for both their antigen specificity and activation. CD4 is expressed on 20%
30% of blood lymphocytes in adult ruminants. Double-negative T cells constitute 15%
30% of the blood T cells in young ruminants, but this may reach 80% in newborn calves. Most of these doublenegative cells are γ/δ1 and WC11. Thus the major circulating T cells in ruminants (γ/δ1, WC11, CD42, CD82) differ from the predominant T cells in humans and mice (α/β1, WC12, CD41, CD82). Cattle possess Th1 and Th2 cells and can mount polarized immune responses. Bovine CD41 cells produce IL-2, IL-4, IL-10, and IFN-γ.

17.7.2 T cell receptor genes The four bovine T cell receptor (TCR) chains are encoded by four gene loci. The TRA/D locus codes for both the α and δ chains since the TRD genes are embedded within the TRA locus. The TRB locus codes only β chains and two TRG loci encode the γ chains. All four TCR loci contain V, J, and C genes, while the TRB and TRD loci also contain D genes (Fig. 17.8). Each of the four TCR loci contains two or more C genes. In the TRA/D locus one C gene codes for the alpha chain and the other for the delta chain. The TRB and two TRG loci, in contrast, contain three and six constant region genes respectively. Cells with α/β TCRs rearrange and express TRA and TRB genes whereas γ/δ T cells express TRG and TRD genes. α/β T cells and γ/δ T cells arise from a common precursor and the TCR class switch is mediated by signals within the thymus. Developing T cells committed to the α/β TCR lineage delete their TRD genes by looping them out and so

268

SECTION | 2 Mammalian orders

switch to using their TRA genes. Some of the V gene segments in the TRA/D locus may be used by either α or δ TCR chains.

17.7.3 TRA/D The bovine TRA/D locus is located on chromosome 10 and contains 337 genes. These include 183 TRAV genes (79 functional, 14 ORFs, and 74 pseudogenes with the remainder undetermined). 39 116 nonlocalized TRDV genes (45 functional, 5 ORF, and 5 pseudogenes). Nine TRDD genes (6 functional and 3 ORF), and 60 TRAJ genes (52 functional, 2 ORF, and I pseudogene). Four TRDJ genes (three functional and one pseudogene). Followed by one functional TRAC gene and one functional TRDC gene [76]. The TRD locus is embedded within the TRA locus and contains up to 100 V genes belonging to four subfamilies depending on the species. Thus cattle have 55, sheep have 40 and goats have 30 TRDV genes [77]. The artiodactyl TRDV1 subfamily is greatly expanded relative to humans or mice. It is also comingled with the TRAV genes. Thus among the subfamilies, there are 52 closely related TRDV1 genes, two TRDV2 genes, two TRDV3, and only one TRDV4 gene [78]. The V genes can be classified into 11 subfamilies. As mentioned previously, DD gene use is optional but cattle may use multiple D segments per delta chain. A single TRD transcript has been shown to contain sequences from all six TRDD genes and generates a very long CDR3 region in its V domain. Four of the TRA/DV genes appear to lack a CDR2 and in some cases may have an expanded CDR3. The TRDV4 gene is located downstream of the C genes and is in inverted orientation [79].

17.7.4 TRB The bovine TRB locus is located on chromosome 4. It contains a cluster of TRBV genes located upstream of two or three D-J-C cassettes, each containing several functional J genes as well as single TRBD and TRBC genes. The TRBD genes are all similar in sequence and length and their use is optional. Any of the 50 TRBV genes may be joined to any of the D-J-C cassettes, and a V gene may join to either a D or a J gene. Cattle have 134 TRBV genes of which 79 are probably functional [80]. This large number of TRBV genes is a result of multiple duplication events involving blocks of DNA [81]. Likewise, duplication has also resulted in the third DJC cluster. The net result is a TRB locus that is considerably larger than that in humans or mice.

17.7.5 TRG There are two, entirely separate bovine TRG loci. Thus TRG1 and TRG2 are located on bovine chromosome four (4q3.1 and 4q1.5-2.2) [82]. The two loci collectively contain 11 TRGV genes in eight subfamilies and six functional TRGC genes [79]. The TRG genes are arranged in six genomic cassettes. The level of identity among the TRGV genes ranges from 34% to 75% of the nucleotide sequences and 17%
66% for their inferred amino acid sequences. The TRGC genes are generally more similar with 55%
90% nucleotide identity [83]. The existence of these six TRGC genes suggests that the TCRs they generate differ in their biological properties. Each of these two loci consists of a set of three cassettes each containing the basic gene arrangement, V-J-J-C. The V-J-J recombination occurs first and the construct is then spliced to the C gene to form mature transcripts. The TRG1 locus is about 250 kb in size and contains TRGC3, 4, and 5. The TRG2 locus is about 190 kb in size and contains TRGC1, 2, and 6. TRGC5 is apparently ancestral. They are all functionally and structurally diverse. The bovine TRG1 locus also contains a fourth, nonfunctional pseudogene, TRGC7, that lacks TRGJ genes [84]. These ruminant gamma chains are unique in that they can closely interact with another cell surface molecule, WC1.

17.7.6 γ/δ T Cell functions The functions of γ/δ T cells differ among mammals. For example, in “γ/δ-low” species such as humans and mice there are relatively few V genes in the TRD and TRG loci, and their combinational repertoire is therefore small. In addition, the cells bearing these receptors use only a few V gene combinations. In contrast, human α/ß T cells show a much wider range of binding specificities. Thus there is a marked difference between the size of the α/ß and γ/δ TCR repertoires in these species. In the γ/δ-low species the γ/δ T cells probably have a limited role in adaptive defense but recognize conserved PAMPs.

Ruminants: cattle, sheep, and goats Chapter | 17

269

The situation in “γ/δ-high” ruminants is very different. In these mammals, γ/δ T cells form a large proportion of the total T cell population. This proportion fluctuates in response to age, management, and stress. In addition, ruminant γ/δ T cells show considerable receptor diversity. For example, in the sheep, γ/δ V-region diversity results from the use of 28 TRDV genes and 13 TRGV genes that contain two distinct hypervariable segments like the CDRs seen in IGHV genes. In addition, there are multiple TCR γ/δ isoforms generated by the association of a single Cδ chain with up to six or eight Cγ chains. All this suggests that γ/δ T cells in ruminants can recognize a very wide diversity of antigens and mount adaptive rather than innate responses. Bovine γ/δT cells home to the skin, the pharyngeal and palatine tonsils, reproductive organs, the intestinal wall, and the mammary gland suggesting that they serve a defensive role on body surfaces. They form the major T cell population at these sites.

17.7.7 γ/δ T cells as innate cells In γ/δ-low species there are two subsets of γ/δ cells. One subset is engaged in innate immunity, has limited γ/δ receptor diversity, and is mainly found in the skin and genital tract. The other subset is engaged in adaptive immunity, has extensive receptor diversity, and is found in secondary lymphoid organs and the digestive tract. The innate γ/δ cells preferentially bind microbial PAMPs, especially heat-shock proteins and phospholigands (carbohydrates or nucleotides with a phosphate group). They also respond to the MHC class Ib molecules, MICA and MICB, produced by stressed cells, cancer cells, and virus-infected cells. These innate γ/δ T cells also respond to lipid antigens presented by CD1 molecules. They respond to stimulation through TLRs, NODs, and NKG2D, as well as to cytokines alone [85]. In response, they secrete IL-10, IL-17, TGF-β, and IFN-γ. Like Th17 cells, innate γ/δ T cells can be activated by IL-23. Bovine γ/δ T cells respond to microbial PAMPs by increasing the expression of the chemokine lymphotactin (XCL1), MIP-1β, TNF-α, and granulocyte-macrophage colony-stimulating factor (GM-CSF). They express TLR3, TLR9, mannose-binding lectin, and CD36. These γ/δ T cells may well be major contributors to innate immunity in cattle especially since many of them have nonpolymorphic TCRs. γ/δ T cells in cattle can also respond to multiple pathogens including Leptospira, Anaplasma, and Mycobacteria [86]. For example, bovine γ/δ T cells but not α/β T cells can recognize the Mycobacterial cell wall component mycolylarabinogalactan-peptidoglycan. These responses require direct contact between the bacterium and the T cell, so it is accepted that their γ/δ TCRs act as pattern recognition receptors detecting bacterial PAMPs. Their role in detecting and responding to microbial glycolipids is unclear since other evidence suggests that cattle and ruminants may lack true NKT cells because they have no functional CD1d genes [87]. The CD1d genes in B. taurus appear to be pseudogenes because of disrupting mutations in their start codon and the donor splice site of the first intron. The same mutation has been found in other bovids such as African buffalo, sheep, and N’dama cattle, in antelope such as bushbuck and bongo, and in roe deer [87]. On the other hand, cattle do have functional CD1a and CD1b genes [88]. Bovine γ/δ T cells do not respond to conventional MHC class I and II antigen presentation.

17.7.8 γ/δ T cells as Th1 cells The adaptive subset of γ/δ T cells can be further subdivided into helper and effector cells. These effector cells are cytotoxic and can destroy cells infected with mycobacteria and some leukemic cells. They are polyclonal at birth, but their diversity decreases with age. As pointed out above, cattle possess multiple gamma chain constant region genes as well as multiple delta chain V genes (Fig. 17.10). As a result, their γ/δ T cell repertoire is highly diverse. In ruminants, CD2 WC11 γ/δ T cells have restricted expression of TCRγ genes since they only use the TRGC5 cassette. In general, WC1.11 γ/δ T cells act as Th1 cells, produce IFN-γ and promote type 1 immune responses. WC1.21 γ/δ T cells tend to act more like Th17 cells and produce greater amounts of IL-17 [89].

17.7.9 γ/δ T cells as Treg cells γ/δ T cells constitute a major regulatory T cell subset in the blood of cattle. These circulating T cells spontaneously secrete the suppressive cytokine, IL-10, and proliferate in response to IL-10, TGF-β, and contact with antigenpresenting cells [90]. These IL-10 secreting γ/δ T cells can inhibit the antigen-specific as well as the nonspecific proliferation of CD41 and CD81 T cells.

270

SECTION | 2 Mammalian orders

FIGURE 17.10 The organization of the bovine TCR locus. Note that there are two different TRG loci located on the same chromosome (chromosome 2).

Bovine

TRA/D

TRB

5'

183 AV

55 9 DV DD

134

3 DJ DC

C1

TRG

5 2 C5

7

C3

C2 2

2

2 C3

C4

C1

TRG1

V

D

J

AC

7

7

7

3'

60 AJ

C2 TRG2

C6

C

17.7.10 Workshop cluster 1 proteins 17.7.10.1 Structure Bovine lymphocytes express several cell major surface proteins not found in either humans or mice. The best defined of these belong to the Workshop cluster 1 (WC1) family. WC1 proteins are single-chain type 1 cell membrane glycoproteins of 220 kDa belonging to the “scavenger receptor cysteine-rich” (SRCR) protein family [91]. Cattle possess a multigene family of WC1 genes, each encoding proteins with six or 11 extracellular SRCR domains. They are expressed exclusively on γ/δ T cells. Between 50% and 99% of ruminant blood γ/δ T cells express the WC1 family proteins on their surface [92]. These WC11 T cells are found in high numbers in ruminant skin and mucus membranes as well as in hemal nodes and the thymus. (WC1 γ/δ T cells predominate in the gut, mammary gland, and uterus). About 13 WC1 gene family members occur in cattle, whereas 50
100 are found in sheep. Humans do not possess WC1, but they do express the closely related SRCR molecules, CD163 and CD163c-α on their γ/δ T cells [93]. Mice express other variants such as SCART1 and SCART2 that are involved in the functional differentiation of γ/δ T cells. Bovine WC1 molecules are very polymorphic and there are differences in their expression between B.t. taurus and B.t. indicus. A gene encoding WCI has been detected in the genome of the extinct aurochs [94].

17.7.10.2 Functions WC1 proteins serve multiple roles. Thus they function as pattern recognition receptors and can bind to surface components of some bacterial pathogens. When the binding specificities of the different WC1 proteins are examined, it is clear that like other PRRs, that they bind to selected bacterial ligands. There is an active, antigen-binding site on one of their SRCR domains [91]. In addition, bovine WC1 proteins can serve as signaling coreceptors with the γ/δ TCRs in a tyrosine phosphorylation-dependent manner. Individual WC1 receptors generate antigen specificity through colligation with the γ/δ TCR. SRCR binding is directly correlated with the γ/δ TCR response.

17.7.10.3 WC11 γ/δ T cells There are two bovine WC11 T cell subpopulations, WC1.11 and WC1.21 generated as a result of variable gene expression. These two subpopulations differ in their ability to respond to specific pathogens. For example, WC1.11 cells can recognize and bind Leptospiral products as well as the BCG vaccine strain of Mycobacterium bovis. WC1.21 cells in contrast can bind Anaplasma marginale and virulent strains of M. bovis [73,95]. Bovine WC1.1, like WC1 in the pig, has only six extracellular SRCR domains. WC1.1 binds Leptospiral antigens through multiple SRCR domains. Thus Leptospira, WC1, and γ/δ TCRs interact directly on the T cell surface [77]. As a result of this binding to WC1, TCRinduced activation is significantly increased so that the cells respond by proliferation and IFN-γ production. WC11 γ/δ T are also found in granulomas surrounding schistosomes and Mycobacteria. In these cases, the initial T cell infiltration is dominated by γ/δ T cells, and this is followed by a wave of α/β T cells. A second wave of γ/δ T cells

Ruminants: cattle, sheep, and goats Chapter | 17

271

may terminate the response. These WC11 cells secrete IL-12 and IFN-γ and may promote a Th1 bias in the immune response. In resting T cells, the WC1 and γ/δ TCR molecules exist as separate entities on the T cell surface. However, once these cells bind antigen, the receptors come together. These cells can then be activated by stimuli acting through their TCR or their WC1. In response, they produce TNF-α, IL-1, IL-12, and IFN-γ. This mixture suggests that they contribute to both inflammation and a Th1 bias in the bovine immune response and thus link the innate and adaptive immune systems. WC1.21 γ/δ T cells can act as Tγ/δ17 cells and produce IL-17. This differentiation appears to be driven by the transcription factor, SOX-13 [96].

17.7.10.4 WC1 γ/δ T cells Not all bovine γ/δ T cells express WC1. WC11 and WC1 cells have a different tissue distribution. In the spleen and uterus, the majority of T cells are WC1. In the blood anywhere between 1% and 50% of the γ/δ T cells can be WC1. This may be related to the parasite burden. The WC1 cell population increases with age while WC11 cell numbers decrease. WC1-negative γ/δ cells may act as regulatory cells. WC11 cells can be considered to be more engaged in innate immunity while WC1 cells are more adaptive. While bovine TRG genes are encoded at two loci with three V-J-C cassettes in each, WC11 T cells only use V genes from one of the six cassettes
the one containing TRGC5. (This cassette has five V genes and two J genes) (Fig. 17.10). In contrast, WC1γ/δ T cells use V genes from all six of the cassettes. The WC1 γ/δ T cells appear to be more myeloid in character whereas the positive cells are associated with the production of IFN-γ [89]. In the absence of WC11 cells, cattle preferentially make IgG1 whereas in their presence they make IgG2. These subpopulations also respond differently to type I IFNs. WC1 cells proliferate in response to IFN-τ, whereas WC11 cells are suppressed by IFN-τ and IFN-α.

17.7.11 The role of CD163 CD163 is also a member of the SRCR family. There are three known types of CD163 molecules in mammals, CD163A, CD163b, and CD163c-α [93]. Their nearest relative is WC1 expressed in cattle, sheep, and pigs. Unlike the CD163 molecules, WC1 possesses an N-terminal SRCR domain which is highly variable. Cattle possess genes encoding CD163A and CD163c-α but not CD163b. CD163A is widely expressed in immune cells. CD163c-α transcripts are enriched in γ/δ T cells. CD163A is encoded by a single gene across eutherian mammals. However, CD163cα is encoded by multiple genes. CD163 molecules may also act as PRRs that serve a similar function to WC1 in other mammals. The platypus also possesses CD163 orthologs.

17.8

Sheep (Ovis aires) and goats (Capra hircus)

While cattle can be considered the archetypal artiodactyls and thus representative of the ruminants, the immune systems in two other species, sheep and goats, have also been intensively studied. Both species are vitally important livestock in many parts of the world where conditions are unsuitable for cattle. They serve as sources of meat, milk, wool, and hides and are the mainstay of many agricultural economies. It will be obvious however that in most respects their immune systems are almost identical to cattle. These similarities are not discussed here. Goats and sheep diverged from their common ancestor about 10 mya. This ancestor diverged from cattle about 30
25 mya. This has provided ample opportunity for their immune systems to diverge as well. The domestic goat was most probably domesticated over 10,000 years ago in Anatolia. Its wild ancestor is believed to have been Capra aegagrus, the common ibex. Since that time, it spread across Eurasia and Africa, and domestic goats have probably hybridized multiple times with their wild relatives. Sheep were likely first domesticated in the same region and around the same time from the wild mouflon (O. orientalis). Kids and lambs were probably first captured for fattening, but this then evolved into breeding and eventually herding.

17.9

Reproduction and lactation

17.9.1 Sheep The gestation period of the ewe is about 145 days. MHC class I-positive cells can be detected in the fetal lamb by day 19, and MHC class II-positive cells can be found by day 25. The thymus and lymph nodes are recognizable by 35- and 50-days post-conception, respectively. Lymphoid follicles appear in the colon at 60 days, in jejunal Peyer’s patches at

272

SECTION | 2 Mammalian orders

about days 75
80, and in ileal Peyer’s patches by days 110
115. Blood lymphocytes appear in fetal lambs by day 32, and CD41 and CD81 cells are present in the thymus by 35
38 days [97]. B cells are detectable at 48 days in the spleen and by that time have already begun to rearrange their IGLV genes. C3 receptors appear by day 120, but Fc receptors do not appear until after a lamb is born. Fetal liver lymphocytes can respond to phytohemagglutinin at 38 days. Lambs can produce antibodies to phage ΦX174 at day 41 and reject skin allografts by day 77. Some fetal lambs can produce antibodies to Akabane virus as early as 50 days post-conception. Antibodies to Cache Valley virus can be provoked by day 76, to SV40 virus by day 90, to T4 phage by day 105, to bluetongue virus by day 122, and to lymphocytic choriomeningitis virus by day 140 [5]. The proportions of α/β and γ/δ T cells in their bloodstream change as lambs mature. Thus 1 month before birth, 18% of blood T cells are γ/δ positive. By 1 month after birth, they constitute 60% of blood T cells. IgE is passively transferred in ewe colostrum [98]. In goat colostrum, the IgG concentration is 2.4
2.8 times greater than in serum. IgG1 accounts for 95%
98% of this colostral immunoglobulin [99].

17.10 Innate immunity 17.10.1 Sheep Sheep TLRs closely resemble the bovine TLR sequences except for an opposite relationship between long and short splice variants when compared to the bovine genes. Sheep possess IFN-ε which has the same functional properties and potent antiviral activities as other type I interferons [29,100]. Sheep have CD161, CD14 NK cells in their blood. More than 80% of these cells also express perforin and NCR1 and are cytotoxic for mouse and sheep cell targets. They produce IFN-γ in response to IL-12 stimulation. DCs derived from sheep peripheral blood monocytes express MHC class II, CD11c, and are CD14 2 negative. Plasmacytoid DCs, cDC1, and cDC2 cells have been identified in this species.

17.11 Lymphoid organs 17.11.1 Tonsils As in other ruminants, the lymphoid tissues of the oropharynx are well developed. Sheep possess six distinct tonsils, lingual, palatine, paraepiglottic, pharyngeal, tubal, and soft palate [39]. The lingual tonsil is not obviously visible since it consists of aggregated lymphoid cells within the connective tissue cores of the gustatory papillae and is entirely covered by stratified squamous epithelium. The palatine tonsil is located on the upper surface of the soft palate and is also not visible macroscopically. Goats also possess these six tonsils.

17.11.2 Peyer’s patches There are a total of 30
40 Peyer’s patches in the sheep intestine. The large 1
2 meter-long ileal patch covers most of the intestinal circumference. It extends for up to 17% of the length of the small intestine and constitutes 1% of the body weight of a 2-month-old lamb [101]. As discussed in Chapter 11, the Peyer’s Patches of the neonatal lamb appear to act as primary lymphoid organs in which immature B cells respond to antigens and metabolites from the commensal microbiota to both diversify and expand their numbers.

17.12 Major histocompatibility complex 17.12.1 Sheep The complete MHC of the sheep (Ovar-MHC) has been sequenced. It is located on chromosome 20. As in cattle, the ancestral chromosome has been inverted to divide the Class II region into IIa and IIb subregions [102]. This same inversion is seen in the MHC of other artiodactyls including the addax antelope (Addax nasomlatus) confirming its ancient origins [103]. Ovar-MHC contains 177 protein-coding genes and ORFs arranged in the same order as in cattle [102] (Fig. 17.11). Of these, 131 are homologous to previously annotated genes in cattle and other mammals while 36 match predicted cattle genes. Eighteen genes encoding micro RNAs are found randomly scattered throughout the Ovar-MHC region [104
106].

Ruminants: cattle, sheep, and goats Chapter | 17

DY DN

A B

A

DO

B

DM

A B

DR DQ DR DQ DR DQ DR DQ DR

A

A1 B1

A2

B2

B1

B3

B2

C4A, C4B C2, Bf TNF HSP70 A

IIb

IIa

III

273

FIGURE 17.11 The structure of sheep major histocompatibility complex (Ovar-MHC). As in cattle, the class II region is divided by a large insertion.

B4

B I

The goat MHC (Cahi-MHC) is located on chromosome 23. It contains about 160 annotated genes including classical and nonclassical class I genes and class II DR and DQ genes. Despite some inversions, the conserved gene sequences are highly colinear with the sheep MHC [107].

17.12.2 Natural killer cell receptors The sheep leukocyte receptor gene complex is inverted on the long arm of chromosome 14. The genes are essentially arranged in the same manner in the goat with some minor differences. It contains two distinct KIR haplotypes [107]. The goat LRC has also been sequenced. It is located on the long arm of chromosome 18. It is also in an inverted configuration. It contains 15 KIR genes, eight LILR genes, and three novel two-domain Ig-like genes. Of these seven KIR, five LILR, and the three novel genes are functional. The LILR genes are arranged in two clusters and in the middle of the cluster is an FCG2R immunoglobulin receptor gene. Like cattle, goats have an expanded KLR repertoire within the NKC. They also possess an expanded complement of KIRs within the LRC as well. The goat NKC region and the LRC appear to be less complex than those in cattle. However, goats and sheep have independently expanded two novel KIR subfamilies and do not appear to possess a functional KIR3DL lineage. LILRs and the novel Ig-like receptors are expressed across a range of diverse tissues [108]. These receptors are differentially expressed on NK cells depending upon the animal’s MHC haplotype.

17.13 B cells and immunoglobulins 17.13.1 Sheep Sheep use both recombination and gene conversion rather than random somatic mutation to generate most of their immunoglobulin diversity [109]. Immature B cells first diversify their V (D) and J genes in central lymphoid tissues such as the spleen or bone marrow. The immature cells then migrate to follicles in the ileal Peyer’s patches where further diversification occurs [110]. Sheep have only seven functional IGHV genes and probably, therefore, use gene conversion to diversify their heavy chains. They have six IGHJ genes, two of which are pseudogenes. One of the active genes, IGHJ1, is used in 90% of expressed heavy chains, suggesting that recombination is minimal. More than 98% of all rearrangement events are in-frame, and there are few N- or P-nucleotides. Unlike rabbits, humans, or mice, stimulation by the intestinal microbiota is not necessary for V gene diversification in sheep. The immunoglobulin subclasses of sheep are similar to those of cattle with IGHG genes coding for IgG1, IgG2, and IgG3. Some sheep have an IgG1a allotype. An IGHD gene has been detected in sheep. Three IgA heavy chain allotypes have been identified, as have three IgE allotypes. Among the light chain genes, five Vλ genes account for .70% of the repertoire [109]. The sheep lambda locus contains more than 90 IGLV genes and a single IGLJ gene, so these are diversified by recombination.

17.13.2 Goats As expected, there is a broad similarity between the antibody loci in cattle, and sheep and those in the domestic goat. The goat heavy chain locus is found on chromosome 21q24. Thirty-four IGHV genes have been identified but only three of these IGHV genes appear to be functional. These functional genes belong to clan II as in cattle and sheep. The

274

SECTION | 2 Mammalian orders

nonfunctional V genes are disrupted by frameshifts, premature stop codons, and disrupted promoters. In addition, they are all nearly identical to one another. Goats also possess four D gene segments (three functional) and six IGHJ gene segments (two functional) and DH-DH fusion is not infrequent. It can be detected in 8% of sequences studied [111]. Interestingly these can be arranged in the usual order 50 DH 2 30 DH, inverted (30 DH 2 50 DH), or even paired identical DH1 gene segments. Neither sheep nor goats express the ultralong heavy chain D gene segments that are present in cattle. Goats possess seven heavy chain constant region genes, IGHM, IGHD, three IGHG genes, IGHE and IGHA. The IGHG genes differ in their hinge regions and share ancestry with cattle and sheep. Their products are predicted to be functionally equivalent to those in cattle. The goat IGL locus spans 460 kb and is located on chromosome 17. The 63 IGLV genes are separated into two clusters as in cattle and sheep. Of these 25 appear to be functional It contains two Jλ-Cλ cassettes. The goat IGK locus is located on chromosome 11. It contains only 15 IGKV genes and four IGKJ genes. Light chain usage is more balanced in the goat with kappa chains constituting 20%
35% of the B cell antigen receptors compared to only 5% in cattle [112].

17.14 T cells and cell-mediated immunity 17.14.1 Sheep Sheep T cells express WC1 (also called T19). The isoform of this molecule expressed on α/β T cells differs from that on γ/δ T cells. In newborn lambs, γ/δ T cells account for 60% of blood T cells, but this drops to 30% by 1 year and 5% by 5 years of age. The genes encoding these γ/δ T cells show much less diversity than the genes encoding their α/β T cells [85].

17.14.2 TRA/D The sheep TRA/D locus is located on chromosome 7 and spans 2882 kb [113]. It contains 277 plus 16 non-localized TRAV genes, more than 70 plus 18 non-localized, TRDV genes, nine TRDD genes, 79 plus 1 non-localized, TRAJ genes, and four TRDJ genes, and one TRAC and TRDC constant gene (Fig. 17.12). As in other mammals, the TRD locus is nested within the TRA locus [114]. The TRA/D locus contains a cluster of TRAV genes interspersed with TRDV genes, followed by seven TRDD gene segments, four TRDJ segments, one TRDC followed by a single TRDV gene in inverted orientation, and at the 30 end, a single TRAC gene segment. Up to four TRDD genes may be used in a single transcript making the δ chain in some receptors significantly longer than is usual. This also supports the idea that γ/δ T cells play a significant role in the immune defenses of sheep.

17.14.3 TRB The sheep TRB locus is located on chromosome 4 and is 506 kb in size. It contains 94 TRBV, three TRBD, 23 TRBJ, and three TRBC genes. It has a similar arrangement to other species with three -D-J-C- clusters downstream of the

TRA/D

TRB

5'

347 TRA/DV 94

9 DD

4 DJ DC

3 7

8 C1

TRG

5 C5

C3 TRG1

C2

2

2 C4

AC

8 C3

2

7

FIGURE 17.12 The structure of the sheep T cell antigen receptor loci.

3'

79 AJ

C1

C2 TRG2

C6

Ruminants: cattle, sheep, and goats Chapter | 17

275

cluster of V genes followed by a single TRBV3 gene in reverse orientation. This is the same organization as seen in the pig [76].

17.14.4 TRG Sheep possess two TRG paralogous loci widely separated on chromosome 4. One locus, TRG1 is in a location homologous to the human TRG locus. The second locus, TRG2 is unique to sheep. Both these TRG loci are functional [115]. This two-locus arrangement in sheep is also present in cattle, water buffalo, and goats. The number of TRGV genes in sheep is 4 or 5.

17.14.5 Goats Both the TRG and TRD loci in goats have a similar organization to those in cattle and their gene sequences are well conserved. The number of functional genes varies slightly as a result of mutations. Just as WC11 γ/δ T cells in cattle use only a single cassette, the same restriction applies to goats. As in cattle, goat γ/δ T cells can incorporate multiple TRDD gene sequences into single transcripts [116]. Analysis of the TCR chain genes in the goat has indicated that there has been a significant expansion in the number of TRB and TRG genes in this species. There are two paralogous TRG loci located at two different positions on the same chromosome as a result of a split in the ancestral locus. Each locus consists of reiterated V-J-J-C cassettes with TRG2 containing an additional cassette relative to the sheep and bovine loci [117].

17.14.6 WC1 Goat WC1 is expressed exclusively on γ/δ T cells. The goat possesses 17 complete WC1 genes and up to 30 a1 or d1 SRCR domains. Thus it has many more than cattle. It also has seven different WC1 gene structures of which four are unique to goats. Two of the unique goat WC1 genes have features found in pig WC1. These differences suggest that goat WC1 may have unique functions [95]. They also suggest that the shared genes may reflect responses to shared pathogens. Goat γ/δ T cells that express WC1 range from B20% to 85% of the total γ/δ T cell population [118]. Fewer than half were classified as WC1.1 or WC1.2 indicating the existence of a third subpopulation. They proliferate in response to Leptospiral and Mycobacterial antigens. However, in these cultures, only WC1 γ/δ T cells produced IL-17 and IFN-γ.

17.15 Other species 17.15.1 Water Buffalo (Bulbalis bulbalis) The MHC of the water buffalo (BuLA) is located on the short arm of autosomal chromosome 2. Its organization is similar to that in cattle. Thus, its class II region is also divided into two separate segments (class IIa and class IIb). The class IIb segment is separate from the other three MHC regions. Water buffalos have two DRA alleles (as have yaks) [49]. DRB3 in this species is polymorphic with 6-11 alleles identified in different populations. Buffalo also have duplicated DQ genes and both DQA and DQB are polymorphic.

17.15.2 Domestic Yaks. (Bos grunniens) Yaks and American bison are believed to have diverged from domestic cattle about 2-1.5 mya. It is estimated that yaks were first domesticated about 7000 years ago. They are important livestock in the Himalayan, Qinghai, and Tibet regions of Central Asia. The immunoglobulin loci of the domestic yak have recently been characterized. Thus, their IGH locus is located on chromosome 17. It spans about 475 kb and contains 42 IGHV genes, 55 IGHD genes, and three IGHJ genes as well as one M, one D, three G, one E, and one A constant region genes. The IGHV genes belong to all three clans and are most closely related to those of cattle [119]. As in cattle, some yak immunoglobulins possess an ultra-long CDR3 region. It contains up to 43 amino acids. The yak IGK locus is found on chromosome 9 and is 175 kb in length. It contains 23 IGKV genes of which nine are functional. It also contains 5 IGKJ genes. The IGL locus is found on chromosome 16 and is 1287 kb in length. It contains 45 IGLV genes of which 18 are functional, nine IGLJ genes and eight IGLC genes. Like cattle, yaks generate their immunoglobulin diversity primarily by somatic hypermutation resulting in junctional variations [119].

276

SECTION | 2 Mammalian orders

References [1] Chen L, Qiu Q, Jiang Y, Wang K, et al. Large-scale ruminant genome sequencing provides insights into their evolution and distinct traits. Science 2019;. Available from: https://doi.org/10.1126/science.aav6202. [2] Edwards EJ, Osborne CP, Stro¨mberg CAE, Smith SA, et al. The origins of C4 grasslands: integrating evolutionary and ecosystem. Science 2010;328:587
91. [3] Cerling LE, Ehleringer JR, Harris JM. Carbon dioxide, starvation, the development of C4 ecosystems, and mammalian evolution. Phil Trans Roy Soc Lond B 1998;353:159
71. [4] Bobe R, Behrensmeyer AK. The expansion of grassland ecosystems in Africa in relation to mammalian evolution and the origin of the genus Homo. Palaeogeog, Palaeoclimat, Palaeoecol 2004;207:399
420. [5] Tizard IR. Veterinary immunology: an introduction. 10th ed. St. Louis: Elsevier; 2017. [6] Cervenak J, Kacskovics I. The neonatal Fc receptor plays a crucial role in the metabolism of IgG in livestock animals. Vet Immunol Immunopathol 2009;128:171
7. [7] Baumrucker CR, Macrina AL, Bruckmaier RM. Clostrogenesis: role and mechanism of the bovine Fc receptor of the neonate (FcRn). J Mamm Gland Biol Neoplasia 2022;. Available from: https://doi.org/10.1007/s0911-021-09506-2. [8] Murphy JM, Hagey JV, Chigerwe M. Comparison of serum immunoglobulin G half-life in dairy calves fed colostrum, colostrum replacer or administered with intravenous bovine plasma. Vet Immunol Immunopathol 2014;158:233
7. [9] Fink K, Zellweger R, Weber J, et al. Long-term maternal imprinting of the specific B cell repertoire by maternal antibodies. Eur J Immunol 2008;38:90
101. [10] Endsley JJ, Roth JA, Ridpath J, Neill J. Maternal antibody blocks humoral but not T cell responses to BVDV. Biologicals 2003;31(2):123
5. [11] Reber AJ, Hippen AR, Hurley DJ. Effects of the ingestion of whole colostrum or cell-free colostrum on the capacity of leukocytes in newborn calves to stimulate or respond in one-way mixed leukocyte cultures. Am J Vet Res 2005;66:1854
60. [12] Reber AJ, Donovan DC, Gabbard J, et al. Transfer of maternal colostral lymphocytes promotes development of the neonatal immune system I. Effects on monocyte lineage cells. Vet Immunol Immunopathol 2008;123:186
96. [13] Hagiwara K, Domi M, Ando J. Bovine colostral CD8-positive cells are potent IFN-γ-producing cells. Vet Immunol Immunopathol 2008;124:93
8. [14] Reber AJ, Donovan DC, Gabbard J, et al. Transfer of maternal colostral lymphocytes promotes development of the neonatal immune system II. Effects on neonatal lymphocytes. Vet Immunol Immunopathol 2008;123:305
13. [15] Donovan DC, Reber AJ, Gabbard JD, et al. Effect of maternal cells transferred with colostrum on cellular responses to pathogen antigens in neonatal calves. Am J Vet Res 2007;68:778
82. [16] George JW, Snipes J, Lane VM. Comparison of bovine hematology reference intervals from 1957 to 2006. Vet Clin Pathol 2010;39(2):138
48. [17] Vlasova AN, Saif LJ. Bovine immunology; Implications for dairy cattle. Front Immunol 2021;. Available from: https://doi.org/10.3389/ fimmu.2021.643206. [18] Kaushik AK, Kandavel H, Nalpathamkalam T, Pasman Y. Bovine neonate is deficient in innate immunity at birth. Mol Immunol 2021;133:101
9. [19] Novak K. Functional polymorphisms in Toll-like receptor genes for innate immunity in farm animals. Vet Immunol Immunopathol 2014;157:1
11. [20] Seabury CM, Seabury PM, Decker JE, Schnabel RD, et al. Diversity and evolution of 11 innate immune genes in Bos taurus taurus and s cattle. Proc Natl Acad Sci 2010;107(1):151
6. [21] Vrentas CE, Schaut RG, Boggiatto PM, Olsen SC, et al. Inflammasomes in livestock and wildlife: insights into the intersection of pathogens and natural host species. Vet Immunol Immunopathol 2018;201:49
56. [22] Elsik CG, Tellam RL, Worley KC, et al. The genome sequence of taurine cattle: a window to ruminant biology and evolution. Science 2009;324:522
7. [23] Bagnicka E, Strzalkowska N, Jozwik A, Krzyzewski J, et al. Expression and polymorphism of defensins in farm animals. Acta Biochem Pol 2010;57(4):487
97. [24] Meade KG, Cormican P, Narciandi F, Lloyd A, O’Farrelly C. Bovine β-defensin gene family: opportunities to improve animal health. Physiol Genomics 2014;46(17):27. [25] Ehrlich P, Sachs H. Ueber den mechanismus der ambocestorenwirking. Berl Klin Wschr 1902;39:492
6. [26] Sliwa-Dominial J, Tokarz-Deptula B, Niedzwied-Rystwej P, Deptula W. Conglutinin
an important element in natural immunity in ruminants (A review). Acta Vet Brno 2010;79:99
104. [27] Dec M, Wernicki A, Puchalski A, Urban-Chmiel R. Conglutinin is not specific to cattle. Vet Med 2011;56(10):510
19. [28] Walker AM, Roberts RM. Characterization of the bovine type I IFN locus: rearrangements, expansions, and novel subfamilies. BMC Genet 2009;. Available from: https://doi.org/10.1186/1471-2164/10/187. [29] Guo Y, Song Z, Cheng X, Wang Y, et al. Molecular and functional characterization of Ovis aries IFN-epsilon. Mol Immunol 2020;119:1
7. [30] Bazer FW, Ying W, Wang X, Dunlap KA, et al. The many faces of interferon tau. Amino Acids 2015;47(3):449
60. [31] Tanikawa N, Seno K, Kawahara-Miki R, Kimura K, et al. Interferon tau regulates cytokine production and cellular function in human trophoblast cell line. J Interferon Cytokine Res 2017;. Available from: https://doi.org/10.1089/jir.2017.0057. [32] Roberts RM. Interferon-tau, a type 1 interferon involved in maternal regulation of pregnancy. Cytokin Growth Factor Rev 2007;18 (5
6):403
8.

Ruminants: cattle, sheep, and goats Chapter | 17

277

[33] Widdison S, Coffey TJ. Cattle and chemokines: evidence for species specific evolution of the bovine chemokine system. Anim Genet 2011;. Available from: https://doi.org/10.1111/j.1365-2052.2011.02200.x. [34] Gangur V, Birmingham NP, Thanesvorakul S. Chemokines in health and disease. Vet Immunol Immunopathol 2002;86:127
36. [35] Sisson S, Grossman JD. The anatomy of the domestic animals. 4th ed. Philadelphia: Saunders; 1953. [36] Udroiu I, Sgura A. The phylogeny of the spleen. Quart Rev Biol 2017;92(4):411
43. [37] Galeotti M, Sarli G, Eleni C, Marcato PS. Identification of cell types present in bovine haemolymph nodes and lymph nodes by immunostaining. Vet Immunol Immunopathol 1993;36:319
31. [38] Thorp BH, Seneque S, Staute K, Kimpton WG. Characterization and distribution of lymphocyte subsets in sheep hemal nodes. Dev Comp Immunol 1991;15:393
400. [39] Casteleyn C, Breugelmans S, Simoens P, Van, den Broeck W. The tonsils revisited: review of the anatomical localization and histologic characterization of the tonsils of domestic and laboratory animals. J Immunol Res 2011;. Available from: https://doi.org/10.1155/2011/472-460. [40] Breugelmans S, Casteleyn C, Simoens P, Van den Broeck W. Distribution of the lingual lymphoid lymphoid tissue in domestic ruminants. Anat Histol Embryol 2011;. Available from: https://doi.org/10.1111/j.1439-0264.2011.01088.x. [41] Liebler-Tenorio EM, Pabst R. MALT structure and function in farm animals. Vet Res 2006;37:257
80. [42] Fouhse JM, Smiegielski L, Tuplon M, Guan LL, Willing BP. Host immune selection of rumen bacteria through salivary secretory IgA. Front Microbiol 2017;. Available from: https://doi.org/10.3389/fmicb.2017.00848. [43] Chuluunbaatar I, Ichii O, Masum MA, Namba T, et al. Genital organ-associated lymphoid tissues arranged in a ring in the mucosa of cow vaginal vestibules. Res Vet Sci 2022;145:147
58. [44] Birch J, Sanjuan CDJ, Guzman E, Ellis SA. Genomic localization and characterization of MHC genes in cattle. Immunogenetics 2008;60:477
83. [45] Ellis SA, Hammond JA. The functional significance of cattle major histocompatibility complex class I genetic diversity. Ann Rev Anim Biosci 2014;. Available from: https://doi.org/10.1146/annurev-animal-022513-114234. [46] Gutierrez SE, Esteban EN, Lutzelschwab CM, Juliarena MA. Major histocompatibility complex-associated resistance to infectious diseases: the case of bovine leukemia virus infection. Trends Adv Veterinary Genet. Available from: https://doi.org/10.5772/65804. [47] Holmes EC, Roberts AFC, Staines KA, Ellis SA. Evolution of major histocompatibility complex class I genes in cetartiodactyls. Immunogenetics 2003;55:193
202. [48] Birch J, Codner G, Guzman E, Ellis SA. Genomic location and characterization of nonclassical MHC class I genes in cattle. Immunogenetics 2008;60:267
73. [49] Behl JD, Verma NK, Tyagi N, Mishra P, et al. The major histocompatibility complex in bovines: a review. ISRN Veterinary Sci 2012; 10.5402;2012/872710. [50] McQueen KL, Wilhelm BT, Harden KD, Mager DL. Evolution of NK receptors: a single Ly49 and multiple KIR genes in the cow. Eur J Immunol 2002;32:810
17. [51] Allan AJ, Sanderson ND, Gubbins S, Ellis SA, Hammond JA. Cattle NK cell heterogeneity and the influence of MHC class I. J Immunol 2015;195:2199
206. [52] Sanderson ND, Norman PJ, Guethlein LA, Ellis SA, et al. Definition of the cattle killer cell Ig-like receptor gene family: comparison with aurochs and human counterparts. J Immunol 2014;193:6016
30. [53] Hao L, Klein J, Nei M. Heterogeneous but conserved natural killer receptor gene complexes m four orders of mammals. Proc Natl Acad Sci USA 2006;103(9):3192
7. [54] Birch J, Ellis SA. Complexity in the cattle CD94/NKG2 gene families. Immunogenetics 2007;59:273
80. [55] Schwartz JC, Gibson MS, Heimeier D, Koren S. The evolution of the natural killer complex: a comparison between mammals using new highquality genome assemblies and targeted annotation. Immunogenetics 2017;69:255
69. [56] Dobromylskyj MJ, Connelley T, Hammond JA, Ellis SA. Cattle Ly49 is polymorphic. Immunogenetics 2009;61:789
95. [57] Boysen P, Storset AK. Bovine natural killer cells. Vet Immunol Immunopathol 2009;130:163
77. [58] Elhmouzi-Younes J, Storset AK, Boysen P, Laurent F, Drouet F. Bovine neonatal natural killer cells are fully functional and highly responsive to interleukin-15 and to NKp46 receptor stimulation. Vet Res 2009;40(6):1
14. [59] Kulberg S, Boysen P, Storset AK. Reference values for relative numbers of natural killer cells in cattle blood. Dev Comp Immunol 2004;28:941
8. [60] Hamilton CA, Mahan S, Bell CR, Villareal-Ramos B, et al. Frequency and phenotype of natural killer cell subsets in bovine lymphoid compartments and blood. Immunology 2017;151:89
97. [61] Chen J, Huddleston J, Buckley RM, Malig M, et al. Bovine NK-lysin: copy number variation and functional diversification. Proc Natl Acad Sci U S A 2015;112:E7223
9. [62] Chen J, Yang C, Tizioto PC, Huang H, et al. Expression of the bovine NK-lysin gene family and activity against respiratory pathogens. PLoS One 2016;11:e0158882. [63] Hope JC, Sopp P, Howard CJ. NK-like CD81 cells in immunologically naive neonatal calves that respond to dendritic cells infected with Mycobacterium bovis BCG. J Leukoc Biol 2002;71:184
94. [64] Summerfield A, Auray G, Ricklin M. Comparative dendritic cell biology of veterinary mammals. Ann Rev Anim Biosci 2015;. Available from: https://doi.org/10.1146/annurev-animal-022114-11109. [65] Ballingall KT, MacHugh ND, Taracha ELN, Mertens B, McKeever DJ. Transcription of the unique ruminant class II major histocompatibility complex-DYA and DIB genes in dendritic cells. Eur J Immunol 2001;31:82
6.

278

SECTION | 2 Mammalian orders

[66] Ma L, Qin T, Chu D, Cheng X, et al. Internal duplications of DH, JH and C region genes create an unusual IgH gene locus in cattle. J Immunol 2016;196:4358
66. [67] Wan Z, Zhao Y, Sun Y. Immunoglobulin D and its encoding genes: an updated review. Dev Comp Immunol 2021;. Available from: https://doi. org/10.1016/j.dci.2021.104198. [68] Zhao Y, Kacskovics I, Pan Q, Liberles DA, et al. Artiodactyl IgD: the missing link. J Immunol 2002;169:4408
16. [69] Rabbani H, Brown WR, Butler JE, Hammarstro¨m L. Polymorphism of the IGHG3 gene in cattle. Immunogenetics 1997;46:326
31. [70] Liljavirta J, Niku M, Pessa-Morikawa T, Ekman A, Livanainen A. Expansion of the preimmune antibody repertoire by junctional diversity in Bos taurus. PLoS One 2014;. Available from: https://doi.org/10.1371/journal.pone.0099808. [71] Oyola SO, Henson SP, Nzau B, Kibwana E, Nene V. Access to ultra-long IgG CSRH3 bovine antibody sequences using short read sequencing technology. Mol Immunol 2021;139:97
105. [72] Deiss TC, Vadnais M, Wang F, Chen PL, et al. Immunogenetic factors driving formation of ultralong VH CDR3 in Bos taurus antibodies. Cell Mol Immunol 2019;16:53
64. [73] Wang F, Herzig CT, Chen C, et al. Scavenger receptor WC1 contributes to the γ/δ T cell response to Leptospira. Mol Immunol 2011;48:801
9. [74] Ekman A, Niku M, Liljavirta J, Livanainen A. Bos taurus genome sequence reveals the assortment of immunoglobulin and surrogate light chains in domestic cattle. BMC Immunol 2009;. Available from: https://doi.org/10.1186/1471-2172-10-22. [75] Pasman Y, Saini SS, Smith E, Kaushik AK. Organization and genomic complexity of bovine λ-chain locus. Vet Immunol Immunopathol 2010;135:306
13. [76] Pe´gorier P, Bertignac M, Ngoune VN, Folch G, et al. IGMT biocuration and comparative analysis of Bos taurus and Ovis aries TRA/TRD loci. Genes 2021;. Available from: https://doi.org/10.3390/genes12010030. [77] Gillispie A, Gervasi MG, Sathiyaseelan T, Connelley T, et al. Gamma delta TCR and the WC1 coreceptor interactions in response to Leptospira using imaging flow cytometry and STORM. Front Immunol 2021;. Available from: https://doi.org/10.3389/fimmu.2021.712123. [78] Reinink P, van Rhijn I. The bovine T cell receptor alpha/delta locus contains over 400 v genes and encodes V genes without CDR2. Immunogenetics 2009;61:541
9. [79] Herzig C, Blumerman S, Lefranc M-P, Baldwin C. Bovine T cell receptor gamma variable and constant genes: combinatorial usage by circulating γδT cells. Immunogenetics 2006;138
51. [80] Conrad ML, Pettman R, Whitehead J, McKinnel L, et al. Genomic sequencing of the bovine T cell receptor beta locus. Vet Immunol Immunopathol 2002;87:439
41. [81] Connelley T, Aerts J, Law A, Morrison WI. Genomic analysis reveals extensive gene duplication within the bovine TRB locus. BMC Genomics 2009;. Available from: https://doi.org/10.1186/1471-2164-10-192. [82] Antonacci R, Vacccarelli G, Di Meo GP, Piccinni B, et al. Molecular in situ hybridization analysis of sheep and goat BAC clones identifies the transcriptional orientation of T cell receptor gamma genes on chromosome 4 in bovids. Vet Res Comm 2007;31:977
83. [83] Herzig CTA, Lefranc M-P, Baldwin CL. Annotation and classification of the bovine T cell receptor delta genes. BMC Genomics 2010;. Available from: https://doi.org/10.3389/1471-2164-/11/100. [84] Vaccarelli G, Miccoli MC, Antonacci R, et al. Genomic organization and recombinational unit duplication-driven evolution of ovine and bovine T cell receptor gamma loci. BMC Genomics 2008;. Available from: https://doi.org/10.1186/1471-2164-9-81. [85] Baldwin CL, Damani-Yokota P, Yirsaw A, Loonie K, et al. Special features of γδT cells in ruminants. Mol Immunol 2021;134:161
9. [86] Baldwin CL, Hsu H, Chen C, Palmer M, et al. The role of bovine γδ T cells and their WC1 co-receptor in response to bacterial pathogens and promoting vaccine efficacy: a model for cattle and humans. Vet Immunol Immunopathol 2014;159:144
55. [87] Van Beeck FAL, Reinink P, Hermsen R, Zajonc D, et al. Functional CD1d and/or NKT invariant chain transcript in horse, pig, African elephant and Guinea pig, but not in ruminants. Mol Immunol 2009;46:1424
31. [88] Nguyen TKH, Reinink P, El Messiaki C, Im JS, et al. Expression profiles of bovine CD1 in vivo and assessment of the specificities of the antibovine CD1 antibodies. PLoS One 2015;. Available from: https://doi.org/10.1371/journal.pone.0121923. [89] Baldwin CL, Telfer JC. The bovine model for elucidating the role of gammadelta T cells in controlling infectious diseases of importance to cattle and humans. Mol Immunol 2015;66(1):35
47. [90] Guzman E, Hope J, Taylor G, Smith AL, et al. Bovine γδ T cells are a major regulatory T cell subset. J Immunol 2014;193:208
22. [91] Hsu H, Chen C, Nenninger A, Holz L, et al. WC1 is a hybrid γδTCR coreceptor and pattern recognition receptor for pathogenic bacteria. J Immunol 2015;. Available from: https://doi.org/10.4049/jimmunol.1402021. [92] Telfer JC, Baldwin CL. Bovine gamma delta T cells and the function of gamma delta T cell specific WC1 co-receptors. Cell immunol 2015;296(1):76
86. [93] Herzig CTA, Waters RW, Baldwin CL, Telfer JC. Evolution of the CD163 family and its relationship to the bovine gamma delta T cell coreceptor WC1. BMC Evol Biol 2010;10:181
C200. [94] Loonie K, Gillespie A, Baldwin C. The WC1 γδ T cell pathogen receptor of ruminants is preserved in the genome of the ancient extinct auroch. Immunogenetics 2021;. Available from: https://doi.org/10.1007/s00251-021-01211-y. [95] Yirsaw AW, Gillespie A, Zhang F, Smith TPL, et al. Defining the caprine γδ T cell WC1 multigenic array and evaluation of its expressed sequences and gene structure among goat breeds and relative to cattle. Immunogenetics 2022;. Available from: https://doi.org/10.1007/s00251022-01254-9.

Ruminants: cattle, sheep, and goats Chapter | 17

279

[96] Damani-Yokota P, Zhang F, Gillespie A, Park H, et al. Transcriptional programing and gene regulation in WC11 γδ T cell populations. Mol Immunol 2022;142:50
62. [97] Cahill RNP, Kimpton W, Dudler L, Trnka Z. Lymphopoiesis in foetal and perinatal sheep. In: Morris B, Miyasaka M, editors. Immunology of the sheep. Edition Roche Basle Switzerland; 1984. [98] Pfeffer A, Shaw RJ, Green RS, Phegan MD. The transfer of maternal IgE and other immunoglobulins specific for Trichostrongylus colubriformis larval excretory/secretory product to the neonatal lamb. Vet Immunol Immunopathol 2005;108:315
23. [99] Micusan VV, Borduas AG. Biological properties of goat immunoglobulins G. Immunology 1977;32:373
81. [100] Guo Y, Song Z, Li C, Yu Y, et al. A novel type-1 interferon family, bovine interferon-chi, is involved in positive feedback regulation of interferon production. Front Immunol 2020;. Available from: https://doi.org/10.3389/fimmu.2020.528854. [101] Kararli TT. Comparison of the gastrointestinal anatomy, physiology, and biochemistry of humans and commonly used laboratory animals. Biopharm Drug Dispos 1995;16:351
80. [102] Gao J, Liu K, Liu H, Blair HT, et al. A complete DNA sequence map of the ovine major histocompatibility complex. BMC Genomics 2010;. Available from: https://doi.org/10.1186/1471-2164/11/466. [103] Li C, Chen L, Liu X, Shi X, et al. A high-density BAC physical map covering the entire MHC region of addax antelope genome. BMC Genomics 2019;. Available from: https://doi.org/10.1186/s12864-019-5790-2. [104] Dukkipati VSR, Blair HT, Garrick DJ, Murray A. “Ovar-MHC”—ovine major histocompatibility complex: structure and gene polymorphisms. Gen Mol Res 2006;5:581
608. [105] Ballingall KT, Miltiadou D, Chai ZW, et al. Genetic and proteomic analysis of the MHC class I repertoire from four ovine haplotypes. Immunogenetics 2008;60:177
84. [106] Herrmann-Hoesing LM, White SN, Kappmeyer LS, et al. Genomic analysis of Ovis aries (Ovar) MHC class IIa loci. Immunogenetics 2008;60:167
76. [107] Dong Y, Xie M, Jiang Y, Xiao N, et al. Sequencing and automated whole-genome optical mapping of the genome of a domestic goat (Capra hircus). Nat Biotech 2013;31(2):135
41. [108] Schwartz JC, Sanderson ND, Bickhart DM, Smith TPL, Hammond JA. The structure, evolution, and gene expression within the caprine leukocyte receptor complex. Front Immunol 2019;. Available from: https://doi.org/10.3389/fimmu.2019.02302. [109] Jenne CN, Kennedy LJ, McCullagh P, Reynolds JD. A new model of sheep Ig diversification: shifting the emphasis toward combinatorial mechanisms and away from hypermutation. J Immunol 2003;170:3739
50. [110] Reynaud C-A, Mackay CR, Mu¨ller RG, Weill J-C. Somatic generation of diversity in a mammalian primary lymphoid organ: the sheep ileal Peyer’s patches. Cell 1991;64:995
1005. [111] Du L, Wang S, Zhu Y, Zhao H, et al. Immunoglobulin heavy chain variable region analysis in dairy goats. Immunobiology 2018;223:599
607. [112] Schwartz JC, Philp RL, Bickhart DM, Smith TPL, Hammond JA. The antibody loci of the domestic goat (Capra hircus). Immunogenetics 2017;. Available from: https://doi.org/10.1007/s00251-017.1033-3. [113] Pe´gorier P, Bertignac M, Chentli I, Ngoune VN, et al. IMGT biocuration and comparative study of the T cell receptor beta locus of veterinary species based on Homo sapiens TRB. Front Immunol 2020;. Available from: https://doi.org/10.3389/fimmu.2020.00821. [114] Piccinni B, Massari S, Jambrenghi AC, Giannico F, et al. Sheep (Ovis aries) T cell receptor alpha (TRA) and delta (TRD) genes and genomic organization of the TRA/TRD locus. BMC Genomics 2015;. Available from: https://doi.org/10.1186/s12864-015-1790-z. [115] Massari S, Lipsi MR, Vonghia G, Antonacci R, Ciccarese S. T cell receptor TCRG1 and TCRG2 clusters map separately in two different regions of sheep chromosome 4. Chromosome Res 1998;6:419
20. [116] Gillespie A, Yorsaw A, Gunasekaran KP, Smith TP, et al. Characterization of the domestic goat γδT cell receptor gene loci and gene usage. Immunogenetics. 2021;73:187
201. [117] Giannico F, Massari S, Jambrenghi AC, Soriano A, et al. The expansion of the TRB and TRG genes in domestic goats (Capra hircus) is characteristic of the ruminant species. BMC Genomics 2020;. Available from: https://doi.org/10.1186/s12864-020-07022-x. [118] Yirsaw AW, Gillespie A, Britton E, Doerle E, et al. Goat γδ T cell subpopulations defined by WC1 expression, responses to pathogens and cytokine production. Dev Comp Immunol 2021. Available from: https://doi.org/10.1016/j.dci.2020.103984. [119] Wu M, Zhao H, Tang X, Zhao W, et al. Organization and complexity of the Yak (Bos grunniens) immunoglobulin loci. Front Immunol, 2022. Available from: https://doi.org/10.3389/fimmu.2022.876509.

This page intentionally left blank

Chapter 18

Chiropterans: the bats

Indian flying fox. Pteropus medius.

Bats, the members of the order Chiroptera, are a large diverse order of nocturnal mammals that constitute about onefifth of extant mammalian species. They are the second-largest mammalian order after the rodents and are among the most geographically dispersed mammals. They consist of more than 1400 different species classified into twenty-one families. They are the only mammals capable of powered flight and have multiple other unique features. These include relatively long lifespans, regular hibernation, and the use of echolocation. They occupy diverse ecological niches in temperate and tropical environments. They have adapted to a great diversity of diets including insectivores, frugivores, nectarivores, carnivores, omnivores, and the blood-drinking sanguivores,
and they have unique immune systems! Bats are members of the superorder Laurasiatheria. There have been disputes as to both their origins and their classification. Recent genetic analysis, however, indicates that they form a sister group with a large clade of carnivores, ungulates, and cetaceans [1]. Bats are believed to have diverged from the other eutherian mammals about 80 mya. They underwent rapid diversification during the Eocene epoch (Fig. 18.1). Bats are classified into two suborders; the Yinpterochiroptera and the Yangochiroptera. The Yinpterochiroptera include the large macrobats such as the flying foxes and fruit bats as well as four microbat families including the echolocating Rhinolophoidea, the Rhinopomatidae, the Hipposideridae, and the Megadermatidae. The Yangochiroptera include all the remaining echolocating microbat families. These two suborders are believed to have diverged about 68 mya. In recent years, bats have received increased attention from immunologists since they are the natural hosts of many important zoonotic viruses. They harbor more viruses than any other group of mammals but usually show no overt signs of disease. Thus, they carry filoviruses such as Marburg and Ebola, coronaviruses such as SARS and MERS, and henipahviruses such as Hendravirus and Nipahvirus. Viruses such as the SARS coronavirus can cause lethal disease in humans and domestic mammals while apparently causing minimal disease in the bats themselves. These viruses are not known to cause mass die-offs or reduce bat longevity. Indeed, bats appear to harbor a huge diversity of zoonotic viruses, even more than rodents, another diverse species-rich order. Egyptian fruit bats (Roussetus aegyptiacus) are the only known reservoir of Marburg viruses and Sosuga virus a pathogenic myxovirus [2]. The only viruses that regularly cause visible disease in bats are rabies and the closely related, bat lyssavirus. The results of studies to determine the reasons for this resistance suggest that multiple alterations have occurred in their antiviral innate immunity pathways [3].

Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00003-4 © 2023 Elsevier Inc. All rights reserved.

281

282

80

SECTION | 2 Mammalian orders

70

60

50

40

30

20

10

0 Million years ago Artiodactyls

FIGURE 18.1 The phylogeny of the major genera of chiropterans discussed in this chapter.

Perissodactyls M. davidii Microbats M. lucifugus Eptesicus Yangochiroptera

Pteropus Megabats Yinpterochiroptera Roussettus

There is growing evidence that bats can serve as very effective virus carriers, not by fighting them but by tolerating them. Two hypotheses have been suggested to explain why bats are more tolerant to viral pathogens than other mammals. One theory suggests that they can mount an especially potent and effective innate defensive response against these viruses. The other theory suggests that they may be immunologically tolerant to the viruses and so mount reduced innate immune responses to viral invasion. The two theories are not mutually incompatible, and mechanisms likely differ between species. Thus, Egyptian fruit bats are carriers of the Marburg virus and mount a detectable antibody response against it, but the antibodies have no neutralizing activity [3]. Recent analyses have indicated that bats show evidence of positive selection for genes whose products are involved in DNA repair and innate immunity. It has also been speculated that adaptation to flight may have had inadvertent consequences for innate immune system pathways. It is well recognized that bats have relatively long life-spans relative to their body size. One possible reason for this is that the power of flight enables bats to escape predators relatively easily compared to ground-dwelling mammals. However, it may also be that modified innate immune responses also enable bats to survive in the face of viral challenges that kill other species. Bats live for very much longer than similarly sized terrestrial mammals such as rodents or shrews. They live for a similar length of time as humans and non-human primates. They are very much K strategists, carefully rearing a limited number of young. Some of this longevity may be attributed to seasonal hibernation but even non-hibernating bat species live three times longer than other mammals of the same weight. Bats also cluster away from other mammals when their microbiota is examined, it most closely resembles that of small songbirds.

18.1

Reproduction and lactation

Bats make a major maternal investment in pregnancy. Some species such as the Lesser Mouse-tailed Bat (Rhinopoma hardwickii) have an endotheliochorial placenta. Most species of bats however have a hemochorial placenta similar to that in humans and lagomorphs [4]. Epitheliochorial placentation has not been described in bats. In view of their hemochorial placentation, it would be anticipated that their developing young would receive the bulk of their passive protection in utero as a result of transplacental immunoglobulin transfer [5]. However, in the case of the well-studied fruit bats (Pteropus spp.), significant amounts of immunoglobulins are passively transferred in milk [5]. For example, following vaccination with canine distemper virus, the Variable Flying Fox, Pteropus hypomelanus transferred antibodies in her milk for an average of 228.6 days. These antibodies had a mean half-life of 96 days. In a related species, the Black Flying Fox (P. alecto), maternal antibodies waned over 255 days with a half-life of 52 days. Female bats tend to have a single offspring each year and neonatal bat pups are relatively large and precocious. While female bats usually undergo lactation and feed their offspring, in some species such as the Dayak Fruit Bat (Dyacopterus spadiceus), the males also lactate [6]. The predominant immunoglobulin in P. alecto milk is IgG and it has been suggested that neonatal fruit bats get the majority of their IgG through lacteal secretions rather than transplacentally [7].

Chiropterans: the bats Chapter | 18

18.2

283

Hematology

White blood cell numbers in the Greater Sac-winged Bat (Saccopteryx bilineata) decrease with age [8]. Total white blood cell counts performed on common neotropical bat species from Costa Rica showed some interesting differences [9]. Bat species in the same taxonomic family had comparable white cell counts. In general, counts were lower in insectivorous, emballonurid, mossolid, and vespertilionod species when compared to the mostly phytophagous phyllostomid bats. One species, the Honduran White Bat (Ectophylla alba) exhibited exceptionally low white cell counts 876 6 166/μL which was less than half that in other bats. These ranged from 1.7 6 0.29 3 103/μL in the Bonda Mastiff Bat (Molossus bondae) to 7.3 6 15 3 103/μL in the Fringe-lipped Bat (Trachops cirrhosis). Blood cell morphology is similar to other mammals but some differences in lymphocyte morphology may be observed, most notably the presence of large granular lymphocytes,
perhaps circulating NK cells [9].

18.3

Innate immunity

In investigating the reasons why bats carry such a remarkable number of viruses, yet remain apparently healthy and longlived, attention has largely focused on their innate immune defenses. During infections, the innate immune response is the first line of defense. It also serves to activate adaptive immune responses. The presence of invading viruses is usually first detected through their binding to diverse pattern recognition receptors. These receptors are located on cell surfaces, in the cytosol, and within endosomal compartments. Among the PRRs some, most notably TLRs 3, 7, 8, and 9 as well as retinoic inducible gene-1 (RIG-1), and melanoma differentiation-associated protein 5 (MDA5), are structured to recognize foreign nucleic acids. Once these receptors bind a foreign nucleic acid, they signal through two major pathways. One pathway triggers inflammation whereas the other pathway triggers an antiviral response by generating diverse interferons. Both pathways differ in bats from other mammals and almost certainly contribute to their unusual response to viral infections.

18.3.1 Pattern recognition receptors Bats have eleven toll-like receptor genes (TLR). Like most other mammals they encode TLR1 thru TLR10. However, they also possess a gene for TLR13 that may be a pseudogene. Jiang et al. have examined the bat nucleic acid-sensing TLR genes, TLR3, -7, -8, and -9 for evidence of positive selection. It is clear that positive selection has occurred in the lineages of TLRs -7 and -8 but it is especially obvious in the TLR9 gene. The most positively selected site is located in the ligand-binding region of TLR9 as shown in seven different bat species [10]. It is probably significant that the sites under positive selection are restricted to those TLRs that are responsible for recognizing intracellular PAMPS. In other words, viral nucleic acid binding sites with the exception of TLR3 are positively selected. All the bat TLRs are very similar to their homologs in other mammals except for TLR9. TLR9 recognizes viral DNA as well as unmethylated CpG motifs. However, the bat TLR9 is very different from all the other mammalian TLRs across multiple genera [8]. Thus, unique nonconservative mutations are found in the ligand-binding site of TLR9. They probably affect its binding specificity or affinity and thus affect innate resistance to some viruses. Examination and analysis of the TLR8 gene in 21 bat species have indicated that it also displays extensive sequence variation and is persistently positively selected in its ligand-binding domain [11]. TLR8 binds single-stranded RNA and is also an important component of antiviral immunity. The TLR pseudogene for TLR13 in P. alecto is about 78% identical to mouse TLR13. A similar pseudogene sequence has been found in the bovine genome with 83% identity at the DNA level [12]. The minor mutations in the TLR13 genes suggest a recent loss of function. Thus, the nearly intact gene is transcribed, spliced, and polyadenylated. When knocked out in the mouse, the animals show increased susceptibility to vesicular stomatitis virus [13]. Two NOD-like pattern recognition receptors have also been found in the P. alecto transcriptome [14]. These are NOD-like receptors, CARD domain-containing 5 (NLRCX5), and NLR family pyrin domain-containing 3 (NLRP3). NLRC5 is a regulator of antiviral immune responses, NLRP3 is activated by diverse DAMPs and environmental irritants. Activated NLRP3 activates caspase 1 in the inflammasome and hence triggers activation of inflammatory cytokines such as IL-1 and TNF-α. RIG-1 and MDA5 from P. alecto have similar structures and tissue expression patterns as their human counterparts [15]. In addition, MHC class I, and interferon regulatory factor 7, have been characterized in P. alecto.

18.3.2 Inflammatory responses It is important to bear in mind that bat species are highly diverse so pathways and genes present or activated in one species may not be relevant in others. While most studies on antiviral immunity have been performed in P. alecto, other

284

SECTION | 2 Mammalian orders

FIGURE 18.2 Some of the unique features of bat innate immune pathways that may affect their responses to viral infections.

Viral infection ds RNA

PRR signaling Elevated c-rel Increased IFND expression

Suppressive micro RNAs Reduced TNF-D expression IL-8 production

Loss of pentraxins CRP and SAP

REDUCED INFLAMMATION

studies on R. agyptiacus or David’s Myotis (Myotis davidii) have provided information on alternative unique pathways and processes (Fig. 18.2). The inflammatory and interferon responses that are activated by double-stranded viral RNA are triggered by the PRRs, TLR3, RIG-1, and MDA5. They act through the NF-κB- and IRF3- mediated pathways. Five members of the NF-κB protein family have been identified in bats. These are RelA, RelB, c-Rel, NF-κB1, and NF-κB2. In a normal resting cell, these inhibitors bind to NF-κB and suppress its activity. They prevent it from moving to the nucleus and activating proinflammatory genes such as TNF-α. When the cells are activated, these inhibitors detach from the NF-κB and permit it to move to the nucleus and activate cytokine genes. Different combinations of these inhibitory factors have different suppressive effects on the NF-κB-mediated inflammatory and immune pathways. When stimulated by the synthetic RNA molecule, polyinosinic: polycytidilic copolymer (poly I:C), cultured cells from the Big Brown Bat (Eptesicus fuscus) and human cells produce elevated levels of interferon β. Poly I:C is known to act by binding to TLRs 3, -7, and -8. However, only the human cells respond by producing transcripts for the proinflammatory cytokines such as interleukin 8 and TNF-α. Bat cells do not. When the bat TNF-α promoter was examined, it was found that it possessed a binding motif for the repressor molecule c-rel. In studies that knocked down c-rel levels in bat cells, TNF-α production increased significantly. Likewise, it was shown that when c-rel bound to this motif, it suppressed TNF-α -mediated inflammation. c-rel could be detected in every major tissue of E. fuscus including its spleen, intestine, kidneys, lungs, and liver [15]. In the case of a very different microbat, M. davidii there are many positively selected genes associated with the DNA damage checkpoint as well as with the NF-κB pathways. These may be related to the metabolic demands placed by flight, but they are also relevant to resistance to virus-mediated disease [16]. Positively selected genes in this bat included c-rel. These bats also appear to use a novel class of NK cell receptors to recognize MHC molecules. These include SIGLECs, LILRs, Carcinoembryonic antigen-related cell adhesion molecules (CEACAMs), and LAIRs that have undergone considerable gene duplication in M. davidii but have failed to expand in P. alecto. The gene for the defensive enzyme RNASE4 has expanded to seven copies and two partial copies in M. davidii but is only a pseudogene in P. alecto. This enzyme plays a role in defense against RNA viruses. However, P. alecto is nectarivorous while M. davidii is an insectivore and perhaps they require different digestive enzymes [16].

18.3.3 Interferon pathways Just as the innate inflammatory pathways are suppressed in bats, so too are the innate antiviral pathways
in some species (Fig. 18.3). The interferons constitute the first line of defense against viral invasion. There are three major types of interferon produced by mammals: types I, II, and III. Type I and type III interferons are induced in direct response to viral infections and they, in turn, induce an antiviral state in infected and neighboring cells. Although most of the bat zoonotic viruses are RNA viruses, they harbor many DNA viruses as well. This foreign DNA is normally detected by cytosolic PRRs such as TLR7, -8, and -9. This in turn leads to inflammasome activation and the production of multiple type I interferons. The pathway that results in interferon production is triggered when viral DNA is sensed by a molecule called cyclic GMP-AMP synthase (cGAS). It responds by generating cyclic GMPAMP (cGAMP). cGAMP in turn binds to STING. (Stimulator of Interferon Genes). As a result, BK1 is recruited to STING, and this causes the phosphorylation of STING and IRF3. This in turn stimulates a potent interferon response. It

Chiropterans: the bats Chapter | 18

FIGURE 18.3 Some of the unique features of bat antiviral pathways that may enable them to act as healthy carriers for many viruses.

VIRUS INFECTION STING-dependent amplification of IFN production is dampened Constitutive interferon activation

Lack of affinity maturation in antibodies

285

Reduced IFN family size Shortened IFN response

INTERFERON RESPONSE

Expansion of antiviral APOBEC family genes

Loss of PYHIN family DNA-binding genes Limited Immune activation

Tolerance to viral infection

has been shown that STING-dependent interferon activation is dampened in some bats as a result of the replacement of an essential serine residue in that molecule [17,18]. PYHIN proteins are characterized by possessing an N-terminal pyrin domain and one or two C-terminal HIN (hematopoietic, interferon, and nuclear) domains. Proteins of the PYHIN family act as central mediators of innate immune responses triggered by cytosolic microbial DNA. PYHIN family members are DNA sensors capable of recognizing viral DNA as well as damaged self-DNA resulting from RNA virus infections. They bind the viral DNA and either generate caspase-1 activating inflammasomes or they drive type 1 interferon production [19]. Normally this recognition of foreign DNA will result in the production of type I interferons through their interaction with STING. The PYHIN family is also the only DNA sensor that can also activate inflammasomes. Significantly, the entire PYHIN gene family including AIM2 is lost in all ten bat genomes sequenced so far [20]. Thus, in the absence of PYHINS, bats will limit the severity of the inflammatory and interferon response triggered by viral invasion. The only remaining trace of this family in the bat genome is a truncated AIM2 gene. Examination of the interferon gene region of P. alecto shows a highly contracted type I IFN family [21]. Thus, it contains only ten genes and only three functional IFN-α genes. However, the three IFN-α genes are constitutively expressed in unstimulated bat cells and tissues, and their production is not enhanced by viral infection [22]. P. vampyrus also have the coding sequence for three type 1 IFNs but P. alecto has only two. The constitutively expressed IFN acts on target cells to induce the subset of interferon-stimulated genes (ISG) associated with antiviral immunity. Again, however, the size of this locus differs between bat species. For example, this type I IFN locus has expanded in the large flying Fox (P. vampyrus) and the Little Brown Bat (Myotis lucifugus) [23]. Infection of bat cells with viruses generates an IFN-ω transcript but at a much lower level than the IFN-β transcripts. This interferon gene expansion is consistent with that also seen in P. vampyricus. Bats also express both type II and III interferons [15]. Bat type III IFNs have similar antiviral activities to those in other mammals. Interestingly, however, experimental infection of P. alecto with the bat paramyxovirus, Tioman virus, resulted in no upregulation of IFN-1 production but did induce a significant type III IFN response [24]. In addition to having a limited number of IFN genes, some bats generate an interferon response with unusual kinetics [25]. Thus, IFN signaling in the cells of P. alecto results in both conserved and unique ISG expression profiles. In IFN-stimulated bat cells, there are two distinct temporal subclusters with similar early induction kinetics but with very different late-phase declines. Human cells in contrast lack this declining phase and so remain elevated for much longer periods. In unstimulated bat cells, ISG background levels are higher than their human counterparts. It appears that the antiviral effector protein 2
5A-dependent endoribonuclease (2
5OA) is not an ISG in humans but is in P. alecto and so contributes to the control of virus infections in that species. Thus, bats and humans use very different mechanisms of antiviral gene regulation. In addition to the loss of pro-inflammatory NF-κB regulators, bats may also have an expanded antiviral APOBEC 3 gene family [26]. These are cytidine deaminases that can disrupt the RNA of invading viruses. The Egyptian fruit bat (R. aegyptiacus) uses multiple antiviral pathways, some of which have not been observed in other bat species [23]. Thus, it has expanded families of KLRC/KLRD NK cell receptors, MHC class I genes, and type 1 interferon. It was believed that there were about 20 type 1 IFN genes in the megachiropteran ancestor and yet there

286

SECTION | 2 Mammalian orders

are 46 functional genes in R. aegyptiacus 2.0. These include twelve IFN-α genes, one each of IFN-β, -ε, and -λ genes, nine IFN-δ genes, and 22 IFN-ω genes. There is only one IFN-ω gene in humans. Another unique feature of R. aegyptiacus is that even healthy bats constitutively produce IFN-α at levels that are physiologically active in humans [27]. This appears to be independent of virus infections and goes far to explain their ability to coexist with many viruses without developing the disease. This “always-on” activity may effectively inhibit early viral replication. Egyptian fruit bats lack several acute-phase proteins including C-reactive protein (CRP) and serum amyloid A (SAA) [3]. This has never been observed in other mammals. In their absence, the bats may generate a weaker inflammatory and acute-phase response. In other bat genomes, the serum amyloid P (SAP) sequences also appear to be pseudogenes. Pentraxins are P-type lectins that act as soluble pattern-recognition receptors (PRRs). Two of the most important are CRP and SAP. CRP is the major acute-phase protein in primates, rabbits, hamsters, and dogs and is important in pigs. CRP has a pentameric structure (five 20-kDa units arranged as a disk) with two faces. One face binds phosphocholine, a common side chain found in all cell membranes and many bacteria and protozoa. The other face binds to the antibody receptors FcγRI and FcγRIIa on neutrophils. CRP thus acts as an opsonin and promotes the phagocytosis and removal of damaged, dying, or dead cells in addition to microorganisms. CRP can bind to bacterial polysaccharides and glycolipids and to necrotic cells, where it activates complement C1q. CRP also has an antiinflammatory role since it inhibits neutrophil superoxide production and degranulation and blocks platelet aggregation. CRP stimulates fibrosis and may promote healing by reducing damage and enhancing the repair of damaged tissue. SAP is also a pentraxin and the major acute-phase protein in rodents. Like CRP it is a PRR, where one face of the molecule can bind nuclear constituents such as DNA, chromatin, and histones as well as cell membrane phospholipids. The other face binds and activates C1q and thus triggers the classical complement pathway. A major function of SAP is to regulate innate immune responses. It interacts with macrophage Fc receptors, reduces the binding of neutrophils to the extracellular matrix, reduces the differentiation of macrophages into fibroblasts so inhibiting fibrosis, and promotes phagocytosis of cell debris. The absence of these short pentraxins in bats suggests evolutionary pressure to reduce inflammation. Bats can tolerate and survive viral infections better than most mammals. This is presumably due to immune adaptations such as those described above. Viral infections leave traces of previous infections in the form of integrated viral sequences in the genome. As a result, bat genomes contain multiple endogenous viral sequences including those from Parvoviruses, Adenoviruses, Bornaviruses, and in some species, Filoviruses. Likewise, as expected, bats possess multiple endogenous retroviral sequences, especially those from alpharetroviruses [26].

18.3.4 MicroRNA Sequencing of the genome of P. alecto has detected 400 microRNAs [28]. These play an important immunoregulatory role by selectively suppressing certain genes. Thus, in bats, they suppress innate immune responses, especially inflammation. The suppressed genes include those that play a key role in antiviral immunity, the response to DNA damage, apoptosis, and autophagy. These microRNAs may also be critically important in regulating antiviral responses in bats that serve as natural carriers of a diverse spectrum of viruses including potential human pathogens. Multiple microRNAs have also been identified in microbats such as the Little Brown Bat (M. lucifugus), the Big Brown Bat (E. fuscus), and the Jamaican Flying Fox (Artebius jamaicensis) [28].

18.3.5 Body temperature and hibernation The internal body temperature of bats can vary greatly, depending upon their degree of activity. For example, under constant temperature conditions, the body temperature of R. leschenaultia ranged from 36 C in the daytime when the bats were resting, to 39 C at night when they were active [29]. Thus, the temperature gap between rest and activity is relatively large. Although many bat species are homeothermic, some such as the vesper bats and the horseshoe bats adjust their body temperature to that of their surroundings
they are, in effect heterothermic. This drop in body temperature results in lethargy but conserves significant amounts of energy. The brain in these bats remains temperature sensitive and can initiate heat production when required for arousal. In addition to developing daily torpor, some bat species also hibernate. This requires the storage of energy in the form of fat reserves. It also requires careful selection of a suitable hibernation site that while cool, will not freeze. Huge numbers of bats may congregate in selected suitable caves. Hibernating animals forgo temperature regulation and at the same time suppress some features of their immune system. Conversely, rapid flight generates large amounts of surplus heat. As a result, during flight, a bat’s

Chiropterans: the bats Chapter | 18

287

internal temperature can rise to 40 C [8]. The “flight as fever” hypothesis suggests that this elevated body temperature during bat flight somehow mimics the protective effect of a febrile response. However, it should be pointed out that a mammalian sickness response is much more complex than simply a rise in body temperature [22].

18.3.6 Immune reconstitution inflammatory syndrome! When microbats such as the Little Brown Bat, M. lucifugus, hibernate, their body temperature drops, and they become immunosuppressed. Under some circumstances, depending upon the environment, this permits the causal agent of white-nose syndrome, a filamentous fungus called Pseudogymnoascus destructans, to colonize and erode the skin of their wings, ears, and muzzle [30]. About a week after infected bats emerge from hibernation, they develop an intense inflammatory response that leads to their death [31]. This response is estimated to have killed a million bats in the north-eastern United States during the first two years of the epizootic. It has been suggested that the sudden reversal of immune suppression that occurs when their body temperature rises, results in the development of immune reconstitution inflammatory syndrome. This syndrome was first reported in humans following the abrupt reversal of immunosuppression. (For example, in AIDS patients following the onset of effective antiretroviral therapy.) It is believed that the recovery of T cell function drives an exaggerated immune response against the underlying infection that causes significant collateral damage. Thus, in white-nose syndrome, cold-loving P. destructans are present in infected tissues. During this time, transcription studies show that the bat mounts an inflammatory cytokine response but neither neutrophils nor T cells are recruited to the lesion [32]. When the body temperature returns to normal, neutrophil and T cell functions are rapidly restored, leading to a massive cellular influx and tissue destruction.

18.4

Lymphoid organs

18.4.1 Thymus The histology of the bat thymus appears to be similar to that of other mammals [29].

18.4.2 Spleen The spleen of fruit bats such as P. vampyrus, the Indonesian Short-nosed Fruit Bat (Cynopterus titthaecheilus), and Leschenault’s Rousette (R. leschenaultia), are surrounded by a thin capsule. The trabeculae are also thin and generally lack smooth muscle cells. This suggests that the spleens of fruit bats cannot contract and are not principally blood storage organs. The red pulp consists of splenic cords and large blood-filled sinuses. Ellipsoids are numerous and form macrophage aggregates. The white pulp consists of periarteriolar lymphoid sheaths with numerous lymphoid follicles and an obvious prominent, marginal zone [33]. There is little evidence of significant splenic hematopoiesis.

18.4.2.1 Lymph nodes The lymph nodes of bats conform to the basic structure seen in other mammals. However, their size and number do vary considerably based upon seasonality, pregnancy, and hibernation status. When bats undergo hibernation, their lymph nodes shrink relative to their size during the active summer season [34].

18.4.3 Peyer’s patches Phyllostomid bats differ in the morphology of their Peyer’s patches. Thus, Seba’s Short-tailed Bat (Carollia perspicillata), the Greater Short-nosed Bat (Phyllostomus hastatus) and Pallas’s Long-tongued Bat (Glossophaga soricine) have Peyer’s patches distributed in the distal portion of their large intestine. Some of these are very small and difficult to identify. In the Common Vampire Bat (Desmodus rotundus) and the Little Yellow-shouldered Bat (Sturnura lilium) there are simply nodular aggregations of lymphoid tissue distributed along the gut [35]. It should be pointed out however that the intestinal morphology does not allow a clear distinction to be made between the large and small intestine in bats since they lack a cecum and appendix [36]. The differences in the abundance, distribution, and morphology of the intestinal lymphoid tissues in Phyllostomids are possibly diet-related. Thus, it has been observed that frugivorous bats have more Peyer’s patches than insectivorous, carnivorous, or nectivorous ones. In insectivorous and carnivorous species, the patches may be small, show little activity, and are usually restricted to the ileal submucosa [36]. No Peyer’s patches have been detected in the Horseshoe Bat

288

SECTION | 2 Mammalian orders

(Rhinolophus hildebrandtii), or in the Common Pipistrelle Bat (Pipistrellus pipistrellus). The submucosa of the Horseshoe Bat lacks lymphoid tissue with the exception of a few aggregated lymphoid nodules in the rectal submucosa [37]. Further studies on the number of Peyer’s Patches in bats with different diets demonstrate some other significant differences [38]. Species of the genus of tropical insect feeders (Molossus spp.) have abundant Peyer’s patches whereas all other tropical and temperate insectivores examined have relatively few patches. In addition, patches in the intestine of insectivores (excluding the free-tailed bats), are usually more restricted in distribution than are those in herbivores and blood feeders. Thus, the Peyer’s patches vary considerably among species, but one consistent feature is that the patches are usually largest at the distal end of the small intestine [38].

18.5

The major histocompatibility complex

The most variable MHCs known are those of passerine birds. This is the result of their wide geographic range resulting from extensive migration as well as their inhabiting a diversity of habitats. If this is the case, then a similar extreme MHC diversity should also be found in bats. Studies on one species, Seba’s Short-tailed Bat, (Carollia perspicillata), a very widely distributed neotropical species, have found a great diversity of both individual and population MHC class I genes that are comparable to that in passerine birds [39]. It appears that in species with an extensive geographic range, high MHC diversity is especially adaptive since it enables an animal to respond to a greater diversity of microbial antigens. This adaptability, diversity, and how they have evolved are of relevance to the evolution of the relationships between bats and their viruses. The first major study of a bat MHC focused on the Australian Black Flying Fox, (Pteropterus alecto), a megabat within the suborder Yinpterochiroptera. As described in previous chapters the MHC region is typically subdivided into three regions encoding the genes for class I, class II and class III genes. As in other mammals, they are organized along a chromosome in a class I-III-II order (Fig. 18.4). The Ptal-MHC-1 region is located on chromosome 1 and contains about 900 kb of contiguous sequence encompassing both the extended and classical subregions [40].

18.5.1 The MHC class I region The bat class I region can be subdivided into two subregions. There is an extended centromeric subregion that contains nonclassical class I genes including numerous butyrophilin, histone, olfactory receptor, and zinc finger loci. The classical subregion typically contains numerous MHC class I genes. It also contains a set of framework genes that are generally highly conserved among mammals. This conserved framework is believed to be ancestral, and its structure is conserved for functional reasons. As discussed in Chapter 7, the MHC class I genes are located at specific sites within this region based on their origins in the three ancestral duplication blocks, alpha, kappa, and beta. The enormous diversity of MHC class I genes is due to changes, duplications, and eliminations in the Class I genes that have occurred within each duplication block without disturbing the overall structure as determined by the framework regions. Much of the class I region is contracted in chiropterans. Thus, the alpha block is contracted and contains no MHC-1 genes. This is similar to the situation in the horse and pig but different from humans. The beta block is also contracted in bats even though it is expanded in humans, horses, and pigs. The functional MHC class I genes of P. alecto appear to be confined to the kappa block. The Ptal-MHC is therefore unusual in that its MHC-I genes are located within only one of the three highly conserved duplication blocks. It is suggested that the MHC-I genes first originated in the beta duplication block. But subsequently duplicated in a stepwise manner across the MHC-I region during the course of evolution. Studies on the genome of the Big Brown Bat, E. fuscus confirm this contraction of the MHC class I region. Epfu-MHC contains only a single class I gene in its beta DR

DP

DO

\

A

DM DO

DQ

FIGURE 18.4 The organization of the major histocompatibility complex in the Black Flying Fox, P. alecto.

DR

B Ptal-

01*1 02 03 05 01*2 06 04

II

III

Ib

Ia

Chiropterans: the bats Chapter | 18

289

block and it is only B1.1 mb in size compared to the human with B1.7 mb and horses with B1.5 mb. The pig is the only other mammal with a contracted class I region of B1 mb. This is consistent with the relatively small genome of P. alecto. (B2.0 gb compared to humans with an average genome size of B3.5 gb.) Seven class I genes have been identified in the P. alecto genome although some may be located outside the class I region. One of these, Ptal-06 appears to be a processed pseudogene [40]. Their protein products have antigen-binding sites located on the α1 and α2 domains that appear to be functional and conserved. These class I gene products show considerable variations in length. Thus, three of the P. alecto proteins contain a five amino acid insertion in the alpha-1 domain. While another three contain a three amino acid insertion in the same region. The α1 domain of Ptal-04 is 26 amino acids longer than other mammalian MHC class-1 α chains while the cytoplasmic domain of Ptal-05 contains a 17 amino acid insertion. Ptal-MHC-1 genes also contain unique insertions within their peptide-binding grooves that could affect the repertoire of peptides recognized by their cytotoxic T cells and hence resistance to infection. An analysis of the antigen-binding specificity of class I genes has identified residues that can bind viral ligands such as those from Hendra virus [41]. In the Egyptian Rousette Bat (E. rousettus), multiple MHC class I genes are found both within and outside the canonical MHC locus [23]. Thus, the expansion of NK cell receptor diversity in this species discussed later is matched by an expansion of classical and nonclassical MHCs. There are 12 MHC class I genes or pseudogenes in its genome. None appear to be related to HLA-E. Only two of these are located within the β block, the rest are dispersed. This appears to be a feature of many bat species. Abduriyim et al. examined the MHC class I region of four bat species [42]; two Yangochiroptera, the Common Vampire Bat (D. rotundus) and the Natal long-fingered Bat, (Miniopterus natalensis); and two Yinpterochiroptera, the Great Roundleaf Bat (Hipposideros armiger) and the Chinese Rufous Horseshoe Bat (Rhinolophus sinicus). In these species, the number of functional MHC class I genes ranges from 1 to 13, and the number of pseudogenes range from one to six [42]. Only the β block is present in these species. In all four species, their encoded proteins also have a three amino acid insertion in the MHC α1 domain resulting in a somewhat larger antigen-binding groove.

18.5.2 The MHC class II region Extreme differences in MHC class II polymorphisms have been reported in bats reflecting differences in pathogen pressure. Thus, the insectivorous Lesser Bulldog Bat (Noctilio albiventris) shows correlations between specific DRB alleles, ectoparasite load, and reproductive state [43]. The Black Flying Fox (P. alecto) also has a relatively small MHC class II region. In most eutherian mammals its size ranges from B0.5 mb in the pig to B1.4 in the horse. In this bat, it is 0.78 mb. There are twelve identified MHCII genes in the Ptal class II region, five encode antigen-presenting genes There are also 30 other genes in a region spanning 360 kb. One class II gene, designated DRB2 is located outside the class II region. Two classical loci, DQ and DR have been identified as have two nonclassical loci DO and DM. Bats lack functional DOA and DOB genes but a classical DP1 pseudogene encoding a partial DP β chain is present. P. alecto also has two functional copies of DQA and DQB. The antigen-processing cluster located within the class II region is relatively well conserved. The bat classical and extended MHC II subregions are bordered by butyrophilin-like protein 2 and kinesin family member C1. The organization of the Ptal-MHC II is highly syntenic with the human, horse, and pig. Thus, although the region is contracted, the gene order is conserved including the entire antigen-processing cluster. 2DOB 2 TAP2 2 PSMB8 2 TAP1 2 PSMB9 2 DMB 2 DMA 2 BRD2 2 DOA 2 :

18.5.3 The natural killer cell receptor complex Unlike other mammals, NKG2A/CD94 receptors encoded by the KLRC (NKG2) and KLRD (CD95) family of genes appear to be the primary NK cell receptors in bats [23]. Multiple studies have shown that the usual mammalian NK receptor genes, both KIRs and Ly49-like are absent from both the transcriptome data set and the genome data of P. alecto and other bat species [14,23]. Likewise BLAST searches of the whole genome sequence of the closely related P. vampyrus have also failed to reveal any KIR or Ly49 gene sequences. A single Ly49 pseudogene has been identified in the genome of M. davidii [14,16]. Two KIR genes have been identified in the genome of E. fucus, but it is unclear whether they are functional. NK cell coreceptor genes that have been identified in P. alecto include both NKG2C and NKG2A while NKG2D has been detected in the pooled transcriptome. Two distinct CD94 contigs have also been detected but one of these is missing two conserved cysteines while the other is missing just one. The significance of

290

SECTION | 2 Mammalian orders

this is unclear. Overall, the genomic evidence suggests that NK cell responses in some bats may be atypical. Other NK cell receptors found in P. alecto include CD16, CD56, and CD244. The microbat, M. davidii also appears to use novel NK cell receptors to recognize MHC molecules. These include SIGLECs, LILRs, CEA-related cell-adhesion molecules (CEACAMs), and LAIRs that have undergone considerable gene duplication in M. davidii but have totally failed to expand in P. alecto [14]. The situation in R. aegypticus is also very different. In this species, there is also no evidence of any functional KIRs. This bat has however undergone a major expansion of its KLRC/KLRD family of receptor genes as well as their ligands, the MHC class I genes. It has 14 NKG2A/B-like genes (NKG2 1
14) including four pseudogenes, one NKG2D-like gene, one NKG2C-like gene, and five CD94-like genes [23]. Six of the functional NKG2A/B genes encode activating and inhibitory interaction motifs. Because there are so many NKG2 and CD94 genes, this permits extreme combinatorial diversity among their encoded receptor heterodimers. Four of the five CD94 molecules lack the cysteines required to form disulfide bonds with NKG2 proteins but they may be able to form noncovalently linked heterodimers. Overall, this major expansion appears to have a net inhibitory effect on NK cell cytotoxicity in this species [23].

18.6

B cells and immunoglobulins

Fruit bats tend to have relatively low levels of immunoglobulins in their bloodstream compared to rodents [44]. They also appear to respond more slowly to injected antigens as measured by a delayed peak in antibody titers [45]. Their cells respond unusually slowly to mitogens such as concanavalin-A [46]. Investigations have been conducted into the serum immunoglobulins of P. alecto. It has proved possible to isolate significant quantities of IgM but only trace amounts of IgA [7]. IgM is the second most abundant serum antibody after IgG. In addition, a survey of mucosal immunoglobulins in this species indicated that IgG is the dominant antibody class
not IgA. Thus, IgA could not be detected in lung lavage fluid or saliva of P. alecto. It appears that high levels of IgG in mucus secretions may serve to compensate for low levels of IgA. The significance of this apparent IgA deficiency remains unclear. Serum IgG concentrations increase with age in S. bilineata. This rise may, in many cases reflect the presence of infections or increased parasitic burdens since bats with higher-than-average IgG levels were more likely to die over the following six months [47]. While T cells predominate in the blood and spleen ( . 50%) of P. alecto, B cell numbers in the bone marrow vary greatly between individuals. This may be due to species differences or seasonality, or to differences between wild and captive bats. Injection of bacterial lipopolysaccharide results in a significant increase in B cell numbers in the blood and spleen. P alecto B cells release significant amounts of calcium on cross-linking of their antigen receptors [48]. The Indian Flying Fox P. medius (giganticus), develops a delayed primary response following immunization with sheep red blood cells [7]. Similarly, the magnitude and duration of the antibody response of the Big Brown Bat (E. fuscus) against bacteriophage φX174 are very much lower than that of rabbits and guinea pigs. In all P. alecto organs examined, most B cells were IgG1 suggesting that they had already been primed.

18.6.1 IGH The IGH locus has been studied in little brown bats (M. lucifugus), the big brown bat (E. fuscus), Seba’s Short-nosed Fruit Bat (C. perspicata) and the Short-nosed Fruit Bat (C. sphinx). All four species are no different from other mammals in that they possess and transcribe genes encoding IgM, IgE, and IgA as well as multiple IgG subclasses [49] (Fig. 18.5). None of these IGH constant genes shows a close similarity to any other mammalian order but their closest association especially IgM and IgD, is with the carnivore immunoglobulins. 5'

66

3

2

5

3'

7 M

\E

M

G1 \E2 G2 \E3 G3 E1 G4 E2

G

E

A

V

D

J

A

FIGURE 18.5 The sequence of the IGH locus in the Egyptian Rousette Bat (E. rousettus) [3] Given the huge diversity of the chiropterans, it is unsurprising that other bat species differ in the number of IgG subclasses. Thus Myotis lucifugus has four Eptesicus fuscus and C. sphinx have two, while P. alecto has only one [49]. Likewise, the precise numbers of V, D and J gene segments also differ between species.

Chiropterans: the bats Chapter | 18

291

18.6.2 IgM IgM is highly expressed in lymph nodes and spleen and moderately expressed in blood lymphocytes, the small intestine, lung, salivary glands, and in young bats [7]. Western blotting has detected two IgM bands in serum suggesting the existence of a second IgM subclass in P. alecto [7].

18.6.3 IgD IGHD transcripts have been recovered from insectivorous bats such as M. lucifugus and E. fuscus spleens but not from frugivorous bats such as P. alecto [14]. They comprise two constant exons corresponding to CH1, and CH3 in other mammals as well as two hinge exons. The second hinge exon is fused with the CH3 exon. Bats, like rodents lack a IgD CH2 exon. (Fig. 7) However, despite this rodent-like feature, sequence studies show that bat IgD is more closely related to that in the dog.

18.6.4 IgG IGHG is the most highly expressed immunoglobulin mRNA in lymph nodes, spleen, and peripheral blood mononuclear cells [7]. As in other mammals, the bat IgG subclasses have diversified after speciation. As a result, the number of these subclasses varies between different species. M. lucifigus has seven different IGHG transcripts, C. sphinx has three, E. fuscus has two, and C. perspicillata only one. Some of these variants are likely to be alleles rather than subclasses. Alignment studies demonstrate that these putative subclasses differ primarily in their hinge region sequences. As in other mammals the hinge regions are rich in cysteines and prolines. Some of the bat IgG heavy chains (IgG2, IgG3, and IgG4) lack a hinge CXXC domain and so may not be able to join the two heavy chains together. They could thus potentially only form half molecules [3].

18.6.5 IgA In all species of bats examined, their IgA resembles human IgA2 and cattle IgA since it is missing the cysteine residue (77) in the CH1 domain that forms a disulfide bridge to the light chain. It is possible therefore that the heavy and light chains in this IgA may be noncovalently linked as in humans. The IGHA gene is highly transcribed in the small intestine, lung, and salivary gland. However, as described above, its product, IgA appears to be present only in trace amounts in P. alecto serum and mucosal secretions [7].

18.6.6 IgE The immunoglobulin heavy chain region of R. aegyptiacus has several unique features. For example, there is an apparent expansion of the IGHV genes associated with protective antiviral immunity. This expansion also includes the presence of two functionally diverse IGHE genes and three IGHE pseudogenes in addition to four functional IGHG genes. As a result, the Rousette Bat is among the very few mammals known to have more than one functional IGHE. (Humans have two but one is a nonfunctional pseudogene. A second IGHE gene is probably also present in the dog) [51]. The two bat IgEs may have different functions. Thus, IgE2 is expressed in the lung and the bone marrow while both IgE1 and IgE2 are expressed in the blood and in the secondary lymphoid organs. IgE2 appears to have conventional functionality while IgE1 lacks a residue required for binding to CD23 suggesting that it may have a different functional role. This residue is present in IgE2 [3]. The Ig Fc region is responsible for the biological activity of immunoglobulins. It acts through diverse Fc receptors on leukocytes. These receptors can bind specific immunoglobulin isotypes with varied affinities. The presence or absence of different Fc receptors in different mammalian species is closely linked to their possession of different immunoglobulin isotypes. The structural changes in P. alecto lack some signal transduction components so that as a result their signaling ability is predicted to be impaired [3]. The IGHC/FcγR gene sequences suggest higher activation and inflammation thresholds. Collectively these features may promote immunological tolerance and decrease inflammation in this species [3].

292

SECTION | 2 Mammalian orders

18.6.7 IGHV The IGHV domain organization in the bat heavy chain loci is similar to other eutherians although the expressed IGHV3 repertoire is unusually diverse [49]. The CDR3 portion of the transcribed VDJs is derived mainly from D gene segments with differences at the 5’ end due to N-region addition and somatic hypermutation. The mean length of the CDR3s is similar to those in mouse, humans, and pigs (around 13 amino acids). M lucifugus has a very diverse IGHV gene repertoire comprised of five IGHV subfamilies with homologs in other mammals [52]. Its IGHV3 subfamily groups with clan III, IGHV1, IGHV5, and IGHV7 group with clan II, and IGHV4 groups with clan I. It has an estimated 236 germline IGHV3 genes. Ninety-five percent of these germline IGHV3 genes differ in framework region 3. Analysis of 67 expressed genes with 75 germline genes revealed a mutation efficiency similar to that in fetal piglets! It also contains at least 13 putative IGHJ segments and a large D segment repertoire. The nature of these germline genes suggests evolution by gene duplication and gene conversion. The IGHV3 transcripts from adult spleens show a low frequency of somatic hypermutation that is not clustered in either CR1 or CR2. Another microbat, M davidii and two megabats Pteropus vampyrus and P. alecto have also had their IGHV genes examined. M davidii has 234 IGHV genes, P alecto has 53, and P vampirus has 314. In addition, 51 VDJ transcripts have been recovered from E. fuscus, C. perspicillata, and C. sphinx. V gene usage has been analyzed in P. alecto and the closely related Pteropid bat P. vampyrus [53]. The representative IGHV genes found belonged to all three mammalian clans (1, II, III) in both species. There is evidence for the use of multiple heavy chain diversity (D) and joining (IGHJ) segments and the generation of diverse VDJ rearrangements in P. alecto. Pteropid bats have V domains that can therefore bind a wide range of epitopes. Many bind weakly but with high specificity [53]. There are multiple D and V gene segments in this species. Thus, the 23 P. alecto IGHV cDNAs share 50%
96% sequence identity. They fall into five distinct subfamilies (IGHV1-IGHV5). Most variation is, as expected, in the CDR3 region. P alecto CDR3 ranges in size from 6 to 18 amino acids, a size comparable to humans and mice. (But they contained fewer tyrosine and more arginine residues
perhaps resulting in weaker binding.) At least five distinct IGHJ gene segments are also present in this species. In the related species, P. vampyrus, 74 unique IGHV sequences have been identified of which eleven are pseudogenes.

18.6.8 Light chains In P. alecto, there are at least 38 IGLV genes and six IGKV sequences [54]. Lee et al. showed that for E. fuscus, most, if not all immunoglobulins contain lambda light chains [55]. At least one microbat, M. lucifugus, does not appear to make any kappa chains at all [56]. On the other hand, other bat species (megabats) probably use kappa chains [14]. This preference for kappa chain usage is a consistent feature of the Laurasiatheria. Most other ungulates other than pigs prefer to use lambda chains.

18.7

T cells and cell-mediated immunity

In the Black Flying Fox, P. alecto, over 40% of the leukocytes present in spleen and bloodstream are T cells [57]. The majority of T cells in the circulation, lymph nodes, and bone marrow are CD41. Conversely CD81 cells predominate in the spleen. As a result, the CD4:CD8 ratio in the bone marrow is 2.0 [57]. Forty percent of the splenic T cells constitutively express IL-17, IL-22, and TNF-β mRNA suggesting a bias towards Th1 and Treg responses. Another feature of the P. alecto immune system is their use of MR1-restricted T cells [58]. Following priming, these cells proliferate rapidly and transiently express Th1 and Th17 cytokines as well as the ability to kill E. coli. These cells likely play an important role in bat antibacterial defenses. There have been relatively few studies on the T cell antigen receptor genes of bats. A notable exception is the Greater Horseshoe Bat, Rhinolophus ferrumequinum. This species has had its TR locus completely annotated [50]. It has been shown that the TRA/D, TRG, and TRB loci contain a total of 128 V segments, three D segments, 85 J segments and 6 C segments (Fig. 18.6).

18.7.1 TRA/D The R. ferrumequinum locus is located on chromosome 6. As in all the other mammals, the TRD locus is embedded within the TRA locus. The length of the TRA locus is approximately 850 kb. It contains 81 TRAV genes that can be classified into 34 subfamilies. (Their nucleotide similarity ranges from 25 to 99%). It also contains 60 TRAJ genes

Chiropterans: the bats Chapter | 18

TRA/D

5'

81 A/DV

TRB

28

TRG

7

18 DV

DD

60 AJ

4 DJ DC

6

293

3'

AC

9

7 4

4

V

D

J

C

FIGURE 18.6 The organization of the T cell antigen receptor loci in the horseshoe bat, (Rhinolophus ferroequinum) [50].

classified into 60 subfamilies, and one TRAC gene. The total length of the TRD locus is approximately 660 kb. It contains 18 TRDV genes classified into eleven subfamilies. Only two of these TRDV genes are pseudogenes, the rest are functional [50]. Many TRAV and TRDV genes are located upstream of these loci. The TRD loci share some of their V genes with TRA. The TRD locus has a complete V-D-J cluster including one TRDD gene, four TRDJ genes and one TRDC gene. There is also a single TRDV gene with an opposite transcription direction and then a TRA locus J-C cluster including the 60 TRAJ genes and the single TRAC gene. This TRA/D locus is relatively small compared to artiodactyls but is a similar size to that in primates and the carnivores. The TRAC and TRDC genes each contain four exons. Exon 1 of TRAC varies greatly between different bat species but exons 2, 3, and 4, are highly conserved. A similar situation applies to TRDC.

18.7.2 TRB The TRB locus of R. ferrumequinum is located on chromosome 26. Its total length is approximately 203 kb. It contains 29 TRBV genes classified into 25 subfamilies; two TRBD genes in two subfamilies; 15 TRBJ genes classified into two subfamilies, and two TRBC genes [50]. Of the 29 TRBV genes, there is one pseudogene, two ORFs and the rest are functional. Thus, there is an upstream V cluster containing 28 TRBV genes followed by two D-J-C clusters. Downstream of the second DRBC gene is a TRBV gene in the opposite transcriptional orientation.

18.7.3 TRG The TRG locus is found on chromosome 20. Its total length is about 150 kb. It contains 14 TRGV genes subdivided into seven subfamilies. Of these, five are pseudogenes, three are open reading frames and six are functional. The six TRGJ genes fall into three subfamilies There are two TRGC genes. Thus, its composed of two V-J-C clusters of equal size. As in other mammals, there is significant diversity in the TRGC genes. This diversity is a result of different intron lengths in the coding junction region [50].

References [1] Tsagkpgeorga G, Parker J, Stupka E, Cotton JA, Rossiter SJ. Phylogenetic analyses elucidate the evolutionary relationships of bats. Curr Biol 2013;23:2262
7. [2] Amman BR, Albarino CG, Bird BH, Nyakarahuka L, et al. A recently discovered pathogenic paramyxovirus, Sosuga virus, is present in Rousettus aegyptiacus fruit bats in multiple locations in Uganda. J Wildl Dis 2015;51(3):774
9. [3] Larson PA, Bartlett ML, Garcia K, Chitty J, et al. Genomic features of humoral immunity support tolerance model in Egyptian rousette bats. Cell Rep 2021;35(7):109140. Available from: https://doi.org/10.1016/j.celrep.2021.109140. [4] Carter AM, Mess A. Evolution of the placenta and associated reproductive characters in bats. J Exp Zool 2008;310B:428
49. [5] Epstein JH, Baker ML, Zambrana-Torrielo C, Middleton D, et al. Duration of maternal antibodies against canine distemper virus and Hendra virus in Pteropid bats. PLoS One 2013;8(6):e67584. Available from: https://doi.org/10.1371/journal.pone.0067584.

294

SECTION | 2 Mammalian orders

[6] Francis CM, Anthony ELP, Brunton J, Kunz TH. Lactation in male fruit bats. Nature 1994;367:691
2. [7] Wynne JW, Di Rubbo A, Shiell BJ, Beddome G, et al. Purification and characterization of immunoglobulins from the Australian Black Flying Fox (Pteropus alecto) using anti-Fab affinity chromatography reveals the low abundance of IgA. PLoS One 2013;8(1):e52930. Available from: https://doi.org/10.1371/journal.pone.0052930. [8] Beltz LA. Chapter 1. Bat immunology. Bats and human health: Ebola, SARS, rabies and beyond. John Wiley; 2018. p. 1
24. [9] Schinnerl M, Aydinonat D, Schwarzenberger F, Voigt CC. Hematological survey of common neotropical bat species from Costa Rica. J Zoo Wildl Med 2011;42(3):382
91. [10] Jiang H, Li J, Li L, Zhang X, et al. Selective evolution of toll-like receptors 3, 7, 8, and 9 in bats. Immunogenetics 2017;69:271
85. [11] Schad J, Voigt CC. Adaptive evolution of virus-sensing toll-like receptor 8 in bats. Immunogenetics 2016;68:783
95. [12] Cowled C, Baker M, Tachedjian M, Zhou P, et al. Molecular characterization of toll-like receptors in the black flying fox Pteropus alecto. Dev Comp Immunol 2011;35:7
18. [13] Baker ML, Schountz T, Wang L-F. Antiviral immune responses of bats: a review. Zoonoses Publ Health 2013;60:104
16. [14] Papenfuss AT, Baker ML, Feng Z-P, Tachedjian M, et al. The immune gene repertoire of an important viral reservoir, the Australian black flying fox. BMC Genomics 2012;13:261. Available from: https://doi.org/10.1186/1471-2164-13-261. [15] Bannerjee A, Baker ML, Kulcsar K, Misra V, et al. Novel insights into immune systems of bats. Front Immunol 2021;11:26. Available from: https://doi.org/10.3389/fimmu.2020.00026. [16] Zhang G, Cowled C, Shi Z, Huang Z, et al. Comparative analysis of bat genomes provides insight into the evolution of flight and immunity. Science 2013;339:456
60. [17] Xie J, Li Y, Shen X, Goh G, et al. Dampened STING-dependent interferon activation in bats. Cell Host Microbe 2018;23:297
301. [18] Banerjee A, Rapin R, Bollinger T, Misra V. Lack of inflammatory gene expression in bats: a unique role for a transcription repressor. Sci Rep 2017;7(1):2232. Available from: https://doi.org/10.1038/s41598-017-01513-w. [19] Schattgen SA, Fitzgerald KA. The PYHIN protein family as mediators of host defenses. Immunol Rev 2011;243:109
18. [20] Ahn M, Cui J, Irving AT, Wang L-F. Unique loss of the PYHIN gene family in bats amongst mammals: implications for inflammasome sensing. Sci Rep 2016;6:21722. Available from: https://doi.org/10.1038/srep21722. [21] Zhou P, Tachedjian M, Wynne JW, Boyd V, et al. Contraction of the type I IFN locus and unusual constitutive expression of IFN-α in bats. Proc Natl Acad Sci U S A 2016;113(10):2696
701. [22] Schountz T, Baker ML, Butler J, Munster V. Immunological control of viral infections in bats and the emergence of viruses highly pathogenic to humans. Front Immunol 2017;8:1098. Available from: https://doi.org/10.3389/fimmu.2017.01098. [23] Pavlovich SS, Lovett SP, Koroleva G, Guito J, et al. The Egyptian rousette genome reveals unexpected features of bat antiviral immunity. Cell 2018;173:1098
110. [24] Zhou P, Cowled C, Todd S, Crameri G, et al. Type III IFNs in Pteropid bats: Differential expression patterns provide evidence for distinct roles in antiviral immunity. J Immunol 2011;186:3138
47. [25] Cruz-Rivera PC, Kanchwala M, Liang H, Kumar A, et al. The IFN response in bats displays distinctive IFN-stimulated gene expression kinetics with atypical RNAESL induction. J Immunol 2018;200:209
17. [26] Jebb D, Huang Z, Pippel M, Hughes GM, et al. Six reference quality genomes reveal evolution of bat adaptations. Nature 2020;583:578
84. [27] Bondet V, Le Baut M, Le Poder S, Le´cu A, et al. Constitutive IFN α protein production in bats. Front Immunol 2021;12:735866. Available from: https://doi.org/10.3389/fimmu.2021.735866. [28] Cowled C, Stewart CR, Likic VA, Friedlander MR, et al. Characterization of novel microRNAs in the Black flying fox (Pteropus alecto) by deep sequencing. BMC Genomics 2014;15(1):682. Available from: https://doi.org/10.1186/1471-2164-15-682. [29] Omatsu T, Watanabe S, Akashi H, Yoshikawa Y. Biological characteristics of bats in relation to natural reservoir of emerging viruses. Comp Immunol Microbiol Inf Dis 2007;30:357
74. [30] Grimaudo AT, Hoyt JR, Yamada SA, Herzog CJ, et al. Host traits and environmental impact interact to determine persistence of bat populations impacted by white-nose syndrome. Ecol Lett 2022;25:483
97. [31] Meteyer CU, Barber D, Mandl JN. Pathology in euthermic bats with white nose syndrome suggests a natural manifestation of immune reconstitution inflammatory syndrome. Virulence 2012;3:10
16. [32] Field KA, Johnson JS, Lillry TM, Reeder SM, et al. The white-nose syndrome transcriptome: Activation of anti-fungal host responses in wing tissue of hibernating little Brown Myotis. PLoS Pathog 2015;11(10):e1005168. Available from: https://doi.org/10.1371/journal.ppat.1005168. [33] Hanadhita D, Rahma A, Prawira AY, Putu NL, et al. The spleen morphophysiology of fruit bats. Anat Histol Embryol 2019;48:315
24. [34] Abel-Galil D.Y., Tawfik A.-R., Saad A.-H. Anatomical and comparative distribution of lymph nodes in some Egyptian chiroptera. 2012. ,http://scholar.cu.edu.eg.. [35] Gadelha-Alves R, Rozensztranch AMS, Rocha-Barbosa O. Comparative intestinal histomorphology of five species of Phyllostomid bats (Phyllostomidae, Microchiroptera): Ecomorphological relations with alimentary habits. Int J Morphol 2008;26(3):591
602. [36] Forman GL. Structure of Peyer’s patches and their associated nodules in relation to food habits in New World bats. J Mammal 1974;55 (4):738
46. [37] Baker ML, Schountz T. Mammalia: chiroptera: immunology of bats. In: Cooper E, editor. Advances in comparative immunology. Cham: Springer; 2018. Available from: https://doi.org/10.1007/978-3-319-76768-0_23. [38] Forman GL. Comparative studies of organized gut-associated lymphoid tissue in mammals with diverse food habits, distribution, size and organization of Peyer’s patches in New World bats. Trans Ill State Acad Sci 1974;67(2):152
62.

Chiropterans: the bats Chapter | 18

295

[39] Qurkhuli T, Schwensow N, Bra¨ndel SD, Tschapka M, Sommer M. Can extreme MHC class I diversity be a feature of a wide geographic range? The example of Seba’s short-tailed bat (Carollia perspicillata). Immunogenetics 2019;71:575
87. [40] Ng JHJ, Tachedjian M, Deakin J, Wynne JW, et al. Evolution and comparative analysis of the bat MHC-1 region. Sci Rep 2016;6:21256. Available from: https://doi.org/10.1038/srep21256. [41] Wynne JW, Woon AP, Dudek NL, Croft NP, et al. Characterization of the antigen processing machinery and endogenous peptide presentation of a bat MHC class I molecule. J Immunol 2016;196:4468
76. [42] Abduriyim S, Zou D-H, Zhao H. Origin and evolution of the major histocompatibility complex class I region in eutherian mammals. Ecol Evol 2019;9:7861
74. [43] Ng JHJ, Tachedjian M, Wang L-F, Baker ML. Insights into the ancestral organization of the mammalian class II region from the genome of the pteropid bat Pteropus alecto. BMC Genomics 2017;18(1):388. Available from: https://doi.org/10.1186/s12864-017-3760-0. [44] Iha K, Omatsu T, Watanabe S, Ueda N, Taniguchi S, et al. Molecular cloning and sequencing of the cDNAs encoding the bat interleukin (IL)2. IL-4, IL-6, IL-10, IL12.p40 and tumor necrosis factor alpha. J Vet Med Sci 2009;71(120):1691
5. [45] Chakraborty AK, Chakravarty AK. Antibody-mediated immune responses in the bat, Pteropus giganteus. Dev Comp Immunol 1984;8:415
23. [46] Paul BN, Chakravarty AK. In vitro analysis of delayed immune response in a bat Pteropus giganteus: Process of Con-A mediated activation. Dev Comp Immunol 1986;10:55
67. [47] Schneeberger K, Courtiol A, Czirja´k GA, Voigt CC. Immune profile predicts survival and reflects senescence in a small, long-lived mammal, the Greater Sac-winged bat (Saccopteryx bilineata). PLoS One 2014;9(9):e108268. Available from: https://doi.org/10.1371/journal. pone.0108268. [48] Periasamy P, Hutchinson PE, Chen J, Bonne I, et al. Studies on B cells in the fruit-eating black flying fox (Pteropus alecto). Front Immunol 2019;10:489. Available from: https://doi.org/10.3389/fimmu.2019.00489. [49] Butler JE, Wertz N, Zhao Y, Zhang Z, et al. The two suborders of chiropterans have the canonical heavy chain immunoglobulin (Ig) gene repertoire of eutherian mammals. Dev Comp Immunol 2011;35:273
84. [50] Zhou H, Ma L, Liu L, Yao X. TR locus annotation and characteristics of Rhinolophus ferrumequinum. Front Immunol 2021;12:741408. Available from: https://doi.org/10.3389/fimmu.2021.741408. [51] Tizard IR. Allergic diseases of domestic animals. St Louis: Elsevier; 2022. [52] Bratsch S, Wertz N, Chaloner K, Kunz TH, Butler JE. The little brown bat, M. lucifugus displays a highly diverse VH, DH and JH repertoires but little evidence of somatic hypermutation. Dev Comp Immunol 2011;35:421
30. [53] Baker ML, Tachedjian M, Wang L-F. Immunoglobulin heavy chain diversity in Pteropid bats: evidence for a diverse and highly specific antigen binding repertoire. Immunogenetics 2010;62:173
84. [54] Butler JE, Wertz N, Baker ML. The Immunoglobulin genes of bats. In: Kaushik AK, Passman Y, editors. Comparative immunoglobulin genetics. Toronto, ON: Apple Academic Press; 2014. [55] Lee WT, Jones DD, Yates JL, Winslow GM, et al. Identification of secreted and membrane-bound bat immunoglobulin using a Microchiropteran-specific mouse monoclonal antibody. Dev Comp Immunol 2016;65:114
23. [56] Das S, Nikolaidis N, Klein J, Nei M. Evolutionary redefinition of immunoglobulin light chain isotypes in tetrapods using molecular markers. Proc Natl Acad Sci U S A 2008;105(43):16647
52. [57] Martinez Gomez JM, Periasamy P, Dutertre C-A, Irving AT, et al. Phenotypic and functional characterization of the major lymphocyte populations in the fruit-eating bat Pteropus alecto. Sci Rep 2016;6:37796. Available from: https://doi.org/10.1038/srep37796. [58] Leeansyah E, Hey YY, Sia WR, Ng JHJ, et al. MR1-restricted T cells with MAIT-like characteristics are functionally conserved in the Pteroptid bat Pteropus alecto. iScience 2020;23(12):101876. Available from: https://doi.org/10.1016/j.sci.2020.101876.

This page intentionally left blank

Chapter 19

Feliformes: The cats and their relatives

African lion. Panthera leo.

The Carnivora, a flesh-eating order of placental mammals, probably diverged from the other placentals, likely the Perissodactyls, about 75 million years ago (mya) during the late Cretaceous period [1]. Like other mammals, they probably began their major expansion and diversification after the dinosaurs had been eliminated by the K-Pg event. The first carnivores that evolved during the Eocene were probably small, weasel-like creodonts. Around 42 mya, however, the crown group split into two clades. As with other paleontological studies, the differences between these two clades are based on bone structure, specifically the structure of the bones that surround the middle ear and other cranial features. In one group, the cat-like Feliformia, the auditory bullae are double-chambered and consist of two bones joined by a septum. The other clade, the dog-like Caniformia, has single chambered or partially divided auditory bullae constructed from a single bone. Feliformia also tend to have shorter rostrums, fewer teeth, and specialized carnassial molars. During the Oligocene, these two clades diversified further to become the primary terrestrial predators [1]. The order Carnivora consists of about 270 extant species [2]. It includes top predators, domestic companion animals, and some iconic species such as polar bears, sea lions, and giant pandas. They are found on all the continents and occupy all major habitat types. The two Carnivora clades are designated superfamilies, the Feliformia and the Caniformia. The Feliformia are the subject of this Chapter. They contain multiple families including the Felidae (cats), the Herpestidae (mongooses), Hyaenidae (hyenas), Priodontidae (linsangs), Nandiniidae (African palm civet), and the Viverridae (genets, civets, and the binturong) [2]. The domestic cat (Felis silvestris catus) is a sister taxon to the Asian wild cat (Felis silvestris ornata). For obvious reasons, the greatest amount of immunological data has been obtained from the domestic cat. Recent evidence suggests that cats were first domesticated about 9
10,000 years ago in the Middle East, possibly from the Libyan wild cat (Felis sylvestris lybica). Unlike dogs, they have retained most of the morphological features of their wild ancestors.

19.1

The evolution of carnivory

The change in diet from an herbivorous diet to eating herbivorous prey resulted in a significant increase in the level of exposure of carnivores to potential microbial pathogens. This was in addition to changes in their intestinal microbiota Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00006-X © 2023 Elsevier Inc. All rights reserved.

297

298

SECTION | 2 Mammalian orders

brought about by a high protein, low fiber diet. Thus grazing herbivores acquire many environmental organisms from the surface of plants. Few of these bacteria are adapted to living in animals. As a result, the greatest risk of dietary infections in grazers is the ingestion of the fecal material from others of their kind. This in turn enabled parasitic gastrointestinal helminths to prosper and spread either indirectly through intermediate hosts or directly by ingestion. The situation in carnivores is very different. On ingesting even freshly killed prey, they will inevitably ingest some of its microbiota as well as any parasites and pathogens present. Any opening in the intestinal tract will release large quantities of bacteria and their accompanying viruses and fungi. In general, most of these will be nonpathogenic commensals. However, there is a significant chance of ingesting potential pathogens and parasites, especially if their prey is freshly killed. Even today in western societies uncooked meat is recognized as a major cause of food poisoning. The immune system of carnivores must adapt accordingly. It must not be assumed that carnivores preferentially eat healthy animals,
quite the contrary. They prey on the young, the old, the weak, and the sick. Animals on the verge of death are easy pickings and require much less risk and energy expenditure to kill. In effect, this results in selective predation on sick and parasitized prey species. When wolves were reintroduced in Yellowstone national park, one immediate effect was an improvement in the overall health of the elk herds as the wolves culled the sick. Wolves preying on wild boar in Spain have resulted in a marked reduction in the prevalence of tuberculosis in the boar population [3]. Some model systems have suggested that in some environments, predators may only survive as a result of the continuing availability of sick prey. Conversely, the selective removal of sick, infected prey species may serve to reduce the spread of infectious diseases and allow prey populations to prosper. Most predators will not care to distinguish between the almost dead and the recently dead. A logical extension of this principle resulted in the rise of carrion eaters. Preying on dead animals or the kills of others, such scavengers will ingest a diet of decomposing meat and bones. While distasteful to us, it must be recognized that the organisms growing in the decomposing body are also simply those of the normal commensal microbiota. Few will be pathogens per se. One exception will however, be the rise of the anaerobes. The decomposing body of a large mammal provides anaerobic bacteria with an ideal environment in which to multiply. Two notable pathogens that are adapted to, and prosper in such circumstances are anthrax (Bacillus anthracis) and tetanus (Clostridium tetani). On the other hand, a decomposing animal body will contain no more viable viruses. Other diseases that may be acquired by carnivores in this way include brucellosis, salmonellosis, botulism, toxoplasmosis, and giardiasis. Parasites may accumulate over the predator’s lifetime [4]. This raises the possibility that the infectious or parasite burden may eventually become clinically significant. Hunting of multiple prey over time may result in infections by both diverse pathogens and parasites, risking superinfections. The consequences of this are potentially catastrophic. For example, the spread of canine distemper from dogs to lions has resulted in mass lion mortality. Likewise, the acquisition of Yersinia pestis from infected Prairie dogs (Cynomys spp.) has resulted in the almost complete elimination of black-footed ferrets (Mustela nigripes). This year (2022), cats, bobcats, and foxes are dying from avian influenza after consuming sick and dying birds. Carnivores must adapt to these increases in the threat from bacterial pathogens. These adaptations can be both behavioral and immunological. Thus an improvement in the sense of smell will enable them to identify sick prey as a result of changes in the spectrum of odors coming from each individual [5]. Other behavioral changes include avoiding eating the carcasses of other related carnivores, reflecting a novel coevolutionary relationship between carnivores and their parasites and pathogens [6,7]. In effect, carnivores eat herbivores whenever possible. Other obvious changes include the maintenance of a relatively low, bactericidal gastric pH, an ability to regurgitate easily, the development of a specialized microbiota, and significant enhancement of the innate immune defenses of the body (Fig. 19.1) [8]. Vomit reflex

Specialized microbiota

Carnivores defenses

Low gastric pH

Avoid carnivore carcasses

Enhanced innate Immunity High neutrophil coumts High serum bactericdal activity

FIGURE 19.1 Some of the enhanced immune defenses of carnivores and carrion eaters.

Feliformes: The cats and their relatives Chapter | 19

299

Carnivores generally have gastrointestinal systems that rapidly expel or pass food, with simple stomachs and relatively short intestines. Such changes may increase the food passage rate and perhaps reduce the opportunities for parasites to be transmitted. Carnivores may also be able to detect parasites by their odors [5]. On the other hand, cats and other carnivores have lost the ability to taste certain flavors. Thus cats have no taste for sweetness, but bears do (They really do like “hunney”) [9]. As a matter of routine, many carnivores also supplement their meat intake with a limited amount of vegetable matter. Some carnivores cache their food for some time before eating it, a process that may also destroy some potential parasites as well as viruses. Another disadvantage of being at the top of the food chain is increased exposure to carcinogens. As a result, cancer risk is significantly greater in Carnivora than in Artiodactyls or primates [10]. Carnivores may accumulate environmental carcinogens derived from their prey. A high-fat, low-fiber diet is also a recognized risk factor for increased cancer risk. There is no evidence that carnivores have any specialized anti-cancer pathways in their immune systems suggesting that there has been minimal selection for anticancer immunity. This is likely because cancers tend to develop late in post-reproductive life.

19.1.1 Reproduction and lactation Cats have an endotheliochorial placenta. Thus their fetal chorionic epithelial cells are in direct contact with maternal endothelial cells. The placenta is of the zonary type that forms a complete band encircling the fetus. There may be hematomas located at the margins of the placenta. In the fetal kitten, lymphocytes are found in the blood at 25 days post-conception. B cells appear in the liver by 42 days post-conception [11]. While endotheliochorial placentas are considered to permit limited immunoglobulin transfer to kittens in utero, in practice the amount transferred is negligible. No IgG or IgA is detectable and only trace amounts of IgM are present in the serum of newborn domestic cats [12]. As a result, neonatal kittens are effectively agammaglobulinemic [13]. Periparturient cats do however, produce colostrum containing a very high concentration of IgG. Thus cat colostrum contains 40
50 g/L of IgG. This IgG concentration drops by 50% within the first 24 hours after birth [14]. The neonatal kitten absorbs IgG from the maternal colostrum for up to 16 hours after birth. Once absorption ceases, their serum IgG drops steadily with a half-life of 4.4 6 3.57 days for IgG and an IgA half-life of 1.93 6 1.94 days [12]. Serum IgM, presumably synthesized by the kitten itself, climbs steadily through the first weeks of life.

19.1.2 Hematology Total white blood cell counts are highly variable in domestic cats and have been reported to range from 4 to 28 cells 3 103/μL. A median value depends on the population studied but is in the region of 15
18 3 103/μL [15]. Within this population, there is also a great range of different cell types (Table 19.1). It has long been believed that carnivores possess relatively high numbers of blood neutrophils to enhance protection. It was believed, as discussed above, that a carnivorous diet increased the risks of acquiring infection or parasites. However, recent studies have indicated that there is also a significant correlation in carnivores between mating TABLE 19.1 The leukocytes of some selected Felidae. Note the different counts in the hyaena relative to the other species [16].

Total WBC 3 103/μL

Domestic cat

Lion

Cheetah

Striped hyena

Crab-eating mongoose

5.1
34.7

2.3
17

3.4
9.7

16.1
17.3

4
4.4

Neutrophils (%)

20
80

60

59
77

71
82

35
47

Lymphocytes (%)

12
76

31

12
35

8
16.5

47
55

Monocytes (%)

0
7

9

3
11

4
7.5

5
7

Eosinophils (%)

0
10

0

0
12

0
3.5

0
4

Basophils (%)

,1

,1

,1

,2

,1

Data from Hawkey CM. Comparative mammalian hematology. Cellular components and blood coagulation of captive wild animals. Heinemann Medical books, London, 1975.

300

SECTION | 2 Mammalian orders

Lymphocyte

Eosinophil

Neutrophil

Basophil

FIGURE 19.2 The leukocytes of cats. They are about 10 μm in diameter. Courtesy of Dr Mark Johnson.

promiscuity and elevated numbers of white blood cells [17]. There is only limited support for an association between the percentage of meat in the diet and white blood cell counts. Baseline WBC counts are higher in more promiscuous species. Many felids such as African lions, exhibit extreme polygyny and promiscuity, and accordingly, neutrophils constitute a very high proportion of their leukocyte counts (Fig. 19.2). While the numbers of neutrophils may not correlate with meat intake, it is clear that the bactericidal activities of feline serum are significantly greater than in other suborders of the Carnivora [18]. Thus analysis of bacterial killing in six free-ranging carnivores; the black-backed jackal (Canis mesomelas), brown hyaena (Hyaena brunnea), caracal (Caracal caracal), cheetah (Acinonyx jubatus), leopard (Panthera pardus ) and the lion (Panthera leo) showed that their bactericidal activity was independent of foraging behavior, body mass, or social organization but reflected their phylogeny. The assay involved exposing a standard number of E. coli to dilutions of the serum of each species. The bactericidal activity of the Feliformes was at least an order of magnitude higher than any other order. Even the cheetah which is believed to be immunologically disadvantaged had this very high activity. While the numbers of blood leukocytes in carnivores may correlate best with sexual promiscuity, it does appear that these cells may also be unusually effective killers of bacteria, especially in the Felidae [18]. Cheetahs appear to have relatively low genetic variability, especially in their major histocompatibility complex (MHC) loci. It was long assumed therefore that this lack of MHC variability would reduce their ability to respond to diverse antigens. As a result, it was also assumed that this was the cause of increased susceptibility to infectious diseases as well as poor reproductive success. However, these problems have only been reported in captive cheetahs. Free-ranging cheetahs show no obvious evidence of increased disease susceptibility or reduced reproductive capabilities. Leopards are sympatric with cheetahs over large areas of Africa. They hunt similar prey, and it is reasonable to assume that they are exposed to very similar pathogens. Unlike cheetahs, however, free-ranging leopards show considerable genetic diversity. For example, they have at least three MHC class I and three DRB class II loci with plenty of polymorphism [19]. Interesting differences have been observed when three immune parameters were measured in cheetahs and leopards from the same area of Namibia [20]. The three measures were adaptive immunity as measured by a hemagglutination/ hemolysis assay and IgG concentrations; induced innate immune-related immune responses as determined by measuring lysozyme and serum amyloid A (SAA); and constitutive innate immunity as determined by a serum bactericidal assay that largely measures complement activity. These studies demonstrated that while cheetahs had reduced induced innate and adaptive immune responses as compared to leopards, their constitutive innate immunity was considerably greater. For example, cheetahs had lower serum IgG levels than leopards, but they had greater bacterial killing activity and higher lysozyme levels. Thus it appears that cheetahs may compensate for a lack of competence in their adaptive immune system as a result of reduced MHC

Feliformes: The cats and their relatives Chapter | 19

301

diversity, by a compensatory increase in their constitutive innate immune responses. Two different investment strategies but each generates significant positive yields.

19.1.3 Innate immunity The acute-phase response to systemic infections in cats appears to differ in several respects from mammals in other orders. For example, C-reactive protein, a major acute phase reactant in humans and dogs is not one in cats. The three major acute phase proteins in cats are serum amyloid A, alpha-1-acid glycoprotein, and haptoglobin [21]. Serum amyloid A serves as a chemoattractant to draw inflammatory cells to sites of microbial invasion. It regulates the intensity of inflammatory processes by inhibiting myeloperoxidase release and inhibiting lymphocyte proliferation and it is also involved in lipid metabolism and transport. Alpha-1-acid glycoprotein has anti-inflammatory and immunomodulatory effects. It inhibits complement activities and promotes the release of IL-1RA from macrophages. Haptoglobin is another major acute-phase protein in cats. In feline infectious peritonitis (FIP) caused by a coronavirus, blood SAA levels increase 10- to 50-fold. They are also elevated in diabetes mellitus, other infectious diseases, traumatic injury, and cancer. α1-Acid glycoprotein also increases in FIP, although this is usually less than 10-fold. It also increases calicivirus, chlamydiosis, feline leukemia, and feline immunodeficiency virus infections. Haptoglobin usually increases 2- to 10fold and is especially high in FIP.

19.1.3.1 Toll-like receptors Feline TLR genes 1
3 and 5
8 have been partially cloned [22]. Sequence homology with human TLRs ranges from 81% for TLR5 to 89% for TLR7. They have also been quantified in normal cat lymphoid tissues. Different expression patterns are found in the spleen, mesenteric lymph nodes, retropharyngeal lymph nodes, thymus, intestinal intraepithelial lymphocytes, and lamina propria lymphocytes. For example, TLR1 was detected in the spleen but not in any other lymphoid organ. TLR6 could not be detected in any of the organs tested. In the thymus, TLR7 and TLR9 predominated. In the small intestine, TLRs 2, -4, -5, and -7 were found, while in the retropharyngeal lymph nodes TLRs 7 and -9 predominated. Expression of TLRs 2, -5, -8, and -9 was very much higher in the mesenteric lymph node than in the retropharyngeal lymph node. Nevertheless, TLRs 1, -4, and -6 could not be detected there. Feline B cells, CD41 and CD81 T cells all express TLRs 2, -5 and 7
9. The relative levels of expression vary among different cell phenotypes. Hyaenas also possess TLRs 1
10 [23].

19.1.3.2 Inflammasomes Species within the order Carnivora demonstrate a progressive downgrading of their pathogen-sensing inflammatory pathways. Thus cats and dogs lack certain nucleotide-oligomerization domain leucine-rich domain receptors (NLRs). In addition, they have acquired the expression of a unique hybrid caspase-1/4 [24]. This caspase processes gasdermin D pore formation without inducing rapid lytic cell death (pyroptosis) which is a normal response in other species. It is a poor mediator of NLRP3 and caspase-4-dependent inflammasome activation. Thus carnivore inflammasomes are driven by caspase-8 rather than by caspase-1/4. Caspase-8 is an inefficient activator of interleukin-1β. These NLRP3 inflammasomes regulate gut immunity so this defect must be compensated for in some way. As in bats, this loss of function of NLRP3 may assist some carnivores in tolerating some persistent virus infections without developing sickness themselves.

19.1.3.3 Natural killer cells Cats, like other mammals, possess natural killer cells. They are present in blood and lymphoid tissues. They are nonadherent, express Fc receptors (CD16), are activated by IL-2, and contain large cytoplasmic granules. Cats possess both NK cells (CD32 CD561) and NKT cells (CD31 CD561). Their numbers, both relative and absolute, are similar to other mammals [25]. Thus in blood, they average 227 NK cells/μL. This is approximately 5.2% 6 2.5% of their total blood lymphocytes. Cat NK cells are significantly larger and more granulated when compared to T cells and are similar to human NK cells in size and granularity. There are however, other clear differences. For example, between 10% and 30% of feline NK cells stain positively for the integrin CD11b and the selectin CD62L.

302

SECTION | 2 Mammalian orders

FIGURE 19.3 A feline dendritic cell cultured in the presence of recombinant human IL-4 and GM-CSF. Note the extensive dendrites so characteristic of these cells. 3 100. From Sprague WS, Pope M, Hoover EA: Culture and comparison of feline myeloid DCs vs macrophages, J Comp Pathol 133:139, 2005. From Tizard Vet Immunology 10 ed.

19.2

Dendritic cells

Feline dendritic cells are readily induced by exposure to GM-CSF and IL-4. Feline Langerhans cells are CD181, MHC class II1, CD1a1, and CD41. Dendritic cells derived from feline blood mononuclear cells are CD11, CD141, and MHC class I and II1 (Fig. 19.3) [26].

19.3

Cytokines

Cytokines such as IL-6 and IL-8 are released into the bloodstream in mammals, including cats suffering from inflammatory diseases such as sepsis or chronic kidney disease or when experimentally challenged with PAMPs such as bacterial lipopolysaccharides. Interestingly, these cytokines also play a role in the regulation of aggression, a marked feature of carnivores such as cats. For example, microinjection of IL-1 into the medial hypothalamus and the periaqueductal gray matter potentiates defensive rage behavior in the cat (hissing)! [27] Interleukin 2 also modulates defensive rage behavior by acting through the substance P and NK1 receptors. Preliminary data suggests that these cytokines may also play a role in mediating predatory attack behavior [28].

19.3.1 Lymphoid organs 19.3.1.1 Thymus The cat thymus is of conventional structure and rich in Hassall’s corpuscles. It atrophies as the animals age. By about 18 weeks-of-age, it is largely replaced by adipose and connective tissue originating from the adventitia of its blood vessels [29].

19.3.1.2 Spleen The spleen of the cat is of high relative weight and both the capsule and trabeculae contain smooth muscle cells [30]. The spleen is rich in red pulp but has relatively less white pulp. Thus the cat’s spleen is a major blood storage organ.

19.4

Bronchus-associated lymphoid tissue

Both nodular and diffuse lymphoid tissues are located along the bronchial tree in healthy cats. These extend distally into the walls of the terminal bronchioles. They are located in the tunica adventitia of the bronchi and bronchioles and form distinct nodules with germinal centers. The diffuse form is found primarily in the submucosa of the bronchial tree [31]. Cats also possess pulmonary intravascular macrophages and hence trap about 80% of blood-borne particulates in their lungs (Fig. 19.4).

Feliformes: The cats and their relatives Chapter | 19

303

6.5% 80%

86%

14%

FIGURE 19.4 Dogs and cats differ greatly in the organ in which blood-borne particulates are trapped. Thus cats possess pulmonary intravascular macrophages whereas dogs do not. As a result, when colloidal carbon is injected intravenously, dogs trap 86% in their liver whereas cats trap 80% in their lungs. DP

B A

DN(O)

A

DM

DO

DR

A B

B

B B A BB A A

BC

II

19.5

FIGURE 19.5 The structure of the feline MHC. Note that an inversion breakpoint has occurred between TRIM 39 and TRIM 26. As a result, the disjointed region is now located adjacent to the chromosome B2 telomere.

E FG A

III

I

I TRIM39

I TRIM26

Mucosal lymphoid tissues

Cats possess four tonsils, lingual, paired palatine, paired paraepiglottic, and the pharyngeal tonsil [32]. The lingual tonsil is small and only detectable histologically. The palatine tonsil is relatively large. It is oval with a central sinus lined by stratified squamous epithelium. It contains many follicles, and the surrounding tissue and epithelium are heavily infiltrated by lymphocytes. The pharyngeal tonsil is located on the roof of the nasopharynx. It has a smooth surface with no tonsillar crypts. As in other mammals, the feline intestinal lymphoid tissues consist of organized structures such as the Peyer’s patches as well as huge numbers of lymphocytes scattered diffusely within the intestinal lamina propria. The Peyer’s patches consist of 4
6 areas of aggregated lymphoid nodules in the jejunum measuring 4
30 mm in length. They also have a single elongated patch close to the ileocecal junction that can reach up to 10 cm in length [33]. The distribution of cells in feline Peyer’s patches appears to be similar to other mammals with a preponderance of B cells over T cells [33]. Cats appear to possess an unusually large number of intraepithelial lymphocytes. These are primarily CD81α/α cells (40%). There are also many double-negative T cells (44%) in the intestinal epithelium [34]. Cats possess especially large numbers of lymphoid aggregates in their anal canal and terminal rectum where feces are temporarily stored.

19.5.1 The major histocompatibility complex The feline MHC complex (FLA) is similar to that of other eutherians and the three gene classes are arranged in the usual order I, III, II on chromosome B2 (Fig. 19.5) [35].

19.5.1.1 The MHC Class I region As in other mammals, the feline class I region consists of multiple functional genes separated by interspersed framework regions. The class I genes encode single polypeptides that dimerize with β2 microglobulin. An inversion breakpoint has occurred between TRIM 39 and TRIM 26 in the cat. As a result, the disjointed region is now located adjacent to the B2 telomere. This has resulted in the development of a gap in the distal class I region. The gap effectively separates B2.85 mb of the MHC that contains the class II and the proximal class I region from B0.50 mb of the extended class I region [35]. In the proximal class I subregion, there are 17 FLA class I genes designated FLA1-A to FLA1-Q occupying about 600 kb on chromosome B2q. (These correspond to HLA-B and -C in humans) [36]. In the 800 kb central subregion there are two additional class I genes, FLA1-rp, a gene fragment, and FLA1-s, a full-length gene. There are no immune system encoded genes in the inverted distal class I region. In both the cat and the dog, since the split of the Carnivora from humans and primates B80 mya the entire HLA-A and HLA-E subregions have been lost

304

SECTION | 2 Mammalian orders

Of the full-length class I FLA loci in the domestic cat, three, FLA1-E, FLA1-H, and FLAI-K, encode products that are highly polymorphic and have all the components needed to be considered classical (class Ia) MHC molecules although there is no direct evidence that they are MHC restricting [36]. They have 26, 27, and 20 different allelic sequences respectively, reported to date. Transcripts of these genes can be found in diverse tissues. The remainder of the full-length genes, FLAI-A, -C, -F, -J, -L, -M, -O, are considered class Ib candidates. Of these, four have been shown to be transcriptionally active, FLA1-A, -J, -L, and -O, and are assumed to be non-classical MHC class Ib genes. FLA1-F is a pseudogene [37]. The polymorphism of these class I gene products is concentrated in three hypervariable regions in exons 2 and 3 as would be expected since these encode the α1 and α2 antigen-binding domains. Thirty-one class Ia alleles have been found in 17 cats. The alleles segregate strongly by loci. Eighteen different class Ib alleles have also been identified. In that case, their diversity has been found to be located outside their antigen-binding sites.

19.5.1.2 The MHC class II region The feline extended class II region is 758 kb in size. It contains 29 genes. Of these, 19 are homologous to known HLA genes. However, when compared to humans and mice there have been major gene gains and losses. The FLA-DR region contains four DRB genes and three DRA genes (compared to humans one DRA and 3
5 DRB genes). However, the cat lacks an entire DQ region and retains only relict DP pseudogenes. On the other hand, this has been compensated for by the expansion of the seven DR genes. The feline DR region is also very different from the human in its gene order suggesting that repeated duplication and inversion events have occurred. About 35% of the feline class II contains multiple repetitive gene families with a very different density and abundance than in other species [38]. Cats have 3
6 times fewer endogenous retroviral long terminal repeats across their class II region compared to humans and mice. The DR subregion in the cat has been rearranged by four gene duplications and one inversion event [38]. Sequence analysis indicates that DRB4 and DRA3 are the most divergent of the cat DR genes. They appear to encode functional β and α chains. Two of the DR gene pairs, DRB1-DRA-1 and DRB3-DRA2 appear to be very closely related and may be the product of a recent duplication event. In total contrast, the regions containing the classical class II genes, DP, DQ, and DR have been reorganized. The cat has only retained the paired DR functional genes. A pair of DPA and DPB genes have been identified in the cat, but they appear to be pseudogenes. It appears that the human and cat DPA pseudogenes are descended from a common ancestral pseudogene. The DQ region is either missing or very much shortened in the cat. The cat appears to be the only mammal analyzed to date that lacks the entire DQ region. There is considerable conservation of gene sequences between the cat and humans. This is most evident in the extended class II region as well as in the classical class II region with the presence of the genes encoding DOA, RING3 kinase, DMA, DMB, LMP2, TAP1, LMP7, TAP2, and DOB genes [38].

19.6

The natural killer cell receptor complex

Feline NK cells are large granular lymphocytes found in the blood and spleen. In general, cat NK cell numbers and properties are similar to those in other domestic mammals. However, unlike other species, 10%
30% of feline NK cells are CD11b1 and selectin CD62L1. They are cytotoxic for feline target cells infected with feline leukemia virus, herpesvirus, or vaccinia virus. In domestic dogs and cats, both species have a single Ly49 gene. The cat Ly49 is functional, but the dog Ly49 product lacks a conserved cysteine in its C-type lectin domain and may therefore be defective.

19.6.1 B cells and immunoglobulins The mean serum concentration of IgG in cat serum is 19.08 6 5.38 mg/mL.; IgM is 2.04 6 0.83; while IgA is 2.6 6 2.16 mg/ml [39].

19.6.1.1 IGH genes The feline IGH locus is located on chromosome B3. Cats have two IGHG genes, one IGHM gene, and possibly two IGHA genes (IgA1 and IgA2), as well as two possible IGHE genes (Fig. 19.6). The two IgG genes appear to be allotypes of one subclass and have been designated IgG1a and IgG1b. Their products account for about 98% of serum IgG.

Feliformes: The cats and their relatives Chapter | 19

IGH

IGK

IGL

5'

3'

64

7 6

V

D J

19

M

G1

G2

E

E

A1

305

FIGURE 19.6 The organization of the feline immunoglobulin heavy and light chain loci.

A2

5

94 11 J-C clusters

M

D

G

E

A

V

D

J

LC

About 2% of serum immunoglobulins belong to a second uncharacterized IgG subclass. Lu et al. have reported that there are two IgG constant chain subclass genes, IGHG1 and IGHG2 with IGHG1 predominating (98%) [40]. There are also two alleles of this gene Cγ1a and Cγ1b, encoding IgG1a and IgG1b proteins respectively. IgG1a constitutes 62% and IgG1b 36% of serum IgG [40]. Both have a high affinity for the FcγR1 receptor and very similar functional properties [41]. As in other species, IgA is the predominant immunoglobulin produced on feline mucosal surfaces, especially in the small intestine where 40%
80% of the plasma cells are producing IgA. IgM-producing intraepithelial lymphocytes have been reported in the feline jejunal lamina propria. If confirmed, this is unusual. However, IgG-producing cells predominate in the cat colon [33].

19.6.1.2 Immunoglobulin heavy chains Cats possess 64 IGHV genes with 42 functional sequences and 22 pseudogenes. The expressed IGHV genes map to chromosomes B3 and D1 [40]. As in the dog, the feline IGHV repertoire is dominated by a preponderance of the IGHV3 subfamily (Clan III) Thus all cats show a similarly skewed IGHV CDR-H3 profile. There are very few orthologs of other species. The dominance of IGHV3 is great with about 99% of the sequences clustering with this subfamily. The remaining 1% consists primarily of IGHV1 subfamily members. Cats have seven IGHD as well as six IGHJ genes, as in dogs. Their use is dominated by the human IGHD3 analog. J gene usage is dominated by IGHJ4 (60%
70%).

19.6.1.3 Immunoglobulin Light chains Cat immunoglobulins employ principally lambda light chains. The ratio of λːκ expression in their immunoglobulins is about 3:1 [11]. Expressed Ig lambda sequences have been mapped to 47 IGLV genes and five IGLC genes on chromosome D3 [40]. In contrast, expressed kappa chain variable sequences map to 13 IGKV genes and one IGKC gene located on chromosome A3. The IGLV CDR3 regions are tightly constrained in their length [42].

19.6.2 T cells and cell-mediated immunity Studies on the phenotypes of peripheral blood leukocytes in domestic cats have demonstrated major differences in lymphocyte populations as a result of age [43]. Thus T cell numbers declined from approximately 35% of total white cells in newborn kittens to about 10% in cats aged 16 years or more. This decrease in their relative levels was associated with a significant increase in CD41 T cells and a smaller decrease in CD81 T cells resulting in a major change in the CD4: CD8 ratio. Thus in newborn kittens, the ratio averages slightly over two but by 12
14 years this had declined to about one [43]. There is also a significant decline in the relative numbers of CD211 cells. (CD21 is a complement receptor that regulates B cell activation.)

306

TRA/D

TRB

SECTION | 2 Mammalian orders

5'

62 AV

11 DV

32

6

5 2 DD DJ DC

64 AJ

FIGURE 19.7 The structure of the feline T cell receptor loci. ψ 5 pseudogene.

3'

AC

6

3

TRG

2

2 5 C\

V

D

J

2 J\ C\

C

Studies on the diffuse T lymphocyte populations in the feline small intestine found that 40% of the CD8α1 T cells expressed CD8α/α chains while 44% were double-negative cells (CD42 CD82). There is also a population of intraepithelial lymphocytes that are CD31 CD11d1 [34]. The CD4:CD8 ratio in these tissues is 0.25. Thus this diffuse lymphocyte population is generally similar to that seen in mice and humans. Intraepithelial lymphocytes (IELs) are located within the intestinal epithelium in close proximity to the basement membrane. There are major differences in these populations between mammals. Thus their numbers range from 12%
20% of the epithelial cells in dogs to 51% in pigs. However, compared to dogs, cats have elevated numbers of intraepithelial lymphocytes. Thus cat IELs are more frequent in villous than crypt epithelium and their relative number rises from about 50% of the epithelial cells in the duodenum to 80% in the ileum [33]. Dogs have similar numbers in their duodenum and ileum. In the large intestine, the proportion of IELs is 4%
5% in both dogs and cats. In both species, these cells are predominantly CD81 (66% in cats). It is tempting to suggest that these populations reflect dietary differences. As in other mammals, cats use two distinct lineages of T cells based on whether their antigen receptors are α/β heterodimers or γ/δ heterodimers. These four chains are encoded by three loci, TRB, TRG, and the combined TRA and TRD locus (Fig. 19.7). The feline TRA/D locus is located on chromosome B3 while the TRB and TRG loci are located on chromosome A2 [44]. Each complete chain is encoded by a variable, (diversity), and joining genes that encode the antigen-binding variable region and attach this to a constant region that is embedded in the T cell membrane in the form of either an alpha-beta heterodimer or as a gamma-delta heterodimer. The joining of the V, D, and J genes is guided by a recombination signal sequence that flanks each of these genes, The T cell-specific endonuclease recombinase activating genes RAG1 and RAG2 initiate the joining process by first breaking the DNA strand between the coding sequences and their adjoining RSS. As described elsewhere, various forms of joining occur with both insertions and deletions, and the ends are rejoined to form a hypervariable region. This region forms the complementarity determining region CDR3 and together with the CDR1 and CDR 2 regions that are germline-encoded, forms the very specific T cell antigen binding site. Alignment of feline T cell receptor T genes forms monophyletic clades with their canine orthologs. It is increasingly clear that human, dog, and cat T cell receptor V genes are extensively intermingled rather than forming separate clades. Clustering of multimember subfamilies frequently occurs in the V genes located at the 5’ ends of the TCR loci [45]. All these results are compatible with the birth and death model of gene evolution.

19.6.2.1 TRA/D The feline TRA/D locus is very similar to that of humans, mice, and canines with only minor differences in gene numbers. It is found in a segment of about 800 kb on chromosome B3. The TRA locus contains 127 genes. These include 62 TRAV gene segments of which 15 are pseudogenes, 64 TRAJ segments of which three are pseudogenes, and one functional TRAC. The 62 TRAV genes belong to 38 subfamilies of which 32 contain only a single gene. The TRAD locus contains 19 genes including 11 TRDV (three pseudogenes), two TRDD (both functional), five TRDJ (one pseudogene), and one functional TRDC. The TRD repertoire is dominated by TRDJ3. A sequence of A and D V genes is followed by a TRD D-J-C cassette and then by an inverted TRDV gene. This is followed by a cluster of TRAJ genes followed by the single TRAC gene.

Feliformes: The cats and their relatives Chapter | 19

307

19.6.2.2 TRB The feline TRB locus is structurally similar to that in humans, dogs, ferrets, and rabbits. It spans 302 kb on chromosome A2. It contains 48 genes; 33 TRBV, two TRBD, 12 TRBJ, and two TRBC [46]. Of the TRBV genes, 20 are functional, four are ORF, and nine are pseudogenes. Both TRBD genes are functional as are eight of the TRBJ genes and both TRBC genes. The V genes are arranged upstream of two D-J-C clusters. TRBV gene usage shows some bias but the TRBJ genes in the second cluster are more commonly rearranged than the J genes in the first cluster. As in many other mammals, one V gene (TRBV30) is located downstream of the second D-J-C cluster and is in inverted transcriptional orientation. Each D-J-C cluster contains one TRBD gene, six TRBJ genes, and one TRBC gene [45]. The functional TRBV genes fall into 27 subfamilies [46].

19.6.2.3 TRG The feline TRG locus spans about 260 kb and is located in the pericentromeric region of chromosome A2 [45]. It most closely resembles that of the dog although cassettes 4 and 5 in the cat are inverted relative to the dog indicating a postspeciation event [47]. It contains 30 genes; 12 TRGV genes of which six are functional and the other six are pseudogenes, 12 TRGJ genes of which four are functional, two have open reading frames, six are pseudogenes, and six TRGC of which four are functional and two are pseudogenes. It is arranged in five complete and one incomplete V-J-(J)-C cassettes. There is preferential intra-cassette over inter-cassette usage and dominant usage of the TRGV2
1 and TRGJ1
2 genes. The TRGV genes are classified into six subfamilies two with four members and the other four having only single members [47]. The six functional TRGV genes encode a conserved amino acid motif IHWY located at the beginning of the framework region 2 [48]. Most of the rearrangements involve the four TRGV2 subfamily genes. Canine TRGC1 and TRGV3 orthologs are absent from the cat genome.

19.7

Other cats

The MHC structure has been investigated in several populations of wild felids to assess their level of genetic diversity. As mentioned above, free-ranging leopards in Namibia demonstrate extensive diversity in both their MHC class I and class II DRB loci [19]. Similar studies on wild and captive tigers in India found significant diversity in their MHC class I alleles but relatively little DRB diversity [49]. There were no significant differences between the captive and wild tigers tested. On the other extreme, studies on the highly endangered leopard cat (Prionailurus bengalensis) in Japan where the wild population totals about 200 individuals, detected relatively low diversity in their DRB loci and implying low resistance to introduced pathogens [50].

19.8

Hyenas

Among the other important feliform species are the Hyaenas. While apparently exposed on a consistent basis to potential pathogens from their diet of dead animals, hyaenas rarely show signs of infectious disease. This presumably reflects some remarkable abilities on the part of their immune systems. One obvious factor that may influence their resistance to otherwise lethal infections is social status. Hyaenas live in complex societies where status is determined by sex, reproductive state, and most importantly social standing. Studies on key immune functions of hyaenas have determined that some aspects of their immunity are directly correlated with their social rank [51,52]. Higher ranking hyaenas generally have better access to food and are thus nutritionally privileged. Two types of immune defenses, complement-mediated bactericidal effects, and total serum IgM levels are correlated with social rank in wild hyaenas. These could be directly related to protein intake and nutritional quality. In addition, if high-status females get first access to carcasses. They may ingest more nutritious organs and less decomposed meat and have lower exposure to potential pathogens. This may also have nutritional benefits. Conversely, total IgM levels are not associated with social rank. As discussed previously, carnivores with promiscuous breeding systems tend to have high WBC counts [17]. Thus spotted hyenas (Crocuta crocuta), tend to live in very large groups of up to 80 individuals. Within these groups, females are dominant over males. All the females within the group mate and maintain massive creches containing multiple litters. Female spotted hyaenas that are socially dominant have a higher serum bactericidal activity and higher IgG and IgM levels than do males. As in other species however, these measures of immunity are lower in lactating than in pregnant females reflecting the high energy costs of lactation. Thus immune defenses are costly and are regulated by multiple social and environmental variables in wild animals [51]. There are extreme interactions between the individuals within the group. Note their WBC counts in Table 19.1.

308

SECTION | 2 Mammalian orders

It is of interest to note that wild spotted hyaenas have significantly higher IgM and IgG concentrations than captive hyaenas. This is a commonly observed feature when captive mammals are compared to free-living ones (Chapter 23). These also include natural antibodies and autoantibodies [52]. Interestingly there is no difference in serum bactericidal activity between the two groups. It is believed that this may simply reflect the higher parasite burden carried by the wild animals. Another feature of the hyaena immune system is their MHC and the forces shaping it, and by extension antigen presentation and adaptive immunity [53]. In both spotted hyaenas (C. crocuta) and striped hyaenas (Hyaena hyaena), there is evidence pointing to strong positive selection favoring MHC diversity. Thus the percentage of positively selected sites was 14.6% for DRB and 13.9% for DQB. These polymorphisms are located within antigen-binding sites. These two species have major differences in their mating behavior and social systems (Striped hyaenas live in small family groups and feed solitarily and basically avoid other carnivores). Yet the MHC diversity and obvious selection pressures are much the same. Thus these appear to relate primarily to their carrion feeding behavior rather than their social status or sexual selection.

References [1] Murphy WJ, Foley NM, Bredemeyer KR, Gatesy J, Springer MS. Phylogenomics and the genetic architecture of the placental mammal radiation. Annu Rev Anim Biosci 2020;. Available from: https://doi.org/10.1146/annurev-animal-061220-023149. [2] Agnarsson I, Kuntner M, May-Collado LJ. Dogs, cats and kin: a molecular species-level phylogeny of carnivora. Mol Phyl Evol 2010;54:726
45. [3] Tanner E, White A, Acevedo P, Balseiro A, et al. Wolves contribute to disease control in a multi-host system. Sci Reps 2019;. Available from: https://doi.org/10.1038/s41598-019-44148-9. [4] Malmberg JL, White LA, VandeWoude S. Bioaccumulation of pathogen exposure in top predators. Trends Ecol Evol 2021;36(5):411
20. [5] Tizard IR, Skow L. The olfactory system: the remote sensing arm of the immune system. Anim Hlth Res Revs 2021;. Available from: https:// doi.org/10.1017/s1466252320000262. [6] Moleo´n M, Martinez-Carrasco C, Muellerklein OC, Getz WM, et al. Carnivore carcasses are avoided by carnivores. J Anim Ecol 2017;. Available from: https://doi.org/10.1111/1365-2656.12714. [7] Beasley DE, Koltz AM, Lambert JE, Fierer N, Dunn RR. The evolution of stomach acidity and its relevance to the human microbiome. PLoS One 2015;. Available from: https://doi.org/10.1371/journal.pone.0134116. [8] Blumstein DT, Rangchi TN, Briggs T, Andrade FS, Natterson-Hprowitz B. A systematic review of carrion eaters’ adaptations to avoid sickness. J Wildl Dis 2017;53(3):577
81. [9] Jiang P, Josue J, Li X, Glaser D, et al. Major taste loss in carnivorous mammals. Proc Natl Acad Sci USA 2012;109(13):4956
61. [10] Vincze O, Colchero F, Lemaitre J-F, Conde DA, et al. Cancer risk across mammals. Nature 2022;601:263
7. [11] Klotz FW, Gathings WE, Cooper MD. Development and distribution of B lineage cells in the domestic cat: analysis with monoclonal antibodies to cat μ 2 , γ 2 , κ 2 , and λ 2 chains and heterologous anti-α antibodies. J Immunol 1985;134(1):95
100. [12] Casal ML, Jezyk PF, Giger U. Transfer of colostral antibodies from queens to their kittens. Am J Vet Res 1996;57(11):1653
8. [13] Claus MA, Levy JK, MacDonald K, Tucker SJ, Crawford PC. Immunoglobulins in feline colostrum and milk, and the requirement of colostrum for passive transfer of immunity to neonatal kittens. J Fel Med Surg 2006;8:184
91. [14] Chastant-Maillard S, Aggouini C, Albaret A, Fournier A, Mila H. Canine and feline colostrum. Reprod Dom Anim 2017;52(Suppl 2):148
52. [15] Osbaldiston GW. Haematological values in healthy cats. Brit Vet J 1978;134:524
36. [16] Hawkey CM. Comparative mammalian hematology. Cellular components and blood coagulation of captive wild animals. London: Heinemann Medical books; 1975. [17] Nunn CL, Gittleman JL, Antonovics J. A comparative study of white blood cell counts and disease risk in carnivores. Proc Roy Soc Lond B 2003;270:347
56. [18] Heinrich SK, Wachter B, Aschenborn OHK, Thalwitzer S, et al. Feliform carnivores have a distinguished constitutive innate immune response. Biol Open 2016;5:550
5. [19] Castro -Prieto A, Wachter B, Melzheimer J, Thalwitzer S, Sommer S. Diversity and evolutionary patterns in free-ranging Namibian leopards (Panthera pardus pardus). J Heredity 2011;102(6):653
65. [20] Heinrich SK, Hofer H, Courtiol A, Melzheimer J, et al. Cheetahs have a stronger constitutive innate immunity than leopards. Sci Rep 2017;7:44837. Available from: https://doi.org/10.1038/srep44837. [21] Ceron JJ, Eckersall PD, Martinez-Subiela S. Aute phase proteins in dogs and cats: current knowledge and future perspectives. Vet Clin Pathol 2005;34(2):85
99. [22] Ignacio G, Nordone S, Howard KE, Dean GA. Toll-like receptor expression in feline lymphoid tissues. Vet Immunol Immunopathol 2005;10:229
37. [23] Flies AS, Maksimoski M, Mansfield LS, Weldele ML, Holekamp KE. Characterization of toll-like receptors 1
10 in spotted hyaenas. Vet Res Commun 2014;38(2):165
70.

Feliformes: The cats and their relatives Chapter | 19

309

[24] Digby Z, Tourlomousis P, Rooney J, Boyle JP, et al. Evolutionary loss of inflammasomes in the Carnivora and implications for the carriage of zoonotic infections. Cell Rep 2021;. Available from: https://doi.org/10.1016/j.celrep.2021.109614. [25] Vermuelen BL, Devriendt B, Olyslaegers DA, Dedeurwaerder A, et al. Natural killer cells: frequency, phenotype and function in healthy cats. Vet Immunol Immunopathol 2012;69
78. [26] Summerfield A, Auray G, Ricklin M. Comparative dendritic cell biology of veterinary mammals. Ann Rev Anim Biosci 2015;3:533
57. [27] Siegel A, Bhatt S, Bhatt R, Zalcman SS. Cytokines and aggressive behavior. In: A. Siegel S.S Zalcman (eds.), The Neuroimmunological basis of behavior and mental disorders. 2008. https://doi.org/10.1007/978-0-387-84851-8-12. Springer Science. [28] Zalcman SS, Siegel A. The neurobiology of aggression and rage: role of cytokines. Brain Behav Immun 2006;20:507
14. [29] Dearth OA. Late development of the thymus in the cat: nature and significance of the corpuscles of Hassall and cystic formations. Am J Anat 1928;41(2):321
51. [30] Udroiu I, Sgura A. The phylogeny of the spleen. Quart Rev Biol 2017;92(4):411
43. [31] Al-Tikriti MS, Khamas WA, Henry RW. Light microscopy of bronchial associated lymphoid tissue of healthy domestic cat with suggested new nomenclature. Anat Physiol 2012;. Available from: https://doi.org/10.4172/2161-0940.1000104. [32] Casteleyn C, Breugelmans S, Simoens P, Van den Broeck W. The tonsils revisited: review of the anatomical localization and histologic characterization of the tonsils of domestic and laboratory animals. J Immunol Res 2011;. Available from: https://doi.org/10.1155/2011/472-460. [33] Stokes C, Waly N. Mucosal defense along the gastrointestinal tract of cats and dogs. Vet Res 2006;37:281
93. [34] Roccabianca P, Woo JC, Moore PF. Characterization of the diffuse mucosal associated lymphoid tissue of feline small intestine. Vet Immunol Immunopathol 2000;75:27
42. [35] Beck TW, Menninger J, Murphy WJ, Nash WG, et al. The feline major histocompatibility complex is rearranged by an inversion with a breakpoint in the distal class I region. Immunogenetics 2005;56:702
9. [36] Holmes JC, Scholl EH, Dickey AN, Hess PR. High-resolution characterization of the structural features and genetic variation of six feline leukocyte antigen class I loci via single molecule real-time (SMRT) sequencing. Immunogenetics 2021;73:381
93. [37] Holmes JC, Holmer SG, Ross P, et al. Polymorphisms and tissue expression of the feline leukocyte antigen class I loci FLAI-E, FLAI-H, and FLAI-K. Immunogenetics 2013;65:675
89. [38] Yuhki N, Beck T, Stephens RM, Nishigaki Y, et al. Comparative genome organization of human, murine and feline MHC class II region. Genome Res 2003;13(6a):1169
79. Available from: https://doi.org/10.1101/gr.976103. [39] Harley R, Gruffydd-Jones TJ, Day MJ. Determination of salivary and serum immunoglobulin concentrations in the cat. Vet Immunol Immunopathol 1998;65:99
112. [40] Lu Z, Tallmadge RL, Callaway HM, Felippe MJB, Parker JSL. Sequence analysis of feline immunoglobulin mRNAs and the development of a felinized monoclonal antibody specific to feline panleukopenia virus. Sci Rep 2017;7(1):12713. Available from: https://doi.org/10.1038/s41598017-12725-5. [41] Strietzel CJ, Bergeron LM, Oliphant T, Mutchler VT, et al. In vitro functional characterization of feline IgGs. Vet Immunol Immunopathol 2014;158:214
23. [42] Steiniger SCJ, Glanville J, Harris DW, Wilson TL, et al. Comparative analysis of the feline immunoglobulin repertoire. Biologicals 2017;46:81
7. [43] Heaton PR, Blount DG, Mann SJ, Devlin P, et al. Assessing age-related changes in peripheral blood leukocyte phenotypes in domestic shorthaired cats using flow cytometry. J Nutr 2002;132:1607S
9S. [44] Cho K-W, Youn H-Y, Okuda M, Satoh H, et al. Cloning and mapping of the cat (Felis catus) immunoglobulin and T cell receptor genes. Immunogenetics 1998;47:226
33. [45] Radtanakatikanon A, Keller SM, Darzentas N, Moore PE, et al. Topology and expressed repertoire of the Felis catus T cell receptor loci. BMC Genomics 2020;. Available from: https://doi.org/10.1186/s12864-019-6431-5. [46] Pe´gorier P, Bertignac M, Chentli I, Ngoune VN, et al. IMGT biocuration and comparative study of the T cell receptor beta locus of veterinary species based on homo sapiens TRB. Front Immunol 2020;11:821. Available from: https://doi.org/10.3389/fimmu.2020.00821. [47] Antonacci R, Massari S, Linguiti G, Jembrenghi AC, et al. Evolution of the T-cell receptor (TR) loci of the adaptive immune response: the tale of the TRG locus in mammals. Genes 2020;11(6):624. Available from: https://doi.org/10.3390/genes11060624. [48] IMGT Repertoire (IG and TR). Domestic cat (Felis catus). ,http://www.imgt.org/IMGTrepertoire/index.php.. [49] Pokorny I, Sharma R, Goyal SP, Mishra S, Tiedemann R. MHC class I and MHC class II DRB gene variability in wild and captive Bengal tigers (Panthera tigris). Immunogenetics 2010;62:667
79. [50] Saka T, Nishita Y, Masuda R. Low genetic variation in the MHC class II DRB gene and MHC-linked microsatellites in endangered island populations of the leopard cat (Prionailurus bengalensis) in Japan. Immunogenetics 2018;70:115
24. [51] Flies AS, Mansfield LS, Flies EJ, Grant CK, Holekamp KE. Socioecological predictors of immune defenses in wild spotted hyaenas. Funct Ecol 2016;30:1549
57. [52] Flies AS, Mansfield LS, Grant CK, Weldele ML, Holekamp KE. Markedly elevated antibody responses in wild vs captive spotted hyaenas show that environmental and ecological factors are important modulators of immunity. PLoS One 2015;10(10):e0137679. Available from: https://doi.org/10.1371/journal.pone.0137679. [53] Califf KJ, Katzloff EK, Wagner AP, Holekamp KE, Williams BL. Forces shaping majpr histocompatibility complex evolution in two hyaena species. J Mammal 2013;94(2):282
94.

This page intentionally left blank

Chapter 20

Caniforms: Dogs, bears, and their relatives

Patagonian Fox. Lycalopex griseus

The Carnivora, probably diverged from the other placentals, most likely the Perissodactyls, about 75 million years ago (mya) during the late Cretaceous period [1]. Like other mammals, they probably began their major expansion and diversification after the dinosaurs had been eliminated by the K-Pg event. The first carnivores that evolved during the Paleocene were probably small and weasel-like creodonts. Around 42 mya, however, the crown group split into two clades, the cat-like Feliforms, and the dog-like Caniforms. During the Oligocene, these two clades diversified further to become the primary terrestrial predators [1]. The order Carnivora consists of about 270 extant species [2]. It includes top predators, companion animals, and some iconic species such as polar bears, sea lions, and giant pandas (Fig. 20.1). They are found on all the continents and occupy all major habitat types. The Caniform families include diverse carnivores such as the Canidae, the dogs, wolves, and foxes; the Mustelidae which include otters, mink, wolverine, and ferrets; the Procyonidae include the raccoons; the Ursidae containing the bears including the giant panda; plus several diverse families of marine pinnipeds. As might be anticipated most immunological studies have focused on the domestic dog. The dog was first domesticated from the Eurasian wolf (Canis lupus) [2]. This likely occurred on multiple occasions and in many different locations. Estimates range from 9000 to 34,000 years ago. It likely began with an association between wolves and human hunter-gatherers. The prime source was probably Middle-eastern gray wolves. The earliest remains of domestic dogs in North America have been dated to 13,100 years ago [3]. Selective breeding has resulted in extreme variations in body size, weight, and skeletal proportions. More than 300 dog breeds are officially recognized but there are many more local variants and races [4].

20.1

The domestic dog (Canis lupus familiaris)

20.1.1 Reproduction and lactation The gestation period of the bitch is about 60 days. The fetal thymus begins to differentiate between days 23 and 33, and fetal puppies can respond to phage φX174 by day 40. Blood lymphocytes can respond to phytohemagglutinin by 45 days postconception, and these cells can be found in lymph nodes by 45 days and in the spleen by 55 days. The ability Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00024-1 © 2023 Elsevier Inc. All rights reserved.

311

312

SECTION | 2 Mammalian orders

70 60 50 40 30 20 10 0 Million years ago Dogs and wolves Seals and sea lions Skunk Mustelids Raccoons Bears Giant panda Feliformes FIGURE 20.1 A simple phylogeny of the caniformes.

to reject allografts also develops around day 45, although rejection is slow at this stage, and fetal puppies may be made tolerant by intrauterine injection of an antigen before day 42. Thymic seeding of T cells to the secondary lymphoid organs and the development of humoral immune responses are therefore relatively late phenomena in the dog compared with the situation in the other mammals. The maternal thymus undergoes involution during pregnancy and this thymic involution is required for a successful pregnancy [5]. There is a reduction in all the major lymphoid and non-lymphoid cell populations, The thymic cortex shrinks while the medulla appears to enlarge. It has been suggested that the pregnant maternal thymus produces more T reg cells that may suppress an immunological attack on the placenta and increase her immunological tolerance of the fetus. The thymus generally recovers by the end of lactation. Dogs have an endotheliochorial placenta that limits the transfer of immunoglobulins to the developing fetus (Chapter 2). In this case, it is a zonary placenta that forms a complete band around the developing puppy. (In other caniforms such as the mustelids and raccoons the placenta takes the form of two half-bands.) In the canine placenta, the chorionic epithelium comes into contact with the endothelium of the maternal capillaries. As a result, some maternal immunoglobulin transfer does occur. About 2%
10% of serum IgG is directly transferred from the mother to the puppy, but most must still be obtained through colostrum. Newborn puppies, while not agammaglobulinemic, have very low IgG levels in their blood (0.3 g/L) as compared to the adults 8
25 g/L [6]. The puppy’s intestinal wall is fully open to the transfer of colostral immunoglobulins for the first 4
6 hours after birth but as the enterocytes mature, the transfer is reduced and eventually terminates by 16 hours postpartum [6]. The closure is due to the replacement of FcRn-bearing enterocytes by cells that do not express this receptor. It also results from the maturation of the enterocyte brush border as well as the tight junctions between epithelial cells. As in other carnivores, canine colostrum contains high concentrations of IgG (60%) with lesser quantities of IgA (35%
40%) and IgM (5%) (Fig. 3.4). As lactation proceeds, IgG levels progressively decline as the colostral IgG is depleted. IgM and IgA levels however remain constant so that once colostrum transitions to milk, IgM and IgA predominate and serve to passively protect the puppy intestine. IgA is therefore the predominant immunoglobulin (90%) in dog milk, with much lesser amounts of IgG and IgM (5% each) [6]. The IgA is produced predominantly within the mammary gland. The IgA-producing plasma cells likely originate in intestinal Peyer’s patches but migrate to the mammary gland where they produce antibodies directed against intestinal commensals and pathogens. As intestinal absorption is taking place, simultaneous proteinuria also occurs. Urine from newborn puppies contains relatively large amounts of IgG, IgM, and IgA. Over the first two weeks of life, the puppy’s glomeruli mature and acquire the ability to retain macromolecules [7].

20.2

Hematology

As in other carnivores, neutrophils are the most abundant leukocytes in dogs, sometimes accounting for 75% of the blood leukocytes (Fig. 20.2). Blood monocytes account for about 5% of these. The proportion of lymphocytes is highly variable but can reach as high as 50%. During the first three months of life, puppies have a higher lymphocyte count

Caniforms: Dogs, bears, and their relatives Chapter | 20

313

LEUKOCYTES ADULT LYMPHOCYTES

3

6-17 x 10 /Pl

Lymphocytes Neutrophils

Eosinophils Monocytes

NK

TDE B TJG

FIGURE 20.2 The white blood cell composition and their adult lymphocyte subpopulations in the domestic dog. Canis lupus familiaris. These numbers may differ somewhat between breeds.

Normal eosinophil

A “gray” eosinophil

FIGURE 20.3 A “gray” eosinophil in a blood smear from a golden retriever compared to a normal canine eosinophil. It is possible that these are simply prematurely degranulated cells. Courtesy of Dr. MC. Johnson. From Tizard IR. Allergies and hypersensitivity diseases of the domestic animals. 1st ed. St Louis: Elsevier, 2022.

than adult dogs. Much of this difference is due to CD211 B cells. The proportion of puppy CD81 T cells is low at birth but gradually climbs to reach adult levels. Thymic involution begins at around six months of age in dogs.

20.2.1 Gray eosinophils Some dogs, especially greyhounds have unusual granules in their eosinophils [8]. They do not stain with eosin and as a result, appear pale gray or clear in blood smears (Fig. 20.3). The eosinophils appear to be vacuolated. Greyhound puppies appear to have normally staining eosinophils, but they lose these as they age so that adult dogs have a high proportion of unstained granules. Ultrastructural studies on greyhound eosinophils reveal no unusual structural features and they stain positively for alkaline phosphatase [9]. The presence of these apparently unusual granules possibly reflects the differential properties of commercially available stains. It may also be an artifact caused by cells that degranulate prematurely while being sampled.

20.3

Innate immunity

20.3.1 Acute-phase proteins The most significant acute-phase proteins in the dog are C-reactive protein (CRP), serum amyloid A, haptoglobin, alpha-1-acid glycoprotein, and ceruloplasmin [10]. In dogs, CRP is the major acute-phase protein, increasing 100-fold in infectious diseases such as pyometraassociated sepsis, babesiosis, leishmaniasis, parvovirus infections, and colibacillosis. It can bind to bacteria and activates complement (Fig. 20.4). It then acts as a potent opsonin. It can also act on the liver to induce the inflammatory cytokines, TNF-α, IL-1β, and IL-6. CRP can also modulate neutrophil function. Haptoglobin binds free hemoglobin and as a result, limits the amount of iron available to growing bacteria. It may inhibit neutrophil phagocytosis and chemotaxis. Free hemoglobin also has a proinflammatory effect, so its removal prevents excessive inflammation. Alpha-1acid glycoprotein is an antiinflammatory protein that promotes the production of IL-1 receptor antagonist (IL-1RA),

314

SECTION | 2 Mammalian orders

40 CRP concentration (mg/L) 30

20

10

0 -24

0 6 16

28 52

76 100 124 Time (Hours)

FIGURE 20.4 The rise in C-reactive protein levels in six dogs following anesthesia and surgery (red line) and in six dogs undergoing anesthesia alone (blue line). From Burton SA, Honor DJ, Mackenzie AL, et al. C-reactive protein concentration in dogs with inflammatory leukograms. Am J Vet Res 1994;55:615.; Tizard IR., Veterinary Immunology. 10th ed. St Louis: Elsevier, 2016.

and thus suppresses excessive inflammation. Ceruloplasmin is a copper-binding protein that out-competes invading bacteria for this essential element. In pregnant dogs, haptoglobin, ceruloplasmin, and fibrinogen levels increase midway through gestation.

20.3.2 Natural killer cells Dog natural killer (NK) cells are medium to large-sized lymphocytes with electron-dense cytoplasmic granules that contain granzyme B and perforin. They account for as many as 15% of blood lymphocytes. They are CD42 and CD202-negative thus distinguishing them from T and B cells [11]. However, up to 30% of these putative NK cells may also express CD8. Dog NK cells are CD32, granzyme B1, CD451, and major histocompatibility complex (MHC) class I1. Both IL-15 and IL-21 stimulate their proliferation, receptor expression, and cytotoxic functions. They can lyse distemper virusinfected target cells as well as cancer cells from thyroid adenocarcinomas, melanomas, osteosarcomas, and mammary carcinomas [12]. NK cell maturity appears to be associated with CD5 receptor expression [11]. Thus, while CD5bright cells appear to be T cells, CD5lo cells appear to be NK cells [13]. Dog NK cells may be isolated by culturing peripheral blood mononuclear cells and removing the T cells with anti-T cell antibodies complexed to magnetic beads. The residual cells can be cultured in the presence of IL-15, and IL-2. The non-T, non-B large granular cells that survive are NK cells (IL-2 is necessary for their survival, and IL-15 is needed for their expansion). Both IL-15 and IL-2 stimulate their proliferation, receptor expression, and cytotoxic functions. An important dog NK cell phenotypic marker is NCR1/NKp46. Widely considered to be a pan-specific NK cell marker in mammals, this is expressed on primate, mouse, and ruminant NK cells (but not in pigs!). NCR1 is expressed on only 2%
3% of canine CD5lo cells—perhaps these are just activated T cells induced by exposure to cytokines. NK cells function through the use of diverse activating or inhibiting receptors encoded by genes in both the leukocyte receptor complex (LRC) and the natural killer complex (NKC). In dogs, the NKC receptor gene complex is restricted to a single region on chromosome 27 [11]. The dog has 22 NKC genes as compared to 29 in humans and 57 in mice. Orthologs of the KIR genes, especially KLRD1 (CD94) are well conserved in humans, dogs, and mice. Dogs have only one copy of the Ly49 gene as is also the case in the cat. It is located on chromosome 27 and possesses an ITIM domain. However, it also has a cysteine to tyrosine conversion in position 168. These conserved cysteines are critical, so the canine Ly49 gene is probably nonfunctional. The single canine KIR gene is found within the LRC on chromosome 1 situated between the LILR genes and the IgA receptor (FCAR) gene. However, this gene also appears to be prematurely truncated or even absent in some dogs [11]. On the other hand, Grondahl et al. have demonstrated that NCR1 is an activating receptor on a subset of canine NK cells [14]. They divided the CD3, granzyme B1 cells into NCR11 and NCR1 subsets. IL-12 induces the expression of NCR1 [15]. These positive cells can kill canine tumor cell lines. They produce IFN-γ in response to IL-12. They also produce TNF-α, IL-8, IL-10, and GM-CSF [15]. CD94 is one chain of the NKG2A heterodimer, and it is present in dogs. Canine NK cells express mRNA for the Fc receptor CD16. Among the expanded cell populations however are

Caniforms: Dogs, bears, and their relatives Chapter | 20

315

FIGURE 20.5 Ultrastructural analysis of a canine monocyte and monocyte-derived dendritic cell by transmission electron microscopy. (A) Monocyte on day one in culture (magnification 2000x). (B) Cultured dendritic cell stimulated with cytokines at seven days in culture showing characteristic cytoplasmic projections (magnification 2000x). (C) Periodical microstructure (arrow head) in the cytoplasm represents a unique feature of canine monocyte-derived dendritic cells (magnification 50,000 x). From Qeska V, et al. Species-specific properties and translational aspects of Canine dendritic cells. Vet Immunol Immunopathol 2013;151:181
192. With permission.

many CD31 and CD81 cells that may indicate the existence of a population of canine NKT cells. Transcriptome studies also show that CD5lo is not specific for canine NK cells but NKp46 is [15]. However, the presence of NKp46 may not identify all NK cell subsets [16].

20.3.3 Dendritic cells Canine dendritic cells are potent T cell stimulators [17]. They can be generated from blood monocytes by treatment with IL-4 and GM-CSF. They can also be derived from CD341 progenitors stimulated by GM-CSF and TNF-α in the bone marrow. As in other species, their state of maturity makes a big difference to their functionality. Dog dendritic cells possess a unique ultrastructure since there are multiple, variable-sized, electron-dense granules with a wasp’s nestlike appearance in their cytoplasm [18] (Fig. 20.5). Dog DCs can be stimulated to differentiate by many different cytokine mixtures but the most widely employed is GM-CSF plus interleukin 4. There are two main populations of canine DCs. One is MHC class II1, CD11c1, CD341, and CD142; the other is MHC class II1, CD341, and CD141. CD40 is expressed on canine DCs but not on monocytes. Canine DCs produce a diverse array of cytokines. The precise mixture produced depends on the stimulus employed but in general resembles those produced by DCs in humans and mice. One unusual feature is however the production of large amounts of IL-10, IL-12, IL-13, and IFN-α by LPS-stimulated canine DCs. Canine DCs also express functional CD1 molecules that can bind and present lipid antigens to lymphocytes. They also express low levels of CD4 and CD8. As in other species, depending upon whether they are DC1 or DC2 cells, they can secrete polarizing cytokines [18].

20.4

Lymphoid organs

20.4.1 Thymus The canine thymus develops in utero after its epithelial precursors appear on postconception day 23. By day 38 the first Hassall’s corpuscles are seen as is the differentiation into cortex and medulla [19]. The thymus grows rapidly in puppies to reach its maximum size at about six months of age. The canine thymus is relatively small and is located almost entirely within the thorax. Once sexual maturity is reached, it involutes, leaving connective tissue cords and cysts together with adipose tissue. Traces of the thymus may still be found in adult dogs [20].

20.4.2 Spleen The spleen of the dog, like that of the cat, is relatively large. The capsule is thin and consists of elastic tissue fibers with abundant smooth muscle cells [21]. Many thick trabeculae are also rich in smooth muscle [22]. The white pulp is abundant, and sinusoids are obvious within the marginal zone. However, there is relatively less white pulp when compared to other species and the spleen is therefore classified as a storage organ. The dog spleen is considered to be of the sinusal type. Ellipsoids are relatively large and prominent in carnivores. These specialized capillary

316

SECTION | 2 Mammalian orders

segments are located around the penicillary arteries within the red pulp. They contain macrophages and serve as highly efficient filters. Thus, they can capture particles such as fat droplets, bacteria, and immune complexes [23]. The arterial capillaries open directly into these sinusoids—a closed circulation. This is a similar structure to that observed in the human spleen.

20.4.3 Mucosal lymphoid tissues Dogs have four tonsils, lingual, paired palatine tonsils, and a pharyngeal tonsil [24]. The lingual tonsil is small and simply consists of an aggregation of lymphocytes at the base of the tongue. The palatine tonsils are located on the lateral walls of the oropharynx. They are smooth tonsils that are partially concealed in a fold of mucosa, the tonsillar fossa. This forms a cavity around the tonsil. Some tonsillar tissue can extend into the wall of the sinus. The overlying stratified squamous epithelium is relatively thin and infiltrated with lymphocytes. The pharyngeal tonsil is located on the dorsal surface of the nasopharynx. It is covered with smooth pseudostratified epithelium [24]. There are 26
29 Peyer’s patches in the canine small intestine [25]. They tend to be randomly distributed between the duodenum and jejunum and there is a single long continuous lymphoid patch in the ileum [26]. The anterior patches are oval or round, but they elongate toward the ileum. The long ileal patch is very much larger since it begins as a 10 mm band in the anterior ileum but broadens progressively so that it involves the entire circumference of the terminal ileum ending at the ileocecal junction. It can be 25
30 cm long and consequently, 80%
90% of the intestinal lymphoid tissue is found in the terminal ileum. The duodenal Peyer’s patches contain large pear-shaped follicles in the submucosa with extensive interfollicular areas. Each follicle is covered by an epithelial dome containing a mixed population of plasma cells, lymphocytes, and macrophages. The follicles decrease in size posteriorly so that those located in the ileal patches are smaller and rounder. Likewise, their domes are small and do not reach the base of the villi and their interfollicular areas are also very much reduced. While surrounded by IgA1 plasma cells there are very few of these cells within the germinal centers. T cells are largely restricted to the interfollicular areas and corona [27]. There are lesser numbers in the dome and germinal centers (Fig. 20.6). The ileal PP contains many fewer T cells than the proximal PP. The domes of the PPs contain some B cells with cytoplasmic IgA and many cytoplasmic IgG1 B cells. Lymphoglandular complexes are present in the wall of the large intestine and cecum in dogs.

20.5

Major histocompatibility complex

The canine MHC is classified as the DLA—Dog Leukocyte Antigen complex. The entire canine MHC spans 3.9 mb compared to 4.6 in humans and 3.3 mb in cats. As in other mammals it is arranged in the order, class II-class III-class I in the pericentromeric region of chromosome 12 (Fig. 20.7) [28]. However, a small number of MHC genes (about 500 bp) are also located on chromosome 35. These consist of four processed pseudogenes but there are no functional class I genes in this region [28]. As noted in the previous chapter, a similar situation is seen in cats where most of their MHC is located on the long arm of chromosome B2 but a small fragment has been inverted and so is found in the subtelomeric region of the short arm of the same chromosome. The break has occurred in the same location in cats and dogs, between TRIM 39 and TRIM 26. This suggests that this break and inversion occurred before the divergence of felids and canids B55 mya. Two additional class I genes are also located on canine chromosomes 7 and 18 [29]. The dog appears to have only retained the class I beta duplication block.

20.5.1 The MHC class I region As discussed, the size of the mammalian MHC class I region varies among species. Humans and rodents have the largest, and pigs have the smallest. The number of class Ia genes also varies among mammals ranging from more than 60 in rats to 11 in pigs. Not all these genes are functional. Thus, dogs have only four functional class I genes, DLA-88, -12, -64, and -79. Only one of these, DLA-88, is highly polymorphic with more than 71 alleles recognized. DLA-79 has 11 alleles, DLA-12 has 15 and DLA-64 has six [30,31]. There is also two class I pseudogenes in this region, DLA-12a and DLA-53. Class I genes are critical for the processing and presentation of exogenous antigens

Caniforms: Dogs, bears, and their relatives Chapter | 20

317

FIGURE 20.6 The structure of canine Peyer’s patches. 1. Duodenal Peyer’s Patch. c. Corona; d. dome; g. Germinal center; v. villus: Arrows indicate intrafollicular invaginations of the dome epithelium. 2. Jejunal Peyer’s Patch. (x 39) 3. Ileal Peyer’s Patch (x 39). From HogenEsch H, Felsburg PJ. Immunohistology of Peyer’s patches in the dog. Vet Immunol Immunopathol 1992;30:147
160. With Permission.

DO

DM

DO

TAP

DQ

DR

A

AB

B

12

BA

BA

FIGURE 20.7 The structure of the canine major histocompatibility complex. Note that the 3’ end has been split between TRIM 39 and TRIM 26 and is now found on a separate chromosome.

88 12a 53 12 64

II

III

I

TRIM39

Chromosome 12 TRIM26

Chromosome 35

318

SECTION | 2 Mammalian orders

20.5.2 The MHC class II region The canine classical and extended class II regions span about 711 kb. In dogs, the MHC class II products are expressed on nearly all resting adult T cells. Their expression is enhanced in rapidly dividing cells and cells activated by exposure to interferon-γ. MHC class II molecules consist of two peptide chains called α and β. The genes for the α-chains are designated A, and the genes for the β-chains are called B. A “complete” MHC class II region, in theory, contains six genes in three paired loci. However, not all mammals possess a complete set of functional genes. For example, cats, dogs, mice, and horses lack functional DPA and DPB genes. (The naked mole-rat does, however!! Chapter 23). Dogs also possess four complete and functional class II genes, DRA1, DRB1, DQA1, and DQB1. These have 1, 241, 39, and 117 alleles respectively [30,32]. Their class II region also contains two, non-functional but closely related (98%) pseudogenes, DRB2 and DQB2 [30]. DRA1, DRB1, and DRB2 are located 195 kb downstream of the other DR loci. The DRB2 pseudogene is not present in all haplotypes [33]. Among the canine class II genes, DRA1 is unique in that it is monomorphic, and its product does not differ between individual dogs. The other three functional class II genes are however highly polymorphic and as a result, they collectively encode over 180 different class II alleles and, of course, many more combinations. Each dog has two sets of these three genes, one set inherited from each parent. Most dog breeds possess four or five major haplotypes. One is usually present at high frequency (50%
70%) with the remaining haplotypes ranging from 20% to 1%. Common breeds with large populations are more diverse and as a result, tend to have more haplotypes [34]. Conversely uncommon dog breeds, especially if they have undergone selective line breeding, or gone through recent population bottlenecks, tend to have fewer. There is however great interbreed variation. Some haplotypes are ubiquitous and found in many breeds while others are limited to just a few breeds. Some haplotypes may be geographically restricted [35]. In addition, the canine class II region also contains pseudogenes encoding DPA1, DPB1, and DPB2, and thus lacks functional DP loci. This is similar to the situation in the cat and mouse, pig, cattle, sheep, and horses. [33] Other genes that are located in this region include TAP, DOB, and LMP-2 [36]. The DO gene encodes a tetramer of two DOα subunits and two DOβ subunits encoded by DOA and DOB respectively. This tetramer does not directly bind antigenic peptides but appears to serve a regulatory function. The DOA locus has been characterized in pinnipeds and the coyote and is highly conserved among the canids suggesting significant stabilizing selection [37].

20.5.3 The MHC class III region The genes within the DLA class III region encode proteins with many diverse functions. Some are important in innate immunity such as the genes for the complement components factors B, C4, and C2. They also include genes that encode cytokines such as tumor necrosis factor-α (TNFA), several lymphotoxins (LTA and LTB), and some NK cell receptors as well as heat shock protein 70 [30]. When DLA alleles are investigated in other canid species it has been found that many class II MHC alleles are shared, not only with the Gray wolf, as expected but also with more distantly related canids including Ethiopian wolves (Canis simensis), Dohle, (Cuon alpinus); African wild dog (Lycaon pictus), and the Bush dog (Speothos venaticus). Many of the DRA alleles are also shared with the maned wolf (Chrysocyon brachyurus) and several species of fox [38]. There is clear evidence of positive selection of the DLA-DQA1 gene in the golden jackal (C. aureus) [39]. In raccoon dogs (Nyctereutes procyonoides) 48 novel class I alleles have been identified with one to six alleles per individual [40]. This level of polymorphism is higher than in many other carnivores. Thus, there have been 17 transcribed alleles (six DLA-12, two DLA 64, two DLA79, and seven DLA-88) reported from four wolf samples; 16 alleles from 26 individuals in the giant panda; 37 alleles from 234 individuals in the brown bear; and 33 alleles from 12 individuals in the domestic cat [40]. The domestic dog is an exception with apparently many more alleles but this may be due to a greater sampling effort.

20.6

B cells and immunoglobulins

20.6.1 Immunoglobulin heavy chains In addition to the four IgG subclasses, dogs make IgA, IgM, IgD, and possibly two subclasses of IgE. An IgM allotype has been described in the dog. The canine IGH locus is located on chromosome 8, just subtelomeric on the antisense strand. This location and orientation are seen in all mammals except monotremes and marsupials [41]. Conversely, light chain gene locations are not strongly conserved (Fig. 20.8).

Caniforms: Dogs, bears, and their relatives Chapter | 20

IGH

5'

89

3'

6

6

M

IGK

IGL

9

D

G1

5

G2

G3

G4

E

E

319

FIGURE 20.8 The organization of the immunoglobulin gene loci in the dog.

A

9

162

M

D

G

E

A

V

D

J

LC

20.6.2 Canine IgD Dogs have a functional IGHD gene, and they produce small amounts of IgD. Most of the body’s IgD is located on the surface of immature B cells where it appears to serve as an antigen receptor. Serum IgD levels in healthy dogs are reported to average 41.4 6 124.27 ng/mL in females and 391.33 6 1873.22 ng/mL in male dogs [42]. The IgD heavy chain constant chains each consist of a 416 amino acid polypeptide encoded by seven exons. These exons are, in order, 50 -CH1-H1-H2-CH2-CH3-M1-M2
30 . Thus, it contains three constant domains and a long hinge region encoded by two exons [43]. The complete genomic sequence spans 8.25 kb. Only two amino acids differ between dog and human IgD heavy chains. Thus, in humans, positions 56 and 63 are methionine and glutamate while in the dog they are lysine and aspartate. The canine sequences are, in general, similar to those in IgD in horses, pigs, cows, and sheep (but not to rodents that lack the CH2 domain). CH1 and the hinge region are the least conserved. Remnants of a CART motif are present in the dog transmembrane domain.

20.6.3 The IgG subclasses Dogs have four IGHG genes and hence make four IgG subclasses, named IgG1, IgG2, IgG3, and IgG4 in order of abundance. Their heavy chain constant domains are very similar with 73%
81% amino acid homology. However, their hinge regions are very different with only 19%
35% amino acid homology. There are no multiple duplications of the hinge region as seen in humans [44]. These four subclasses differ in their biological activities [45]. These include halflives, complement activation, and antibody-dependent cell-mediated cytotoxicity (ADCC). Many of these activities are determined by whether their Fc regions interact with the appropriate Fc receptors. Canine subclasses IgG2 and IgG4 appear to have minimal effector functions. Subclasses IgG1 and IgG3 bind to Fc gamma receptors and mediate ADCC.

20.6.4 The IgE subclasses The canine IGHE constant region is encoded by four exons extending over 1.7 kb [42] each encoding a heavy chain domain. Overall, it shares 49% identity with mouse IgE and 57% identity with the human gene [46]. The CH3 domain is the most conserved while CH2 is the least. Studies on polyclonal canine IgE have identified two biochemically, physically, and biologically distinct IgE subclasses, IgE1 and IgE2 [47]. These can be distinguished by their reactivity with monoclonal IgE antiglobulins and by binding to protein A (A staphylococcal immunoglobulin-binding protein). While sharing the major characteristics of IgE such as heat lability, molecular weight, mast cell binding, and reactivity with polyclonal anti-IgE, they differ in their reactivity with monoclonal antibodies and their isoelectric point. These differences are not due to altered glycosylation. They also appear to have different biological properties in that IgE2 levels are highly variable when compared to IgE1 in ragweed sensitized dogs.

320

SECTION | 2 Mammalian orders

Examination of the canine genome focusing on the IgE heavy chain gene locus confirms the existence of two functional IgE heavy chain genes, IGHE1 and IGHE2. The biological significance of this gene duplication remains is unclear. Perhaps the existence of two IgE subclasses may account in part, for the relatively high levels of IgE in dogs when compared to other mammals (Table 9.2).

20.6.5 Canine IgA The canine IGHA gene sequence is 1.5 kb in length and includes all three constant region coding exons [46]. The dog IgA heavy chain shares 65% and 70% amino acid sequence identity with mice and humans respectively. The CH3 domain is the most highly conserved. Unlike other mammalian IgA genes, the hinge region of canine IgA is fused with the 50 end of the CH2 domain and is 10 amino acids long. A cysteine residue at position 311 in the CH2 domain is responsible for linkage to a secretory component as in the human. Four allelic variants of the canine IGHA gene have been identified [48]. These variants are all located within the hinge region. Their functional significance is unknown.

20.6.5.1 IGHV At present, the canine IGHV region is known to contain at least 89 IGHV genes, six D genes, and six IGHJ genes [49]. These are subdivided into four subfamilies (IGHV1
4). There is a considerable bias toward using IGHV3. There is also preferential pairing of certain V and J genes. The canine IGH locus has a gene/pseudogene ratio of about one, which is consistent with other IGH loci and quite different from the 4: 1 ratio in the light chains. Dogs have 43 functional and 37 pseudogenes in their IGHV gene segments [50]. All the IGHV functional segments are transcribed in the same direction as the constant genes with the exception of one reversely transcribed pseudogene (IGHV3
4). When compared with human and mouse VH regions, canine V domain CDR3 regions are shorter than in humans but longer than in mice [51]. The CDR3 length is normally distributed with a mean of 13
14 amino acids. The amino acids at the base of the CDR3 loop are strictly conserved. In general, the canine CDR3 sequence is biased toward the use of small, negatively charged amino acids.

20.6.6 Immunoglobulin light chains 20.6.6.1 Lambda light chains The canine IGL locus is located on chromosome 26 and is very large (2.6 mb). It contains at least 162 IGLV genes of which 78 are functional. They are classified into seven subfamilies. IGLV1 is the largest subfamily with 86 members. As in other IGL loci, there are nine pairs of J and C genes [50]. Many of these IGLV genes are inverted with respect to the J-C cluster. Dog immunoglobulins use up to 90% lambda chains.

20.6.6.2 Kappa light chains The canine IGK locus is located on chromosome 17 and is only 400 kb in size. It has a somewhat unusual structure. It contains 18 IGKV genes, nine of which are functional, located upstream of the J and C genes, and an additional cluster of nine V genes downstream of which eight are functional and inverted in relation to the J and C genes. Fourteen of the 17 functional genes are members of the IGJV2 subfamily while two belong to the IGKV4 subfamily. IGKV4 subfamily V genes are located in the downstream cluster [50]. This is similar to the situation in horses and seems to make no functional difference. There are also five IGKJ and a single IGKC gene. Inversions and block duplications appear to be characteristic of the canine IGK loci. Humans, pigs, mice, horses, and dogs all contain IGKV genes in reverse orientation to their C gene, and dogs, humans, and pigs have had inversional duplication of entire blocks [50].

20.7

T cells and cell-mediated immunity

In the blood of newborn puppies, neutrophil numbers are three times greater than lymphocytes. But this changes by the end of their first week when lymphocytes begin to predominate [52]. The percentage of B cells (CD211) in newborn puppies is 40% in newborn puppies and decreases as the animals get older [53]. The percentage of CD81 T cells in newborn puppies averages 7.7% after birth but also increases as the animals age. These reach normal levels (46%
84% T cells, 7%
30% B cells) by three months. Some breed differences are also seen. Thus, lymphocyte percentages are higher in Beagles and Dachshunds than in Dalmatians and German Shepherd dogs [53]. German shepherd dogs have the lowest absolute lymphocyte counts and the greatest neutrophil: lymphocyte ratio compared to other breeds.

Caniforms: Dogs, bears, and their relatives Chapter | 20

321

In dogs, unlike the artiodactyls, α/β T cells far outnumber γ/δ T cells in peripheral blood and lymphoid organs. On average, newborn puppies have 1.7%
2.5% γ/δ T cells and this does not change over time [53]. In older crossbred puppies γ/δ T cells account for 2.92% 6 2.27% of white blood cells (133 6 86 cells/μL) [54]. The numbers of γ/δ cells drop further in old dogs, 10
13 years of age to 0.42% 6 0.08% (10 6 3 cells/μL). The role of CD4 and CD8 on the surface of T cells is well recognized since they determine whether T cells are helper cells or effector cells. The roles of double-negative CD42CD82 T cells are much less clear, but they are believed to be involved in the interactions between the innate and adaptive immune responses. In dogs, CD4 is expressed on neutrophils and macrophages but not on monocytes, whereas in cats, CD4 is found on only a subset of T cells and their precursors. In dogs, there is a high proportion of double-negative (DN) α/β T cells in both blood and lymphoid organs [55]. Some of these DN cells are FoxP3 positive suggesting that they play a regulatory role. Thus 80% of FoxP1, α/β1 CD4, CD8 T cells co-express CD25, another Treg marker. Some also express the transcription factor GATA3. These double-negative cells also produce IFN-γ on stimulation. Therefore, there are two double-negative cell subsets in the dog. One is a large population of T cells with members showing an activated phenotype, high expression of FoxP3, or GATA-3, and produce IFNγ or IL-17A. The second is a small γ/δ T cell subset expressing GATA-3 that does not produce these cytokines. Wu et al. have characterized the CD41, CD25high T cells and shown that they have a regulatory function, and a transcriptome that resembles human and mouse cells and are enriched for FoxP3 [56].

20.7.1 T cell antigen receptor genes 20.7.1.1 TRA/D genes The TRA/D locus in the dog is located on chromosome 8. It has a similar structure to that of other mammals (Fig. 20.9). The TRDD, TRDJ, and TRDC genes are flanked by TRDV genes and the downstream TRDV gene is inverted. This block lies upstream of a cluster of TRAJ genes as well as the single TRAC gene. The C-distal end of the locus appears to have undergone block duplications. In this locus, there are 56 TRV genes, of which five are clearly TRDV genes. There are two TRDD, four TRDJ, one TRDC, and one TRAC gene. Dogs have 59 TRAJ genes (compared to 60 in the mouse and 61 in humans) [50].

20.7.1.2 TRB genes The canine TRB locus on chromosome 16 is relatively small (271 kb) compared to 302 kb in the cat and 620 kb in the human [57]. It contains only a single cluster of D-J-C segments whereas humans and mice have two [58]. There are 38 TRBV segments (22 functional) followed by two TRBD genes, twelve TRBJ genes (nine functional), and two functional TRBC genes [57]. When dog tissues are examined, 31 of these TRBV genes can be detected in the thymus and spleen transcriptome suggesting that they are functional and subject to editing and splicing. As in humans, there is a single TRBV gene located downstream from the C segment in the opposite orientation. This arrangement is seen in many other mammals and is at least 93 million years old. TRA/D

TRB

TRG

5'

56 TRAV 38

5 2 DV DD

4 DJ DC

59 AJ

6

6

16 16J

8C

V

D

J

C

3'

AC

FIGURE 20.9 The organization of the canine T cell receptor loci.

322

SECTION | 2 Mammalian orders

20.7.1.3 TRG genes The canine TRG locus is about 460 kb in length. It consists of 40 genes with 21 of them functional and 19 pseudogenes. It has multiple V, J, and C genes arranged in eight cassettes aligned in tandem and the same transcriptional orientation and separated from each other by a 10
35 kb gap [59]. Each cassette consists of V-J-J-C genes except for a single J-JC cassette located at the 30 end of the locus that lacks the V gene segment [60]. However, the TRG locus also contains LINE1 elements evenly spaced along the entire sequence. These account for about 20% of its size. Eight of the 16 TRGV genes are functional and belong to four different subfamilies. Only seven out of the 16 TRGJ genes are functional. Six of the eight C genes are functional. TRGC1 is an open reading frame and TRGC6 is a pseudogene. The structure of this locus suggests that it has developed as a result of repeated duplication events.

20.8

MUSTELIDS

20.8.1 Major histocompatibility complex Sea otter (Enhydra lutris) MHC class II genes have been characterized and are similar to the dog with at least one DQA, DQB, and DRA locus and at least two DRB loci [61].The wolverine (Gulo gulo) as an aggressive carnivore has been ruthlessly extirpated by humans. As a result, it has undergone a severe population bottleneck with an accompanying significant loss of diversity in exon 2 of its DRB locus. This locus contains 11 functional alleles and three pseudogenes. In the northern Finnish population almost half the animals tested had only a single allele. This loss implies a significant reduction in immune fitness [62].

20.8.2 Immunoglobulins The immunoglobulin genes of the ferret (Mustela putorius faro) have been characterized [63]. As in the dog, ferrets make five heavy chain isotypes M, D, G, E, and A as well as the two light chain families, kappa and lambda. Their IGHV genes primarily belong to clan III (118 sequences) with only two sequences from clan II and one from clan I. 18
20 IGHV genes have been identified. There are five predicted IGHJ gene segments and seven D segments. The ferret CDR3 segments range from five to 25 amino acids in length which are comparable to canines. Overall, these heavy chains are highly homologous to human and canine sequences. The exception is in the IgD gene where the CH1 and hinge regions show high sequence divergence from the dog [43]. About 20% of ferret light chains are functional kappa chains while 37% are functional lambda chains. The length of the CDR3 in these light chains ranges from 5
11 amino acids for kappa chains and 8
13 for the lambda locus.

20.8.3 TCRs The ferret TRB locus contains 34 TRBV genes, two TRBD genes, 12 TRBJ genes, and two TRBC genes [57].

20.9

PROCYONIDS

20.9.1 Lymphoid tissues The immune cell populations have been phenotyped in the raccoon (Procyon lotor) [64].

20.10 URSIDS Bears, especially giant pandas, produce highly altricial cubs. They have the lowest ratio of neonatal to maternal body mass of any eutherians. Bears have endotheliochorial discoidal zonary placentas and consequently, their cubs are dependent on colostrum for maternal immunoglobulins. As a result, they require prolonged maternal care and nursing. These cubs have a relatively underdeveloped immune system when born and thus require prolonged passive immunity from maternal colostrum. In the giant panda, the transition from colostrum to milk is a relatively slow process taking about 30 days. Colostral IgG is present over a long time period implying that intestinal absorption also persists. Likewise, the secretion of milk oligosaccharides with antibacterial functions also persists for at least 20
30 days [65]. Hematology has been performed on many species of bear including the giant panda (Ailuropoda melanoleuca) [66]. Their total white cell counts are generally in the order of 6
8 3 103/μL. Neutrophils account for about 64%
76% of these with lymphocytes ranging from 18% to 29%, monocytes 2%
3%, and eosinophils 1%
4%. This can be

Caniforms: Dogs, bears, and their relatives Chapter | 20

323

compared with another tropical bear, the Indian sloth bear (Melursus ursinus ursinus) which has a blood white cell count of 12.9 6 4 3 103/μL [67]. Of these, neutrophils account for about 66%, lymphocytes are 22%, monocytes 2%, and eosinophils 6%. It is important to point out however that eosinophil counts often reflect an animal’s parasite burden. An interesting feature of the blood leukocytes in the giant panda is the presence of a structurally unique mononuclear cell [66]. These cells account for 1%
4% of all leukocytes. They range in size from 12 to 18 μm in diameter. They have a moderately indented nucleus and shared cytochemical and ultrastructural characteristics of both lymphocytes and monocytes. For example, they stain diffusely for alpha-naphthyl-butyrate esterase. It is unclear whether they are lymphocytes or monocytes. The histology of ursid lymphoid organs shows no remarkable differences from other carnivores [68].

20.10.1 Hibernation It is also appropriate to point out that bears adapt to the winter and a lack of available food resources by hibernating for several months. Thus, they double their fat deposits over the summer when they eat well and then survive on these when hibernating [69]. They reduce their metabolic rate and energy demand, reduce their body temperature slightly (30-36 C), and make numerous physiologic adjustments. Interestingly hibernation is associated with consistent changes in their gut microbiota. Hibernating bears have reduced levels of Firmicutes and Actinobacteria and increased levels of Bacterioidetes. The summer microbiota appears to promote adiposity without affecting glucose tolerance [69]. When American black bears (Ursus americanus) hibernate, they increase the expression of numerous immunerelated genes. As a result, their serum composition changes. Upregulated proteins include IgM heavy and J chains and complement components such as C1s and C4. These are generally increased about 1.5-fold. Other upregulated proteins include α2-macroglobulin, α1B-glycoprotein, α1-antitrypsin, clusterin, and haptoglobin. Downregulated proteins include C4 binding protein α chain, transferrin, kininogen 1, α2HS-glycoprotein, as well as apolipoproteins A-I and A-IV. [70]. These changes may assist the bear in maintaining immune health during hibernation.

20.10.2 Climate change and immunity Climate change is having a significant effect on the distribution of some mammals. This is especially the case in the Arctic, where warming temperatures are facilitating the northward expansion of pathogens [71]. Historically, while polar bears (Ursus maritimus) spent the winter hibernating on land, in the summer they moved north onto the pack ice where they could indulge in their specialized diet of seals. The premature melting of the arctic sea ice means that some bears are unable to go north and as a result are forced to spend their summer foraging on land. There is a great difference in the microbiota found in seals in floating ice floes and that found on land. Thus, terrestrial bears are exposed to much greater microbial challenges than those at sea. If such is the case, then the land-bound bears would be expected to exhibit greater immune system activity than those at sea. After accounting for body condition, (land-bound bears are food-deprived) it is clear that bears stranded onshore have significantly higher white blood cell counts, neutrophils, and monocytes than bears living on the ice. Thus, in October, bears that have summered onshore have a mean white cell count of about 9 3 103 /μL while bears that summered on the ice had mean counts of about 5 3 103/μL. This suggests that they may have suffered more infections. Interestingly there is no significant difference in lymphocyte, eosinophil, basophil numbers, blood globulins, or CRP. The long-term consequences of this are unclear [71]. It is also of relevance to note that Canadian polar bears have low MHC diversity [69]. This is consistent with longstanding exposure to relatively low pathogen loads. Thus, all of the polar bears studied showed maximum linkage disequilibrium between the DRB/DQB loci. Twelve of the 98 individuals sampled had a duplicated DQB haplotype. These results may reflect balancing selection at these loci, a response to a previous infectious disease challenge, or simply reflect the original polar bear/brown bear divergence. Thus, one DQA allele in polar bears is also present in brown bears. This low MHC diversity in association with the increased risks of infectious disease poses a significant threat to polar bears in a rapidly changing environment [72].

20.10.3 Major histocompatibility complex The MHC of the Scandinavian brown bear (Ursus arctos) has been sequenced and analyzed [73]. Thirty-seven MHC class I alleles were detected among 224 bears, 16 MHC class II DRB among 234 individuals, four DQB, and two DQA alleles. Several loci were clearly expressed. These included three MHC class I, two DRB, two DQB, and one DQA

324

SECTION | 2 Mammalian orders

locus. The MHC class I region also contained two clusters of non-expressed loci. Class I and DRB sequence frequencies differ between the northern and southern populations of these bears. There is evidence of positive selection since the numbers of nonsynonymous substitutions (DN) greatly exceed the numbers of synonymous substitutions (DS) in the antigen-binding sites of the DRB and DQB loci as well as in the class I loci. Both class I and class II showed orthology to the gene clusters found in the giant panda (A. melanoleuca). While some carnivores have lost their taste receptors for sweetness, bears have not [74].

20.11 PINNIPEDS The pinnipeds are marine mammals that belong to three superfamilies: The Phocidae, the true seals; the Otariidae, the sea lions and fur seals; and the Odobenidae, the walruses. All have evolved from land-based carnivores. They are often the subjects of mass mortalities as a result of infections, especially viruses, derived from terrestrial carnivores. The young seals and sea lions, living in a cold marine environment, require high-calorie food beginning immediately after birth. This is generally provided by maternal colostrum and milk with a very high-fat content (B50%). Like other mammalian young, they also require passive immunity in the form of colostral immunoglobulins. Thus, in the northern fur seal (C. ursinus), transplacental immunoglobulin transfer is relatively unimportant and the young seal cubs are born having received less than 5% of the level of maternal IgG [75]. Seal colostrum contains IgG at levels similar to those in her serum. However relatively little is transferred to her young and by one week of age pup IgG levels are only slightly higher than at birth. By five weeks of age, it only reached 70% of the adult level. The rate of appearance of IgG and IgA in pup serum is therefore significantly less than in terrestrial carnivores. In terrestrial mammals, this low level of IgG would be predicted to result in increased disease susceptibility. Mortality data does not show this in seals! It appears that the young seal cub can mount a very rapid IgM-mediated immune response on its own. This is of special interest since fur seal rookeries are very densely populated during the breeding season and the shore is very contaminated with fecal material. Despite this, localized disease outbreaks appear to be rare [77].

20.11.1 Lymphoid tissues There are no significant structural differences between the lymph nodes of seals and the lymph nodes of terrestrial mammals [75]. Studies on the lymphoid organs of the carnivorous Leopard seal (Hydryrga leptonyx) show some interesting adaptations in some lymphoid organs [76]. Thus obvious Peyer’s patches have not been identified. Instead, there are single to multiple lymphoid nodules in the lamina propria, and some contain germinal centers. Lymph nodes appeared to have a conventional structure similar to that seen in other pinnipeds and terrestrial carnivores. Eosinophils are commonly seen in pinniped lymph nodes reflecting the presence of parasites. Many pinnipeds often house large populations of intestinal nematodes. Their spleen has a thick capsule and prominent trabeculae reflecting its function as a blood storage organ. (These features tend to be better developed in deeper diving seal species.) These contain abundant smooth muscle cells with collagen and elastic fibers. Sheathed capillaries are readily seen but sinuses are not apparent [76].

20.11.2 Major histocompatibility complex The pinniped MHC has an overall organization similar to that seen in other carnivores. Thus, they share the main features of the dog MHC [78]. They lack a functional pair of DPA/DPB genes and have an inverted DRB locus between the DQ and DO subregions. This is believed to be functional. The only DRA locus from the Hawaiian monk seal (Monachus schauinslandi) has a single base pair deletion in exon2 leading to premature termination and so is presumed to be a pseudogene [79]. This species also appears to have a functional DRB gene. While most pinniped DP genes are pseudogenes, there is one functional DPB gene in the Weddell seal (Leptonychotes weddellii), and one DPA gene in the northern fur seal (Callorhinus ursinus). Pinniped genomes have variable numbers of DR loci but only a single pair of presumed DQ genes. These differences determine the extent of polymorphism that occurs in each species. Thus, the southern elephant seal (Mirounga leonina), has very limited receptor polymorphism compared to terrestrial mammals—a feature of other marine mammals [80]. Likewise, the Northern elephant seal (M. angustirostris) has few alleles at its class II loci and is significantly inbred [81].

Caniforms: Dogs, bears, and their relatives Chapter | 20

325

20.11.3 Natural killer cells and receptors In examining the genomes of four species of wild marine carnivore, three seals, and one sea lion, Hammond et al. found that Ly49 and KIR were each represented by a single orthologous gene [82]. They show little polymorphism, and their products are expressed on the NK cell surface. Thus, in pinnipeds, neither Ly49 nor KIR is polygenic. It appears that pinniped Ly49 has been subjected to purifying selection. But there has been a positive selection for KIR3DL in pinnipeds. This suggests that this is a functional receptor and the reason why KIR3DO has been lost. Thus, Ly49 and KIR have been stable for the B33 my since the last common ancestor of seals and sea lions [82].

References [1] Murphy WJ, Foley NM, Bredemeyer KR, Gatesy J, Springer MS. Phylogenomics and the genetic architecture of the placental mammal radiation. Annu Rev Anim Biosci 2020;. Available from: https://doi.org/10.1146/annurev-animal-061220-023149. [2] Agnarsson I, Kuntner M, May-Collado LJ. Dogs, cats and kin: a molecular species-level phylogeny of carnivora. Mol Phyl Evol 2010;54:726
45. [3] Fedje D, Mackie Q, McLaren D, Wigen B, Southon J. Karst caves in Haida Gwaii: archeology and paleontology at the Pleistocene-Holocene transition. Quat Sci Rev 2021;. Available from: https://doi.org/10.1016/j.quascirev.2021.107221. [4] VonHoldt BM, Pollinger JP, Lohmueller KE, Han E, et al. Genome-wide SNP and haplotype analysis reveal a rich history underlying dog domestication. Nature 2010;. Available from: https://doi.org/10.1038/nature08837. [5] Pereira M, Valerio-Bolas A, Saraiva-Marques C, Alexandre-Pires G, et al. Development of dog immune system: from in uterus to elderly. Vet Sci 2019;. Available from: https://doi.org/10.3390/vetsci6040083. [6] Chastant S, Mila H. Passive immune transfer in puppies. Anim Repro Sci 2019;207:162
70. [7] Schaefer-Somi S, Baer-Schadler S, Aurich JE. Proteinuria and immunoglobulinuria in neonatal dogs. Vet Rec 2005;157:378
82. [8] Giori L, Gironi S, Scarpa P, et al. Grey eosinophils in greyhounds: frequency in three breeds and comparison of eosinophil counts determined manually and with three hematology analyzers. Vet Clin Pathol 2011;40:475
83. [9] Lazbik MC, Couto CG. Morphological characterization of specific granules in Greyhound eosinophils. Vet Clin Pathol 2005;34:140
3. [10] Ceron JJ, Eckersall PD, Martinez-Subiela S. Acute phase proteins in dogs and cats: current knowledge and future perspectives. Vet Clin Pathol 2005;34(2):85
99. [11] Gingrich AA, Modiano JF, Canter RJ. Characterization and potential applications of dog natural killer cells in cancer immunotherapy. J Clin Med 2019;. Available from: https://doi.org/10.3390/jcm8111802. [12] Michael HT, Ito D, McCullar V, Zhang B, et al. Isolation and characterization of canine natural killer cells. Vet Immunol Immunopathol 2013;155:211
17. [13] Huang Y-C, Hung S-W, Jan T-R, et al. CD5-low expression lymphocytes in canine peripheral blood show characteristics of natural killer cells. J Leukoc Biol 2008;84:1501
10. [14] Grøndahl-Rosado C, Boysen P, Johansen GM, Brun-Hansen H. NCR1 is an activating receptor expressed on a subset of canine NK cells. Vet Immunol Immunopathol 2016;177:7
15. [15] Foltz JA, Somanchi SS, Yang Y, Aquino-Lopez A, et al. NCR1 expression identifies canine natural killer cell subsets with phenotypic similarity to human natural killer cells. Front Immunol 2016;. Available from: https://doi.org/10.3389/fimmu.2016.00521. [16] Gingrich AA, Reiter TE, Judge SJ, York D, et al. Comparative immunogenomics of canine natural killer cells as immunotherapy target. Front Immunol 2021;. Available from: https://doi.org/10.3389/fimmu.2021.670309. [17] Qeseka V, Baumgartner W, Beineke A. Species-specific properties and translational aspects of canine dendritic cells. Vet Immunol Immunopathol 2013;151(3
4):181
92. [18] Isotani M, Katsuma K, Tamura K, Yamada M, et al. Efficient generation of canine bone marrow-derived dendritic cells. J Vet Med Sci 2006;68 (8):809
14. [19] Bodey B, Calvo W, Prummer O, Fliedner TM, Borysenko M. Development and histogenesis of the thymus in dog: a light and electron microscope study. Dev Comp Immunol 1987;11:227
38. [20] Sisson S, Grossman JD. The anatomy of the domestic animals. 4th ed. Philadelphia, PA: Saunders; 1953. [21] Onkar DP, Govardhan SA. Comparative histology of human and dog spleen. J Morphol Sci 2013;30(1):16
20. [22] Udroiu I, Sgura A. The phylogeny of the spleen. Quart Rev Biol 2017;92(4):411
43. [23] Sørby R, Wein TN, Husby G, Espenes A, Landsverk T. Filter function and immune complex trapping in splenic ellipsoids. J Comp Pathol 2005;132:313
21. [24] Casteleyn C, Breugelmans S, Simoens P, Van den Broeck W. The tonsils revisited: review of the anatomical localization and histologic characterization of the tonsils of domestic and laboratory animals. J Immunol Res 2011;. Available from: https://doi.org/10.1155/2011/472-460. [25] HogenEsch H, Housman JM, Felsburg PJ. Canine Peyer’s patches: macroscopic, light microscopic, scanning electron microscopic and immunohistochemical investigations. In: Mestecky J, et al., editors. Recent advances in mucosal immunology. New York: Plenum; 1987. [26] Kararli TT. Comparison of the gastrointestinal anatomy, physiology, and biochemistry of humans and commonly used laboratory animals. Biopharm Drug Dispos 1995;16:351
80. [27] HogenEsch H, Felsburg PJ. Immunohistology of Peyer’s patches in the dog. Vet Immunol Immunopathol 1992;30:147
60.

326

SECTION | 2 Mammalian orders

[28] Yuhki N, Beck T, Stephens R, Neelam B, O’Brien SJ. Comparative genomic structure of humans, dog and cat MHC: HLA, DLA, and FLA. J Heredity 2007;98(5):390
9. [29] Wagner JL. Molecular organization of the canine major histocompatibility complex. J Heredity 2003;94(1):23
6. [30] Wagner JL, Sarmiento UM, Storb R. Cellular, serological, and molecular polymorphism of the class I and class II loci of the canine major histocompatibility complex. Tissue Antigens 2002;59:205
10. [31] Miyamae J, Suzuki S, Katakura F, Uno S, et al. Identification of novel polymorphisms and two distinct haplotype structures in dog leukocyte antigen class I genes: DLA-88, DLA-12 and DLA-64. Immunogenetics 2017;. Available from: https://doi.org/10.1007/s00251-017-1031-5. [32] Kennedy LJ, Ollier WER, Marti E, Wagner JL, Storb RF. Canine immunogenetics. In: Ostrander EA, Ruvinsky A, editors. The genetics of the dog. 2nd ed. CAB International; 2012. [33] Debenham SL, Hart EA, Ashurst JL, Howe KL, et al. Genomic sequence of the class II region of the canine MHC: comparison with the MHC of other mammalian species. Genomics 2005;85:48
59. [34] Kennedy LJ, Barnes A, Happ GM, et al. Extensive interbreed, but minimal intrabreed, variation of DLA class II alleles and haplotypes in dogs. Tissue Antigens 2002;59:194
9. [35] Kennedy LJ, Barnes A, Happ GM, et al. Evidence for extensive DLA polymorphism in different dog populations. Tissue Antigens 2002;60:43
52. [36] Gojanovich GS, Ross P, Holmer SG, Holmes JC, Hess PR. Characterization and allelic variation of the transporters associated with antigen processing (TAP) genes in the domestic dog (Canis lupus familiaris). Dev Comp Immunol 2013;41(4):578
86. [37] Soll SJ, Stewart BS, Lehman N. Conservation of MHC class II DOA sequences among carnivores. Tissue Antigens 2005;65:283
6. [38] Kennedy LJ, Kitchener A, Laurenson K, Radford AD, et al. MHC allele sharing between canid species including Ethiopian wolves and African Wild dogs. In: Proc. Tufts canine and feline breeding and genetics conference; 2009. [39] Stefanovic M, Cirovic D, Bogdanovic N, Knauer F, et al. Positive selection of the MHC class II DLA-DQA1 gene in golden jackals (Canis aureus) from their recent expansion range in Europe and its effect on their body mass index. BMC Eco Evol 2021;. Available from: https://doi. org/10.1186/s12862-021-01856-z. [40] Bartocillo AMF, Nishita Y, Abramov AV, Masuda R. Evolution of MHC class I genes in Japanese and Russian raccoon dogs, Nyctereutes procyonoides (Carnivora: Canidae). Mammal Res 2021;66:3710383. [41] Das S, Nikolaidis N, Klein J, Nei M. Evolutionary redefinition of immunoglobulin light chain isotypes in tetrapods using molecular markers. Proc Natl Acad Sci USA 2008;105(43):16647
52. [42] Martinez-Orellana P, Maristany C, Baxarias M, Alvarez-Fernandez A, et al. Total serum IgD from healthy and sick dogs with leishmaniosis. Parasit Vectors 2019;. Available from: https://doi.org/10.1186/s13071-019-3384-0. [43] Rogers KA, Richardson JP, Scinicariello F, Attanasio R. Molecular characterization of immunoglobulin D in mammals: immunoglobulin heavy constant delta genes in dogs, chimpanzees and four old world monkey species. Immunology 2006;118:88
100. [44] Tang L, Sampson C, Dreitz MJ, McCall C. Cloning and characterization of cDNAs encoding four different canine immunoglobulin γ chains. Vet Immunol Immunopathol 2001;80:259
70. [45] Bergeron LM, McCandless EE, Dunham S, Dunkle W, et al. Comparative functional characterization of canine IgG subclasses. Vet Immunol Immunopathol 2014;157:31
41. [46] Patel M, Selinger D, Mark GE, Hickey GJ, Hollis GF. Sequence of the dog immunoglobulin alpha and epsilon constant region genes. Immunogenetics. 1995;41:282
6. [47] Peng Z, Arthur G, Rector ES, Kierek-Jaszczul D, et al. Heterogeneity of polyclonal IgE characterized by differential charge, affinity to protein A, and antigenicity. J Allergy Clin Immunol 1997;100:87
95. [48] Peters IR, Helps CR, Calvert EL, Hall EJ, Day MJ. Identification of four allelic variants of the dog IGHA gene. Immunogenetics 2004;56:254
60. [49] Hwang M-H, Darzentas N, Bienzle, Moore PF, et al. Characterization of the canine immunoglobulin heavy chain repertoire by next generation sequencing. Vet Immunol Immunopathol 2018;202:181
90. [50] Martin J, Ponstingl H, Lefranc M-P, Archer J, et al. Comprehensive annotation and evolutionary insights into the canine (Canis lupus familiaris) antigen receptor loci. Immunogenetics 2018;70:223
36. [51] Steininiger SCJ, Dunkle WE, Bammert GF, Wilson TL, et al. Fundamental characteristics of the expressed immunoglobulin VH and VL in different canine breeds in comparison with those of humans and mice. Mol Immunol 2014;59:71
8. [52] Toman M, Faldyna M, Knotigova P, Pokorova D, Sinkora J. Postnatal development of leukocyte subset composition and activity in dogs. Vet Immunol Immunopathol 2002;87:321
6. [53] Faldyna M, Leva L, Knotigova P, Toman M. Lymphocyte subsets in peripheral blood of dogs—a flow cytometric study. Vet Immunol Immunopathol 2001;82:23
37. [54] Marchetti C, Borghetti P, Cacchioli A, Ferrari L, et al. Profile of gamma-delta (γδ) T lymphocytes in the peripheral blood of crossbred dogs during stages of life and implication in aging. BMC Vet Res 2020;. Available from: https://doi.org/10.1186/s12917-020-02504-2. [55] Rabiger FV, Rothe K, von Buttlar H, Bismarck D, et al. Distinct features of canine non-conventional CD4-CD8a- double negative TCRαβ 1 vs TCT γδ 1 T cells. Front Immunol 2019;. Available from: https://doi.org/10.3389/fimmu.2019.02748. [56] Wu Y, Chang Y-M, Stell AJ, Priestnall SL, et al. Phenotypic characterization of regulatory T cells in dogs reveals signature transcripts conserved in humans and mice. Sci Rep 2019;. Available from: https://doi.org/10.1038/s41598-019-50065-8. [57] Pegorier P, Bertignac M, Chentli I, Ngoune VN, et al. IMGT biocuration and comparative study of the T cell receptor beta locus of veterinary species based on Homo sapiens TRB. Front Immunol 2020;. Available from: https://doi.org/10.3389/fimmu.2020.00821.

Caniforms: Dogs, bears, and their relatives Chapter | 20

327

[58] Matiasovic J, Andrysikova R, Karaskova D, Toman M, Faldyna M. The structure and functional analysis of canine T-cell receptor beta region. Vet Immunol Immunopathol 2009;132:282
7. [59] Antonacci R, Massari S, Linguiti G, Jembrenghi AC, et al. Evolution of the T-cell receptor (TR) loci of the adaptive immune response: the tale of the TRG locus in mammals. Genes 2020;. Available from: https://doi.org/10.3390/genes11060624. [60] Massari S, Bellahcene F, Vaccarelli G, Carelli G, et al. The deduced structure of the T cell receptor gamma locus in Canis lupus familiaris. Mol Immunol 2009;46:2728
36. [61] Bowen L, Aldridge BM, Miles AK, Stott JL. Expressed MHC class Ii genes in sea otters (Enhydra lutris) from geographically separate populations. Tissue Antigens 2006;. Available from: https://doi.org/10.1111/j.1399-0039.2006.00559.x. [62] Sugiyama Y, Nitisha Y, Lansink GMJ, Holmala K et al.. Diversity of the MHC class II DRB gene in the wolverine (Carnivora: Mustelidae: Gulo gulo) in FinlandAvailable from: https://doi.org/10.1371/journal.pone.0267609. [63] Wong J, Tai CM, Hurt AC, Tan H-X. Sequencing B cell receptors from ferrets (Mustela putorius faro). PLoS One 2020;. Available from: https://doi.org/10.1371/journal.pone.0233794. [64] Heinrich F, Jungwirth N, Carlson R, Tipold A, et al. Immunophenotyping of immune cell populations in the raccoon (Procyon lotor). Vet Immunol Immunopathol 2013;168:140
6. [65] Griffiths K, Hou R, Wang H, Zgang Z, et al. Prolonged transition time between colostrum and mature milk in a bear, the giant panda, Ailuropoda melanoleuca. Roy Soc Open Sci 2015;. Available from: https://doi.org/10.1098/rsos.150395. [66] Kehoe SP, Stacy NI, Frasca SJr, Stokol T, et al. Leukocyte and platelet characteristics of the giant panda (Ailuropoda melenoleuca): morphological, cytochemical and ultrastructural features. Front Vet Sci 2020;. Available from: https://doi.org/10.3389/fvets.2020.00156. [67] Kirkegaard M, Sonne C, Leifsson PS, Dietz R, et al. Istology of selected lymphoid organs in polar bear (Ursus maritimus) from East Greenland in relation to concentrations of organohalogen contaminants. Sci Total Environ 2005;341:119
32. [68] Sommer F, Stahlman M, Ilkayeva O, Arnemo JM, et al. The gut microbiota modulates energy metabolism in the hibernating brown bear Ursus arctos. Cell Rep 2016;14:1
7. [69] Shanmugham AA, Kumar JK, Selvaraj I, Selvaraj V. Hematology of sloth bears (Melursis ursinus) from two locations in India. J Wildl Dis 2008;44(2):509
18. [70] Chow BA, Donahue SW, Vaughan MR, McConkey B, Vijayan MM. Serum immune-related proteins are differentially expressed during hibernation in the American black bear. PLoS One 2013. Available from: https://doi.org/10.1371/journal.pone.0066119. [71] Whiteman JP, Harlow HJ, Durner GM, Regehr EV, et al. Heightened immune system function in polar bears using terrestrial habitats. Physiol Biochem Zool 2019;92(1):1
11. [72] Weber DS, De Groot CPJ, Peacock E, Schrenzel MD, et al. Low MHC variation in the polar bear: implications in the face of Arctic warming. Anim Conser 2013;16:671
83. [73] Kuduk K, Babik W, Bojarska K, Sliwinska EB, et al. Evolution of major histocompatibility complex class I and class II genes in the brown bear. BMC Evol Biol 2012;12:197. [74] Jiang P, Josue J, Li X, Glaser D, et al. Major taste loss in carnivorous mammals. Proc Natl Acad Sci USA 2012;109(13):4956
61. [75] Cavagnolo RZ, Vedros NA. Serum and colostrum immunoglobulin levels in the Northern fur seal Callorhinus ursinus. Dev Comp Immunol 1979;3:139
46. [76] Welsch U, Schwertfirm S, Skirnisson K, Schumacher U. Histological, histochemical and fine structural observations on the lymph node of the common seal (Phoca vitulina) and the grey seal (Halichoerus grypus). Anat Rec, 247. 1997. p. 225
42. [77] Gray R, Canfield P, Rogers T. Histology of selected tissues of the leopard seal and implications for functional adaptations to an aquatic lifestyle. J Anat 2006;209:179
99. [78] Sa´ ALA, Breaux B, Buriamqui T, Deiss TC, et al. The marine mammal class II major histocompatibility complex organization. Front Immunol 2019;. Available from: https://doi.org/10.3389/fimmu.2019.00696. [79] Stott J, Aldridge B, Bowen L, Johnson M, et al. Diversity of immune response (Major Histocompatibility complex, MHC) genes in free ranging pinnipeds. Available from: https://www.vin.com/doc/?id 5 3865036. [80] Slade RW. Limited MHC polymorphism in the southern elephant seal: implications for MHC evolution and marine mammal population biology. Proc Roy Soc B Lond 1992;249:163
71. [81] Weber DS, Stewart BS, Scgeinman J, Lehman N. Major histocompatibility complex variation at three class II loci in the northern elephant seal. Mol Ecol 2004;13:711
18. [82] Hammond JA, Guethlein LA, Abi-Rached L, Moesta AK, Parham P. Evolution and survival of marine carnivores did not require a diversity of killer cell Ig-like receptors or Ly49 NK cell receptors. J Immunol 2009;182:3618
27.

This page intentionally left blank

Chapter 21

The perissodactyls: horses and their relatives

Plains zebra: Equus quagga. Courtesy of Fiona Tizard-Meyer.

The Perissodactyls diverged from the other mammalian orders in the great radiation that followed the K-Pg event around 65 mya. They diversified across what is now Eurasia and North America and the fossil record shows many extinct forms. They thrived during the Eocene about 38 mya years ago when climate change resulted in the massive expansion of grasslands. At that time, they were probably the most abundant herbivores in Laurasia. However, while they prospered for a time, they were eventually out-competed by the more efficient ruminants. In the late Eocene and early Oligocene about 35
30 mya, many failed to adapt to climate change during the cooler and dryer Oligocene epoch. Some subsequently migrated to Africa during the Miocene giving rise to modern rhinos and zebras. Thus, the great perissodactyl populations of North America including rhinoceroses and horses eventually died out. Some species survived however until the end of the ice ages, about 11,000 years ago. Their final extinction has been attributed to climate change but possibly also resulted, in part, from hunting by humans. The remaining species persisted over a much-reduced range in Asia and Africa. The tapirs remained jungle species and some migrated to South America where three species survive. The remaining wild perissodactyls are all endangered as a result of hunting and habitat destruction. The odd-toed ungulates have reduced their weight-bearing digits to three as in rhinos, or one as in horses. Their nonweight-bearing toes are vestigial or absent. They are therefore classified as perissodactyls. Perissodactyls are exclusively herbivores. Unlike the ruminants, they retained their single stomachs but became hindgut fermenters to exploit the nutritional properties of the grasses. In effect, therefore, food that has already passed through the intestine is stored and further digested in the cecum. Thus, ideally extracting every possible calorie before it is excreted. The order includes 17 species classified into three families, the Equidae including horses, asses, and zebras, the Rhinocerotidae (rhinoceroses), and the Tapiridae (tapirs).

Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00007-1 © 2023 Elsevier Inc. All rights reserved.

329

330

SECTION | 2 Mammalian orders

The earliest evidence of horse domestication comes from the Volga-Don region of Western Eurasia around 5500 years ago. It is clear however that there were multiple subsequent domestication events and even feralization of some breeds [1]. As might be anticipated, there has been very little immunological research conducted on species other than the domestic horse. Surprisingly also, little is known about the immune system of the donkey, despite its importance as a mode of transport and beast of burden in many countries.

21.1

Reproduction and lactation

Horses use an epitheliochorial placenta. Six tissue layers separate the mother and her fetus. As a result, in order to provide sufficient oxygen and nutrition for the fetus, the entire surface of the placenta must be functional. In addition, only one fetus can usually be supported at a time. As another consequence, no immunoglobulins can cross the placenta, and foals are therefore born agammaglobulinemic. The gestation period of the mare is about 340 days. Lymphocytes are seen first in the fetal thymus at about 60
80 days post-conception. They are found in the mesenteric lymph node and intestinal lamina propria at 90 days and the spleen at 175 days. Blood lymphocytes appear at about 120 days. A few plasma cells may be seen at 240 days. Graftvs-host disease, a cell-mediated response, has developed in immunodeficient foals transplanted with tissues from a 79day-old fetus. The equine fetus can respond to coliphage T2 by 200 days post-conception and Venezuelan equine encephalitis virus by 230 days. Like other large herbivores, the neonatal foal has a well-developed ileal Peyer’s patch that may serve as a primary lymphoid organ and eventually involutes. Major B cell markers are detectable by 90
120 days gestation. IGHM and IGLC transcripts are expressed in the liver, bone marrow, and spleen at all ages. Immunoglobulin V region sequence diversity progressively increases as the fetus develops and as the foal develops into an adult. Thus, gene recombination and immunoglobulin class switching occur during equine fetal life despite the lack of antigenic stimulation [2,3]. As a result, newborn foals have detectable quantities of IgM and IgG in their serum, but IgE production in the horse does not begin until foals are 9
11 months of age [4]. In animals with a long gestation period, such as the horse, the adaptive immune system is fully developed at birth but cannot function at adult levels for several months. Thus, the production of IgG1, IgG3, IgG5, and IgA begin before or at birth and reach maturity at three months. Other antibody responses are much slower to develop. IgG4, and IgG7 production start shortly after birth and develop slowly over the first year of life. Similar delays are also seen in their Tcell responses [5]. For example, IFNγ production by CD41 helper T cells and CD81 cytotoxic T cells starts slowly, shortly after birth and gradually increases over the first year. In contrast, IL-4 is almost undetectable for the first three months of life. These findings, especially the slow development of type two responses, begin to explain how foals are very susceptible to some bacterial infections during their first months of life. When a foal is born, it emerges from the sterile uterus into an environment where it is immediately exposed to a host of microorganisms. The complete development of adaptive immunity depends on antigenic stimulation. Thus, newborn foals are vulnerable to infection for the first few weeks of life. They need assistance in defending themselves at this time. This temporary help is provided by the mare’s colostrum in the form of antibodies, cytokines, and possibly also T cells. The passive transfer of immunity from mare to foal is essential for survival. Normal equine colostrum contains, on average, about 7000 mg/dL IgG but this can range from 3000 to 12,000 mg/ dL. It also contains about 150 mg/dL 1gM and 300 mg/dL IgA. An average of about 100 g of IgG is produced per lactation in the mare [6]. Equine colostrum also contains about 20% IgA. There are significant amounts of IL-4 in a mare’s colostrum which may serve to compensate for the very low level of IL-4 production in the young foal [5]. IL-13 has not been detected on the colostrum but IFN-γ is present in both colostrum and foal serum at birth. The chemokine IL-8 (CXCL8) is also present in significant amounts [7]. Equine colostrum is also rich in activated T cells. These may have a similar immune-stimulating function to that seen in pigs and cattle [8]. It is of interest to note that donkey foals are not totally agammaglobulinemic at birth. When born they have a significant concentration of IgG (8.97 6 0.5 mg/mL) in their serum. In contrast, IgG concentrations in the serum of neonatal horse foals average 0.3 mg/mL [9].

21.2

Hematology

The leukocytes of the perissodactyls are generally unremarkable. The only exception is their eosinophils which contain many large granules (Fig. 21.1; Table 21.1) [10,11].

The perissodactyls: horses and their relatives Chapter | 21

LEUKOCYTES 3 5.4-14.3 x 10 /Pl

FIGURE 21.1 The blood leukocytes of the horse and the composition of adult equine lymphocyte populations. Note the large granules within the eosinophil. Courtesy Dr. Mark Johnson.

ADULT LYMPHOCYTES

CD8+ Lymphocytes

331

T

B

Neutrophils

CD4+ Monocytes Eosinophils

NK?

TABLE 21.1 Blood leukocyte counts in selected equine species. Horse*

Donkeys [10]

White rhino [11]

Total WBC 3 103/μL

4.3
15

10.1

14.5
16.1

Neutrophils (%)

28
83

60

49

Lymphocytes (%)

20
59

54

28

Monocytes (%)

1.5
10.5

3

12

Eosinophils (%)

0
9

3.7

15

Basophils (%)

0
2

,1

,1

*Courtesy of Dr. Karen Russell.

21.3

Innate immunity

21.3.1 Toll-like receptors Equine toll-like receptors, TLR2, 3, 4, 5, 7, and 8 have been fully sequenced. Relative to human gene sequences they show 65%
77% nucleotide homology [12]. Thirteen SNPs have been found in equine TLRs 3, 7, and 8 while TLR4 shows restricted polymorphism.

21.3.2 Antimicrobial peptides Horses are equipped with a diverse variety of bactericidal peptides. These include at least one lysozyme, five cathelicidins, nine defensins, one psoriasin, one hepcidin, and five possible equinins [13]. Equinins are a group of closely related proteinase inhibitors found in the granules of equine neutrophils as well as in their tracheobronchial secretions. They can inhibit microbial proteinase K and subtilisin and some have antibacterial and antiviral functions [14]. For example, they can kill Streptococcus zooepidemicus, Escherichia coli, and Pseudomonas aeruginosa. They are also active against equine herpesvirus 2.

332

SECTION | 2 Mammalian orders

FIGURE 21.2 Paneth cells in the equine small intestine are a major source of intestinal defensins in this species. They are filled with large eosinophilic granules. Original magnification x 60. Courtesy Dr. Brian Porter. From Tizard IR. Veterinary immunology. 10th ed. Saint Louis: Elsevier, 2016.

Horses possess multiple beta-defensins that are B60% identical to those in other mammals [15]. For example, equine beta-defensin-1 is 69.5% identical to porcine BD-1. They are expressed in many different tissues including the heart, liver, lungs, pancreas, and especially the Paneth cells of the gastrointestinal tract. (Fig. 21.2) A defensin gene cluster is located on equine chromosome 27q17. Analysis has identified nine possible defensin genes [16]. Six are very similar in sequence to human BD-4. Ten beta-defensin pseudogenes have also been identified. Horses also possess multiple alpha-defensin genes. Thus, there are 38 gene transcripts of alpha-defensins in equine intestinal tissues and at least twenty may act as functional peptides [17]. However, no transcripts have been detected yet. If these transcripts are expressed, the horse would be the first eutherian outside the primates, lagomorphs, and rodents known to possess alpha-defensins [17].

21.3.3 Cytokines Foals are interferon-gamma deficient at birth [18]. This may explain their marked susceptibility to certain bacteria such as Rhodococcus equi as well as viruses such as equine herpesvirus 1.

21.3.4 Interleukin-26 Interleukin-26 (IL-26) is a member of the IL-10 family. Its gene is located in a cluster bounded by the interferongamma gene on one side and the IL-22 gene on the other. The IL-26 gene is inactivated in horses and related equids such as zebras and donkeys as a result of a single base-pair frameshift. Since IL-26 is a stimulator of T cells and promotes inflammation and autoimmunity in other mammals, this inactivation may explain, in part, why some aspects of acute inflammation differ in horses. It is also of interest to note that the IL-26 gene has been inactivated independently in other mammals such as the African elephant (Loxidonta Africana) and the European hedgehog (Erinaceus europaeus) as well as in rats and mice [19].

21.3.5 Natural killer cells Horses possess natural killer (NK) cells that are cytotoxic for MHC- class I deficient target cells [20]. As in humans, equine NK cells play an important role in remodeling the placenta as it grows, and the fetus develops. As discussed in Chapter 2, the interaction zone between the placenta is a site both of immune activity and rigorous immune regulation. The horse is no exception and there are dense infiltrations of lymphocytes surrounding the endometrial cup structures in the horse placenta during early pregnancy [21]. It has been shown that there is a significant enrichment in the uNK cell populations in the mare’s placenta during pregnancy.

The perissodactyls: horses and their relatives Chapter | 21

21.4

333

Lymphoid organs

21.4.1 Thymus Most of the thymus in the newborn foal is located in the anterior mediastinum ventral to the trachea and large blood vessels. It consists of two lobes divided into lobules held together by connective tissue together with some adipose tissue. The thymic lobes may extend as a chain of lobules up the neck as far as the thyroid gland. However, the cervical lobes are very variable. One lobe may have no cervical projection while the other may give off a bifurcating one. The two lobes are in contact. Only the thoracic part may persist into adulthood. When the thymus atrophies, a thin remnant remains in the anterior mediastinum. Other perissodactyls such as rhinos have a thoracic thymus only [22]. The overall histologic structure of the thymus is similar to that seen in other mammals with a distinct cortex and medulla. As the animal ages, the thymic parenchyma progressively shrinks but the extra-parenchymal component does not. An unusual feature is the presence of prominent nonlymphoid hematopoiesis including eosinopoiesis, erythropoiesis, mastocytopoiesis, and plasmacytopoiesis in the equine thymus [23]. Lymphatic vessels full of lymphocytes are especially prominent and presumably serve as the routes by which selected T cells to leave the organ once fully functional. The thymic structure has been well described in rhinos [22].

21.4.2 Spleen The equine spleen has the largest blood storage capacity among mammals. In addition to an extensive red pulp, it has a very muscular capsule and trabeculae. The capsule is bilayered with the inner layer and the trabeculae composed of elastic fibers and smooth muscle cells. The red pulp is of the nonsinusal type [24].

21.4.3 Mucosal lymphoid tissues The horse has well-developed lymphoid tissues in its pharyngeal region. Thus, there is a lingual tonsil located at the root of the tongue. There is also a prominent tonsil in the soft palate located on the oral side. This soft-palate tonsil consists of a central crypt surrounded by primary and secondary lymphoid follicles and interfollicular lymphoid tissue [25]. Horses have two pharyngeal tonsils in the dorsal nasopharynx at the end of the nasal septum. These are not well defined and difficult to see macroscopically. As in other species, there is diffuse lymphoid tissue and isolated follicles located around the opening of the auditory tube—the tubal tonsils. These are continuous with the pharyngeal tonsils [26]. The follicle-associated epithelium overlying these tonsils contains M cells. The central follicular areas are composed largely of B cells but the dome and parafollicular areas contain both CD41 and CD81 T cells [27]. The tonsils of several species of rhinoceros have been examined in detail [28]. In addition to a large faucial tonsil, they possess a laryngopharyngeal tonsil located within the pyriform fossa that appears to develop from the third pharyngeal pouch in association with the thymus. Because of its large size and the high volume of air that passes through the airways with every breath (100,00 L/ 24 hours) the horse may be considered more vulnerable to inhaled respiratory pathogens than smaller mammals. Bronchus-associated lymphoid tissue (BALT) is not present before birth in the foal. It has been suggested that species such as horses with a well-developed Waldeyer’s ring in the pharynx and larynx have less need for bronchial lymphoid tissues. Surveys have shown that BALT is present in less than half of the horses surveyed. This of course may depend on the cleanliness of the inhaled air. Air from a dusty barn is very different from the air in an open pasture. When BALT does develop it is located in bronchioles at bifurcations where inhaled material is most likely to impact [26]. Peyer’s patches are present in both the jejunum and ileum and there are isolated lymphoid follicles in the large intestine. The Peyer’s patches develop during gestation and 245
320 have been counted in the jejunum of the newborn foal. Many disappear as the animal ages and their numbers drop to 100
200. In the ileum, however, there is a large 20
35 cm lymph node present in the newborn. It increases in size until sexual maturity but eventually disappears in older horses.

21.5

The major histocompatibility complex

The equine MHC locus, as in other mammals, is divided into three regions encoding Class II, Class III, and Class I in the usual order. The equine MHC, designated ELA, is located on chromosome 20q14-q22 (Fig. 21.3).

334

DO

A

SECTION | 2 Mammalian orders

DM

B A

TAP1 DM TAP2DO DR

\

B\

B

BB

II

III

DQ

DR

A BAB

B A

FIGURE 21.3 The organization of the equine major histocompatibility complex (ELA). Ψ 5 pseudogene.

II

21.5.1 The MHC class I region Fifteen different MHC class I genes have been identified in the horse class I region. Of these seven appear to be functional [29]. Each of the class I genes encodes a protein chain with a leader sequence, three constant domains (α1
α3), a transmembrane domain, and two cytoplasmic domains. Most mammalian class I genes originated in the three duplication blocks, alpha, beta, and kappa, usually located within the class I region. However, while the equine class I genes are located within the three blocks, they are distributed throughout the MHC region. The alpha block is located at the BNTL2 end of the class II region, The beta block is located at the junction of class III and I regions, while the kappa block is located between GNL1 and TRIM26 within the class I region. The horse beta block contains two putative nonclassical genes and four pseudogenes. In the horse kappa block, there are four classical genes, one putative nonclassical gene, and two pseudogenes. (In general, the horse class I region tends to resemble the pig class I except for the alpha block. This alpha block appears to be unique to the horse) [30].

21.5.2 The MHC class II region The equine class II region is about 1.2 mb in size. It contains 35 gene loci. Of these, 20 have undisrupted open reading frames (ORF) [31]. Of the classical MHC class II genes in the ELA, there are six DRB loci of which three are functional. There are four pairs of DQA-DQB genes. Of the nonclassical MHC class II genes, four are functional—DOA, DOB, DMA, and DMB. The DRB2, DRB3, and DQB3 genes have high sequence conservation [32]. In contrast, DRB1 and DQB1 are highly polymorphic. When comparing these genes to other mammals, significant differences emerge. Thus, horses have four DQA and DQB loci while the cat has none. In the horse, the DRB genes are spread out across a region spanning 0.7 mb whereas, in the human, mouse, cat, pig, and cattle, they are clustered. The organization of the ELA class II subregion of equids is similar across all species of the family tested. However, DOB2 is not present in the genomes of the Asiatic asses, Equus hemionus kulan, and E. h. onager [32]. As in other related species, there is no evidence of any functional DP genes, although many, except the pig, have DP pseudogenes. The dog shows some similarities to the horse with a single functional DRB gene located on the reverse strand in a cluster with the DRA gene and an inverted DRB pseudogene. All the other laurasiatheria—dog, cat, pig, and bovine, have a single inverted DRB gene but the horse has five of them. Note that this inverted DRB gene is not present in humans or mice. This suggests that the inversion event occurred B80
100 mya following the separation of the Euarchontoglires but before the divergence of the laurasiatherian species. This eventually evolved into a separate inverted ELA-DRB gene cluster containing both functional genes and pseudogenes [31]. There are at least 17 DRA alleles, 14 DRB alleles, 27 DQA alleles, and more than 17 DQB alleles in the equine class II region [33]. The expression of MHC class II antigens on horse lymphocytes is determined by their MHC haplotype [34]. Thus, horses with haplotype H3 have lower levels of class II expression regardless of their age. (Foals express less MHC class II on their T cells regardless of haplotype). These class II antigens are expressed on both B and T cells. At the boundary between the ELA class I and class III regions, there is a large, segmental duplication of about 710 kb. It consists of 11 repeated blocks, 10 of which contain an MHC class I-like sequence. The remaining block contains a full-length BAT1 gene. A similar duplication is found in other Perissodactyls indicating that it is of ancient origin, probably over 20
55 my. Its persistence and transcription suggest that it is probably functionally important [35].

The perissodactyls: horses and their relatives Chapter | 21

335

21.5.3 The natural killer receptor complex In marked contrast to the other nonrodent mammals, the domestic horse and other members of the family Equidae employ multiple Ly49-like receptors as a result of significant expansion of the LY49 gene locus, presumably by duplication [36,37]. Genomic analysis has shown the presence of six highly conserved polymorphic LY49 genes in the NKC of horses, asses, and zebras [36]. Five of their products have a conserved immune-tyrosine-based inhibition motif (ITIM) and one has arginine in its transmembrane region. All five of these functional Ly49 genes are present across the entire perissodactyl family including not only the equids but also rhinos and tapirs [36]. Several KIR-like sequences have also been detected in the equine genome although only one appears likely to encode a functional NK cell receptor [36]. There is also a nonfunctional KIR-immunoglobulin-like transcript fusion gene (KIR-ILTA) present in horses, donkeys, and plains zebra (E. quagga). There are two probable pseudogenes (KIRP1 and KIRP2) with premature stop codons or frameshift mutations, and a KIR3DL-like sequence has also been identified. This protein contains a single nonmutated ITIM. This is a potentially functional KIR gene. These gene arrangements also appear to be present in tapirs and rhinos as well. Equine peripheral blood NK cells are active against diverse xenogeneic and allogeneic target cells. Thus, they are active against equine herpesvirus-1-infected equine embryonic kidney and embryonic lung cells [38].

21.5.4 Dendritic cells Equine DCs have typical dendritic cell morphology—irregularly shaped large cells and prominent cytoplasmic processes. They express MHC class II, CD11, EqWC1, LFA1, and EqWC2 as would be anticipated from a cell type whose function is to present antigens to T cells. Different subsets have been identified based on their expression of MHC class II and other markers. Thus, dendritic cells from lymph nodes are CD1b positive but blood DCs are not [39]. DCs are present in equine lungs. They too express high levels of MHC I and CD44 but appear to be at an earlier developmental stage than blood DCs and have a significantly greater ability to take up antigens [40]. Equine pDCs have been characterized and shown to produce large amounts of IFN-α on stimulation with TLR9 agonists. Dendritic cells can be readily derived from equine monocytes. They too are both MHC class II positive and CD14 positive. There appear to be two subpopulations that differ in size, but it is not known if they are functionally different [41]. Single-cell RNA sequencing has confirmed that three distinct dendritic cell subsets are present in equine blood including cDC1, cDC2, and plasmacytoid DCs [42].

21.6

B cells and immunoglobulins

Equine B cells express CD20 and MHC class II molecules as well as immunoglobulin antigen receptors. The majority of equine peripheral blood B cells are unusual in that they also express T-bet (T box expressed in T cells). T-bet is a T cell transcription factor. T-bet expression evolved long before the emergence of the adaptive immune system. It is normally expressed in helper T cells where it plays a role in regulating the expression of IFN-γ [43]. T-bet expression is usually associated with chronic infections and inflammation in humans and mice. In humans, when T cells proliferate and mutate in response to antigenic stimulation T-bet1 memory B cells develop and promote B cell survival by regulating the transcription of the mature B cell receptor and thus regulate immunoglobulin class switching. It also drives the migration of some B cells to sites of inflammation. In horses, about half of these T-bet1 cells express IgM, a quarter express IgG while the remainder express neither [43]. A significant increase in immunoglobulin diversity occurs during the long fetal life in the horse, especially through the last two-thirds of gestation despite the absence of stimulation by exogenous antigens. In the fetus, this diversity begins increasing about day 120 gestation and is focused on the third heavy chain complementarity determining region (CDR) within their V domains. In neonatal foals, the diversity increases in the second CDR as well. In adult animals, variation in the first CDR increases belatedly, although it is usually much less diverse than are CDRs 2 and 3. Some V, D, and J genes are used predominantly throughout life, and changes in relative gene usage do not occur as horses age [6].

336

SECTION | 2 Mammalian orders

21.6.1 Immunoglobulin heavy chains The horse-heavy chain gene locus is located on chromosome 24q. This corresponds to human chromosome 14 where the IGH locus is located. Repeated duplication of the IGHG locus has resulted in the production of seven functional IgG subclasses in the horse (Fig. 21.4). All seven of these genes are expressed, in addition to the other four heavy chain genes, M, D, A, and E [44]. The seven IgG subclasses are designated IgG1 thru IgG7. (The previous nomenclature for IgG1 thru IgG4 was IgGa, IgGc, IgG(T), and IgGb. IgG6 was previously called IgG[B]). The former IgG(T) of the horse is now known to be a mixture of two subclasses IgG3 and IgG5. IgG7 is closely related to IgG4 and likely resulted from a recent duplication of the IGHG4 gene. The relative expression ratio of IgG7 to IgG4 is 1:1.6. This duplication event has been detected in all horse breeds examined to date including both Thoroughbred horses and the isolated population of Icelandic horses [45]. Thus, the IGHG4/7 duplication event occurred more than 1000 years ago. IgG1, IgG4, and IgG7 are generally produced in response to intracellular infections while IgG3 and IgG5 are mainly produced in response to extracellular invaders. Horses have two IgG4 alleles (IgG4a and IgG4b) and four IgE alleles (IgE1
4). Horses also express IgM, IgD, IgA, and IgE. The equine IgM locus spans 1472 bp and encodes 451 amino acids. Its CH3 and CH4 exons are highly conserved. The horse IGHD gene is located 5 kb downstream from IGHM. It spans about 9.1 kb and contains eight exons that are very similar in sequence to human IgD, namely—CH1-H1-H2—CH2-CH3- S-M1-M2-. As in humans, no switch region has been detected upstream of the IGHD gene. IGHD is located very close to the IGHM gene, and it is possible that long pro-messenger transcripts might first form and are then subjected to alternative splicing or processing. IgD appears to be expressed at least at the mRNA level in equine B cells. The equine IgE locus is similar to IgM in that there is no separate hinge region for IgE, and flexibility is conferred by CH2. Interestingly, IgE concentrations in the equine bloodstream are about 1000-fold higher in normal horses than in normal humans [45] (Table 9.2). There is only one isotype of equine IgA. The horse IgA locus contains three exons. These encode two constant domains and a single hinge region. The hinge region is an extension of the Cα2 domain. IgA is found in horse serum as well as in their mucosal secretions [44]. The IgA in mucosal secretions is produced in the form of secretory IgA. That is a dimer connected by a 17 kDa J chain. It is transported onto mucosal surfaces by the polymeric Ig receptor (pIgR), a type I transmembrane glycoprotein. The extracellular component of the pIgR is cleaved off to form secretory component which then protects the IgA against bacterial proteases. Both the equine J chain and pIgR have been characterized. The J chain contains four exons while pIgR contains 11. They have been mapped to equine chromosomes three and five respectively [46]. Many of the functions of IgA are mediated through its specific receptor, FcαR, otherwise called CD89. This receptor has been identified and characterized in the horse where it is located on chromosome 10 [47]. The genes encoding other classical Fc receptors are located on equine chromosome 5.

IGH

5'

52

40

3’

8 M

IGK

IGL

60

D

G1 G2 G3 G7 G4 G6 G5 E

5

34

110 7 J-C cassettes

M

D

G

E

A

V

D

J

LC

A

FIGURE 21.4 The organization of the equine immunoglobulin gene loci.

The perissodactyls: horses and their relatives Chapter | 21

337

21.6.2 IGHV genes Overall, the horse has 280 IGH genes while the white rhino (Ceratotherium simum) has 266. Rhinos have 69 IGHV genes. In the equine IGH locus, there are 52 IGHV, 40 IGHD, eight IGHJ, and 11 IGHC genes. The 40 identified D genes are among the largest number recorded in mammals, although guinea pigs have 41 and elephants have 87 [6]. The horse D segments range from 18 to 48 bp in length. This is longer than in humans (11
37 bp) but significantly shorter than in cattle. Horses therefore predominantly employ gene recombination. The horse IGHV subfamilies belong to all three mammalian clans [48]. Thus subfamilies 1 and 5 belong to clan I; subfamilies 2, 4, 6, and 7, to clan II and subfamily 3 belongs to clan III. Note that clan III IGHV segments are found in all mammals Including camels, rabbits, platypus opossums, and pigs. (Fig. 9.6). In humans and mice, clan III is preferentially expressed during fetal life so these IGHV genes may play a role in the development of the foal’s immune system. While the horse IGHV3 subfamily belongs to clan III and has at least ten members, none appear to be functional. Examination of the equine B cell repertoire indicates that horses prefer to use the IGHV2 subfamily found in 92.49% of B cells. The CDR3 extended loop structure also predominates in the horse [49].

21.6.3 Immunoglobulin Light chains The equine kappa light chain locus contains 60 IGKV, five IGKJ, and one IGKC gene. It is arranged in a single cluster of 820 kb. The ratio of Vλ to Vκ genes agrees with the preponderance of lambda chain use in horse serum although the ratio in serum is 13:1. Thus lambda light chains constitute B92% of the antibody repertoire in horses [6]. The lambda light chain locus IGL, contains seven IGLC genes, each preceded by a single IGLJ gene and 144 IGLV genes divided into two clusters both upstream and downstream of the J-C cluster. They extend for 1310 kb on chromosome 8. Each of the J segments is associated with a constant gene [6]. Horses preferentially use IGLV8 which is used in 82.5% of B cells. The two clusters of IGLV genes are arranged in opposite transcriptional orientations. Within each cluster, there are both functional genes and pseudogenes. There are many more ORFs and pseudogenes than functional IGLV genes. However, allotypic and allelic variants are seen in IGLC1, IGLC5, and IGLC6/7. Two IGLV pseudogenes are also transcribed. The pseudogenes may also be used for gene conversion. V gene usage appears to not change throughout the animal’s life although sequence diversity increases as animals develop from fetuses to adults [3].

21.7

T cells and cell-mediated immunity

There are two major T cell populations in the horse, CD31 perforin1 (cytotoxic) cells, and CD31 perforin (noncytotoxic) T cells [42]. Some of these cytotoxic cells are γ/δ positive. Horse lymphocytes express two species-specific proteins. EqWC1 is found on 70% of equine T cells, 30% of B cells, and 50% of granulocytes and could be a homolog of human CD90. A second specific protein, EqWC2 is expressed on granulocytes and most T cells. Both α/β and γ/δ T cells are widely distributed in multiple organs. The two cell types have a similar distribution, especially in epithelial and lymphoid tissues. Treg cells expressing FoxP3 are present in the peripheral blood of horses. Their numbers are higher in horses under 6 years of age (2.7%) compared to older horses (1.5%) [50]. The proportion of FoxP31 cells is also much higher in foals than in their mothers. Not only that but they are functionally more suppressive [51]. In equine CD81 T cells both the α and β chains of the CD8 heterodimer are expressed in the cells from blood, thymus, spleen, mesenteric lymph node, and ileal intraepithelial lymphocytes [52]. However, there is also a subpopulation of CD8α/α cells present in the intestinal mucosa that account for up to a third of their intraepithelial lymphocytes.

21.7.1 T cell receptor genes 21.7.1.1 TRA/D genes The equine TRA/D locus contains five unique TRAV gene segments in five different subfamilies, five distinct TRAJ genes that have conserved 30 residues but diverse residues at their 50 end, and a single TRAC gene (Fig. 21.5). The TRAC gene, when compared to other mammals, has highly conserved transmembrane and cytoplasmic domains. This conservation is almost certainly required to retain its ability to interact with the CD3 signaling peptides. The equine TRD locus contains eight unique TRDV genes from seven subfamilies. These result in significant junctional diversity in the V-J region. It is of interest to note that the equine TRDV gene segments most closely resemble

338

TRA/D

TRB

SECTION | 2 Mammalian orders

5'

5 AV 16

8 DV

6 DD DJ DC

5 AJ

3'

AC

FIGURE 21.5 The equine TCR gene loci. No data is available as yet on the organization of the TRG locus except for the identification of two TRGC genes.

7

7

V

D

J

C

those in the sheep. Given that both are grazers exposed to a similar spectrum of intestinal parasites it may be speculated that these TRDV segments have been selected by the same antigenic stimuli. Horses have six, highly conserved TRDJ genes of which three are functional. The locus also contains a single TRDC gene that is homologous to that in other mammals except for its connecting peptide region [53].

21.7.1.2 TRB genes The equine TRB gene locus contains 16 TRBV, 14 TRBJ, and two constant region genes. The TRBV genes are grouped into nine subfamilies based on 75% or greater similarities to the human sequences. The most common subfamily is TRBV2 which has four unique members out of five horses tested. Ten of the TRBV genes are full length while six are truncated pseudogenes. There are also obvious TRBV polymorphisms among the five different horse breeds tested (Thoroughbred, Arabian, Standardbred Quarter horse, and American miniature horse) [54]. Each of the constant genes TRBC1 and TRBC2 encode a peptide chain of 177 amino acids. The chains differ in nine amino acid residues of which four are located in the transmembrane domain and one in the cytoplasmic domain.

21.7.1.3 TRG genes The horse has at least two TRGC genes. These have 77% nucleotide identity including four conserved cysteine residues and stop codons that are also found in humans, cattle, sheep, and mice. The duplication occurred a very long time ago. The two TRGC gene products differ most in their hinge region where TRGC1 has 18 more amino acids than TRGC2. It is speculated that these differences in their antigen receptors may affect the specific tissue distribution of equine γ/δ T cell populations [54]. Equine TRGV and TRGJ genes have yet to be characterized

21.7.2 Natural killer T cells and CD1 CD1d-restricted invariant natural killer T cells (NKT cells) are present in humans and mice, pigs and horses [55]. Using an invariant T cell α chain, these cells can recognize the characteristic glycolipid ligand α-galactosylceramide (α-GalCer) presented by a CD1d molecule. (Fig. 10.7). The presence of a functional CD1 system appears to be critical for horse health. As discussed in other chapters, the CD1 gene family is a group of nonpolymorphic MHC class I molecules that specifically present lipid antigens to T cells. In the horse, for example, they have been shown to present lipid antigens from the major equine pathogen, R. equi to cytotoxic T cells. The equine genome has been shown to contain 13 complete CD1 genes located in a 918 kb region on chromosome 5. Of these seven are homologous to human CD1a, two to CD1b, one each to CD1c and CD1d, and two to CD1e. It also contains five nonfunctional pseudogenes. This is one of the largest CD1 gene families yet recorded in mammals. Twelve of these CD1 molecules are expressed on antigen-presenting cells, including monocytes, macrophages, and dendritic cells. The polymorphism within the CD1 family is restricted to the antigen-binding α1 and α2 domains suggesting that multiple different lipid antigens can be presented and recognized. Lacking a tyrosine sorting motif, the eqCD1 molecules are predicted to colocalize with R. equi in intracellular vesicles [56]. Although horses possess functional NKT cells as well as a large CD1 gene family, they also differ in some respects from other mammals. For example, while α-GalCer is a potent NKT cell stimulator in most species examined, it is not however, an immunostimulatory NKT cell agonist in the horse [57].

The perissodactyls: horses and their relatives Chapter | 21

339

References [1] Brunson K, Reich D. The promise of paleogenomics beyond our own species. Trends Genet 2019;. Available from: https://doi.org/10.1016/j. tig.2019.02.006. [2] Talmadge RL, McLaughlin K, Secor E, et al. Expression of essential B cell genes and immunoglobulin isotypes suggests active development and gene recombination during equine gestation. Dev Comp Immunol 2009;33:1027
38. [3] Tallmadge RL, Tseng CT, King RA, Felippe MJB. Developmental progression of equine immunoglobulin heavy chain variable region diversity. Dev Comp Immunol 2013;41(1):33
43. [4] Marti E, Ehrensperger F, Burger D, et al. Maternal transfer of IgE and subsequent development of IgE responses in the horse (Equus caballus). Vet Immunol Immunopathol 2009;127:203
11. [5] Perkins GA, Wagner B. The development of equine immunity: current knowledge on immunology of the young horse. Equine Vet J 2015;47:267
74. [6] Walther S, Rusitzka TV, Diesterbeck US, Czarny C-P. Equine immunoglobulins and organization of immunoglobulin genes. Dev Comp Immunol 2015;53:303
19. [7] Mariella J, Castagnetti, Prosperi A, Scagliarini A, Peli A. Cytokine levels in colostrum and in foals’ serum pre-and post-suckling. Vet Immunol Immunopathol 2017;185:34
7. [8] Perkins GA, Goodman LB, Wimer C, Freer H, et al. Maternal T-lymphocytes in equine colostrum express a primarily inflammatory phenotype. Vet Immunol Immunopathol 2014;161:141
50. [9] Turini L, Bonelli F, Nocera I, Battaglia F, et al. Evaluation of Jennies’ colostrum: IgG concentrations and absorption in the donkey foals. A preliminary study. Heliyon 2020;. Available from: https://doi.org/10.1016/j.heliyon.2020.e04598. [10] Burden FA, Hazell-Smith E, Mulugeta G, Patrick V, et al. Reference intervals for biochemical and haematological parameters in mature domestic donkeys (Equus asinus) in the UK. Equine Vet Educ 2016;28(3):134
9. [11] Miller M, Buss P, Wanty R, Parsons S, et al. Baseline hematological results for free-ranging white rhinosceros (Ceratotherium simum) in Kruger National Park, South Africa. J Wildlife Dis 2015;51(4):916
22. [12] Jungi TW, Farhat K, Burgener IA, Werling D. Toll-like receptors in domestic animals. Cell Tissue Res 2011;343:107
20. [13] Bruhn O, Grotzinger J, Cascorbi I, Jung S. Antimicrobial peptides and proteins of the horse
insights into a well-armed organism. BMC Vet Res 2011;. Available from: https://doi.org/10.1186/1297-9716-42-98. [14] Pellegrini A, Kalking M, Hermann M, Gru¨nig B, et al. Equinins in equine neutrophils: quantification in tracheobronchial secretions as an aid in the diagnosis of chronic pulmonary disease. Vet J 1998;155:257
62. [15] Bagnicka E, Strazlkowska N, Jo´zwik A, Krzyzewsli J, et al. Expression and polymorphism of defensins in farm animals. Acta Biochim Polon. 2010;57(4):487
97. [16] Looft C, Paul S, Regenhard P, Kuiper H, et al. Sequence analysis of a 212 kb defensin gene cluster on ECA 27q17. Gene 2006;376:192
8. [17] Bruhn O, Paul S, Tetens J, Thaller G. The repertoire of equine intestinal α-defensins. BMC Genomics 2009;. Available from: https://doi.org/ 10.1186/1471-2164-10-631. [18] Breathnach CC, Sturgill-Wright T, Stiltner JL, et al. Foals are interferon gamma-deficient at birth. Vet Immunol Immunopathol 2006;112:199
209. [19] Shakhsi-Niaei M, Drogemuller M, Jagannathan V, Gerber V, Leeb T. IL-26 gene inactivation in Equidae. Anim Genet 2013;44:770
2. [20] Vivieros MM, Antczak DF. Characterization of equine natural killer and IL-2 stimulated lymphokine activated killer cell populations. Dev Comp Immunol 1999;23:521
32. [21] Noronha LE, Huggler KE, de Mestre AM, Miller DC, Antczak DF. Molecular evidence for natural killer-like cells in equine endometrial cups. Placenta 2012;33:379
86. [22] Cave AJE. The thymus gland in three genera of rhinoceros. J Zool 2009;142(1):73
84. [23] Contreiras EC, Lenzi HL, Meirelles MNL, Caputo LFG, et al. The equine thymus microenvironment: a morphological and immunochemical analysis. Dev Comp Immunol 2004;28:251
64. [24] Udroiu I. Evolution of sinusal and non-sinusal spleens of mammals. Hystrix It J Mamm 2006;17(2):99
116. [25] Casteleyn C, Breugelmans S, Simoens P, Van den Broeck W. The tonsils revisited: review of the anatomical localization and histological characteristics of the tonsils of domestic and laboratory animals. Clin Dev Immunol 2011;. Available from: https://doi.org/10.1155/2011/4/472460. [26] Liebler-Torino E, Pabst R. MALT structure and function in farm animals. BMC Vet Res 2006;37(3):257
80. [27] Kumar P, Timoney JF, Sheoran AS. M cells and associated lymphoid tissue of the equine nasopharyngeal tonsil. Equine Vet J 2001;33 (3):224
30. [28] Cave AJE. The rhinoceros faucial and laryngopharyngeal tonsils. J Zool Lond 1979;187:471
503. [29] Tallmadge RL, Lear TL, Antczak DF. Genomic characterization of MHC class I genes in the horse. Immunogenetics 2005;57:763
74. [30] Tallmadge RL, Campbell JA, Miller DC, Antczak DF. Analysis of MHC class I genes across horse MHC haplotypes. Immunogenetics 2010;62:159
72. [31] Viluma A, Mikko S, Hahn D, Skow L, Andersson G, Bergstrom TF. Genomic structure of the horse major histocompatibility complex class II region resolved using PacBio long-read sequencing technology. Sci Rep 7:45518; doi:10.1038/srep45518 [32] Klumplerova M, Splichalova P, Oppelt J, Futas J, et al. Genetic diversity, evolution and selection in the major histocompatibility complex DRB and DQB loci in the family Equidae. BMC Genomics 2020;. Available from: https://doi.org/10.1186/s12864-020-07089-6. [33] Fraser DG, Bailey E. Polymorphism and multiple loci for the horse DQA gene. Immunogenetics 1998;47:487
90.

340

SECTION | 2 Mammalian orders

[34] Barbis DP, Bainbridge D, Crump AL, Zhang CH, Antczak DF. Variation in expression of MHC class II antigens on horse lymphocytes determined by MHC haplotype. Vet Immunol Immunopathol 1994;42:103
14. [35] Brinkmeyer-Langford CL, Murphy WJ, Childers CP, Slow LC. A conserved segmental duplication within ELA. Anim Genet 2010;. Available from: https://doi.org/10.1111/j.1365.2052.2010.02137.x. [36] Futas J, Horin P. Natural killer cell receptor genes in the family Equidae: Not only Ly49. PLoS One 2013;. Available from: https://doi.org/ 10.1371/journal.pone.0064736. [37] Takahashi T, Yawata M, Raudsepp T, et al. Natural killer cell receptors in the horse: evidence for the existence of multiple transcribed LY49. Eur J Immunol 2004;34:773
84. [38] Chong YC, Duffus WPH, Hannant D. Natural killer cells in normal horses and specific-pathogen-free foals infected with equine herpesvirus. Vet Immunol Immunopathol 1992;33:103
13. [39] Siedek E, Little S, Mayall S, Edington N, Hamblin A. Isolation and characterization of equine dendritic cells. Vet Immunol Immunopathol 1997;60:15
31. [40] Lee Y, Kiupei M, Soboll Hussey G. Characterization of respiratory dendritic cells from equine lung tissues. BMC Vet Res 2017;. Available from: https://doi.org/10.1186/s12917-017-1240-z. [41] Lopez BS, Hurley DJ, Giancola S, Gigue`re S, et al. Phenotypic characterization of equine monocyte-derived dendritic cells generated ex vivo using commercially available serum-free medium. Vet Immunol Immunopathol 2020;. Available from: https://doi.org/10.1016/j. vetimm.2020.110036. [42] Patel RS, Tomlinson JE, Divers TJ, Van der Walle GR, Rosenberg BR. Single-cell resolution landscape of equine peripheral blood mononuclear cells reveals diverse cell types including T-bet 1 B cells. BMC Biol; doi:10.1186/s12915-020-00947-5 [43] Lazarevic V, Glimcher LH, Lord GM. T-bet: a bridge between innate and adaptive immunity. Nature Rev Immunol 2013;13:777
89. [44] Wagner B. IgE in horses: occurrence in health and disease. Vet Immunol Immunopathol 2009;132:21
30. [45] Wagner B, Miller DC, Lear TL, Antczak DF. The complete map of the Ig heavy chain constant gene region reveals evidence for seven IgG isotypes and for IgD in the horse. J Immunol 2004;173:3230
42. [46] Lewis MJ, Wagner B, Irvine RM, Woof JM. IgA in the horse: cloning of equine polymeric Ig receptor and J chain and characterization of recombinant forms of equine IgA. Mucosal Immunol 2010;3(6):610
21. [47] Morton HC, Pleass RJ, Storset AK, Brandtzaeg, et al. Cloning and characterization of equine CD89 and identification of the CD89 gene in chimpanzees and rhesus macaques. Immunology 2005;115:74
84. [48] Sun Y, Wang C, Wang Y, Zhang T, et al. A comprehensive analysis of germline and expressed immunoglobulin repertoire in the horse. Dev Comp Immunol 2010;34:1009
20. [49] Manso TC, Groenner-Penna M, Minozzo JC, Antunes BC, et al. Next-generation sequencing reveals new insights into gene usage and CDR-H3 composition in the horse antibody repertoire. Mol Immunol 2019;105:251
9. [50] Robbin MG, Wagner B, Noronha LE, Antczak DF, de Mestre AM. Subpopulations of equine blood lymphocytes expressing regulatory T cell markers. Vet Immunol Immunopathol 2011;140:90
101. [51] Hamza E, Mirkovitch J, Steinbach Marti. Regulatory T cells in early life: comparative study of CD4 1 CD25 high T cells from foals and adult horses. PLoS One 2015; 1371/journal.pone.0120661. [52] Tschetter JR, Davis WC, Perryman LE, McGuire TC. CD8 dimer usage on ab and gd T lymphocytes from equine lymphoid tissues. Immunobiology 1997;198:424
38 98. [53] Schrenzel MD, Ferrick DA. Horse (Equus caballus) T cell receptor alpha, gamma and delta chain genes: nucleotide sequences and tissue specific gene expression. Immunogenetics 1995;42:112
22. [54] Schrenzel MD, Watson JL, Ferrick DA. Characterization of horse (Equus caballus) T-cell receptor beta chain genes. Immunogenetics 1994;40:135
44. [55] Van Beeck FAL, Reinink P, Hermsen R, Zajonc D, et al. Functional CD1d and/or NKT invariant chain transcript in horse, pig, African elephant and Guinea pig, but not in ruminants. Mol Immunol 2009;46:1424
31. [56] Dossa RG, Alperin DC, Hines MT, Hines SA. The equine CD1 gene family is the largest and most diverse yet identified. Immunogenetics 2014;66:33
42. [57] Dossa RG, Alperin DC, Garzon D, Mealey RH, et al. In contrast to other species, α-galactosylceramide (α-GalCer) is not an immunostimulatory NKT cell agonist in horses. Dev Comp Immunol 2015;49:49
58.

Chapter 22

The Lagomorpha: rabbits, hares, and picas

Cape hare. Lepus capensis

Rabbits, hares, pikas, and rodents are diverse mammals with some superficial similarities. They are collectively classified as members of the clade Glires. The rabbits, hares, and pikas belong to the order Lagomorpha whereas the rodents such as rats and mice belong to the order Rodentia. The orders Rodentia and Lagomorpha together constitute almost half of all species of extant mammals. Despite belonging to the same clade, however, these two orders have had a long and divergent history. They probably diverged in the northern part of Gondwana around 70 mya. The first lagomorph fossils are dated to the late Paleocene and early Eocene about 57
55 mya in Eurasia and North America. The early Lagomorphs likely originated in Eurasia and spread into North America. The climatic changes that occurred in the early Miocene facilitated their expansion into newly expanded grasslands. The lagomorphs are classified into two families. One contains a single genus, the Pikas—the Ochotonidae while the other contains eleven genera including the rabbits and hares—the Leporidae. Fossil data suggests that the two diverged between 30 and 40 mya whereas molecular evidence suggests that it was considerably earlier, probably around 50
55 mya [1]. The divergence of the major genera, the hares, and the rabbits, likely occurred around 14 mya but there are differences of opinion on this. The Lagomorphs declined after the Miocene, most likely due to competition from Artiodactyls and of course, the rise of the carnivores [2]. Rodents and Lagomorphs superficially resemble each other. Members of both orders possess gnawing incisors, no canine teeth, and a gap (diastema) between the incisors and the cheek teeth. However, Lagomorphs possess four incisors on their upper jaw, while rodents have only two. Thus, lagomorphs have two pairs of upper and one pair of lower incisors. The second pair of upper incisors are reduced in size, lacks a cutting edge, and is situated behind the first pair. The upper and lower incisors grow continuously and as in rodents, develop sharp, chisel-like edges. Lagomorphs are exclusively herbivorous while some rodents are omnivorous.

Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00020-4 © 2023 Elsevier Inc. All rights reserved.

341

342

SECTION | 2 Mammalian orders

During the Ice Age, the European rabbit was restricted to the Iberian Peninsula where two subspecies currently exist, Oryctolagus cuniculus algirus in the southwest and O.c. cuniculus in the northeast. Once the ice had gone and the climate warmed, about 11,000 years ago, rabbits moved north to occupy the rest of Europe. As in other chapters, one species, the European rabbit, O.c. cuniculus, has been most intensively studied by immunologists. Unless stated otherwise this is the rabbit described in this chapter. This is an incredibly successful species that, with human assistance, has colonized most suitable habitats worldwide. One reason for this is its adaptability reflected by its ability to switch, depending upon circumstances, from an r strategy to a K strategy or back as needed.

22.1

Reproduction and lactation

The rabbit placenta is of the hemodichorial type. As in rodents, it consists of a persistent bidiscoid yolk sac placenta with two trophoblast layers [3]. The transfer of immunoglobulins across the rabbit yolk sac placenta occurs in phases [4]. Thus, for the first 8 days of pregnancy antibodies are simply transferred from the uterine lumen to the developing yolk sac. From 8 to 13 days, the yolk sac partially inverts and as a result, no immunoglobulins are transferred. By 13 days however the yolk sac endoderm comes into close contact with the uterine epithelium and immunoglobulin transport resumes. IgG is readily transferred with the rate of transfer increasing as pregnancy proceeds. As a result, IgG concentrations in newborn rabbits are approximately the same as their mother’s [5]. Unlike other mammals, IgM antibodies in the rabbit are also transferred in significant amounts through the yolk sac splanchnopleure to the fetal bloodstream. Rabbits possess effective and functional Fc receptors in their placentas that facilitate this immunoglobulin transfer [6]. Rabbits (Oryctolagus), have a gestation period of only 31
32 days. As a result, newborn rabbits are very altricial and only open their eyes at 10 days of age. Their spleen is small and has no follicles or germinal centers at birth. They are therefore born with very immature immune systems. However, when immunized with tetanus toxoid; even the B cells of one-day-old rabbits can respond by producing specific IgM and IgG immunoglobulins although the amounts produced are low and their responses very slow [7]. Other lagomorphs such as hares (Lepus) are very precocial while cottontails (Silvilagus) are intermediate. Interestingly, the milk composition in all three species is very similar [8]. Immunization of pregnant does using respiratory syncytial virus administered by the oral or intratracheal routes results in the appearance of antiviral IgA in their colostrum and milk. On the other hand, administration of bovine serum albumin results in the production of IgG antibodies in colostrum and milk [8]. Studies on normal rabbit mammary glands indicate that of the cells that stain positively for immunoglobulins, 66% stain with anti-IgA, 21% with anti-IgG, and 12% with anti-IgM [9].

22.2

Hematology

As in many mammals, blood leukocyte numbers tend to be lower in newborn and juvenile rabbits under 1 year of age. As a result, the neutrophil to lymphocyte ratio which is about 1:2 in 2-month-old rabbits climbs to 1:1 in adults. Rabbit lymphocytes have conventional morphology [10]. Neutrophils however have large cytoplasmic granules that stain dark pink with Romanowsky stains. As a result, they are designated heterophils. Despite this different staining pattern, rabbit heterophils are functionally the same as neutrophils in other mammals. True eosinophils have larger and more spherical granules (Fig. 22.1) that stain with eosin-based stains. Basophils are relatively common in rabbits and may account for up to 5% of blood leukocytes. Rabbits normally have about 3000
9000 leukocytes per μL. Between 38% and 66% are heterophils, 26%
51% are lymphocytes, 4%
16% monocytes and up to 3% eosinophils [11].

22.3

Innate immunity

Evidence suggests that young rabbits have heightened innate immune responses when compared to adult rabbits. This is reflected in the increased expression of Major histocompatibility complex (MHC) class II molecules, and the activities of their natural killer (NK) cells and macrophages (as measured by RNA-expression profiles). As a result, young rabbits can respond more effectively than adults to infection with some variants of the rabbit hemorrhagic disease virus [12].

The Lagomorpha: rabbits, hares, and picas Chapter | 22

343

H H

E E FIGURE 22.1 The top two images are rabbit heterophils (H) showing the densely packed small eosinophilic cytoplasmic granules. Compare them with rabbit eosinophils (E) from the same blood smear. Their cytoplasmic granules are very much larger. Courtesy Dr. Mark Johnson.

22.3.1 Toll-like receptors Toll-like receptors 1
6 and 8
10 have been characterized in the rabbit. They appear to have similar ligands and functions as their homologs in other mammals. Studies on the phylogenetic tree of TLR2 showed longer branches in Lagomorphs. Further analysis has indicated that TLR2 has undergone mutation at a much higher rate than in other mammals [13]. This suggests that rabbit TLR2 has been strongly selected as a result of pathogen-mediated selection. TLR7 is also absent from rabbits while TLR8 has very low activity [1]. There is no evidence for the presence of TLR7 genes within the rabbit genome. In most mammals, both TLR7 and TLR8 are encoded by genes located within a highly conserved syntenic region that is absent in the rabbit. Both TLR7 and TLR8 in other mammals have similar functional activities and are activated by similar ligands. It is unclear why both are retained but presumably, each has some unique feature that ensures their retention. The mechanisms behind the loss of TLR7 in rabbits are currently unknown. The other PRRs including TRIM5a, RLRs, MDA5, and LGP 2 are all present in rabbits. RIG-1 plays an important role in recognizing the myxomatosis virus [1].

22.3.2 Cytokines While the family of interleukins keeps growing, the identification of many of these proteins in the rabbit has tended to lag. They clearly possess the key cytokines, IL-2, IL-4, IL-6, and IL-10. Likewise, the genes for IL-4, -5, and -13 have been mapped to rabbit chromosome 3. There are however some species-specific differences. Thus, a mutation in the stop codon of the IL-6 gene results in a product with an elongated tail of 27 amino acids. This mutation is absent from Cottontail (Sylvilagus floridanus) and Jackrabbits (Lepus californicus) [14]. Chemokine receptor genes tend to be relatively well conserved among mammals. Rabbits are different. For example, the leporid CCR5 gene is distinctly different from other mammals since it has undergone a gene conversion event with the paralogous CCR2 gene. Thus one, six amino acid motif from CCR5, is replaced by a six amino acid motif from CCR2. Rabbit CXCR1 and CXCR 2 appear to be inactivated as a result of frameshifts although it is likely that one remains functional [15]. The CC chemokine ligand 16 (CCL16) is a potent proinflammatory molecule in most mammals. However, it is absent in European rabbits as well as rodents such as rats and mice because its gene is mutated at its Cys-Cys motif creating premature stop codons and frameshifts [1]. These mutations are also present in eleven species of Leporids studied but the CCL16 gene is functional in seven species of Ochotonidae (pikas)! It is also functional in rodents such as the thirteen-lined ground squirrel (Icidomys tridecemlineatus). Thus, it appears that pseudogenization has been random in both branches of the Glires clade [16].

344

SECTION | 2 Mammalian orders

22.3.3 Defensins Two important rabbit defensins are the macrophage cationic peptides, MCP-1 and MCP-2. These two peptides are of similar length and are closely linked suggesting recent gene duplication [17]. They have potent antibacterial and antifungal activity. MCP mRNA is present in lung macrophages, but this mRNA has not been detected in any other macrophage populations suggesting that lung macrophages are its prime source. These defensins are also present in large amounts in rabbit granulocytes. Both MCP1 and MCP-2 can neutralize herpes simplex-2 virus, vesicular stomatitis virus, and influenza virus [18].

22.3.4 Natural killer cells Such evidence as is available suggests that rabbits either do not possess NK cells or possess an NK cell system that differs significantly from that of other species [19]. Thus, the use of classical approaches in rabbit peripheral blood and lymph node cells has failed to detect either NK or natural cytotoxic cell activity or even ADCC under conditions where NK cells from other species were highly effective. Nor have any large granular lymphocytes, the typical NK cells, been seen in rabbit blood. That said, there is a single published report regarding the anti-tumor cell activity of rabbit NK cells [20]. In this report, it was determined that in rabbits undergoing radiofrequency ablation of transplanted liver tumors, the levels of IFN-γ and NK cells climbed significantly following thermal coagulation. It was also reported that NKG2D expression and NK cell antitumor activity directed against hepatic cancer cells were enhanced in treated rabbits. In this study, positive staining with anti-rabbit CD56 was the criterion used to identify and separate the NK cell population by cell sorting [20].

22.3.5 Acute-phase proteins In response to inflammatory stimuli, rabbits produce the same diverse range of acute-phase proteins as other mammals [21]. These include C-reactive protein, serum amyloid A, haptoglobin, α1-macroglobulin, and transferrin [22]. Conversely, serum albumin is a negative acute-phase protein. The precise mixture of APPs generated depends upon the nature of the inflammatory stimulus, the pattern recognition molecules activated, and the inducing cytokine mixture.

22.3.6 Necroptosis Necroptosis is a form of cell death triggered by receptors such as the tumor necrosis factor receptor, nucleic acid sensors, or by TLR-signaling. They activate receptor kinases and form molecular complexes called necrosomes. Together with proteases and kinases, they generate pore-forming proteins that insert themselves into the cell walls and allow the escape of cellular contents. Mitochondria are not involved in necroptosis. Like pyroptosis, necroptosis causes inflammation because of the escape of intracellular DAMPs such as HMGB-1 and interleukin 33 (IL-33). Necroptosis pathways involve two major components. a receptor-interacting protein kinase (RIPK3) that phosphorylates a mixed-lineage kinase-like protein (MLKL) so causing cell membrane rupture and death. However, among some mammalian lineages, the genes encoding RIPK3 and MLKL are deleted or are pseudogenes. Frameshifts or premature stop codons are found in all studied species of cetaceans and leporids (rabbits and hares) (but not other lagomorphs such as the pika) [23]. The MLKL gene is deleted in the European rabbit as well as in carnivores. In some Afrotheria and rodents, there are premature stop codons in either RIPK3 or MLKL. The fact that the necroptosis pathway has been lost several times during mammalian evolution suggests that its presence is not mandatory. It is however still present in humans and mice.

22.4

Lymphoid organs

22.4.1 Thymus As in other mammals, the rabbit thymus is derived from the third pharyngeal pouch and its surrounding mesenchyme. The thymus of the newborn rabbit is bilobed, covered by a thin capsule, and subdivided into separate lobules [22]. It consists of two parts, a large thoracic part and a much smaller cervical part that are continuous with each other. Hassall’s corpuscles develop late in gestation but are obvious in the neonatal thymus.

The Lagomorpha: rabbits, hares, and picas Chapter | 22

345

22.4.2 Spleen As in the rodents, the rabbit spleen has an open circulation, an increased quantity of white pulp, obvious pulp sinuses, and shows no evidence of erythropoiesis [24]. Rabbit spleens also lack the specialized capillary segments known as ellipsoids. As a result, they trap immune complexes in the marginal zones of the white pulp.

22.4.3 Mucosal lymphoid tissues Rabbits possess only a palatine tonsil that forms a bilateral oval protrusion on the dorsolateral wall of the oropharynx. There are no tonsils present in the nasopharynx, but aggregates of lymphoid tissue and lymphoid follicles are found within much of the nasal mucosa [25]. The rabbit’s small intestine contains oval Peyer’s patches located on the antimesenteric border of the intestine. They increase progressively in size and number from the duodenum to the ileum. There are no obvious patches in newborn rabbits, but they become apparent around fifteen days of age. The number of patches in adult rabbits varies from two to ten and averages about four. The covering epithelium contains multiple M cells [26]. There are also two large lymphoid patches located at the beginning of the large intestine. Each patch has the usual structure with multiple lymphoid follicles and a specialized overlying epithelium containing M cells. Rabbits possess two other sites containing organized lymphoid tissues. Thus, the sacculus rotundus is located at the ileocecal junction in a location consistent with the terminal ileal Peyer’s patch in other species. It is however much larger than any Peyer’s patch and contains large numbers of lymphoid follicles as well as M cells in the follicle-associated epithelium [27]. Rabbits also possess a prominent appendix at the distal end of the cecum that is also packed with lymphoid follicles [28]. The functions of these specialized lymphoid organs are discussed below. The appendix develops independently of the thymus.

22.5

Major histocompatibility complex

The rabbit MHC is also called the Rabbit Leukocyte Antigen (RLA) system [1]. It has been mapped to chromosome 12q1.1. Its size is greater than 2 mb. Its overall organization of class I-class III-class II is the same as seen in other mammals. As might be anticipated, the European rabbit appears to have originated in what is now the Iberian peninsula and as a result, MHC diversity is greatest among Spanish rabbits (Fig. 22.2).

22.5.1 The MHC Class I region The rabbit MHC class I region is relatively small and contains about eight to 14 genes per haplotype but only one (pR27), is transcribed in T cell lines and lymphoid tissues containing T cells. Studies of the rabbit MHC proteins have suggested that rabbits (actually a rabbit cell line) may express only a single class I antigen [29]. Analysis of the expressed protein encoded by this single class I gene shows high homology with both the introns and exons of human HLA class I genes. The gene encodes an alpha chain that associates with β2M. There are however significant differences in the size of the introns and the 30 sequences between rabbits and humans or mice. Rabbits also possess three other class I RLA gene clusters in their MHC, but their sequences suggest that while transcribed, they may not be expressed on cell membranes [30]. DP

DN

DM

DO

DQ

DR

BA

A

A B

B

B A

B B B B B A Class I 8-14 genes

II

III

FIGURE 22.2 The organization of the rabbit major histocompatibility complex gene locus.

I

346

SECTION | 2 Mammalian orders

22.5.2 The MHC class II region Eleven MHC class II genes have been identified in the rabbit genome. Rabbits possess one each of DQA and DQB, 1 DRA, 5 DRB, 2 DPA genes, and 1 DPB gene [30,31]. Their relative order is DP, DQ, and DR. The class II region may also contain genes homologous to DO and DN. All appear to be transcribed but there are differences in levels of expression, polymorphism, tissue distribution, and transcript size of their proteins [32]. As expected, the highest levels of DRα, DQα, and DPα are expressed in cells found within lymphoid tissues. The divergences in the DQA alleles appear to have preceded the divergence of the leporid genera.

22.5.3 The MHC class III region It has been claimed that in rabbits, the MHC class I genes are positioned close to the DR region without an intervening Class III region [33]. It is more likely however that the class III region is present but relatively small. Rogel-Gaillard et al. have shown that it too is located on 12q1.12 in close association with the Class I and II regions [34].

22.5.4 Natural killer cell receptors Given the difficulties encountered in identifying and characterizing rabbit NK cells, it is unsurprising that there is very little data available on NK cell receptor function in this species.

22.6

B cells and immunoglobulins

The European rabbit, Oryctolagus cuniculus, has the usual four classes of immunoglobulins; IgM, IgG, IgA, and IgE. They do not possess any exons encoding any IgD domains (Fig. 22.3). They have a single isotype each of IgM, IgG, and IgE but, most unusually, have fifteen IgA genes. The IGH locus is located on the q-telomeric region of rabbit chromosome 20 [35]. 372

5'

11

6

3’

IGH M

IGK

(D)

E

G

>100

A4 5

6 1 2 3 7 8

94

5

5 CN2

CN1 IGL

9 10 11 12 13 14 15

43 6 CO

M

D

G

E

A

V

D

J

LC

FIGURE 22.3 The organization of the rabbit immunoglobulin gene loci. Note that the remnant IgD gene lacks a CH1 domain hinge and has truncated CH2, CH3, and transmembrane exons. There are 13 functional IgA heavy chain genes but only one IgG gene. The empty boxes denote nonfunctional genes.

The Lagomorpha: rabbits, hares, and picas Chapter | 22

347

22.6.1 IGHM As in other mammals, transmembrane IgM is usually expressed first during B cell development. The IGHM exons are relatively well conserved with about 98% sequence similarity among the leporids. The Cμ1 exon is the most diverse while Cμ5 is the most conserved.

22.6.2 Immunoglobulin Heavy chains 22.6.2.1 IGHD The rabbit does not have a functional IGHD locus. However, about 21 kb downstream of the IGHM gene the remnants of IGHD have been found [36]. These remnants lack the Cδ1 and hinge exons but contain truncated Cδ2 and Cδ3 exons. The IGHD locus appears to have become nonfunctional prior to the divergence of rabbits and hares B14 mya.

22.6.2.2 IGHG Unusually for a mammal, the IgG heavy chain constant region is encoded by a single IGHG gene in the rabbit. Like the other heavy chain constant regions, its domain structure is relatively conserved with Cγ1 being most conserved while Cγ3 is least. Complement C1q binds to sites on the Cγ2 domain. Eleven Lepus species share exactly the same hinge motif suggesting that it has been positively selected.

22.6.2.3 IGHE The rabbit IGHE exons are also relatively well-conserved so the Cε2 and Cε3 domains are most conserved and Cε1, the least [37]. Like IgM, the most conserved domains are those that interact with the major IgE receptor, FcεRI on mast cells and basophils.

22.6.2.4 IGHA What is most unusual about rabbit immunoglobulins is that the European rabbit has at least 15 IGHA genes and this is therefore the most complex IgA system known in mammals [38]. Mice and most other mammals as described previously have a single IGHA gene, although hominoid primates, with the exception of the orangutan, have two that code for their IgA1 and A2 subclasses. These two IgA subclasses clearly arose as a result of an ancestral duplication. Thirteen of the fifteen rabbit IGHA genes are expressed and show differential tissue expression [39]. It appears that the repeated duplication of the rabbit IGHA genes likely began at least 35 mya since multiple IgA subclasses are also present in other leporid genera such as Silvilagus (Cottontail), Lepus (jackrabbit), Pentalagus (Amami rabbit) and most interestingly Ochotona (Pika) [38]. Presumably, they all have equally complex mucosal surface immune systems. Rabbit IGHA genes are clustered in a single subregion of the IGH locus spanning at least 100 kb and with intervals of 8
18 kb between each. All these IGHA genes have the same sequence orientation and switch sequences are located 50 of at least 12 of the 13 functional genes. The IGHA3 and IGHA8 genes have defective Ia promoter regions and consequently are not expressed in vivo. The first 13 IGHA chains sequenced show between 72% to 85% amino acid sequence identity. Sequence comparisons show that the Cα2 and Cα3 domains of these α chains are very highly conserved. As a result, their Fc regions can all activate complement and form bonds with secretory component. However, their Cα1 domains have double the number of variable sites and their hinge regions are highly diverse. This suggests that each of these Cα1 domains has evolved independently. As in most mammalian immunoglobulins, the rabbit immunoglobulin hinge regions are rich in proline, serine, and threonine. Most European rabbit IgA subclasses also contain an extended hinge (IgA2, 3, 4, 5, 6, 8, 10, 13, and 14). Unlike mouse and human IgA hinge regions, they also contain multiple cysteine residues. These range from one (IGHA1 and IGHA12) to five (IGHA8 and IGHA13). Thus, all the rabbit hinge regions can form the disulfide bonds needed for interchain binding. They also possess a cysteine at position 471, the penultimate amino acid that is required for binding to the J chain as well as for interchain bonding. However, IGHA13 does not bind secretory component covalently [37]. IGHA15 possesses a unique hinge region that is serine-rich with a sequence of nine consecutive serine residues and no proline or threonine [39]. It is believed that there are functional differences among the IgA isotypes. Given the extensive variability of their Cα1 and hinge regions, it is suggested that this is the part of the molecules where such functional differences would be expressed. One significant difference is probably susceptibility to bacterial proteolytic cleavage. Given the importance

348

SECTION | 2 Mammalian orders

of the gut microbiota in rabbits, the microbiota may have played a significant role in the evolution and persistence of the multiple IgA isotypes. Alternatively, the expansion of the IgA subclasses may be a mechanism to compensate for the lack of diversity among the IgG subclasses.

22.6.3 Locations The IgA heavy chain genes are differentially expressed [40]. They can be grouped into those that are consistently expressed at high levels including IGHA4, 5, 6, 9, 10, 12, and 13. While others (IGHA1, 2, 7, and 11) are expressed at very low levels. Two of them, IGHA3 and IGHA8, do not appear to be expressed at all [40]. At least ten of the IGHA genes are expressed in the small intestine as well as in the appendix, mesenteric lymph nodes, and mammary gland [40]. A high percentage of IgAproducing plasma cells in the duodenum and upper jejunum produce IgA4 whereas in the ileum the opposite is the case. (This is a similar pattern of expression to that seen with IgA1 and IgA2 in humans). The IgA subclass switching is likely determined by the local microenvironment. This might also reflect the presence of the microbiota and its proteases. Only IgA4 is expressed in all lymphoid tissues analyzed, although its level in the spleen is very low. IgA1, 6, 9, 10, and 12, are expressed in most tissues but not in the lungs, spleen, or tonsils. IgA5 is similar but is expressed in low levels in the lungs. In the submandibular salivary gland, only seven IgA isotypes are expressed at significant levels. In the lung and tonsil, only IgA4 is expressed. Analysis of Peyer’s patch transcripts also shows differences in the level of IGHA gene expression between different individuals and even between different Peyer’s patches in the same rabbit. For example, in some animals, most of the IGHA genes are expressed, but in others, only IGHA4 is expressed at a significant level. Since IGHA4 is the most 50 of the C alpha genes, it has been suggested that IgA-producing cells are derived from B cells that have initially undergone isotype switching to IgA4 and subsequently switched to one of the more 30 genes [40]. To investigate the reasons for these differences in tissue expression, the activities of the Iα promoters have been analyzed following stimulation of B cells with TGF-β. It has been found that the expression of IgA3 and IgA8 could not be induced. However, all isotypes are stimulated to a similar degree except that the Iα4 promoter expresses much higher levels of IgA4. This suggests that the B cell switch to producing a single IgA isotype is likely due to a second step following germline transcription rather than due to the activities of the Iα promoter [41]. The ability of lagomorphs to produce so many diverse IgA subclasses likely has a significant effect on the diversity of their microbiota. In some ways, these may function as innate IgAs that control microbial entrance, especially in the large rabbit cecum. IgA diversity may encourage diversification of the microbiota.

22.6.4 IGHV Analysis of the heavy chain immunoglobulin locus has shown, in addition to the constant region genes, that it contains at least 372 IGHV genes, 11 D genes, and six IGHJ genes [35,42]. Twenty-one of these IGHV genes appears to be pseudogenes. All the IGHV genes belong to a single-family. In addition, genes IGHV1 to IGHV10 have 98%
100% identity suggesting that gene conversion must be the major mechanism of generating V region diversity in this species. Despite the presence of this huge number of IGHV genes, the European rabbit uses the closest, IGHV1, the most 30 IGHV gene to the IGHD and IGHJ gene clusters, to generate 80%
90% of its antibody repertoire and this also encodes the VHa allotype [42]. This allotype group consists of three allotypes IGHV1-a1, -a2, and -a3. These IGHV1 allotypes are a result of sequence differences among the 16 amino acids encoded in FR1 and FR2. As rabbits age, based on their antigen exposure they may also use IGHV genes other than IGHV1. The same three allotypes are also present in snowshoe hare (Silvilagus ssp) populations—a different genus. All three belong to Clan III. Phylogenetic analysis and comparison with related human and mouse IGHV genes suggest that this polymorphism has been maintained for about 50 my. The specific reasons for its prolonged persistence are unclear [43]. The remaining 10%
20% of the rabbit immunoglobulin V domains are encoded by three VHn genes; IGHVx, IGHVy, and IGHVz whose location is uncertain but map at least 100 kb upstream from the IGH locus. Normal rabbits utilize both somatic mutation and templated mutations through gene conversion to diversify their rearranged heavy chain V, D, and light chain VJ genes. This occurs in both the appendix and in the gut-associated lymphoid tissue of young rabbits during their pre-immune immune response. The maintenance of this VHn usage at low frequency has been suggested to be a relic of an ancestral immune response to pathogens. The situation in hares (Lepus) is identical to that in rabbits. Thus, hares use the VHn genes to encode only 5%
10% of their VDJ rearrangements. The VHn genes may be an ancestral polymorphism that has persisted in the Leporid genome. Unlike humans and mice where immunoglobulin gene rearrangement usually starts with D-J joining and is subsequently followed by V-D-J recombination; rabbits may join V to D first before adding J [42].

The Lagomorpha: rabbits, hares, and picas Chapter | 22

349

22.6.5 Immunoglobulin Light Chains 22.6.5.1 Kappa chains Rabbit kappa chain genes are located on chromosome 2 [35]. Unusually, rabbits show a duplication of their kappa locus, IGK1, and IGK2. The IGK2 locus is normally poorly expressed since the intronic enhancer, and the surrounding matrix-associated region have been deleted [35]. The IGK2 region is also located on chromosome 2 but in the opposite transcriptional orientation (negative strand). As a result, rabbits can make two different kappa light chains. This seems to have resulted from a duplication since the kappa chain order is -Vκ-Cκ2-Jκ-Vκ-

and -Vκ-Jκ-Cκ1-

In the IGKC1 chain, there is also an unusual extra interdomain disulfide bond that connects either the IGKV or IGKJ gene segment directly to an extra cysteine in the IGKC1 domain. There are at least 94 different IGKV genes and pseudogenes located between IGKC1 and IGKC2. These sequences are therefore available to contribute to gene conversion. Some of these IGKV genes and IGK2 are located about two mb away from IGK1. In wild rabbits, IGK1 is preferentially expressed. Rabbit IGK chain genes also contain two clusters of five IGKJ segments. Only one of these appears to be functional [44]. This significantly decreases their available recombination potential. Nevertheless, sequencing has demonstrated significant heterogeneity in the size of the rabbit CDR3. This is not due to the presence of D segments. Instead, rabbit IGKV genes have up to ten additional nucleotides attached at their 30 ends. As a result of this N-region addition, high CDR3 diversity can be generated. This also increases their potential for junctional deletions.

22.6.5.2 Lambda chains About 10%
30% of the light chains used by rabbit immunoglobulins are lambda light chains. The genes encoding IGL chains are located on chromosome 21. The region contains 43 IGLV genes, of which about 20 are functional. There are four IGLJ genes, two of which are functional. There are also six functional IGLC genes of which only two appear to be expressed at any time [1].

22.6.6 The rabbit B cell antibody repertoire Mammals employ multiple strategies to generate their diverse antibody repertoire in developing B cells. In the case of rabbits, it is clear from the above discussion that simple combinatorial joining of the heavy and light VDJ genes in rabbits can only generate a limited diversity of antigen-binding V domains. The predominant use of the IGHV1 gene in heavy chains severely limits this diversity. As a result, antibody diversity in the rabbit is also generated by V gene rearrangements, somatic hypermutation, and gene conversion. It is also generated by base deletion and N-region addition.

22.6.7 Fetal liver and bone marrow As mammals develop in utero, the earliest sites of lymphocyte production are the fetal liver and the bone marrow. During this early development stage, the B cells undergo rearrangements of their IGH and IGL loci so that when they are born, they possess diverse surface antigen receptors—immunoglobulins. As noted above, while rabbits possess 372 IGHV genes and at least 137 light chain V genes, they use very few of them They preferentially rearrange only a few V genes so these newly generated B cells can recognize only a limited diversity of antigens. In addition, evidence suggests that bone marrow lymphopoiesis does not in fact persist throughout a rabbit’s lifetime. The bone marrow eventually stops producing B cells in adult rabbits. These animals then become dependent on intestinal lymphopoiesis to maintain their large and diverse B cell population [45]. There is, in effect, a two-stage process of B cell lymphopoiesis. If B cell production in the bone marrow is experimentally stopped, it does not resume spontaneously! The number of progenitor B cells declines progressively, and they may become undetectable in the marrow by 16 weeks. Studies have shown that the microenvironment within the bone marrow is effectively suppressive for B cell production in older rabbits while, at the same time, the red marrow is replaced by adipocytes. As a result, first, of their limited V region diversity and second, of a loss of marrow lymphopoiesis, B cells must look elsewhere for sites to develop and diversify. It appears that these sites are located in the gastrointestinal lymphoid tissues [45].

350

SECTION | 2 Mammalian orders

STEP 1

Initial diversification within the bone marrow

STEP 2

B cell recruitment to nascent follicles in the intestinal lymphoid organs

Diversification of primary STEP 3 antibody repertoire Somatic mutation and gene conversion STEP 4

Driven by signals from commensals

Intrafollicular B cell proliferation

FIGURE 22.4 The multistage process of generating immunoglobulin diversity in rabbits. After initial diversification in the bone marrow, a second stage occurs in the intestinal tract, specifically the rabbit appendix. This second stage is driven by-products generated by the intestinal microbiota.

22.7

Appendix

Since the B cells that emerge from the initial lymphopoiesis in the rabbit bone marrow have limited diversity, they must undergo a second diversity-generating step. This second step takes place in their well-developed intestinal lymphoid tissues, especially the appendix [28] (Fig. 22.4). Rabbits have large amounts of lymphoid tissues in the intestinal tract. These include not only their Peyer’s patches but also the sacculus rotundus and the appendix. All are important sites for B cell development. The B cells newly released from the bone marrow are trapped as they pass through the appendix, probably bound by specialized addressins such as the peripheral lymph node addressin (PLNAd) that binds to CD62L on the B cells. Interactions between chemokine receptors such as CCR7 on B cells and its ligand CCL21 in the appendix also result in activation of integrins such as LFA-1 and α4/β1 resulting in even firmer B cell binding [28]. Once lodged within the appendix, the B cells continue both development and diversification. This B cell development within the rabbit appendix is believed to be somewhat similar to the functions of the Bursa of Fabricius in birds [28]. The rabbit appendix has a similar structure to the Bursa—consisting of densely packed lymphoid follicles full of B cells. These follicles develop in two phases. There is an initial phase of B cell recruitment to nascent follicles. Then there is the second phase of intrafollicular B cell proliferation driven by the intestinal commensals [46]. By about 8 weeks of age, the rabbit immunoglobulin heavy chain gene segments, V, D, and J have already undergone limited somatic diversification. Surgical removal of the appendix, the sacculus rotundus, and the Peyer’s patches effectively reduces this diversification by about 90% at 10
12 weeks [47]. However, by 28 weeks of age, there was no difference in the degree of antibody diversity. This later diversification, it is suggested, probably occurred in the diffuse intestinal lymphoid aggregates that were not surgically removed. The occurrence of somatic hypermutation and gene conversion in their B cells has been detected in the appendices of rabbits 4
6 weeks of age [45]. Within the appendix and other intestinal lymphoid tissues, developing B cells are exposed to a diverse universe of microbial products generated by the commensal intestinal microbiota. It is believed that this complex mixture promotes B cell diversification and furthers the development of the pre-immune repertoire. The intestinal lymphoid tissues in conventional animals are in intimate contact with the billions of bacteria that make up the gut microbiota. It is clear that products derived from the microbiota are required for proper development not only of the gut-associated lymphoid tissue (GALT), but also of the antibody system itself. Germ-free, sterile rabbits fail to develop B cell follicles in their GALT. They only make small follicles and produce low levels of antibodies. In addition, surgical ligation of the appendix at birth prevents bacterial colonization and the development of B cell follicles. Thus, products from the intestinal bacteria, possibly acting through pattern recognition receptors activate the developing B cells and promote gene diversification. This diversification may also depend upon the presence of selected bacterial species such as Bacillus subtilis, Bacteroides fragilis, or certain Clostridia. Bacterial structural proteins such as lipoarabinomannans, lipopolysaccharides, some superantigens such as Staphylococcal protein A, or even metabolites such as fatty acids and retinoic acid may also influence the B cell diversification process [48]. Likewise endogenous stimulatory molecules from helper cells

The Lagomorpha: rabbits, hares, and picas Chapter | 22

351

such as IL-4, IL-13, or IL-33 or even complement components may also be required. Two very obvious cytokines that promote B cell development are BAFF (B cell-activating factor) and APRIL (a proliferation-inducing ligand) are upregulated by bacterial products and promote B cell proliferation in an autocrine fashion. Other stimuli may include CD40-CD40L and B7-CD28 interactions. Both gene conversion and somatic hypermutation of B cell immunoglobulin heavy and light chains continue to occur within the appendix. This is in addition to clonal expansion and diversification promoted by components from the intestinal microbiota [28].

22.8

Other mammals

Rabbits are not the only mammals to utilize the intestine and its microbiota as a stimulus for B cell development and diversification. A pre-immune repertoire also develops in sheep and cattle, and this is subsequently diversified and regulated in the intestinal lymphoid tissues in the same way that the thymus regulates T cell production and diversity. In these species, the Peyer’s patches may be considered to be primary lymphoid organs where developing B cells finish their development and expand their antigen-binding repertoire by somatic hypermutation before colonizing the secondary lymphoid organs. It is also important to reiterate that in germ-free animals (animals such as mice that are delivered under aseptic conditions by caesarian section and maintained in sterile isolators), their immune system fails to develop fully. While the diversification of B cells in the rabbit GALT is well established. The situation in other species, especially the artiodactyls is less clear. Thus, as discussed in Chapter 17, it has long been believed that the Peyer’s patches in the sheep serve as primary lymphoid organs where B cells undergo further diversification. However, in the pig, this does not appear to be the case since pigs still produce B cells and antibodies even after the surgical removal of their Peyer’s patches [49]. The significance of the microbiota is readily seen when germ-free mice are reared under aseptic conditions. Their intestinal lymphoid tissues fail to grow, and the germ-free animals are hypogammaglobulinemic and effectively immunosuppressed (Fig. 5.1). They fail to develop a functioning B cell system or may develop an imbalanced system where type 2 immune responses predominate. Germ-free animals develop elevated levels of IL-4, increased basophil numbers, and high levels of IgE. As a result, these mice are very much skewed toward a Th2 phenotype. They can develop severe allergies to food or airborne antigens. This Th2 skewing can be prevented by colonizing their intestines with a normally mixed microbiota or some selected bacteria from conventional animals. However, this colonization needs to be performed within 3 weeks of birth to be fully effective. The key to successful accommodation with the microbiota, therefore, depends on the regulation of immune responses in the gut wall. Intestinal helper T cell phenotypes are “plastic” and Th0 precursor cells can differentiate into many other types of helper T cells. This differentiation is regulated by signals from two major sources, the microbiota and other tissue cells such as endothelial cells. In effect, the microbiota manages these helper T cell populations with the help of cytokines. Local immune stability is achieved by adjusting T cell populations so that a balance is maintained between each of the pro-inflammatory helper T cell types (Th1, Th2, and Th17) and anti-inflammatory Treg cells.

22.9

T cells and cell-mediated immunity

As in other mammals, rabbit T cells employ both α/β and γ/δ antigen receptors. Rabbits are considered a γ/δ-high species since γ/δ TCRs may be found on 20%
50% of their circulating T cells (Fig. 8.9) [50]. Neonatal rabbits have relatively few CD81T cells in their bloodstream and spleen (6.8%). However, as they develop, the CD41/CD81 ratio declines as the numbers of CD81 cells increase to reach about 10% in adults, presumably as a result of antigenic stimulation. The number of double-positive cells averages about 2% of T cells irrespective of age. The number of circulating B cells also increases in the blood as well as in mesenteric lymph nodes. Absolute numbers of lymphocytes also increase from about 1.04 x109/l to 3.45x109/l in adults [51]. As in other mammals, the four rabbit T cell receptor peptide chains are encoded by three gene loci, TRA/D, TRB, and TRG.

22.9.1 TRA/D The TRA/D locus encoding both alpha and delta chains is situated on rabbit chromosome 17. It has the usual mammalian organization. Rabbits have a unique TRAC gene containing three exons as well as a unique TRDC gene containing four exons. The two TRDD genes are flanked by functional recombination signal sequences. There are also three functional TRDJ genes and a cluster of 61 TRAJ genes located between the usual downstream inverted TRDV3 and the TRAC gene (Fig. 22.5).

352

SECTION | 2 Mammalian orders

TRA/D

TRB

5'

61 AV

4 DV

77

6

2 3 DD DJ DC

22

3'

AC

6 C1

TRG

61 AJ

C2

2

V

D

J

C

FIGURE 22.5 The organization of the rabbit T cell antigen receptor gene loci. Arrows denote transcriptional orientation.

The TRA locus contains 61 TRAV genes with 48 functional genes, two ORF, and eleven pseudogenes. They fall into 31 subfamilies. Of these, 20 functional V genes and seven pseudogenes are conserved between humans and rabbits [52]. An additional four TRDV genes are present; TRDV 1, 2, and 4 are located within the TRAV cluster while the TRDV3 gene is positioned downstream of TRAC and in inverted orientation. TRDV4 appears to be a pseudogene. Compared to humans, the 50 terminus of the rabbit TRA/D locus is missing five V genes, TRAV1, 2, 3, 6, and 7. The mouse has also lost four of these genes but has retained TRAV1. All the carnivores including the dog also appear to have lost TRAV1. TRAV1 is an important gene since it encodes an α chain that plays a key role in the functioning of invariant T cells [52]. As a result, rabbit lymphocytes do not carry any of the invariant TRAV-TRAJ receptor rearrangements that are characteristic of mucosal-associated invariant T cells (MAIT) or natural killer T (NKT) cells [52]. NKT cells are also not detectable in rabbit blood. Thus, rabbits appear to lack the populations of invariant T cells that are found in other species. These invariant T cells have receptors that can bind unstable pyrimidine adducts derived from 5-A-RU (5Amino-6-(D-ribitylamino)uracil), a precursor of bacterial riboflavin, that are usually presented by MR1 (MAIT) cells in addition to the alpha-glycosyl ceramides presented by CD1d (NKT) cells. The TRAV1 gene that encodes the evolutionary conserved invariant TRA chain of MAITs is not only missing in the rabbit but is also absent in carnivores and the armadillo. The MAIT restricting MHC class I-like MR1 is also absent from these species and lagomorphs. However, the MHC class1-like molecule, CD1d that normally restricts NKT cells is present in the rabbit [52].

22.9.2 TRB The rabbit beta-chain locus, TRB has yet to be assigned a chromosome location. It spans B600 kb and contains 77 TRBV genes, two TRBD genes, 12 TRBJ genes, and two TRBC genes. The TRBV genes are located upstream of two D-J-C clusters in tandem followed by one TRBV gene with an inverted transcriptional orientation at the 30 end. The 77 TRBV genes are arranged in 24 subfamilies and 16 of these genes are pseudogenes [1].

22.9.3 TCRG The rabbit gamma chain locus (TRG) is located on chromosome 10 and spans B70 kb. It contains 22 TRGV, two TRGJ, and one TRGC gene [53]. The TRGV genes belong to two different subfamilies. All except one belong to the TRGV1 subfamily. The remaining V gene belongs to the TRGV2 subfamily. They appear to have arisen as a result of repeated duplication. There is evidence of a gene conversion event occurring between TRGV1.1 and TRGV1.6. All the TRGV genes appear to be functional. The two TRGJ genes are about 4 kb apart and are located about 4 kb upstream of a single TRGC gene. Diversity is generated at the VJ junctions together with N-region addition as well as some deletions. Like humans but unlike mice, TRGV gene usage is similar in all tissues studied. However, as rabbits age, there is more frequent usage of TRGJ2, greater junctional diversity, and expression of TRGV2 in their γ/δ receptors [53].

The Lagomorpha: rabbits, hares, and picas Chapter | 22

353

References [1] Pinheiro A, Neves F, Lemos de Matos A, Abrantes J, et al. An overview of the lagomorph immune system and its genetic diversity. Immunogenetics 2015;. Available from: https://doi.org/10.1007/s00251-015-0868-8. [2] Lacher TE, Murphy WJ, Rogan J, Smith AT, Upham NS. Evolution, phylogeny, ecology and conservation of the clade Glires. In: Lagomorpha and Rodentia. Handbook of the mammals of the world, vol. 6. Lynx Edicions; 2020. p. 15
26. [3] Furukawa S, Kuroda Y, Sugiyama A. A comparison of the histologic structure of the placenta in experimental animals. J Toxicol Pathol 2014;27:11
18. [4] Borghesi J, Mario LC, Rodrigues MN, Favaron PO, Miglino MA. Immunoglobulin transport during gestation in domestic animals and humans—a review. Open J Anim Sci 2014;4:323
36. [5] DeSesso JM, Williams AL, Ahuja A, Bowman CJ, Hurtt ME. The placenta transfer of immunoglobulins, and safety assessment of biopharmaceuticals in pregnancy. Crit Rev Toxicol 2012;42(3):185
210. [6] Szikora B, Hiripi L, Bender B, Kacskovics I, Ilias A. Characterization of the interactions of rabbit neonatal Fc receptor (FcRn) with rabbit and human IgG isotypes. PLoS One 2017;. Available from: https://doi.org/10.1371/journal.pone.0185662. [7] Jeklova E, Leva L, Faldyna M. Lymphoid organ development in rabbits: major lymphocyte subsets. Dev Comp Immunol 2007;31:632
44. [8] Jenness R. Lactational performance in various mammalian species. J Dairy Sci 1986;69:869
85. [9] Peri BA, Theodore CM, Losonsky GA, Fishaut JM, et al. Antibody content of rabbit milk and serum following inhalation or ingestion of respiratory syncytial virus and bovine serum albumin. Clin Exp Immunol 1982;48:91
101. [10] Siegel A, Walton RM. Hematology and biochemistry of small mammals Chapter 39. Ferrets, rabbits, and rodents. Elsevier Inc; 2020. p. 569
82. [11] Hawkey CM. Comparative mammalian hematology. Cellular components and blood coagulation of captive wild animals. London: Heinemann Medical Books; 1975. [12] Neave MJ, Hall RN, Huang N, McCall KA, et al. Robust innate immunity in young rabbits mediates resistance to rabbit hemorrhagic disease caused by Lagovirus Europaeus G1.1 but not G1.2. Viruses 2018;. Available from: https://doi.org/10.3390/v10090512. [13] Neves F, Agueda-Pinto A, Pinheiro A, Abrantes J, Esteves PJ. Strong selection of the TLR2 coding region among the lagomorpha suggests an evolutionary history that differs from other mammals. Immunogenetics 2019;71:437
43. [14] Perkins HD, van Leeuwen BH, Hardy CM, Kerr PJ. The complete cDNA sequences of IL-2, IL-4, IL-6, and IL-10 from the European rabbit (Oryctolagus cuniculus). Cytokine 2000;12:555
65. [15] Shibata K, Nomiyama H, Yoshie O, Tanase S. Genome diversification mechanism of rodent and lagomorpha chemokine genes. Biomed Res Intern 2013;2013:856265. Available from: https://doi.org/10.1155/2013/856265. [16] Neves F, Abrantes J, Lopes AM, Fusinatto LA. Evolution of CCL16 in Glires (Rodentia and lagomorpha) shows an unusual random pseudogenization pattern. BMC Evol Biol 2019;19:59. Available from: https://doi.org/10.1186/s12862-019-1390-7. [17] Ganz T, Rayner JR, Valore EV, Tumolo A, et al. The structure of the rabbit macrophage defensin genes and their organ-specific expression. J Immunol 1989;143(4):1358
65. [18] Lehrer RJ, Daher K, Ganz T, Selsted ME. Direct inactivation of viruses by MCP-1 and MCP-2, natural peptide antibiotics from rabbit leukocytes. J Virol 1985;54(2):467
72. [19] Laybourn KA, Hiserodt JC, Hunt JM, Sell S, Strayer DS. Natural cell-mediated immunity in the rabbit. Cell Immunol 1990;129:519
27. [20] Mo Z, Lu H, Mo S, Fu X, et al. Ultrasound-guided radiofrequency ablation enhances natural killer-mediated antitumor immunity against liver cancer. Oncol Lett 2018;15:7014
20. [21] Kostro K, Wojicika-Lorenowicz R, Glinski Z. Acute phase proteins in rabbits. Medy Weter 2003;59(10):864
7. [22] Gaber W. Morphogenesis of the thymus in rabbit during prenatal and early postnatal periods. Open Access text 2017;. Available from: https:// doi.org/10.15761/HCE.1000118. ´ gueda-Pinto A, Alves LQ, Neves F, McFadden G, et al. Convergent loss of the necroptosis pathway in disparate mammalian lineages shapes [23] A virus countermeasures. Front Immunol 2021;. Available from: https://doi.org/10.3389/fimmu.2021.747737. [24] Udroiu I, Sgura A. The phylogeny of the spleen. Quart Rev Biol 2017;92(4):411
43. [25] Casteleyn C, Breugelmans S, Simoens P, Van, den Broeck W. The tonsils revisited: review of the anatomical localization and histologic characterization of the tonsils of domestic and laboratory animals. J Immunol Res 2011. Available from: https://doi.org/10.1155/2011/472-460. [26] Kararli TT. Comparison of the gastrointestinal anatomy, physiology, and biochemistry of humans and commonly used laboratory animals. Biopharm Drug Dispos 1995;16:351
80. [27] Beyaz F, Ergu¨n E, Bayraktaroglu AG, Ergun L. The identification of intestinal M cells in the sacculus rotundus and appendix of the Angora rabbit. Vet Res Commun 2010;34:255
65. [28] Mage RG, Lanning D, Knight Kl. B cell and antibody development in rabbits: the requirement of gut associated lymphoid tissues. Dev Comp Immunol 2006;30:137
53. [29] Marche PN, Tykocinski ML, Max EE, Kindt TJ. Structure of a functional rabbit class I MHC gene with similarity to human class I genes. Immunogenetics 1985;21:71
82. [30] Marche PN, Rebiere MC, Leguern C, Kindt TJ. Structure and expression of rabbit MHC genes. Annu Rev Vet 1989;20(3):361
5. [31] Spieker-Polet H, Sittisombut N, Yam P-C, Knight KL. Rabbit major histocompatibility complex. IV. Expression of major histocompatibility complex class II genes. J Immunogenet 1990;117:123
32.

354

SECTION | 2 Mammalian orders

[32] Kulaga H, Sogn JA, Weossman JD, Marche PN, et al. Expression patterns of MHC classss II genes in rabbit tissues indicate close homology to human counterparts. J Immunol 1987;139:587
92. [33] Chouchane L, Kindt TJ. Mapping of the rabbit MHC reveals that class I genes are adjacent to the DR subregion and defines an insertion/deletion-related polymorphism in the class II region. J Immunol 1992;149(4):1216
22. [34] Rogel-Gaillard C, Piumi F, Billault A, Bourgeaux N. Construction of a bacterial artificiak chromosome (BAC) library: application to the mapping of the major histocompatibility complex to position 12q1.2. Mammal Genome 2001;12:253
5. [35] Gertz EM, Scha¨ffer AA, Agarwala R, Bonnet-Garnier A, et al. Accuracy and coverage assessment of Oryctolagus cuniculus (rabbit) genes encoding immunoglobulins in the whole genome sequence assembly (OryCun 2.0) and localization of the IGH locus to chromosome 20. Immunogenetics 2013;65:749
62. [36] Lanning DK, Esteves PJ, Knight KL. The remnant of the European rabbit (Oryctolagus cuniculus) IgD gene. PLoS One 2017;. Available from: https://doi.org/10.1371/journal.pone.0182029. [37] Pinheiro A, Almeida T, Esteves PJ. Genetic diversity of IGHM and IGHE in the Leporids revealed different patterns of diversity in the two European rabbit subspecies, Cuniculus algirus and O. c. cuniculus. Animals 2019;. Available from: https://doi.org/10.3390/ani9110955. [38] Burnett RC, Hanly WC, Zhai SK, Knight KL. The IgA heavy chain gene family in rabbit: cloning and sequence analysis of 13 Cα genes. EMBO J 1989;8(13):4041
7. [39] Pinheiro A, Sousa-Pereira S, Strive T, Knight KL, et al. Identification of a new rabbit IgA with a serine rich hinge region. PLoS One 2018;. Available from: https://doi.org/10.1371/journal.pone.0201567. [40] Spieker-Polet H, Yam P-C, Knight KL. Differential expression of 13 IgA-heavy chain genes in rabbit lymphoid tissues. J Immunol 1993;150 (12):5457
65. [41] Spieker-Polet H, Yam P-C, Knight KL. Functional analysis of Ia promoter regions of multiple IgA heavy chain genes. J Immunol 2002;168:3360
8. [42] Ros F, Puels J, Reichenberger N, Van Schooten W, et al. Sequence analysis of 0.5 Mb of the rabbit germline immunoglobulin heavy chain locus. Gene 2004;330:49
59. [43] Su C, Nei M. Fifty-million-year-old polymorphism at an immunoglobulin variable region gene locus in the rabbit evolutionary lineage. Proc Natl Acad Sci USA 1999;96:710
9715. [44] Heidmann O, Rougeon F. Immunoglobulin κ light chain diversity in rabbit is based on the 30 length heterogeneity of germ-line variable genes. Nature 1984;311:74
6. [45] Severson KM, Knight KL. Generation of the antibody repertoire in rabbits: role of gut-associated lymphoid tissues. In: Kaushik AJ, Pasman Y, editors. Comparative immunoglobulin genetics. Toronto: Apple Academic Press; 2014. [46] Hanson NB, Lanning DK. Microbial induction of B and T cell areas in rabbit appendix. Dev Comp Immunol 2008;32:980
91. [47] Vajdy M, Sethupathi P, Knight KC. Dependence of antibody somatic diversification on gut-associated lymphoid tissue in rabbits. J Immunol 1998;160:2725
9. [48] Rhee K-J, Jasper PJ, Sethupathi P, Shanmugam M, et al. Positive selection of peripheral B cell repertoire in gut-associated lymphoid tissues. J Exp Med 2005;201(1):55
62. [49] Butler JE, Sinkora M. The enigma of the lower gut-associated lymphoid tissue (GALT). J Leukocyte Biol 2013;94:259
70. [50] Porcelli SA, Morita CT, Modlin RL. T-cell recognition of non-peptide antigens. Curr Opin Immunol 1996;8(4):510
16. [51] Jeklova E, Leva L, Kudlackova H, Faldyna M. Functional development of immune responses in rabbits. Vet Immunol Immunopathol 2007;118:221
8. [52] Mondot S, Lantz O, Lefranc M-P, Boudinot P. The T cell receptor (TRA) locus in the rabbit (Oryctolagus cuniculus): genomic features and consequences for invariant T cells. Eur J Immunol 2019;49:2146
58. [53] Cho KS, Zhai S-K, Esteves PJ, Knight KL. Characterization of the T-cell receptor gamma locus and analysis of the variable gene segment expression in rabbit. Immunogenetics 2005;57:352
63.

Chapter 23

The rodents: mice, rats, and their relatives

Indian palm squirrel. Funambulus palmarum.

Like the lagomorphs, rodents are members of the clade Glires and probably first emerged around 70 mya. The first rodent fossils in Eurasia and North America are dated between 55 and 57 mya. These early rodents had a skull morphology similar to modern beavers and squirrels. The order soon diversified into the Myomorpha (modern rats and mice), and the Sciuromorpha (modern squirrels, marmots, and chipmunks). A third suborder, the Ctenohystrica (Hystricomorphs), (porcupines, cavies, and agouti) first appeared in the Eocene, 40
50 mya. The modern families that make up the Myomorpha appeared about 25 mya in the late Oligocene and diversified rapidly. They prospered under the dry conditions of the Miocene, 20
10 mya when grasslands spread, and conditions favored these small herbivorous animals. There is probably more known and published about the immunology of some rodents than any other mammals. However, the vast majority have never been investigated from an immunological perspective. There are about 2400 different rodent species, and they account for about 45% of all living mammalian species. There are 34 living families of rodents. The order Rodentia is conveniently subdivided into three suborders based on their skull morphology. The Sciuromorpha; the Myomorpha; and the Ctenohystrica (Fig. 23.1) [1]. Rodents are important pests being responsible for significant crop losses as well as serving as vectors of infectious diseases such as the plague, leptospirosis, Lyme disease, hantaviruses, and murine typhus. It is obvious too, that the habitats occupied by small rodents such as mice are different from those encountered by large mammals including humans. Not only that but among the rodents, there are massive size differences as well as an enormous difference in lifespan—up to 50-fold. Larger mammals must defend themselves against microbial invasion for very much longer and immunological memory in turn becomes critical to living a longer lifespan. However, there are exceptions as seen in the naked mole-rat [2]. Another critical feature is the need for a defense against somatic mutations that can result in the development of cancers. These defenses must be very much more effective in larger, longer-lived mammals.

Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00005-8 © 2023 Elsevier Inc. All rights reserved.

355

356

SECTION | 2 Mammalian orders

80

70

60

50

40

30

20

10

CTENOHYSTRICA

0 Million years ago Guinea pigs Naked mole rats Porcupines

SCIUROMORPHA

Squirrels Rats Mice

MYOMORPHA Deer mice Hamsters CASTORIMORPHA

Beavers

FIGURE 23.1 A simple phylogeny of the rodents mentioned in the text.

23.1

Wild rodents versus laboratory rodents

Multiple studies have compared the immune systems of laboratory and wild mice—they are very different. Laboratory mice tend to be genetically homogeneous and protected against many of the pathogens that mice encounter in the wild. Indeed, wild mice carry a relatively large burden of infections. They may also carry intestinal nematodes (Syphacia spp) as well as ectoparasites such as mites and ticks. These differences are reflected in the functioning of their immune systems (Fig. 23.2). As a result of their exposure to infectious agents, wild mice have serum concentrations of IgG that can be 20-fold higher than laboratory mice. Because of their ectoparasite burden, they may have a 200-fold increase in IgE levels [3]. The spleens of wild mice are also much smaller than those of laboratory mice and contain many fewer mononuclear cells. Flow cytometry indicates that all the cell populations engaged in the adaptive immune systems are significantly reduced in numbers in wild mice. On the other hand, studies on the spleen cell populations in mice immunized with sheep red cells or keyhole limpet hemocyanin showed that the wild mice produce higher antibody titers against both antigens. Complement-mediated hemolysis is also greater in wild animals. Studies on the spleen cell populations of these animals showed that wild mice had proportionately more CD4 1 T cells and that their CD4 1 T cells, B cells, dendritic cells, macrophages, and natural killer (NK) cells had a more activated phenotype. However, the stimulated spleen cells of both wild and laboratory mice produced equivalent amounts of IFN-γ [4]. Wild mice also possess a population of highly activated myeloid cells termed hypergranulocytic myeloid cells [3]. There are also differences in the quantity of cytokines produced by T cells in wild and laboratory mice. Laboratory mice produce more IL-1, -4, -6, -10, -12p40, and IL-13. This reduction in cytokine production by wild mice may reflect an effort to maintain immune homeostasis. Thus several of these cytokines trigger sickness behaviors such as fever and lethargy. Wild animals must of necessity hide any such signs of sickness and vulnerability to avoid attracting predators. Thus in wild mice, cellular immune systems appear to be in a highly active state. Given that they likely encounter many more potential pathogens than their counterparts housed in clean laboratory animal facilities, this is unsurprising. There are also interesting interactions with other factors that determine the immune status of wild mice [5]. Thus like humans, mice within a specific sampling site tend to be more similar immunologically to each other than to mice at other distant sites, perhaps reflecting the local nature of their microbial challenges. In addition, the status of both an animal’s innate and adaptive immune responses is highly positively correlated with their body condition score and negatively correlated with their age. These two factors are interconnected, especially in females. Body condition scores are heavily dependent on available food supplies and stored fat reserves. Similar studies have been performed on another species of wild rodent, the field vole (Microtus agrestis) [6]. These studies have demonstrated significant seasonality in blood lymphocyte, neutrophil, and monocyte counts with lower numbers during the winter but significant peaks in neutrophil and monocyte counts in the spring. Animals in poorer body conditions also have low lymphocyte, neutrophil, and monocyte counts implying declines in immune and inflammatory functions. These hematological parameters are also adversely affected by high population densities [6].

The rodents: mice, rats, and their relatives Chapter | 23

WILD MICE VS Smaller spleens Lower white cell numbers ( T, B and NK cells) Higher T:B cell ratio Reduced cytokine responses to PAMPs Unusual myeloid cells (hypergranulocytic myeloid cells) Activated NK cells

23.2

LABORATORY MICE

357

FIGURE 23.2 Immunity in wild and laboratory mice compared [3,4].

Higher concentrations of serum proteins Elevated antibody responses Higher immunoglobulin levels (IgG, IgE) Higher Acute Phase Proteins (CRP, haptoglobin)

Myomorpha (rats and mice)

The Myomorpha are divided into two superfamilies, the Dipodoidea which contains over 50 species of small to medium-sized rodents that include the jumping mice such as jerboas; and the Muroidea, an enormous superfamily that contains at least 1750 species. It includes mice, rats, voles, hamsters, and gerbils among others. These two superfamilies collectively account for about two-thirds of all rodent species.

23.3

Reproduction and lactation

Muroid rodents are classical r strategists with high fecundity, short gestation periods, altricial young, and large litter sizes. They are adaptable and have high survivability despite high predation rates. The common brown rat (Rattus norvegicus) can reproduce up to 15 times a year while their offspring can reach sexual maturity and begin to breed by 3
4 months of age. Their longest recorded lifespans are four years for mice and five years for rats [7]. Other rodents, on the other hand, are very different. The Eastern Gray squirrel (Sciurus carolinensis) has a maximum lifetime of 24 years. Likewise, guinea pigs give birth to only one or two very precocious young but may live for up to seven years. Rats and mice have a hemotrichorial and discoid type of placenta [8]. Thus three trophoblast layers separate the maternal blood from the fetal blood vessels. It contains a relatively persistent yolk sac that degrades in mid-gestation so that its interior is exposed to the intrauterine cavity. Hematopoiesis begins in the fetal mouse liver by day 10 of gestation and the spleen by day 13. The mouse thymus develops from the third and fourth thymic pouches on day 10 and is colonized by hematopoietic stem cells the next day. Immigration of these stem cells into the bone marrow occurs on gestational day 17.5 and it rapidly assumes the role of a primary hematopoietic organ [9]. B cells with immunoglobulin rearrangements are found in the liver by day 11, but cells with surface immunoglobulins are not detectable until day 18. The gestation period of the house mouse is 20 days. Because of this very short gestation period, their young are very altricial and mouse lymphoid organs such as the spleen, Peyer’s patches, and lymph nodes are not colonized by T or B lymphocytes until after birth. IgM and IgA cannot cross the placental barrier in rats and mice whereas IgG does by binding to and transporting by FcRn. By three days after birth, the main immunoglobulin in both rat plasma and milk is IgG. It accounts for about 76% of the total serum immunoglobulins. IgA accounts for 15% while IgM constitutes 9%. Over the first three weeks of lactation, the proportion of IgA in the milk gradually increases to 30% while IgG drops to 66% and IgM drops to 4% [10]. Total plasma immunoglobulins stay relatively constant in the range of 1000
1500 μg/mL during this period [11].

23.4

Hematology

Unlike many other mammals such as humans, in which neutrophils are the predominant blood leukocyte, mouse blood and bone marrow leukocytes consist predominantly of lymphocytes [12,13]. In general, the typical white blood cell count in mice is highly variable ranging from 2000 to 10,000 per μL. These numbers vary depending upon the animal’s gender, wildness, and time of day. Lymphocytes account for 70%
80% of these white cells. They are 10
15 μm in diameter with a typical scanty cytoplasm and a round or indented central nucleus. However, even in healthy mice, their morphology varies, and natural killer cells tend to be larger with multiple azurophilic cytoplasmic granules [13]. Neutrophils account for 20%
30% of mouse blood leukocytes. Their half-life in the bloodstream is 8
11 hours. They have a pale granular cytoplasm with very faint pink granules and a complex segmented nucleus with uneven chromatin staining. Increased neutrophil counts occur as a result of excitement or other stresses, in addition to infections. Unlike many other mammals, mouse granulocytes do not contain alkaline phosphatase [14].

358

SECTION | 2 Mammalian orders

Monocytes are the largest of the white cells and usually make up less than 2% of the total white count. Their halflife in the bloodstream is between 24 and 60 hours. In mice, about 40% are in the circulating pool while the remainder is in the marginating pool. While round or horseshoe-shaped nuclei predominate, they may also contain ring-shaped nuclei. Their extensive cytoplasm may be vacuolated in addition to containing granules. Eosinophils, account for up to 7% of the white cell count depending upon their parasite burden. They have a bandshaped nucleus and contain uniform large round orange-staining granules [14]. Basophils are uncommon in mice [13]. There are significant differences in the leukocyte populations between different inbred mouse strains [15]. For example, total neutrophils and splenic macrophages are significantly higher in C57BL/6NCr mice, whereas total B cells are lower in female mice of the same strain. In this same strain, NK cell numbers are relatively higher in both the bone marrow and spleen. Splenic CD41 T reg cells and splenic CD81 T cells are reduced in male BALB/c mice when compared to females [15].

23.5

Innate immunity

Muroid rodents account for more than 10% of existing mammal species. This diversity is a result of recent radiation beginning around 12 mya. These repeated adaptive radiations and recurring colonization of new habitats have resulted in enormous phylogenetic diversity [16]. Analysis of their genomes has demonstrated that among their pervasive positively selected gene families are those associated with the immune systems, especially innate immunity. Their adaptive immune system is also subjected to constant pressure from rapidly evolving pathogens that have much shorter generation times than their hosts. Muroid rodents interact with diverse pathogens, both endemic and introduced, and responses to these pressures require rapid adaptation to pathogen threats [16].

23.5.1 Pattern recognition receptors Mice possess twelve different toll-like receptors. TLR1, 2, 4, 5, 6, and 11 are primarily expressed on the plasma membranes of sentinel cells. They recognize microbial components such as lipids, polysaccharides, and proteins such as flagellin. TLR12 plays a role in the defense of the urogenital system. The other TLRs, TLR3, 7, 8, and 9 are expressed on the membranes of endosomes and primarily recognize microbial nucleic acids. Until recently TLR8 was considered to be inactive in mice but this has proven to be incorrect. It is triggered by ligation of bacterial RNA released from phagosomal vacuoles and in response induces both NF-κB- dependent cytokines as well as the production of type I IFNs such as IFN-β [17]. Resistance to influenza in mice is controlled by alleles of the Mx gene locus. The Mx1 allele is present in wild mice as well as some laboratory strains. However, most laboratory mice are susceptible since they possess an Mx gene with either a large deletion or a missense mutation [18]. Among the many differences in TLR expression between humans and mice include such details as TLR2 expression on human peripheral blood leukocytes but not on T cells is constitutive. In contrast, in mice, TLR2 expression on leukocytes is low including T cells. TLR3 in humans is induced by LPS but not in mice; TLR9 in humans is expressed only on B cells, pDCs, and neutrophils, whereas it is expressed on all myeloid cells in mice; TLR10 in humans is expressed on many different human cells while it is a pseudogene in mice [12].

23.5.2 Chemokines As in other mammals, not all the enzymes or immune mediators are present in all species. For example, mice lack caspase 10. They also possess proteins not found in other species. This is especially the case in the large family of chemokines and their receptors. For example, CCL6, CCL9, lungkine (CXCL15), and MCP-5 (CCL12) have been identified in mice but not in humans. Conversely, CXCL8 (IL-8), CXCL7, CXCL11, CCL13, CCL18, CCL23, and eotaxin 2/3 (CCL24/CCL26) have been identified in humans but not in mice [12,19].

23.5.3 Antibacterial peptides As discussed in previous chapters, the antibacterial peptides, the defensins, have a long and important evolutionary history. They are major players in the innate defenses of many mammals. Neutrophils are the prime source of defensins in humans and most other mammals, and are largely responsible for the oxygen-independent antibacterial functions of these cells. The defensins are stored within the neutrophil granules and delivered to the phagolysosomes as required.

The rodents: mice, rats, and their relatives Chapter | 23

359

They comprise 30%
50% of the total weight of human neutrophil granules. Studies on neutrophils from multiple mouse strains have failed to detect any significant defensin content [20]. Mouse neutrophils do not contain defensins [12]. They are not lacking in these peptides, however. The Paneth cells of the small intestine are a major source of defensins in the mouse, producing more than twenty different ones. Presumably, they play a role in regulating the intestinal microbiota [12]. Human Paneth cells produce only two defensins.

23.5.4 Acute-phase responses The major acute-phase proteins in mice include C-reactive protein, haptoglobin, hemopexin, serum amyloid P, apolipoprotein A-1, and serum amyloid A [21]. Other proteins whose levels are affected by acute inflammation include α1antitrypsin, α1 acid glycoprotein, α2 macroglobulins, and transferrin [22]. In addition, apolipoproteins A-IV and A-V are positive acute-phase proteins whose levels in serum increase during infection whereas apolipoprotein A-II is a negative acute-phase protein in mice. These apolipoproteins are believed to regulate lipid metabolism during the acute-phase response [23].

23.6

Lymphoid organs

23.6.1 Thymus While the structure of the rodent thymus is similar to that of other species, it may have some morphologic peculiarities. For example, there are two types of Hassall’s corpuscles in the guinea pig thymus. One type is conventional and develops in the normal manner when reticular cells accumulate concentrically around a central blood vessel. The second type differs from the conventional type in its wall, its contents, and how it degenerates. Its functional significance is unclear [24].

23.6.2 Spleen The mouse spleen is a major hematopoietic organ in contrast to most other mammals. Hematopoietic progenitor cells are detectable in the fetal mouse spleen from day 13 post-conception [9]. However, splenic hematopoiesis ceases soon after birth and only restarts at about six months of age. Splenic extramedullary hematopoiesis accounts for about 30% of red cell production. It continues throughout adulthood. As a result, splenomegaly is a common consequence of this increased cell production. It is suggested that the spleen is the primary responding tissue when increased red cell production is needed whereas the bone marrow is the main source of the maintenance level of blood cells [25]. Stem cells (CD341, c-kit1) are highly enriched in the mouse spleen but found at low frequency. It is likely that the bone marrow and spleen stem cells may readily interchange via the bloodstream [25]. The stem cells in the spleen also differ from the bone marrow pool because they contain no osteoblasts. It should however, be pointed out that there is no evidence that lymphopoiesis occurs in the spleen of rodents after birth [9]. Rodent and rabbit spleens also lack ellipsoids. As a result, blood-borne particulates such as immune complexes are captured in the marginal zone of the spleen with subsequent transport to the germinal centers.

23.6.3 Mucosal tissues Muroid rodents appear to lack tonsils. Instead, they have well-developed nasal lymphoid tissue located at the bottom of the ventral nasal meatus and the nasopharyngeal meatus. It consists of lymphocytic aggregations with lymphoid follicles [26]. Mice do however, have significant amounts of bronchus-associated lymphoid tissue. It has been suggested that this reflects a much greater inhalation of dust particles, both organic and inorganic, in animals whose nostrils are very close to the ground [12]. Rodents are also obligatory nasal breathers. Mice possess 6
12 Peyer’s patches of variable size, distributed along the small intestine [27]. Small patches may contain only two or three lymphoid nodules while the largest patches may contain up to nine. Each can be up to 3 mm in length. These patches are covered by an epithelial layer consisting of mixed epithelial cells and M cells.

23.7

Major histocompatibility complex

The mouse genome contains about 20,210 genes compared to 19,052 in humans. Of these just over 15,000 of the genes in both species are functionally related. Presumably, they would have been present in the common ancestor of rodents and primates about 90 mya. Thus about 75% of mouse genes are orthologs. The number of duplicated genes in the

360

SECTION | 2 Mammalian orders

mouse, (3767) is higher than in humans (2941). These duplicated mouse genes encode mainly olfactory and vomeronasal proteins and some are associated with predator and pathogen avoidance [28]. The mouse major histocompatibility complex (MHC) is located on chromosome 17 and is designated H-2. It is one of the most polymorphic gene complexes known (Fig. 23.3). Its name, H-2, was derived from two research groups each studying graft rejection, When it was identified, one group called the complex “histocompatibility” while the other called it “antigen II.” Normally, each mammalian MHC contains about 200 expressed genes arranged in three classes (I, II, and III). In the case of mice, however, the regions are arranged Ia, II, III, and Ib. Thus the H-2 class I region is unusual since it is divided into two separate regions separated by classes II and III. The genes that encode the first cluster of class I genes are designated as K, class II are designated I, the class III genes are designated as S, while the second set of class I genes are designated as D. The MHC class I genes encode protein receptors expressed on most nucleated cells. Their function is to trigger immune responses by presenting peptide antigens from intracellular proteins to T lymphocytes. Class I genes can be subdivided into those that are highly polymorphic (class Ia genes) and those that show very little polymorphism (class Ib, Ic, or Id genes). The products of the classical H-2 class Ia genes such as H2-K and H2-D are expressed on many different cell types and present antigens to CD81 cytotoxic T cells. Genes classified as class Ib are considered nonclassical. Their products are structurally similar to the class Ia molecules but generally have limited polymorphism and restricted tissue expression. Some of them have functions other than antigen presentation to CD81 T cells. Class Ic genes have limited polymorphism and are found within the MHC. Their products include MICA and MICB, specialized proteins that are involved in signaling to NK cells but do not bind antigenic peptides. Class Id genes are located outside the MHC on different chromosomes. Genes in class II regions also encode polymorphic MHC molecules usually restricted to professional antigenpresenting cells (dendritic cells, macrophages, and B cells). Genes within the MHC class III region code for a mixture of proteins, many of which are important in innate immunity such as complement components. Although each MHCH2 contains all three gene regions, their gene content, and arrangement vary markedly between different haplotypes and inbred strains of mice [29]. Thus there are also diverse haplotypes within mouse populations that are designated by an arbitrary italic superscript such as H-2a, or H-2b. A typical outbred mouse population can have many individuals with multiple alleles at each class II locus. However, the levels of allelic variation vary greatly between each locus and among the different chains of the class II genes. While the human and mouse MHC class I genes exhibit extreme allelic polymorphism this is not the case in other mammals. In other species, MHC diversity is generated by variations in the number of MHC class I genes expressed. If some MHC genes are expressed in some MHC haplotypes but not in others, the effect will be to generate even more diversity than alternative combinations of alleles of a fixed gene. Gene content variation and allelic polymorphism can therefore be considered as two alternative strategies to diversify MHC haplotypes. Mammals use both these strategies for maintaining high levels of MHC class I diversity. Mice, dogs, cats, and humans simply use a small number of highly polymorphic genes. In other primates, horses, pigs, and ruminants as well as rats, however, MHC diversity is generated by varying the number and combinations of their class I gene loci [30].

23.7.1 The MHC class Ia region Mouse class Ia molecules are expressed on most nucleated cells. They have been detected on lymphocytes, platelets, granulocytes, hepatocytes, kidney cells, and sperm. Some cells, such as myocardium and skeletal muscle, may express very few class Ia molecules. FIGURE 23.3 The organization of the H-2 locus of mice. Note that the class Ia L region is not present in all mouse strains. Likewise, the long, extended class Ib region (Q, T, M) is a unique rodent feature.

IAB IAA IEB IEB2 IEA

K

I

IA

IE

II

S

III

D L

Ia

Q

T

Ib

M

The rodents: mice, rats, and their relatives Chapter | 23

361

Class Ia molecules consist of two linked glycoprotein chains. An α-chain (45 kDa) is associated with a much smaller chain called ß2-microglobulin (ß2M) (12 kDa). The α-chain is inserted in the cell membrane (Fig. 7.2). It consists of five domains: three extracellular domains, α1, α2, and α3, each about 100 amino acids in length; a transmembrane domain; and a cytoplasmic domain. The antigen-binding site is formed by the α1 and α2 domains. The β2M chain consists of a single domain and stabilizes the structure. Each MHC class I region has a common framework of non-MHC genes, and any size differences are mainly due to variations in these framework genes. In mice, only two (or three) class I loci contain expressed genes. The remainder consist of pseudogenes. The expressed genes are called H2-K, H2-D, and H2-L. The H2-L locus is present in laboratory mouse strains such as A/J and BALB/c but is absent in others such as C3H/Bl and C57Bl/10. It apparently arose by duplication of H2-D and has a similar function. A lack of H2-L does not appear to significantly affect mouse antiviral immunity [31].

23.7.2 Polymorphism Mouse MHC class Ia gene products show extreme polymorphism as a result of variations in the amino acid sequences in their α1 and α2 domains. On average, there are more than 100 alleles per locus. The most extreme polymorphism is restricted to three to four small regions within the α1 and α2 domains. In these variable regions, two or three alternative amino acids can occur at each position. The other domains of MHC class Ia molecules show little variation. The MHC α1 and α2 domains fold together to form an open-ended groove. A flat ß sheet forms the floor of this groove, and its walls are formed by two α helices (Fig. 7.4). This groove can bind antigenic peptides that are eight to ten amino acids long. The variable regions located along the walls of this groove determine its shape. The shape of the groove in turn determines which peptides can be bound and thus trigger immune responses. Polymorphism in the α1 and α2 domains results from variations in the nucleotide sequences between MHC alleles. These sequence variations result from point mutations, reciprocal recombination, and gene conversion. Point mutations are simply changes in individual nucleotides. Reciprocal recombination involves crossing over between two chromosomes. In gene conversion, small blocks of DNA are exchanged between different class I genes in a nonreciprocal fashion. The donated DNA blocks may come from nearby nonpolymorphic class I genes, nonfunctional pseudogenes, or other polymorphic class I genes. Class I MHC genes have the highest mutation rate of any germline genes yet studied (1023 mutations per gene per generation in mice). This high mutation rate implies that there are significant advantages to be gained by having very polymorphic MHC proteins available to present antigens to T cells.

23.7.3 Nonpolymorphic major histocompatibility complex class Ib molecules The class Ib region of the MHC-H2 contains more than 60 genes and pseudogenes within a region of about 2 kb. Class Ib proteins show reduced expression and tissue distribution when compared with class Ia molecules. They have limited polymorphism and they probably originated from class Ia precursors by duplication. These class Ib molecules serve specialized roles. Class Ib genes code for proteins on the surface of regulatory and immature lymphocytes and hematopoietic cells. They also consist of a membrane-bound α-chain associated with ß2-microglobulin, so their overall shape and antigen-binding groove are similar to those in MHC class Ia molecules. Since they are not polymorphic, however, MHC class Ib molecules bind a limited range of ligands. In effect, they act as pattern recognition receptors for commonly encountered, microbial PAMPs. Most of these nonpolymorphic genes are located in clusters at the telomeric end of the 2 mb H-2 region within the H2-Q, -T, and -M sub-regions. Thus there is a mosaic of orthologous class I genes separated from each other by regions of species-specific or paralogous class I genes. Peptide presentation by the Qa-1 protein plays a role in regulating CD41 T cell functions [32]. Qa-1 is involved in the suppression of CD41 T cell responses by signaling through the CD94/NKG2A or NKG2C receptors [33]. Some of these proteins have nonimmunologic functions. For example, the Q7 and Q9 proteins act on the reproductive system to influence the rate of preimplantation embryonic development and subsequent embryonic survival. The H2-TL gene product is involved in the generation of CD81 memory T cells and the regulation of intestinal immune responses by interacting with homodimeric CD8 receptors on intraepithelial lymphocytes. The H2-M region is 500 kb long and contains nine class I genes and four pseudogenes that fall into two subfamilies, M1 and M10. This class I gene cluster is separated from the other class I genes by nonclass I genes. The centromeric 1 mb of the H2-M region is enriched in olfactory receptor genes and contains just two functional class I genes, H2-M3 and H2-M2. Between the last 50 kb of the H2-T region and the H2-M region up to GabbrI, the sequence contains 26 genes, of which two are class I genes and three are class I pseudogenes [34]. The H2-M3 molecules can activate CD81

362

SECTION | 2 Mammalian orders

T cells by presenting them with N-formylated bacterial peptides. The H2-M1 and H2-M10 molecules specifically interact with the V2R class of pheromone receptors in the vomeronasal organ [34]. Class Ic genes have limited polymorphism and are also found within the MHC. The most significant of their products are probably MICA and MICB, which serve to signal cell stress to NK cells (Chapter 10). Class Id genes are nonpolymorphic class I
related genes not located on the MHC chromosome. Many of their products contribute to innate immunity since they bind PAMPs. For example, CD1 molecules are antigen-presenting receptors that bind lipid antigens (Chapter 9). The neonatal Fc receptor, FcRn, is a class Id MHC molecule that acts as an antibody (Fc) receptor on epithelial cells. It is expressed on the mammary gland epithelium and on the enterocytes of newborns where it serves as an immunoglobulin transporter (Fig. 2.3).

23.7.4 The MHC class II region In rodents, MHC class II expression is restricted to the professional antigen-presenting cells (dendritic cells, macrophages, and B cells) but can be induced on others such as T cells, keratinocytes, and vascular endothelial cells. Their expression is enhanced in rapidly dividing cells and in cells treated with interferon-γ (IFN-γ). The mouse class II region is relatively small compared to other species occupying only 495 kb while in humans it is 995 kb in size. In the mouse, there is one gene every 13.7 kb compared to humans which have one gene per 18.1 kb [34]. The mouse extended class II region is homologous to that in humans and it contains eleven genes from Pb to Tap2. Beyond Tap2 there is very limited homology, except for Eb1, Ec2, and Ea, which are distantly similar to DR while Ab2, Ab1, and Aa are distantly related to DO/DQ genes in humans [34].

23.7.5 Gene arrangement MHC class II molecules each consisting of two peptide chains called α and β. Each chain has two extracellular domains (one constant and one variable), a connecting peptide, a transmembrane domain, and a cytoplasmic domain. A third, class II chain, called the invariant or γ-chain, is used within antigen processing cells to serve as a surrogate antigen during MHC class II assembly. A “complete” MHC class II region in mice contains two paired loci. These are AA and AB, EA and EB. (The genes for the α-chains are designated A and the genes for the β-chains are called B.) Some of these genes are polymorphic. There may also be additional nonpolymorphic loci present such as DM and DO in humans. DM and DO gene products regulate the loading of antigen fragments into the MHC groove. Not all loci contain genes for both chains, and some contain many pseudogenes. These pseudogenes serve as DNA donors that can be used to generate class II polymorphism by gene conversion. Other genes within the class II region code for molecules involved in antigen processing. These include the transporter proteins TAP1 and TAP2 as well as some proteasome components. The measurement of MHC class II diversity is a convenient method to determine the overall genetic diversity in an animal population. For example, the decline in the red squirrel (Sciurus vulgaris) in the United Kingdom has been associated with a loss of genetic diversity [35]. By using the class II DRB locus as a marker of diversity across the MHC region, twenty-four Scvu-DRV alleles were identified at two functional loci, Scvu-DRB1 and Scvu-DRB2. Studies of squirrel populations in continental Europe showed high levels of diversity. Conversely, there was no diversity identified at the Scvu-DRB2 locus in the UK population and a high level of homozygosity at the Scvu-DRB1 locus.

23.7.6 Major histocompatibility complex class III molecules The remaining genes found within the H-2 complex are located within the class III region. In many cases, the structure or function of their products is poorly defined. Unlike the class I and class II molecules, these proteins are generally produced by hepatocytes or macrophages. They have many diverse functions (Fig. 7.6). Some are important in the defense of the body such as the complement components C4, factor B, and C2. The class III region also includes genes that encode tumor necrosis factor-α (TNF-α), several lymphotoxins, heat shock proteins, and some NK cell receptors. While generally similar, not all these genes are widespread across the Eutheria. For example, human NCR3, MIC, and MCCD1 genes are absent in mice. Conversely, every human structural gene present in the MHC class III region has an ortholog in the mouse and the gene configurations with respect to each other are identical between the species [36]. Thus the complement factor genes BF (factor B) and C2 have clearly resulted from an ancestral gene duplication event and so the two proteins share about 40% sequence identity. Likewise, four consecutive genes RP, C4, CYP21, and TNX are duplicated as modules (cassettes).

The rodents: mice, rats, and their relatives Chapter | 23

363

23.7.7 The natural killer cell receptors The NK cell receptor complex (NKC) is expanded in rats and mice. As noted elsewhere, rodents are unique in that they use multiple Ly49 genes. These genes are absent in humans and reduced to a single member in most other mammals such as cattle and dogs [12]. The mouse NKC contains 16 KLRA (Ly49) genes distributed over 8.7 mb whereas the rat has 36 KLRA genes spread over 10.3 mb. Thus most NK cell receptors in mice are glycoproteins belonging to the C-type lectin superfamily and are encoded by genes in the NKC on mouse chromosome 6 (and rat chromosome 4). These KLRA genes take up almost half of the NKC. They encode multiple Ly49 disulfide-linked type II transmembrane proteins that act as receptors for MHC class I molecules. In the mouse, at least fourteen members of the Ly49 family of C-type lectin homodimers are expressed on NK and natural killer T (NKT) cells. Most of these act as inhibitory receptors when they bind their ligands, the MHC class I molecules. These receptors possess immunoreceptor tyrosine-based inhibitory motifs (ITIMs) that become phosphorylated when the receptors bind their ligands. In turn, the phosphotyrosines activate the cytoplasmic SHP-1 tyrosine phosphatase that dephosphorylates components of the NK cell-activating pathway. As a result, target cells are protected against destruction by NK cells. This explains the missing-self hypothesis (Chapter 10). Another protein encoded by genes within the mouse NKC is NKG2A (CD159). NKG2A is one of a cluster of NKG2-like genes. It pairs with CD94 to act as a receptor for the nonclassical MHC molecule, H2-Qa-1b. This is the mouse ortholog of HLA-E. This binding also results in inhibition of target cell lysis [33]. The NKG2A gene maps close to CD94 within the NK complex. Both NKG2A and CD94 are class II transmembrane glycoproteins of the C-type lectin-like receptor superfamily [37]. NKG2A contains two ITIMs and readily transmits inhibitory signals thus preventing target cell lysis (Fig. 23.4). The ligands for another NK cell-activating receptor NKG2D also differ between mice and humans. Thus in humans, NKG2D binds to MICA, MICB, and the UL-16-binding protein family expressed on damaged, transformed, or infected cells. In the mouse, however, NKG2D binds to members of the H-60 and RAE protein families [12]. These are cell surface molecules distantly related to MHC class I proteins that are expressed abundantly on stressed and cancerous cells. The cytoplasmic domain of CD94 is only ten amino acids long so presumably the NKG2 peptide serves as the signaling component of this heterodimer. It possesses an ITAM motif and therefore delivers an activating signal. NKG2C and NKG2E also lack ITIMS and probably activate NK cells with the assistance of the small signaling transmembrane adapter protein, DAP12 that possesses an ITAM.

FIGURE 23.4 The activating and inhibiting NKG2 natural killer cell receptors in mice. Their ligand is a nonpolymorphic major histocompatibility complex class Ib surface receptor.

Invariant MHC

RAE H60

H2-Qa-1

NKG2C NKG2D CD94

CD94

NKG2A

DAP 12

ITAM

ITAM

ACTIVATING

ITIM INHIBITING

ITIM

364

SECTION | 2 Mammalian orders

Mouse NK1.1, also called KLRB1 is a cell surface glycoprotein of the C-type lectin family. It plays a role in NK cell activation and cytokine release. It is also involved in the generation of Th2 cells. Like other such proteins, it is a disulfide-linked homodimer. It is not expressed in all mouse strains. Thus it is present in C57Bl/6 and NZB mice but not in BALB/c, CBA/J, or NOD strains [37]. The mouse leukocyte receptor complex (LRC) is located on chromosome 7 and contains orthologs of LILRs, KIRs, and KIR-like genes, as well as genes encoding other receptors such as NKp46, Ly94, and the leukocyte-associated Iglike receptors (LAIRs). It also contains twelve paired Ig-activating receptor genes (PIRs) that share sequence identity with the human LILRs [38]. (Fig. 10.4) The PIRs are arranged in two clusters like LILRs, and their products can interact with MHC class I molecules on target cells [39]. Two mouse KIR-like sequences have been detected outside the LRC on the mouse X chromosome. In rats, the LRC is located on chromosome 1. It contains one PIR and one KIR sequence as well as FCAR, the IgA receptor gene that is absent in mice [39].

23.8

B cells and immunoglobulins

23.8.1 B cell subsets In mice, there are two subpopulations of B cells called B1 and B2, that develop from different precursor stem cells. B2 cells are conventional B cells that are central to the adaptive antibody responses. They appear late in neonatal life. They are the predominant population in the adult bone marrow and produce most of the body’s IgG. B1 cells, in contrast, originate early in fetal development from stem cells in the liver or omentum rather than from the bone marrow. They are innate-like cells that share some features with macrophages. B1 cells are, for example, are phagocytic and microbicidal (they produce reactive oxygen species) and can present antigen to CD41 T cells [40,41]. There are two subpopulations of B1 cells termed B1a and B1b. B1a cells develop exclusively in the neonate, are self-replenishing, and are responsible for most “natural” IgM in serum. They thus participate in innate immunity. B1a cells also express CD5, an adhesion, and receptor molecule. (CD5 is the receptor for CD72.) They recognize common bacterial molecules such as phosphoryl-choline. They produce antibodies in a T-independent manner. B1a cells also differ from conventional B2 cells in that they are found in the peritoneal and pleural cavities and can renew themselves. B1b cells are distinguished from B1a cells by lacking CD5 expression. They are, however, required for protection against some parasites and bacteria. B1b cells are produced throughout adult life. Many of the IgA-producing cells in the mouse intestine originate from B1 cells. B1 cells have been identified in humans, mice, rabbits, guinea pigs, pigs, sheep, and cattle. It is unclear, however, whether the B1-B2 classification applies to all these species [40].

23.8.2 Immunoglobulin Heavy chains The immunoglobulin heavy chain locus (IGH) is located on mouse chromosome 12 F2. The number of known genes in the complex is 182
185 and the predicted number is about 200. It spans about 2.3 mb. The genes are arranged in order, from 5’ to 3’; 152 IGHV genes (predicted number 170), 17
20 IGHD (17 in Balb/C and 20 in C57/Bl); four IGHJ, and eight or nine different IGHC genes depending on the strain. The functional IGH repertoire consists of 119
124 IGHV genes, 10
14 IGHD genes, and four IGHJ genes. The IGHV genes belong to 15 subfamilies. The mouse CDR3 region is relatively short and ranges from 5 to 15 amino acids in length. Unlike cattle, sheep, and rabbits, mice and humans rely solely on untemplated somatic point mutations in generating V region diversity. They do not employ gene conversion [42].

23.8.3 IGHG genes Downstream from the cluster of IGHJ genes are a series of constant region genes (Fig. 23.5). Mice have four or five functional IGHG genes that encode IgG1, IgG2a, IgG2b, IgG2c, and IgG3. Except for IgG2a and IgG2c, these IgG subclasses are shared among inbred mouse strains as well as wild mouse populations. As a result, BALB/cJ and C3H/HeJ mice express IgG2a while C57BL/6, C57BL/10, SJL, and NOD mice do not make IgG2a, but make IgG2c instead. These different subclasses have different affinities for specific Fcγ receptors as well as for the first component of complement C1q and, as a result, have different biological functions. The production of these isotypes also depends on the presence of specific cytokines. Thus IFN-γ induces switching to IgG2a/c and IgG3 whereas IL-4 promotes switching to IgG1 and IgE, and TGF-β promotes switching to IgG2b and IgA [43]. Inbred mice use either an Igh-1a or Igh-1b allele in their heavy chain IgG1 locus [43].

The rodents: mice, rats, and their relatives Chapter | 23

IGH

5’

20

152

3’

4 M

IGK

IGL

D

G3

G1

G2b G2c

G2a

E

A

365

FIGURE 23.5 The Immunoglobulin gene loci in mice. In wild mice, the number of known functional IGLV genes is eight [79].

5

174

2

2 C2

C4

M

C3

C1

D

G

E

A

V

D

J

LC

23.8.4 IGHD genes Mice possess a functional IGHD gene and are capable of producing IgD. In this species, IgM and IgD are derived from an alternatively spliced transcript. They both share IGHV gene segments and as a result, their products share antigen-binding specificities. They are also expressed simultaneously by antigen-naı¨ve B cells. After exposure to an antigen, these B cell genes undergo a class switch recombination event mediated by the usual DNA double-strand break repair enzymes. As a result, their IGHM or IGHD exons are replaced by IGHG, IGHE, or IGHA. They rarely switch from IGHM to IGHD since this would involve activation of both the Cμ and Cδ switch regions. It is unclear what signals trigger IGHD class switching. Recently however, it has been shown that IGHD class switch recombination depends upon the presence of an intact microbiota and is anatomically confined to B cells situated in the mucosal-associated lymphoid tissues, especially those located in the nasal mucosa. The microbiota appear to signal to these B cells through TLRs and MyD88 to trigger the switch [44]. Serum IgD levels in mice vary with the strain and with the age of the animal. Thus young adult BALB/c, A/J, and DBA1 mice have 200
300 ng/mL while C57BL/6 mice have 400
500 ng/mL [45].

23.8.5 Immunoglobulin Light Chains 23.8.5.1 Kappa chains The mouse light chain kappa locus is located on chromosome 6C2 and spans 3.2 mb. From 5’ to 3’ it contains 174 IGKV genes, 5 IGKJ genes, and a single IGKC gene. The IGKV genes fall into 19 subfamilies and belong to three clans. In practice, mice use 95 functional IGKV genes, 4 functional IGKJ, and the single IGKC gene.

23.8.5.2 Lambda chains The mouse light chain lambda locus (IGL) is located on chromosome 16B1 and occupies 240 kb. In laboratory mice, it contains twelve genes but in wild mice, 17 have been detected. In laboratory mouse strains it contains three IGLV genes belonging to two subfamilies. There are also five IGLJ genes and four IGLC genes that each belong to two subfamilies. The IGHJ and IGHC genes form two clusters, -J2-C2-J4-C4- and -J3-J3P-C3-J1-C1. Each is preceded by two and one IGLV gene respectively. The functional repertoire in laboratory mice thus consists of three IGLV genes, three IGLJ genes, and 2
3 IGLC genes. Wild mice have eight functional IGLV genes belonging to three subfamilies [3]. In adult mice, the ratio of kappa to lambda light chains in serum immunoglobulins approaches 20:1. However, further analysis shows that in neonatal mice the κ:λ ratio is close to one. As mice age, this ratio progressively increases in the spleen with a major shift occurring between four and six weeks of age. A similar shift also occurs in the bone marrow. Since kappa chains very much predominate in adults, it appears that their rise is a result of antigen-driven clonal expansion [46]. It is also apparent that the recombination signal sequences (RSS) exert this differential effect [47]. Thus

366

SECTION | 2 Mammalian orders

the kappa RSS supports six- to ninefold higher levels of light chain transcription than does the lambda RSS. This in turn is due to enhanced pair-wise cleavage of the kappa RSS by RAG-1/2.

23.8.6 Fc receptors Mice, unlike most mammals, do not possess the IgA receptor CD89 (FcαR1). The gene region involved has been broken up and the fragments relocated on two different chromosomes. In addition, the genes encoding FcγR1, FcRL5, FcRL1, and FcRls are located on chromosome 3 whereas the remaining Fc receptors are found on chromosome 1 [48]. Likewise, humans have two receptors for IgG (FcγRIIA and FcγRIIC) that mice lack. Thus mouse and human FcγRs are quite different in their expression patterns and binding abilities [12]. The three major differences are first, that all the human activating FcγRs bind IgG1 whereas only one mouse activating receptor FcγRIII binds mouse IgG1 [49]. Second, human inhibitory FcγRs have a lower affinity for the IgG subclasses but not so in mice. Third, mouse FcγRs can bind IgE which the human FcγRs cannot [49].

23.9

T cells and cell-mediated immunity

In addition, to antigen-presenting cells, activated T cells from many mammals, with the exception of mice, synthesize and express MHC class II molecules [50]. In cattle and horses, expression of MHC class II is associated with T cell activation. The lack of MHC class II expression on mouse T cells is a result of inhibited transcription following hypermethylation of the class II transactivator. In other mammals activated T cells can present antigens to other T cells. In the mouse, they cannot. The mouse circulating T cell repertoire contains more than 95% α/β T cells (and 0.5%
2% γ/δ T cells) as does the rat [51]. Guinea pigs are also a γ/δ -low species. The percentage of γ/δ positive T cells among guinea pig peripheral blood mononuclear cells is about 9% [52]. The genes encoding the TCRs in the mouse are in general, very similar to those described in other mammals (Fig. 23.6). γ/δT cells are present in the skin of rats where they are the predominant T cell type, but they are only minor components in peripheral lymphoid organs, the lungs, or the gastrointestinal tract. This contrasts with mice where γ/δ T cells account for up to 50% of the intestinal intraepithelial lymphocytes [53].

23.9.1 TRA/D The mouse combined TRA and TRD gene locus occupies 1650 kb on chromosome 14. From 50 to 30 the TRAV genes are organized into two clusters upstream of 60 TRAJ genes and a single TRAC gene. There are 98 TRAV genes per haploid genome belonging to 23 subfamilies [54]. These include 10 TRA/DV genes. There are up to 73
84 functional TRAV genes (including the TRA/DV genes), and 38 functional TRAJ genes. The mouse TRD locus occupies 275 kb embedded within the TRA locus. It consists of six TRDV plus ten upstream TRA/DV genes. In addition, it contains two TRDD, two TRDJ, and a single TRDC. As in other mammals, a single TRDV gene is located in an inverted orientation at the 30 end of TRDC. The potential TRD repertoire consists of five functional TRDV genes, nine or ten TRA/DV, and two each of TRDD and TRDJ. TRA/D

TRB

5'

98

5

2

AV

DV

DD

34

7

DJ DC

AJ

C2

\

4 C1

V

C2

D

C3

J

3'

61

7 C1

TRG

2

C

C4

AC

FIGURE 23.6 The organization of the T cell antigen receptor genes in mice.

The rodents: mice, rats, and their relatives Chapter | 23

367

23.9.2 TRB The mouse TRB locus is located on chromosome 6B2. It contains 34 TRBV genes in a single cluster followed by two duplicated D-J-C cassettes. The first cassette contains one D gene, seven J genes, and one C gene. The second cassette contains the same. This is followed, as usual, by a single TRBV31 gene in inverted orientation. The entire locus occupies 700 kb. The TRBV genes belong to 31 subfamilies. Of the 35 TRBV genes, 21
22 are considered functional, and of the 14 TRBJ genes, 11 are functional.

23.9.3 TRG The mouse TRG locus is located on chromosome 13A3.1. It occupies about 200 kb and is bounded by AMPH at the 50 end and STARD3NL at the 30 end. The mouse locus closely resembles the arrangement in the artiodactyls and carnivores insofar as it is organized into four V-J-C cassettes. However, it has both fewer cassettes and fewer genes. Thus it contains seven TRGV genes belonging to five subfamilies with four TRGJ genes and four functional TRGC genes organized into four V-J-C cassettes. The most 50 cassette, TRGC1 occupies 41 kb and contains four genes, TRGV7, -4, -6, and -5 followed by one TRGJ and one TRGC gene. This is followed by the TRGC2, TRGC3, and TRGC4 cassettes. Each of these contains a single TRGV gene [54,55]. The TRGC3 cassette is not functional because TRGC3 is a pseudogene and TRGC2 is in inverted orientation relative to the other three cassettes. Enhancer elements are present at the 30 end of all four V-J-C cassettes.

23.9.4 Natural killer T cells NKT cells are defined as cells that express both NK and T cell markers but express a very limited TCR repertoire. NKT cells can be activated by strong receptor agonists such as the lipid α-GalCer without the need for any additional stimulation. Some TLR ligands such as bacterial lipopolysaccharides can also activate them indirectly by stimulating dendritic cells to produce interleukin 12. The IL-12 amplifies weak stimulants such as those triggered by CD1 ligands. Mouse NKT cells can also be activated by IL-12 alone. Of the five CD1 genes found in humans (a-e) only C1d is expressed in mice [12]. Lipid antigens, together with interactions with the costimulatory molecules CD28 and CD40 and their ligands, trigger NKT activation [56]. The NKT cells are activated by glycolipids presented by CD1d. These lipid antigens are recognized by TCRs that use an invariant TCR α-chain (Vα14) in mice. This invariant chain pairs with Vβ8.2, Vβ7, or Vβ2. Because they are present in unsensitized animals, NKT cells can mount a very rapid response against lipidbearing microorganisms. (Fig. 10.7). NKT cells are not, however, present in all mammals. In many ruminants including cattle, sheep, and buffalo, the CD1d gene is a pseudogene and thus nonfunctional. However, it is present and functional in horses, pigs, rabbits, elephants, and primates [57].

23.9.5 Thy-1 Thy-1 (CD90) is a cell surface, GPI-linked, glycoprotein of 25
37 kDa belonging to the immunoglobulin superfamily (Fig. 23.7). It is expressed in large amounts on mouse thymocytes and peripheral blood T cells. It is probably the most FIGURE 23.7 The Thy-1 in the mouse.

Most abundant glycoprotein 1 million mols/cell Cover 10-20% of T cell surface GPI anchored protein

T cell activation thru CD28

Promotes T cell signaling from APCs

Thy 1

Crosslinking

? ligand

Stimulates TCR signaling Null Thy1 blocks TCR signaling

May trigger cytolysis in the presence of MHC and CD28

functions

of

368

SECTION | 2 Mammalian orders

abundant glycoprotein on murine thymocytes with an estimated million molecules expressed per cell. These are estimated to cover about 10%
20% of the T cell surface! [58] If Thy-1 is cross-linked by specific monoclonal antibodies it triggers T cell proliferation and cytokine synthesis [59]. However, its physiological ligands have yet to be determined. It is clear that in association with appropriate co-stimulation by CD28, Thy-1 cross-linking can partially substitute for TCR signaling. It also acts as a potent co-stimulator of TCR signaling. In humans, unlike mice, Thy-1 is expressed by endothelial cells, fibroblasts, and a subpopulation of neurons and glial cells in the brain. Its expression is limited to a small population of cortical thymocytes, and it is not expressed on mature human T cells.

23.10 Rats (Rattus norvegicus) 23.10.1 RT1: the rat major histocompatibility complex Molecular data suggest that rats and mice diverged about 20
40 mya [60]. The major histocompatibility complex of the rat (R. norvegicus) termed RT1, is located in the telomeric region of the short arm of chromosome 20. Its orientation with respect to the centromere is the same as in the mouse. It appears to be about 3
4 mb in length, - similar in size to that of the mouse. As in other species, RT1 consists of a group of closely linked class I and class II genes together with a mixture of other genes that do not play a role in immune functions. The major regions are RT1-A, the centromeric class I region; RT1-B/D, the class II region; the class III region, and RT1-C/E/M, the telomeric class I region. Thus these genes are arranged in the same order as the mouse, and except for the rat and mouse-specific telomeric class I region are very similar to the human HLA system as well. Among over 200 different rat strains there are a limited number of haplotypes [60]. Studies on the RT1 polymorphism of wild rats suggest that their allelic diversity is also restricted in a manner similar to that seen in standard inbred strains. The RT1 class I genes occur in clusters embedded within groups of framework genes. This is seen both in the RT1-A region and in the RT1-C/E/M region. In the latter, there are at least four class I clusters with framework subregions between them. Again, this is a similar situation to that in the mouse. The number of class Ia genes also varies between rodents; rats have more than 60. Rat class Ia genes are highly polymorphic, and their products are expressed on most nucleated cells, especially lymphoid cells. As is normal, polymorphism is mainly restricted to the peptide binding region. So far, class Ia functions have been assigned only to the genes in the RT1-A region. No genes that correspond functionally to the mouse D/L genes have been found in the rat. The number of class I genes in the RT1-A region can vary between one and three depending on the rat haplotype. The fact that some rat haplotypes have only a single class Ia gene is quite different from the mouse where a minimum of two, H2-K and H2-D are present. In humans at least three are present. A large number (45
62) of class Ib genes have been mapped to the RT1-C/E/M region [60]. RT1-H genes are orthologous to HLA-DP and H2-P; RT1-B genes to HLA-DQ and H2-A; and RT1-D genes to HLA-DR and H2-E. While the order of these genes is the same as in mice and humans, the gene copy number and their functional status are different in the rat. For example, RT1-HB is a pseudogene. In RT1-B there is only one copy each of BA and BB. In RT1-D there is a single DA gene and two DB genes, one of which is a pseudogene. The rat class III region located between the class II and the telomeric class I regions contains the usual genes such as those encoding complement C4, factor B, and C2, as well as the tumor necrosis factor gene family. All the class III genes in the rat are present in mice and humans and are present in the same order. These are present in multiple copies in mice and humans but the situation in the rat is unclear [60].

23.10.2 Rat natural killer cell receptor complex Rat NK cells can recognize MHC class Ia molecules on normal cells and are thereby inhibited from killing them [60]. The inhibitory MHC receptors in this species have been identified as members of the Ly49 family, 110 kDa, disulfidelinked dimers whose genes are located within the NKC. (They have been identified by monoclonal antibodies called STOK1 and STOK2.) Their expression is controlled by the RT1 complex [61].

23.10.3 Rat immunoglobulins The rat IGH locus is about 220 kb in size and is located on chromosome 6 q32
33 [62]. It encodes eight immunoglobulin isotypes in the following order: -M-D-G2c-G2a-G1-G2b-E-A-

The rodents: mice, rats, and their relatives Chapter | 23

369

Thus the number of IGHC genes in the rat is similar to that in the mouse. Neither contains any conserved pseudogenes, unlike humans. Each of the IGHC genes except IGHA, encodes six exons, four for the constant domains and two transmembrane domains. With the exception of IgD, they have greater than 90% sequence identity with the mouse. The rat IGHD locus is twice the size of the other IGHC loci. In addition, it has only about 55% identity with mouse IGHD. Rat IGHG1 and IGHG2 share 94% nucleotide identity, and both resemble mouse IGHG1. The rat IGHV region is about 4.8 mb in length, twice the size of the mouse IGHV locus. This reflects the fact that it contains twice as many V genes as the mouse. The rat has 363 of which 136 are functional. The mouse has 152 IGHV genes of which 110
124 are functional [62]. The rat also has 21 IGHD segment genes of which 14 are functional and five IGHJ genes of which four are functional. The vast majority of IGHV genes are orientated towards the D gene cluster so permitting deletional joining. Thus despite its size, the total number of functional IGHV genes in the rat and mouse are not very different. Nearly all the nonfunctional V genes in the rat are pseudogenes that lack an open reading frame (ORF). (It should however, be noted that not all genes with an ORF are functional. Alterations in their regulatory or RSS signal sequence or splice sequences can render them nonfunctional.-They may also be orphons in the wrong chromosomal position). The rat IGHV genes can be grouped into 13 subfamilies. The largest such subfamily is IGHV2 which contains 99 functional and nonfunctional members. As in mice, B90% of expressed immunoglobulins in the rat express kappa light chains. The IGK locus is located on rat chromosome 4. It contains 163 IGKV genes of which 135 are functional. (The mouse has 174 of which 95 are functional). They are divided into 21 subfamilies. It has seven IGKJ genes (six functional) and one IGKV gene. The IGL locus is located on rat chromosome 11. Three IGLC genes and one IGLV gene have been described. Each of the C genes is preceded by two J genes. Thus their order is, -Vn-J-J-C1-J-J-C2-: However, IGLC1 is not expressed as a result of the two J genes having aberrant RSS and splice donor sites. One of the JL2 genes is also nonfunctional so only IGLJ2-IGLC2 are functional and expressed [62].

23.11 Other rodents 23.11.1 Prairie voles (Microtus ochrogaster) The prairie vole IGH locus is located on chromosome 1 and extends over 1600 kb. It contains at least 79 IGHV gene segments of which 28 are likely functional while two are ORFs and 49 are pseudogenes. There are also seven D and four J gene segments [63]. They have six IGHC genes, one each of M, E, and A and three of G. The three IGHG genes share about 70% homology. There are also gene remnants encoding two transmembrane regions of an IGHD gene. The IGK locus contains 124 IGKV segments of which 47 appear to be functional, one is an ORF while 76 are pseudogenes. There are five IGKJ segments and a single IGKC. The IGL locus consists of 21 IGLV genes (14 functional, one ORF, and six pseudogenes), these are followed by two J-C clusters. However, the IGLV1 gene appears to be a pseudogene. The presence of so many pseudogenes likely provides a resource for extensive gene conversion [63].

23.11.2 Great gerbils (Rhombomys opimus) The genome of the great gerbil has been sequenced and shows some interesting differences in its immune system [64]. For example, it, and the related Mongolian gerbil (Meriones unguiculatus) have lost the genes encoding TLR8, TLR10, and all three members of the TLR11 subfamily. They also show evidence of diversifying selection in TLR7 and TLR9 genes. In addition, the great gerbil has a species-specific duplication of its DRB1 gene in its MHC class II region. It is speculated that this duplication may assist the great gerbil in resisting infection by the plague bacillus, Yersinia pestis.

23.11.3 Guinea pigs (Cavia porcellus) 23.11.3.1 Hematology The major granulocyte population in guinea pigs are termed heterophils based on their staining properties (Fig. 23.8). Thus in addition, to the azurophil and specific granules found in the neutrophils of other mammals, they also possess a third type of granule called a nucleated granule [65].

370

SECTION | 2 Mammalian orders

K K

H FIGURE 23.8 Guinea pig Kurloff cells (K). Also, notice the heterophil (H) with its dark staining granules. Courtesy Dr. Mark Johnson.

The guinea pig also has a population of atypical mononuclear cells, called Kurloff cells, (also known as Foa`-Kurloff cells), with both monocyte and lymphocyte-like characteristics (Fig. 23.8). Kurloff cells are also found in the blood and spleen of other Hystricomorph rodents such as capybaras, pacas, and agoutis. They are mononuclear cells about the size of a large lymphocyte that contain a characteristic intracytoplasmic, slightly granular inclusion known as a Kurloff body. They amount to 3%
4% of the leukocyte differential count. They are more common in females than in males and are especially prominent during pregnancy. It appears that Kurloff cells possess NK cell functional activity as well as antibody-dependent cellular cytotoxicity and may be the guinea pig equivalent of the large granular lymphocyte recognized in other rodent species [66].

23.11.3.2 Major histocompatibility complex Guinea pigs have retained all three of their ancestral Ly49 genes as well as six functional MHC class I genes and 19 MHC class I pseudogenes [67].

23.11.3.3 Immunoglobulins The guinea pig IGH locus extends over 6 mb [68]. It contains five IGHC genes, one each of M, E, and A as well as two IGHG and the remnants of reversely orientated CH1 and CH2 IGHD genes. Guinea pigs possess two IgG isotypes G1 and IgG2, but IgG1 is preferentially expressed. The IGH locus also contains at least 507 IGHV segments of which 94 are likely functional and 413 are pseudogenes. There are also 41 D gene segments located 500 kb downstream from the last IGHV segment. and six IGHJ gene segments [68]. This is the largest number of IGHV genes found in any of the mammals studied to date! The pseudogenes account for 81% of the total V genes. They may be nonfunctional, but they may also act as a source of DNA sequences to generate V region diversity through gene conversion. These IGHV genes belong to clans II and III. The guinea pig IGK locus extends over 4029 kb and contains 349 IGKV genes of which 111 appear to be functional, while 238 are pseudogenes. Downstream of the V genes there are three IGKJ segments and four kb downstream of that there is a single IGKC. The IGL locus spans 1642 kb and consists of 142 IGLV genes (58 functional, and 84 pseudogenes), these are followed by eleven J-C clusters [68].

23.11.4 Capybaras (Hydrochoerus hydrochaeris) The capybara is the largest living rodent. It is 60 times more massive than its closest relative, the guinea pig, and B2000 times more massive than the house mouse. The growth of such large animals is a result of an increased rate of cell proliferation. It would be anticipated therefore that this increased proliferation would also increase the accumulation of mutations and the risks of developing cancer. This does not happen. Large species do not suffer from an increased cancer rate. Cancer is rarely reported in this species. This phenomenon is known as Peto’s paradox (Box 25
1). Analysis of the capybara genome has identified a novel anticancer adaptation that involves T-cell-mediated tumor suppression [69]. Three gene families are expanded in this species. They are TPT1 (Tumor protein, translationally controlled 1), MAGEB5 (Melanoma antigen family B5), and GZMB (Granzyme B). TPT1 plays a major role in regulating cell proliferation and in tumor cell reversion. MAGEB5 proteins expressed on proliferating neoplastic cell surfaces are recognized and attacked by cytotoxic T cells. Granzyme B plays a key role in T cell-mediated cytotoxicity and the

The rodents: mice, rats, and their relatives Chapter | 23

371

induction of caspase-dependent apoptosis. Thus collectively, these three gene families mediate enhanced tumor suppression by T cells and appear to reduce any increased risk in capybaras despite their large body size.

23.11.5 Hamsters (Mesocricetus ssp) There are several species of hamsters found in the wild and used in the laboratory. The most widely used experimental species is the Syrian hamster (Mesocricetus aureus). Early studies on this species indicated that it was polymorphic at its MHC class II locus but monomorphic at its MHC class I locus. Subsequent investigations have however, demonstrated that the Syrian hamster is, in fact very polymorphic at its class I locus [70]. As in other mammals, the nucleotide sequences encoding the antigen-binding class I α1 and α2 domains, are highly variable and show evidence of selective pressures. The probable reason for this error is most likely the fact that many laboratory populations are highly inbred. It is also clear that some wild populations, especially the European hamster (Cricetus cricetus) are under severe pressure from modern agricultural practices and as a result have lost much of their MHC diversity [71].

23.11.6 Mole-rats (Heterocephalus glaberi, Spalax ehrenbergi) Naked mole-rats (Heterocephalus glaberi) are mouse-sized (B35 g) African rodents that have an unusually long life span in relation to their size. Mole-rats are eusocial subterranean mammals found in the arid and semiarid regions in the Horn of Africa. They live in large complex underground colonies with a single breeding queen and a few breeding males. Within a colony, there is low genetic diversity and a high inbreeding rate. Mole rats live on a diet of underground plant food storage organs such as tubers. Their metabolic rate is approximately that which would be predicted based on their size [7]. They can, however, maintain homeostasis until well over 20 years of age [67]. Humans and other mammals have a progressively greater risk of death as they grow older, (Gompertz Law) [2]. Mole-rats defy Gompertz Law. Their chances of dying do not increase progressively as they age. Even at ages 25-fold greater than their time to reproductive maturity they do not show age-related mortality effects. They do not appear to suffer a decline in health with increasing age and show few degenerative changes. They can live for more than 30 years which is about eight times longer than Mus musculus. In addition, mole rats appear to be highly resistant to developing cancer. These animals show remarkable resistance to both spontaneous tumors and induced tumor formation. Using single-cell RNA sequencing, Hilton et al. have shown that the mole-rat immune system, especially their spleens, is characterized by a high myeloid-to-lymphoid cell ratio that includes a novel LPS-responsive granulocytic cell subset [67]. Mole rats also appear to have no NK cells since their genome lacks the expanded KLR gene family that controls NK cell function. The mole-rat NKC complex, unlike the mouse, contains only a single Ly49 gene. This is expressed at low levels on mole-rat CD81 T cells. They appear to compensate for a lack of NK cells with enhanced myeloid cell activity. They do, however, express a high level of telomerase that prevents telomere shortening and hence cellular growth arrest [7]. In mice, the immune cell population in the spleen consists of about 90% lymphoid cells. B cells constitute 60% of the total spleen immune cell population, T cells account for 28%, and NK cells for 3%. Myeloid cells constitute only 8%, most of which are dendritic cells and macrophages. In contrast, in mole-rats, lymphoid cells account for only 44% of the splenic immune cells, T cells account for 30%, and B cells for 14%. The remaining 54% are myeloid cells with 16% macrophages and 37% neutrophils. Dendritic cells account for only 2%. The resistance of mole rats to cancer is associated with the use of a pathway termed “concerted cell death” that is triggered by the production of interferon-β. It effectively kills cells when they get too crowded by triggering contact inhibition. Cells and tissues of these mole-rats express very low levels of n-methyltransferase 1. When cancerous cells undergo hyperplasia, their genome becomes demethylated. This results in the activation of retrotransposable elements. These upregulated RTEs form cytoplasmic RNA-DNA hybrids that activate the cGAS-STING pathway cyclic GMP/ AMP synthase (cGAS)—Stimulator of interferon genes (STING) to induce cell death [72]. As a result, hyperplastic cells such as tumor cells are rapidly and automatically destroyed immediately, they appear in mole-rats. The second pathway of cancer resistance in the naked mole-rat may involve the production of high molecular weight hyaluronan [73]. Mole-rat fibroblasts secrete extremely high molecular weight hyaluronan (HA)—which is over five times larger than human or mouse HA. They have decreased HA-degrading enzymes and unique hyaluronan synthase 2. As a result, the HA accumulates in mole-rat tissues. It has been suggested that the presence of this increased HA may confer additional tumor resistance on these rodents. Another species of mole-rat, Spalax ehrenbergi has lost two complete gene families from its MHC class II region. These deleted families are DO and DR leaving them with only DP and DQ [74]. The DP family however, contains at least three α- and four β-genes. One α- and one β-gene are found in its DQ subregion. Nucleotide sequence analysis

372

SECTION | 2 Mammalian orders

indicates that the DPB genes are very closely related to human DQB genes [75,76]. At least some of these DP loci are both polymorphic and fully functional. Thus during the evolution of the mole-rat, the DO and DR loci have been deleted and their role was taken over by an expanded DP locus. In effect, these MHC loci are functionally interchangeable. Conversely, S. ehrenbergi has also undergone a massive expansion of its MHC class I loci [77]. Thus it has B65 class I genes arranged in 14 clusters. Most of these clusters contain two genes and the remainder contain only one. Further analysis has shown that three of these class I genes are homologous to the mouse K, D, and L genes. The gut microbiome of H. glaberi also possesses some features associated with human centenarians such as a high load of Spirochetaceae [78]. The gut contents are also enriched in short-chain fatty acids and mono- and disaccharides— carbohydrate degradation products when compared to the human and house mice.

References [1] Lacher TE, Murphy WJ, Rogan J, Smith AT, Upham NS. Evolution, phylogeny, ecology and conservation of the clade Glires: lagomorpha and rodentia. Handbook of the Mammals of the World, Vol. 6. Lynx Edicions; 2020. p. 15
26. [2] Ruby JG, Smith M, Buffenstein R. Naked mole-rat mortality rates defy Gombertzian laws by not increasing with age. eLife 2018. Available from: https://doi.org/10.7554/eLife.31157. [3] Abolins S, King EC, Lazarou L, Weldon L, et al. The comparative immunology of wild and laboratory mice, Mus musculus domesticus. Nat Commun 2017. Available from: https://doi.org/10.1038/ncomms14811. [4] Viney M, Riley EM. The immunology of wild rodents: current status and future prospects. Front Immunol 2017. Available from: https://doi. org/10.3389/fimmu.2017.01481. [5] Abolins S, Lazarou L, Weldon L, Hughes L, et al. The ecology of immune state in a wild mammal, Mus musculus domesticus. PLoS Biol 2018. Available from: https://doi.org/10.1371/journal.pbio.2003538. [6] Beldomenico PM, Telfer S, Gebert S, Lukomski L, et al. The dynamics of health in wild field vole populations: a haematological perspective. J Anim Ecol 2008;77:984
97. [7] Gorbunova V, Bozzella MJ, Seluanov A. Rodents for comparative aging studies: from mice to beavers. Age 2008;30:111
19. [8] Furukawa S, Kuroda Y, Sugiyama A. A comparison of the histologic structure of the placenta in experimental animals. J Toxicol Pathol 2014;27:11
18. [9] Landreth KS. Critical windows of development of the rodent immune system. Hum Exp Toxicol 2002;21:493
8. [10] Grases-Pinto B, Abril-Gil M, Torres-Castro P, Castell M, et al. Rat milk and plasma immunoglobulin profile throughout lactation. Nutrients 2021. Available from: https://doi.org/10.3390/nu13041257. [11] Parr EL, Bozzola JJ, Parr MB. Purification and measurement of secretory IgA in mouse milk. J Immunol Meth 1995;180:147
57. [12] Mestas J, Hughes CCW. Of mice and not men: differences between mouse and human immunology. J Immunol 2004;172:2731
8. [13] O’Connell KE, Mikkola AM, Stepanik AM, Vernet A, et al. Practical murine hematopathology: a comparative review and implications for research. Comp Med 2015;65(2):96
113. [14] Everds NE. Hematology of the Laboratory mouse. The mouse in biomedical research. 2nd ed. Elsevier; 2007. [15] Hensel JA, Khasttar V, Ashton R, Ponnazhagan S. Characterization of immune cell subtypes in three commonly used mouse strains reveals gender and strain-specific variations. Lab Invest 2019;99(1):93
106. [16] Roycroft E, Achmadi A, Callahan CM, Esselstyn JA, et al. Molecular evolution and ecological specialization: Genomic insights from the diversification of murine rodents. Genome Biol Evol 2021. Available from: https://doi.org/10.1093/gbe/evab103. [17] Cervantes JL, Weinerman B, Basole C, Salazar JC. TLR8: the forgotten relative revindicated. Cell Mol Immunol 2012;9:434
8. [18] Staeheli P, Grob R, Meier E, Sutcliffe J, Haller O. Influenza virus-susceptible mice carry Mx genes with a large deletion or a nonsense mutation. Mol Cell Biol 1988;8(10):4518
23. [19] Zlotnik A, Yoshie O. The chemokine superfamily revisited. Immunity 2012;36:705
16. [20] Eisenhauer PB, Lehrer RI. Mouse neutrophils lack defensins. Infect Immun 1992;60(8):3446
7. [21] Wait R, Chiesa G, Parolini C, Miller I, et al. Reference maps of mouse serum acute-phase proteins: challenges with LPS-induced inflammation and lipoprotein A-I and A-II transgenes. Proteomics 2005;5:4245
53. [22] Cray C, Besselsen DG, Hart JL, Yoon D, et al. Quantitation of acute phase proteins and protein electrophoresis in monitoring the acute inflammatory process in experimentally and naturally infected mice. Comp Med 2010;60(4):263
71. [23] Khovidhunkit W, Duchateau PN, Medzihradszky KF, Moser AH, et al. Apolipoproteins A-IV and A-V are acute-phase proteins in mouse HDL. Atherosclerosis 2004;176:37
44. [24] Kostowiecki M. Development and degeneration of the second type of Hassall’s corpuscles in the thymus of guinea pig. Anat Rec 1962;142:195
203. [25] Isek A, Morita Y, Nakauchi H, Ema H. Hematopoietic stem cells in the mouse spleen. Blood 2008;112(11):2421
6. [26] Casteleyn C, Breugelmans S, Simoens P, Van, den Broeck W. The tonsils revisited: review of the anatomical localization and histologic characterization of the tonsils of domestic and laboratory animals. J Immunol Res 2011. Available from: https://doi.org/10.1155/2011/472-460.

The rodents: mice, rats, and their relatives Chapter | 23

373

[27] Kararli TT. Comparison of the gastrointestinal anatomy, physiology, and biochemistry of humans and commonly used laboratory animals. Biopharm Drug Dispos 1995;16:351
80. [28] Tizard IR, Skow L. The olfactory system: The remote sensing arm of the immune system. Anim Hlth Res Revs 2021. Available from: https:// doi.org/10.1017/s1466252320000262. [29] Shiina T, Blancher A, Inoko H, Kulski JK. Comparative genomics of the human, macaque and mouse major histocompatibility complex. Immunology 2016;150(2):127
38. [30] Tizard IR. Veterinary immunology: an introduction. 10th ed. St. Louis: Elsevier; 2017. [31] Biddison WE, Hansen TH, Levy RB, Doherty PC. Involvement of H-2L gene products in virus-immune T-cell recognition. Evidence for an H2L-restricted T-cell response. J Exp Med 1978;146(6):1678
86. [32] Ohtsuka M, Inoko H, Kulski JK, Yoshimura S. Major histocompatibility complex (MHC) class Ib gene duplications, organization and expression patterns in mouse strain C57BL/6. BMC Genomics 2008. Available from: https://doi.org/10.1186/1471-2164-9-178. [33] Vance RE, Kraft JR, Altman JD, Jensen PE, Raulet DH. Mouse CD94/NKG2A is a natural killer cell receptor for the non—classical major histocompatibility complex (MHC) Class I molecule Qa-1b. J Exp Med 1998;188(10):1841
8. [34] Takada T, Kumanovics A, Amadou C, Yoshino M, et al. Species-specific class I gene expansions formed the telomeric I Mb of the mouse major histocompatibility complex. Genome Res 2003;13:589
600. [35] Ballingall KT, McIntyre A, Lin Zm, Timmerman N, et al. Limited diversity associated with duplicated class II MHC-DRB genes in the red squirrel population in the United Kingdom compared with continental Europe. Conserv Genet 2016;17:1171
82. [36] Yu CY, Yang Z, Blanchong CA, Miller W. The human and mouse MHC class III region: a parade of 21 genes at the centromeric segment. Immunol Today 2000;21(7):320
8. [37] Smith HRC, Idris AH, Yokoyama WM. Murine natural killer cell activation receptors. Immunol Rev 2001;181:115
25. [38] Martin AM, Kulski JK, Witt C, Pontarotti P, et al. Leukocyte Ig-like receptor complex (LRC) in mice and men. Trends Immunol 2002;23(2): 81
8. [39] Kelley J, Lutz W, Trowsdale J. Comparative genomics of natural killer cell receptor gene clusters. PLoS Genet 2005. Available from: https:// doi.org/10.1371/journal.pgen.0010027. [40] Gao J, Ma X, Gu W, Fu M, et al. Novel functions of murine B1 cells: active phagocytic and microbicidal abilities. Eur J Immunol 2012;42: 982
92. [41] Cancro MP. Phagocytic B cells: De´ja` vu all over again. J Leukocyte Biol 2012;91:519
21. [42] Dale GA, Wilkins DJ, Bohannon CD, Dilernia D, et al. Clustered mutations at the murine and human IgH locus exhibit significant linkage consistent with templated mutagenesis. J Immunol 2019;203:1252
64. [43] Morgado MG, Cam P, Gris-Liebe C, Cazenhave P-A, Jouvin-Marche E. Further evidence that BALB/c and C57BL/6 γ2a genes originate from two distinct isotypes. EMBO J 1989;8(11):3245
51. [44] Choi JH, Wang K, Zhang D, Wang T, et al. IgD class switching is initiated by microbiota and limited to mucosa-associated lymphoid tissue in mice. Proc Natl Acad Sci USA 2017. Available from: https://doi.org/10.1073/pnas.1621258114. [45] Van Vollenhoven RF, Swenson CD, Soriano A, Goidl EA, et al. Serum IgD levels in mice: effects of strain, age and autoimmune disease. J. Autoimmunity. 1989;259
67. [46] Haughton G, Lanier LL, Babcock GF. The murine kappa light chain shift. Nature 1978;275:154
7. [47] Larijani M, Chen S, Cunningham LA, Volpe JM, et al. The recombination difference between mouse κ and γ segments is mediated by a pairwise regulation mechanism. Mol Immunol 2006;43:870
81. [48] Akula S, Mohammadamin S, Hellman L. Fc receptors for immunoglobulins and their appearance during vertebrate evolution. PLoS One 2015;. Available from: https://doi.org/10.1371/journal.pone.0124530. [49] Bruhns P. Properties of mouse and human IgG receptors and their contribution to disease models. Blood 2012. Available from: https://doi.org/ 10.1182/blood-2012-01-380121. [50] Holling TM, Schooten E, van Den, Elsen PJ. Function and regulation of MHC class II molecules in T-lymphocytes: of mice and men. Human Immunol 2004;65:282
90. [51] Haas W, Peirera P, Tonegawa S. Gamma/delta cells. Annu Rev Immunol 1993;11:637
85. [52] Xiong X, Morita CT, Bukowski JF, Brenner MB, Dascher CC. Identification of guinea pig γδ T cells and characterization during pulmonary tuberculosis. Vet Immunol Immunopathol 2004;102:33
44. [53] Ku¨hnlein F, Vicente A, Varas A, Hu¨nig T, Zapata A. γ/δ T cells in fetal, neonatal, and adult rat lymphoid organs. Dev Immunol 1995;4:181
8. [54] Bosc N, Lefranc M-P. The mouse (Mus musculus) T cell receptor alpha (TRA) and delta (TRD) variable genes. Dev Comp Immunol 2003;27:465
97. [55] Antonacci R, Massari S, Linguiti G, Jembrenghi C, et al. Evolution of the T cell receptor (TR) loci in the adaptive immune response: the tale of the TRG locus in mammals. Genes 2020. Available from: https://doi.org/10.3390/genes11060624. [56] Kumar A, Suryadevara N, Hill TM, Bezbradica JS, et al. Natural killer T cells: an ecological evolutionary developmental biology perspective. Front Immunol 2017. Available from: https://doi.org/10.3389/fimmu.2017.01858. [57] Van Beeck FAL, Reinink P, Hermsen R, Zajonc DM, et al. Functional CD1d and/or invariant chain transcript in horse, pig, African elephant and guinea pig but not in ruminants. Mol Immunol 2009;46(7):1424
31. [58] Killeen N. T-cell regulation: Thy-1 hiding in full view. Curr Biol 1997;7(12):774
7. [59] Haeryfar SMM, Hoskin DW. Thy-1: more than a mouse pan-T cell marker. J Immunol 2004;173:3581
8.

374

SECTION | 2 Mammalian orders

[60] Gu¨nther E, Walter L. The major histocompatibility complex of the rat (Rattus norvegicus). Immunogenetics 2001;53:520
42. [61] Naper C, Ryan JC, Nakamura MC, Lambracht D, et al. Identification of an inhibitory MHC receptor for alloreactive rat natural killer cells. J Immunol 1998;160:219
24. [62] Dammers PM, Hendricks J, Terpstra P, Bos NA, Kroese FGM. Organization of the immunoglobulin heavy and light chain loci in the rat. Comparative immunoglobulin genetics. Apple Academic Press, Toronto, ON. [63] Qin T, Zhao H, Zhu H, Wang D, et al. Immunoglobulin genomics in the prairie vole (Microtus ochrogaster). Immunol Lett 2015;166:79
86. [64] Nilsson P, Solbakken MH, Schmid BV, Orr RJS, et al. The genome of the great gerbil reveals species-specific duplication of an MHCII gene. Genome Biol Evol 2020;12(2):3832
49. [65] Brederoo P, Daens WT. The ultrastructure of guinea pig heterophil granulocytes and the heterogeneity of the granules. Cell Tiss Res 1978;194: 183
205. [66] Debout C, Quillec M, Izard J. Natural killer activity of Kurloff cells: a direct demonstration on purified Kurloff cell suspensions. Cell Immunol 1984;87:674
7. [67] Hilton HG, Rubinstein ND, Janki P, Ireland AT, et al. Single -cell transcriptomics of the naked-mole-rat reveals unexpected features of mammalian immunity. Sci Rep 2001. Available from: https://doi.org/10.1371/journal.pbio.3000528. [68] Guo Y, Bao Y, Meng Q, Hu X, et al. Immunoglobulin genomics in the guinea pig (Cavia porcellus). PLoS One 2012. Available from: https:// doi.org/10.1371/journal.pone.0039298. [69] Herrera-Alvarez S, Karlsson E, Ryder OA, Lindblad-Toh K, et al. How to make a rodent giant: Genomic basis and tradeoffs of gigantism in the capybara, the world’s largest rodent. bioRxiv 2018. Available from: https://doi.org/10.1101/424606. [70] Watkins DI, Chen ZW, Hughes AL, Lagos A, et al. Syrian hamsters express diverse MHC class I gene products. J Immunol 1990;145(10): 3483
90. [71] Smulders MJM, Snoek LB, Booy G, Vosman B. Complete loss of MHC genetic diversity in the common hamster (Cricetus cricetus) population in the Netherlands. Consequences for conservation strategies. Conserv Genet 2003;4:441
51. [72] Zhao Y, Oreskovic E, Zhang Q, Lu Q, et al. Transoposon-triggered innate immune response confers cancer resistance to the blind mole-rat. Nat Immunol 2021;22:1219
30. [73] Tian X, Azpurua J, Hine C, Vaidya A, et al. High molecular weight hyaluronan mediates the cancer resistance of the naked mole-rat. Nature 2013;499:346
9. [74] Scho¨pfer R, Figueroa F, Niˇzetı´c D, Nevo E, Klein J. Evolutionary diversification of Class II P loci in the MHC of the mole-rat Spalax ehrenbergi. Mol Biol Evol 1987;4(3):287
99. [75] Niˇzetic D, Figueroa F, Dembic Z, Nevo E, Klein J. Major histocompatibility gene complex gene organization in the mole-rat Spalax ehrenbergi: evidence for transfer of function between class II genes. Proc Nat Acad Sci USA 1987;84:5828
32. [76] Yuhki N, Beck T, Stephens RM, Nishigaki Y, et al. Comparative genome organization of human, murine and feline MHC class II region. Genome Res 2003. Available from: https://doi.org/10.1101/gr.976103. [77] Vincek V, Niˇzetic D, Golubic M, Figueroa F, et al. Evolutionary expansion of the MHC class I loci in the mole-rat Spalax ehrenbergi. Mol Biol Evol 1987;4(5):483
91. [78] Debebe T, Biagi E, Soverini M, Holtze S, et al. Unravelling the gut microbiota of the long-lived naked mole-rat. Sci Rep 2017. Available from: https://doi.org/10.1038/s41598-017-10287-0. [79] Folch G, Giudicelli V, Jean C, Ginestoux C, et al. 2005, IGMT overview: the mouse immunoglobulin heavy IGH genes. Available from: http:// imgt.cines.fr.

Chapter 24

The primates: humans and their relatives

Ring-tailed lemur: Lemur catta

Primates appear to have emerged before the K-Pg event around 65 mya. The oldest known primate fossils date from the late Paleocene around 57 mya, but molecular clock studies suggest that they diverged from the other mammalian clades closer to 87 mya. Like other mammals, once the dinosaurs were eliminated, the early primates began to expand and diversify. They probably originated as small terrestrial creatures that eventually adapted to an arboreal existence in the trees of the tropical rain forests of Africa. This adaptation would have involved the evolution of shoulder joints to permit extensive movement, manual dexterity to hold onto branches, increased visual acuity rather than olfaction, and a larger brain. The first primate divergences likely occurred about 70 mya when the lemurs, lorises, and galagoes split off to constitute the Strepsirrhines. They prospered on the island continent of Madagascar. (Madagascar separated from Africa about 160 mya so the lemur ancestors must have inadvertently rafted there!) The rest of the primates, the Haplorrhines, then progressively diverged with the tarsiers emerging around 58 mya. Some of the remaining simians succeeded in reaching South America from Africa across a relatively narrow Atlantic ocean about 35
43 mya, again, either by rafting or possibly by island hopping, and so gave rise to the New-World Monkeys. These are classified as Platyrrhines since their nostrils face sideways. Those that remained in Africa (the Catarrhines) have nostrils that face downward. These eventually diverged into the apes (Hominoidea) and the Old-World Monkeys (Cercopithecoidea) that include the macaques, the gibbons, and the baboons, about 27
30 mya (Fig. 24.1). The Hominoidea subsequently diverged about 8 mya into the gorillas, chimpanzees, and hominids. About 150,000
200,000 years ago modern humanids (H. sapiens) developed in Africa and thereafter migrated to eventually populate all the major continents. They progressively adapted bipedal locomotion, gradually enlarged their brains, and progressively lost their snouts and with this their olfactory skills. Their legs grew longer, their feet became adapted to walking long distances, they developed long fingers that could work precisely, and eventually, they learned to make tools and study immunology.

Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00017-4 © 2023 Elsevier Inc. All rights reserved.

375

376

SECTION | 2 Mammalian orders

70

Million years ago 60 50 40 30 20

10

0

Prosimians

Lemurs Tarsiers

70 Platyrrhines 37

Catarrhines

New World Monkeys Old World Monkeys Gibbons

30

Orangutans 13

Gorillas

Hominidae

Chimpanzees 5-7

Humans

FIGURE 24.1 A simplified phylogeny of the primates discussed in this chapter. The numbers indicate estimated divergence times in millions of years.

24.1

Infectious disease history

Like all mammals, humans have evolved in a microbial world. Pathogens such as tuberculosis have infected them every step of the way. As humans spread across the globe, they took their pathogens with them. The spread and significance of infectious diseases grew appreciably as agriculture developed, and human activities changed from simply being wandering hungry huntergatherers to sedentary, well-fed, village and city dwellers. Heirloom diseases such as tuberculosis and malaria and more recent pandemics such as smallpox and the plague have killed millions and exerted enormous selection pressure on human populations. Until very recently, infectious diseases were the major cause of death in all societies, including developing western societies. Thus in modern hunter-gatherers and forager-farmers without access to modern medicine, 73% of deaths were caused by infectious diseases [1]. The normal human lifespan in the Greco-Roman world was around 20
35 years
just long enough to reach puberty, breed, and raise children. Thus current human life expectancy figures in the developed West are a historical (and biological) anomaly. However, as a result, infections have become relatively less important (with the exception of coronavirus), but other aspects of the innate and adaptive immune systems such as immune-mediated diseases have become relatively more important. One feature that is largely unique to humans is the constant change in infectious disease threats associated with alterations in lifestyle and the growth in population size. Thus the effects of infectious diseases on widely scattered, sparse, populations of hunter-gatherers were not overwhelming, especially in the young. As a result, epidemic disease outbreaks likely exerted minimal selective pressure on such populations. For example, in Native Americans prior to the European invasions, the lack of such pressure left these populations highly vulnerable. With the introduction of common European diseases such as smallpox and measles, mass mortality ensued. Most Europeans, having grown up in filthy, overcrowded cities, were immune to these diseases. The human immunome has been subjected to enormous selective pressure, perhaps more than in species with much lower population densities [2]. Markers of such selection have been revealed in a number of studies since it became possible to sequence and characterize ancient DNA. For example, positive selection for the TLR1-TLR6-TLR10 gene cluster occurred in Europeans around the time of the onset of agriculture. This appears to be associated with resistance to tuberculosis and other Mycobacteria [3]. Using RNA seq, it is possible to characterize the response of cells such as monocytes to bacterial and viral stimuli. Thus the effects of TLR ligands can be measured and quantitated. For example, Quach et al. measured human monocyte responses to the ligands for TLR1/2, TLR4, TLR7/8, and influenza virus [4]. They compared the responses in Europeans and Africans and found that Europeans possessed a gene that acted to suppress some NF-κB-mediated inflammatory responses triggered by TLR1 ligation. Further analysis showed that this variant was introduced into European genomes by Neanderthals [4]. This also provides support for the concept that different human populations may exhibit very different inflammatory responses, especially if they are geographically isolated.

The primates: humans and their relatives Chapter | 24

377

One feature of immune system functionality that is very obvious in humans and other primates is its susceptibility to social status. Social status has effects on immune cell populations and cell-specific gene expression levels as well as the gene expression response to immune stimulation. For example, in Rhesus macaques, animals with high status tend to have increased CD81 T cell numbers, decreased neutrophils, and higher natural killer (NK) cell numbers [5]. NK cell numbers are especially sensitive to social status. Low-status macaques tend to develop a proinflammatory antibacterial phenotype whereas high-status animals tend to develop an antiviral phenotype as reflected by increased interferon responses [5].

24.2

Reproduction and lactation

The human placenta is of the hemochorial type in which the extravillous fetal trophoblast cells invade the uterine wall. This invasion occurs by two routes. The trophoblast cells can invade endometrial tissues directly, but they can also invade through blood vessels, - the endovascular route. They migrate through the lumen of the spiral arteries. This same process also occurs in great apes such as chimpanzees and gorillas. In contrast, trophoblast invasion of the endometrium in Old-world monkeys such as the gibbons only occurs via the endometrial route and never penetrates as deeply as that in humans and apes. This difference in the degree of trophoblast invasion is associated with the expression of the major histocompatibility complex (MHC) class I antigen, human leukocyte antigen (HLA)-C on trophoblast cells [6]. Maternal and paternal HLA-C are expressed and polymorphic in the great apes and humans but are absent in the Old-world monkeys. The interactions between HLA-C1 and HLA-C2 expressed on fetal trophoblast cells and killer cell immunoglobulin-like receptors (KIRs) expressed on maternal uterine NK cells are key to the maternal immune response and the degree of trophoblast invasion. Thus the presence of HLA-C appears to permit deeper trophoblast invasion and so increases the efficiency of placental nutrient transfer. On the other hand, this HLA-C/KIR response also appears to predispose humans and the great apes to pre-eclampsia and possibly to an increased risk of recurrent abortion [6].

24.2.1 Immunoglobulin transfer The placental barrier between mother and fetus consists of two cell layers, syncytiotrophoblast cells, and fetal capillary endothelial cells. Of the maternal immunoglobulins, only IgG can cross this barrier in significant amounts. This is mediated by the neonatal Fc receptor FcRn. The maternal IgG binds to FcRn. It is transported through the syncytiotrophoblasts within endosomes and carried to the fetal side where the physiological pH releases it into the fetal circulation. The amount of IgG transferred depends upon maternal IgG levels and the specific IgG subclass involved. Thus transfer efficiency is greatest for IgG1 followed by IgG3, IgG4, and IgG2 in that order [7]. As a result, beginning around 13 weeks of gestation, immunoglobulin G can pass to the developing fetus. The amount transferred progressively increases, eventually reaching 50% of maternal concentrations around 30 weeks post-conception [8].

24.2.2 Human colostrum and milk IgA is the predominant immunoglobulin in both human colostrum and milk. Its concentration varies widely, ranging from 1.5 to 83.7 g/L with a mean value of 32 g/L. The concentration of IgM averages about 1.13 g/L while that of IgG averages 0.53 g/L. Although immunoglobulin levels drop over time, these relative proportions do not vary significantly during lactation [9]. The level of serum immunoglobulins in newborn infants depends upon their gestational age and birth weight. However, IgG levels in full-term newborns are in the region of 750 mg/dl for IgG, and 10 mg/dl for IgM [10].

24.3

Hematology

In general, total white cell counts in monkeys are higher than those in humans, but this depends largely upon the health of the population, as well as the amount of stress an animal is under. They also depend upon female mating promiscuity (Chapter 19). In adult primates, including humans, neutrophils generally outnumber lymphocytes. Typically, the white cell count is 4
11 3 103/μL consisting of 50%
70% neutrophils and 30%
50% lymphocytes in human peripheral blood. In juvenile rhesus macaques and baboons, however, the lymphocytes may outnumber the neutrophils [11]. The blood leukocytes in owl monkeys (Aotus trivirgatus) are also reportedly different, with high numbers of neutrophils (5
14.3 3 103/μL) even in healthy animals. Healthy owl monkeys of some karyotypes have blood eosinophil counts tenfold those of other karyotypes, even in the absence of parasites (3.7 6 1.2 3 103/μL compared to 0.4 6 0.3 3 103/μL). The eosinophils of this species are reported to have cigar-shaped eosinophilic granules [12].

378

SECTION | 2 Mammalian orders

24.4

Innate immunity

Humans respond differently to infections than do other primates. Despite being related, there are great differences among the primates in their sensitivity to bacterial and viral invasion [13]. For example, humans are highly sensitive to stimulation by bacterial acetylated lipopolysaccharides. Doses of 2
4 μg/kg of these lipopolysaccharides administered to a human can provoke malaise and a fever while a slightly higher dose (15 μg/kg) can induce septic shock [13]. In contrast, baboons and macaques require almost a tenfold higher dose to induce similar responses. In these cases, pattern recognition receptors play a key role. These systematic responses suggest that the differences between species may relate to the functions of their signaling pathways. Humans generally have higher numbers of circulating blood neutrophils. In response to lipopolysaccharides in vitro, these cells show stronger phagocytic and apoptotic responses while producing fewer NETs [14]. Phylogenetic studies show that the apes mount a significantly stronger transcriptional response to both viral and bacterial infections. Presumably, this strong initial innate immune response may be more effective in preventing infection but at the cost of increased energy expenditure. Thus apes transcribe a greater number of genes in response to immune stimulation and release a very similar battery of inflammatory mediators regardless of pathogen type. This powerful response however, shifts dramatically after 24 h, perhaps to limit bystander collateral damage. This rapid but extreme response may be related to their K strategy and the longer life history adopted by the larger primates in contrast to the r strategy used by smaller primates such as macaques and baboons. With respect to the toll-like receptors, human TLR2 expression on peripheral blood leukocytes is constitutive but not on T cells. Human TLR9 expression is limited to B cells, plasmacytoid DCs, and neutrophils. Humans also have a functional TLR10 gene. Defensins are abundant in primate neutrophils [15]. They can account for up to 50% of the total weight of the neutrophil granules.

24.4.1 Acute-phase proteins Given the intensity of research into human immunology, it is unsurprising that at least 30 different proteins have been shown to increase their serum levels during acute inflammatory episodes. These include multiple complement components, proteins involved in the coagulation and fibrinolytic systems, antiproteases, transport proteins, and inflammatory mediators. Several proteins such as albumin, transferrin, and transthyretin drop during these episodes. These changes are in addition, to multiple other alterations including neuroendocrine, hematopoietic, and metabolic changes. Not all acute-phase responses have identical dynamics. Thus C-reactive protein and serum amyloid A can rapidly increase by 30,000%! However, most of the others such as C3, haptoglobin, and fibrinogen, merely double or triple their levels (Fig. 24.2) [16].

FIGURE 24.2 An example of the acute-phase response in humans. From Gitlin JD, Colton HR. Molecular biology of the acute-phase plasma proteins. In Piclk E, Landy M, editors. Lymphokines, vol. 14. San Diego, CA: Academic Press; 1987, p. 123
53. With Permission.

30,100 30,000 700 % change in 600 concentration 500

Serum Amyloid A

C-reactive protein 400 300 Haptoglobin

200

Fibrinogen

100 0

Albumin Transferrin 0

I 7

I 14 21 Days after inflammatory stimulus

The primates: humans and their relatives Chapter | 24

24.5

379

Lymphoid organs

24.5.1 Thymus The human thymus is located in the same position, the anterior mediastinum, and serves the same functions as in other mammals, namely the production of self-tolerant T cells. As in other mammals, it reaches its maximum relative size at birth and maximum absolute size at about 10 years of age, prior to puberty. Thereafter it undergoes progressive involution. There is a direct correlation between thymic function and naı¨ve T cell numbers [17]. However, while involuting, it continues to produce hormones and plays a role in maintaining immune efficiency in the adult human. While it has long been assumed that thymic atrophy was caused by the sex hormones generated at puberty, studies have shown that thymic decline is phasic. Thus it shrinks in the first few weeks after birth and this then proceeds at a steady rate of about 3%/ year until middle age. After that, the atrophy slows somewhat, occurring at about 1%/year [17]. The human thymus, in contrast to the mouse thymus, does not decrease in volume during aging [18]. It actually maintains its apparent size but inside it is a different situation. The thymic epithelium gradually shrinks to be replaced by an enlarged perivascular space occupied by adipocytes and mature T cells. While the normal human thymus is almost exclusively populated by immature T cells undergoing both negative and positive selection, it also contains a population of mature CD81 effector cells within its perivascular space. These cells are exceedingly rare in newborn infants, but their numbers grow as an individual ages and the thymus atrophies [19]. Thus the human thymus may be considered to be a chimeric organ with both peripheral T cells and central T cells
the latter form the true thymus. The thymic tissues shrink with age, but they are surrounded by infiltrating cytotoxic T cells and adipocytes. It is entirely possible that these cytotoxic T cells may play a role in reducing thymopoiesis and so promoting thymic atrophy.

24.5.2 Spleen As in other mammals, the human spleen consists of white pulp embedded in red pulp [20]. The large white pulp consists of lymphocyte accumulations that form periarteriolar lymphoid sheaths with attached follicles surrounding the central arteries. The red pulp consists of reticular connective tissue full of diverse cell types but especially erythrocytes. Unlike laboratory rodents, however, there is no marginal zone in the human spleen. Splenic secondary follicles consist of three zones, a central germinal center surrounded by a mantle zone and a superficial zone. The arterioles and sheathed capillaries of the red pulp are surrounded by predominantly B cells. The human sheathed capillaries are related to the splenic ellipsoids seen in other mammals. The sheaths consist of endothelial cells, pericytes, and special stromal sheath cells as well as macrophages and B cells. As discussed in Chapter 11, these ellipsoids are very effective at removing particulates such as immune complexes from the blood. The splenic circulation system in the human is totally open since there are no direct connections from the capillaries to the sinuses. Hematopoiesis in the human spleen ends prior to birth. While the human spleen (and the spleens of Old-world monkeys) as described above is of the sinusal type with venous sinuses and has a predominantly defensive role, this is not the case with all primates. Thus the spleens of New World monkeys are nonsinusal and they have smaller, nonanastomosing veins [21].

24.5.3 Mucosal lymphoid tissues Humans have a “ring” of lymphoid tissue, called Waldeyer’s ring surrounding their pharynx. It consists of the nasopharyngeal tonsil (or adenoid) on the roof of the pharynx, paired tubal tonsils at the pharyngeal openings of the Eustachian tubes, paired palatine tonsils in the oropharynx, and a lingual tonsil in the posterior third of the tongue. Each consists of aggregations of lymphocytes located in the lamina propria of the pharyngeal wall, as well as subepithelial lymphoid follicles (Fig. 24.3). They are also covered by epithelium consisting of narrow crypts that considerably increase the surface area of the epithelium. The tonsils may be polycryptic or monocryptic. The nasopharyngeal and tubal tonsils are covered by ciliated respiratory epithelium whereas the palatine and lingual tonsils are covered by stratified squamous epithelium [22]. The Peyer’s patches of baboons and humans are very similar morphologically to those in rodents [23]. The number of Peyer’s patches containing more than five lymphoid follicles varies from about 240 at puberty to 60 by the age of 30. The number of very large patches containing over 25 follicles ranges from over 40 at birth to more than 100 at puberty [24]. In general, the patches are larger and most numerous in the ileum where some may reach 10 cm in length. As in other mammals, their domes are covered by epithelium containing numerous M cells. However, B cells surround the mantle zone in all three species. In rodents they express IgM and IgD, in baboons and humans they express IgM or

380

SECTION | 2 Mammalian orders

squamous epithelium

Tonsillar crypt

FIGURE 24.3 The structure of the palatine and lingual tonsils. Lymphocytes can infiltrate the epithelium either by vascular or non-vascular routes. Note how close these lymphoid tissues are to the epithelial surface, so they readily interact with infiltrating antigens. However, the crypt with its relatively thin epithelium also represents a vulnerable point to microbial invasion.

Germinal center

IgA. Thus in rodents all the cells surrounding the central follicle including the dome express IgD. In baboons and humans, there are very few IgD-positive cells in these locations. As in other mammals, there are many more lymphocytes in the primate intestine than in the peripheral lymphoid system. For example, in the rhesus macaque (Macca mulatta), most of the intraepithelial and lamina propria lymphocytes are B cells with relatively few T cells. 63%
80% of their intraepithelial lymphocytes (IELs) are CD81 T cells with very few CD41 cells. CD41 cells are present in greater numbers in the lamina propria but CD81 cells still predominate. As many as 38% of the IELs are γ/δ-positive T cells but these cells are rare in the lamina propria and mesenteric lymph nodes. The gut-associated lymphoid tissues of the rhesus macaque are thus very similar to those of humans [25]. As discussed in Chapter 11, both humans and great apes, possess a vermiform appendix. There is however, some difficulty in defining this organ. While it is typically thin and narrow bored with a thick wall, other primates also possess lymphoid tissues in the terminal portion of their cecum although it may not be as thin as the typical human organ [26]. Such cecal appendages have been described in vervet monkeys (Chlorocebus aethiops) and subadult macaques. Thus the definition of an appendix is somewhat subjective. Appendix-like structures have also been described in some strepsirrhines. For example, common marmosets (C. jacchus) have cecal diverticula that contain lymphoid follicles.

24.6

Major histocompatibility complex

24.6.1 Humans Collectively known as the HLA system, the human MHC is a 3 mb gene complex located on the long arm of chromosome 6 (p21.3). As in other mammals, it is divided into three regions arranged in the order class I, class III, and class II. The class I and class II regions contain the genes encoding the key antigen-presenting molecules that trigger adaptive immune responses. The class III region contains diverse genes, many of which encode products involved in immunity and inflammation (Fig. 24.4).

24.6.2 The MHC class I region HLA class I molecules are heterodimers that consist of an antigen-binding alpha chain associated with a β2-microglobulin chain (β2-M). β2-M is encoded by a separate gene found on chromosome 15. They are divided into classical and nonclassical molecules based on their functions and degree of polymorphism. The three major human classical polymorphic class I gene loci are HLA-A, HLA-B, and HLA-C. They encode receptors of about 41 kDa expressed on most nucleated cells. These receptors bind to small peptides 8
10 amino acids in length, produced within a cell by the processing of antigenic proteins. They are found on almost all nucleated cells and the processed peptides they present are derived from endogenous antigens such as those made by viruses. These peptides are recognized by the class I MHC molecules expressed on CD81 cytotoxic T cells. In addition, to being antigen receptors, these class I molecules can

The primates: humans and their relatives Chapter | 24

DP

B A B A

Centromere

II

DN

DM

A

A B

DO

B

II

DQ

FIGURE 24.4 A simplified view of the organization of the MHC-HLA system with specific reference to the MHC class I and class II regions. The empty boxes are pseudogenes.

DR

B A B A

381

B B B A

I

III

B C

E

A G F

also act as ligands for the KIRs found on NK cells and some innate T cells. HLA-C is not only a major determinant of NK cell activity but it also plays a key role in successful pregnancy and placentation (Chapter 2). The nonclassical class I HLA loci in humans are designated HLA-E, -F, and -G. While their products also form heterodimers between the alpha chains and β2-M and are expressed on cell surfaces, they have restricted tissue expression and specialized functions. They primarily act as ligands for NK cell receptors. They are not highly polymorphic but show a pattern of variation similar to other immune system genes.

24.6.3 The MHC class II region HLA class II molecules are also heterodimeric cell-surface receptors formed by paired alpha and beta chains. They also act as receptors for processed antigenic peptides. Both of these peptide chains are bound to the cell membrane unlike the class I genes. They bind processed peptides that are then presented to CD41 helper T cells and so trigger antibody responses. These peptides are usually between 9 and 30 amino acids in length. The three major class II loci expressed in humans are designated HLA-DP, HLA-DQ, and HLA-DR. In addition, the class II region contains genes encoding the minor class II-like molecules HLA-DO and HLA-DM. These minor proteins are involved in the transportation of processed peptides to cell surfaces DO is primarily expressed in B cells where it acts as an inhibitor of DM. The HLA class II region also contains genes encoding HLA-DPA and HLA-DPB. These form the DP heterodimer; as well as HLA-DQA and HLA-DQB genes that encode the DQ heterodimer. The DPA and B loci are located at the centromeric end of the HLA locus, distal from the DR and DQ-encoding loci. HLA-DRA is different from the other genes since it has very limited polymorphism. However, it can pair with one of the β-chains encoded by DRB1, DRB3, DRB4, or DRB5. (There can be no more than three functional DRB loci in any individual and sometimes an individual may possess two copies from the same locus). DRB1 is especially ubiquitous. All of these HLA class I and II genes encode highly polymorphic antigen receptors. There are no dominant or even wild-type alleles. These genes are by far the most diverse in the human genome reflecting their roles in antigen presentation and the importance of balancing selection. They are among the fastest evolving genes in the human genome as well as in other mammals. The three classical class I genes encode over 4000 identified allelic proteins each resulting from multiple amino acid substitutions. These substitutions are concentrated in the antigen-binding sites of the α1 and α2 receptor domains since this is what determines the antigen-binding specificity of these MHC molecules. Among the class I genes, there are about 2500 known HLA-A alleles that encode about 1750 different functional transmembrane proteins. HLA-B has more than 2000 different alleles. Over 5600 alleles and 3400 protein variants have been identified for HLA-C. Among the class II genes, at least 1000 alleles have each been identified for DPB and DQB while about 2600 have been identified for DRB. The number of potential alpha chain alleles is much smaller ranging from 95 in DQA to a mere 7 in DRA. (While these numbers of alleles appear very large with respect to the other mammals discussed in this book, it must be pointed out that this is very much determined by how hard we look. Much more effort has been put into identifying HLA alleles than in other mammals. Even in humans, about 40% of these alleles are unique and have only been identified in single individuals).

382

SECTION | 2 Mammalian orders

24.6.4 The MHC class III region Multiple HLA class III genes are located in the 330 kb region located between the class I and class II regions. It is the most gene-dense region of the human genome [27]. 60 genes have been identified in the region but only a few have been defined and their function elucidated. It contains 21 structural genes, one pseudogene, and two partially duplicated gene fragments [28]. All the structural genes in humans have orthologs in mice. They have diverse functions, not all of which are related to the immune system. Thus from 50 to 30 there are gene clusters encoding regulatory receptor proteins such as RAGE, complement components (C2, C4, and factor B), several heat-shock proteins, and inflammatory cytokines such as TNF-α and the lymphotoxins. Many of these proteins are expressed by hepatocytes and macrophages. There are also interesting patterns of genetic variation in this region. For example, most of the C4 genes contain a 6.36 kb endogenous retrovirus sequence, HERV-K(C4). The four consecutive genes RP, C4, CYP21, and TNX also form a module. Two-thirds of human HLA haplotypes have two modules, but the remainder may have either one or three [28]. A similar pattern of modular duplication is seen in the mouse.

24.6.5 Great apes 24.6.5.1 MHC class I While humans possess the three classical Class I gene loci, HLA-A, -B, and -C, orthologs are present in our closest relatives, the chimpanzees (Pan troglodytes) and bonobos (Pan paniscus). These are also designated -A, -B, and -C with the prefix either Patr- or Papa-. Some chimpanzees have a fourth class I locus designated Patr-AL. This may encode a nonclassical MHC protein. Gorillas (Gorilla gorilla) also have three class I loci, Gogo-A, Gogo-B, and GogoC. Some gorilla haplotypes lack Gogo-A but have a gene related to Patr-AL in the same location [29]. Some may also have two copies of the Gogo-B gene, a duplication that occurred around the same time as that of the C gene [30]. Orangutans (Pongo pygmaeus) also have Popy-A, -B, and -C MHC class I genes. Their polymorphic Popy-A gene is related to Patr-AL. They may also have two or more copies of the Popy-B gene, and the Popy-C gene may be present or absent. This Popy-C gene clearly arose by duplication of a Popy-B gene in the ancestor of orangutans and humans about 21
28 mya [31].

24.6.5.2 MHC class II The equivalent of HLA-DP, -DQ, and -DR class II loci are present in all the great apes. Thus they each possess two DPA, DPB, DQA, and DQB genes in each haplotype. Functional transcripts of DPA/DPB and DQA/DQB have both been detected. The great apes also show marked DR variation, for example, chimpanzees have nine different DR haplotype configurations. Thus they can have one DRA gene and two to five DRB genes or pseudogenes.

24.6.6 Old World monkeys 24.6.6.1 MHC Class I The complete MHC class I genome of rhesus macaques (M. mulatta)has been sequenced and 51 class I genes have been identified. In most of the Old World monkeys studied, homologs of HLA-A and -B are present but not -C. However, their class I region tends to be much larger and more complex than the HLA class I region. Thus in the rhesus macaque, one haplotype can contain two major class I genes and up to five minor transcribed genes [29]. In this species, they are named Mamu-A1 to -A7. The cynomolgus macaque (M. fasicularis) has eight expressed class I genes. Generally, the Mafa-A gene is most polymorphic. The minor genes show much lower levels of transcription and expression compared to Mafa-A. A similar situation occurs in the Mamu-B genes where there may be up to six major and ten minor transcribed genes. These different gene products may also show different expression patterns among different cell types. Fukami et al. have determined that rhesus monkeys do not have paired MHC-B and -C genes but many repeated genes that are very similar to MHC-B [31].

24.6.6.2 MHC Class II Old World monkeys possess DP, DQ, and DR loci. Likewise, they possess two DPA, and DPB genes but only single DQA and DQB genes. DQA2 and DQB2 genes are deleted [29]. Monkeys such as baboons and macaques have only one DRA gene but can have variable numbers (2
6) of DRB genes or pseudogenes.

The primates: humans and their relatives Chapter | 24

383

24.6.7 New World monkeys 24.6.7.1 MHC Class I Evidence suggests that homologs of HLA-A are absent from New World monkeys. They do possess however, an equivalent to HLA-B, and this shows different degrees of expansion in different species [29]. On the other hand, most of these primates show expansion of their HLA-G-like genes. This is a nonclassical gene in humans but in the New World monkeys appears to have taken on the role of a functional class Ia gene.

24.6.7.2 MHC Class II New World monkeys possess two DPA, DPB, DQA, and DQB genes in each haplotype. However, in common marmosets (Callithrix jaccus) only a single DR haplotype configuration has been reported
one Caja-DRA and three DRB genes, one of which is a pseudogene.

24.6.8 Nonclassical MHC class I genes 24.6.8.1 Humans The genes encoding HLA- E, -F, and -G are located on the telomeric side of HLA-A. There are 84 HLA-E alleles, 44 HLA-F alleles, and 69 HLA-G alleles. Together they encode 15 HLA-E, 6 HLA-F, and 19 HLA-G proteins [29]. These nonclassical HLA genes, HLA-E, -F, and -G are present in the great apes as well as in Old- and New World monkeys. As in humans, these are less polymorphic than the class Ia genes.

24.6.8.2 Human leukocyte antigen -E The HLA-E gene is the most conserved and oldest of all the primate class I genes [32]. Its product has very limited polymorphism and low levels of cell-surface expression, but it is expressed on all cell types examined. Studies of its nucleotide sequences from cotton-top tamarins (Sanguinis oedipus) show that synonymous substitutions greatly outnumber nonsynonymous substitutions within the exons that form its antigen-binding site. There must have been significant evolutionary pressures to retain the structure of what is a very ancient protein and suggests that whatever its function is, it is critical [32]. HLA-E shares many characteristics with the mouse nonclassical H2 gene, Qa-1. Both genes encode molecules with extensive tissue distribution, low cell-surface expression, and limited polymorphism. These proteins also share structural features within their antigen-binding groove. The role of the HLA-E protein is to present peptides cleaved from MHC class Ia leader sequences. They can also present pathogen-derived peptide sequences to certain CD81 T cells. As a result, HLA-E has a very highly conserved antigen-binding groove. The HLA-E protein is expressed on cell surfaces and functions as the ligand for the inhibitory NGG2A/B receptors or their activating NKG2 C receptors located on NK cells [33]. As a result, this surface HLA-E reflects the amount of MHC synthesis occurring within a cell and enables NK cells to detect changes caused by infections or other stressors. These NK cells can also detect an absence of MHC expression and react accordingly. There are two MHC-E genes in gorillas and chimpanzees and one in orangutans and macaques. These genes are expressed at both the population and individual levels. Despite these differences, humans, and macaques express MHCE proteins at similar levels across immune cell subsets and present identical viral peptides to CD81 T cells. Thus this protein is very much functionally conserved across all these species.

24.6.8.3 Human leukocyte antigen -F HLA-F is also a ligand of NK cell receptors. It interacts with leukocyte immunoglobulin-like receptors 1 and 2 (LILR1, LILR2). In at least some Old World monkeys, MHC-F is a pseudogene and its functions have been taken over by a gene designated MHC-AG. Gorillas also have three Gogo-F genes encoding three proteins. There are six Papa-F genes in chimpanzees also encoding three proteins. In the old World monkeys, the number of MHC-F genes ranges from 33 in the rhesus macaque to four in the olive baboon (P. anubis), and one in the yellow baboon (Papio cyanocephalus) [33].

24.6.8.4 Human leukocyte antigen -G HLA-G is a receptor protein expressed on the extra villous trophoblast cells of human and gorilla placentas where it interacts with uterine NK cells. Gorillas and chimpanzees also have a single MHC-G/-AG gene. Its product plays a role

384

SECTION | 2 Mammalian orders

in preventing immune rejection of the fetus during pregnancy by acting as a potent inhibitory receptor on trophoblast cells. In the common marmoset, however, MHC-G appears to have retained its antigen-presenting role with 109 Caja-G alleles distributed through 19 different lineages. In this species, it acts as a ligand for LILRB1 and B2 as well as KIR2DL4, an NK cell receptor. The numbers of MHC-G/-AG genes vary from 10/36 in the Cynomolgus monkey to 0/5 in the olive baboon. The New World monkeys are somewhat unusual in that they have 4
11 MHC-F genes and from 1 to 19 MHC-G genes. A notable exception is the common marmoset that has 3 MHC-E, 11 MHC-F, and 109 MHC-G genes. The latter encode 99 proteins.

24.6.9 Natural killer cell receptors As in other mammals, primate NK cell receptors are encoded by genes located within two gene complexes, the NK complex (NKC) or the leukocyte receptor complex (LRC). The human LRC is located on chromosome 19q13.4 and is by far the most important of the two. It contains about 30 Ig-superfamily genes. Of these 15 encode KIRs, 13 encode LILRs, and two code for LAIRs. Their products, KIRs have two or three Ig domains, LILRs have two or four, and LAIRs have a single Ig domain [34]. The human NKC is located on chromosome 12p13.1. It spans 2.8 mb and contains 15 C-type lectin receptor genes. Eight of these appear to be functional [35]. NK cell functions in humans are primarily regulated by the KIRs whose ligands are polymorphic MHC class I molecules on potential target cells [36]. The human KIR cluster consists of a family of 15 KIR genes within the LRC (Fig. 24.5). Some of these are highly polymorphic while others are less so. The KIR locus is flanked on its centromeric side by the LILR gene family and on the telomeric side by the FCAR gene that encodes the IgA Fc receptor and by NCR1, the natural cytotoxicity receptor 1 of NK cells (NKp46). The proximity of the KIR genes means that crossingover is facilitated and this can generate new KIR haplotypes. In the human genome, the acquisition of the KIR gene family has been accompanied by the inactivation of the single human Ly49 gene. Primate KIR genes can be classified into four phylogenetic lineages (1, II, III, and V). In humans, there are two lineage 1 genes, two lineage II genes, nine lineage III genes, and a single lineage V gene. All of these are highly polymorphic ranging from over 100 alleles in some to as few as 16 in others [37]. In the human KIR locus are two KIR haplotypes that are functionally different and so subject to balancing selection. The group A KIR haplotypes resemble chimpanzee KIR and are enriched with genes that bind HLA class I strongly. Conversely, group B KIR haplotypes are enriched with genes that bind HLA class I poorly. Type B haplotypes appear to favor reproductive success whereas type A haplotypes favor successful immune defenses [38].

24.6.10 Killer cell immunoglobulin-like receptor ligands The expression of KIRs on their surface enables NK cells to detect perturbations in MHC class I expression by body cells [39]. While only a subset of HLA-A and -B allotypes can function as KIR ligands, all HLA-C allotypes can do so. HLA-C is therefore unique in that it appears to be the most significant of the KIR ligands. For example, HLA-C1 and -C2 are recognized by the inhibitory KIR2DL2/3 and KIR2DL1 NK cell receptors respectively
both belonging to lineage III. These interactions between the KIR genes enable NK cells to respond to changes in cell surface expression of HLA-C [37]. Their ligands, HLA-C molecules can be divided into two groups depending on which of their epitopes binds a KIR. Thus C1 and C2 proteins are recognized through a dimorphism at position 80 in their alpha1 domain. Evidence currently suggests that the C gene evolved as a result of duplication of a B allele that already had that C1 epitope, so it probably was present in the ancestral primate. Some human KIRs can also recognize viral ligands. For example, the activating receptor KIR2DS2 can recognize peptides derived from the hepatitis C virus as well as from flaviviruses such as dengue and zika [40]. Some members of the other MHC class I families, HLA-A, and -B may also be recognized by the highly polymorphic KIR receptors. These KIRs have coevolved with the epitopes on the MHC class I molecules. This is especially obvious in the chimpanzee. Some human NK cell receptors are also encoded by genes located within the NKC (Fig. 24.5). These include nine KLR and five CLEC (calcium dependent C-type lectin) receptors that specifically recognize HLA-E. Thus the recognition of HLA-E by NK cells is also highly conserved. This class Ib cell-surface protein is the ligand of the NKG2A/ CD94 heterodimer.

The primates: humans and their relatives Chapter | 24

385

FIGURE 24.5 The two clusters of NK cell receptor genes in humans. Note that the human Ly49 gene is a pseudogene. Some interactions between NK cells and their targets are highly specific while others recognize diverse ligands.

Target cell

Target cell

HLA-A, -B, -C Diverse KIRs

HLA-E Conserved CD94:NKG2A

NK cell

HUMAN LRC LILRA4 LAIR1 LAIR2 KIR3DX1 LILRA2 LILRA1 LILRB1 LILRB4 LILRP1 LILRP2

KIR GROUP

FCAR NCR1

HUMAN NKC GABARAPL1 CD94 NKG2D 3DL3 2DS2 2DL2 2DL5 2DLS3 2DP1 2DL1 3DP1 2DL4 3DS1 2DL5 2DS3 2DS1 3DL2

NKG2F NKG2E NKG2C NKG2A Ly49L MAGOHB

24.6.11 Great apes As pointed out above, the duplication of the MHC-C locus in the great apes that generated MHC-C1 and -C2 probably occurred about 21
28 mya. The duplication of the MHC-B locus probably occurred in macaques around the same time. Because of this close relationship, the expressed C1 structure is also present on some MHC-B allotypes. Consequently, MHC-C and some cross-reactive B proteins are present in the great apes. As C1-bearing MHC-C products diversified from their partner MHC-B, the prime target of NK cells changed to MHC-C using its cognate receptors, lineage III KIR. Evolution to MHC-C2 from -C1 and fixation of MHC-C drove further changes in the NK cell MHC-C-specific KIRs. The presence of MHC-C1 in hominids has driven the subsequent expansion of lineage III KIR in these species. Half the orangutans tested lack a Popy-C gene. The others possess a single C1 gene. Orangutans also lack both C2 genes as well as C2-specific KIRs. All humans and chimpanzees express HLA-C suggesting that the orangutan may be an evolutionary intermediate between the C-negative catarrhines and the C-positive chimpanzees. This difference is reflected in the KIR genes of orangutans. They possess only two lineage II genes in their genome, but they have four lineage III genes. These four gene products recognize MHC-C while the lineage II gene products recognize MHC-A and -B. The paired KIRs have complementary functions, one being activating and the other inhibitory. As pointed out previously, the expression of MHC-C on trophoblast cells may permit deeper penetration by fetal villous trophoblast cells into the uterine myometrium and hence improve the nutrition of the developing fetus in these species [6]. Nine lineage III KIR receptors are expressed on chimpanzee NK cells. Five are specific for C2 while three are specific for C1. On the other hand, chimpanzees have only a single lineage II KIR receptor that binds to MHC-A and -B.

386

SECTION | 2 Mammalian orders

Interestingly, the bonobo lacks an MHC-C1 gene. It has one lineage I, one lineage II and two lineage III KIR genes. Gorillas have one lineage II KIR gene, and seven lineage III KIR genes (encoding one activating and six inhibitory receptors) [37]. Analysis suggests that this diversity in KIR genes and their products among different primates is the result of recombination in which whole domains are shuffled [41]. Within the complex, the primate LRC is largely conserved. One exception is however, the common marmoset where there has been a substantial contraction in the number of LILR genes. There is also only a single copy of the KIR3DX1 gene in primates and great apes. This has been duplicated in Old- and New World monkey species [42].

24.6.12 Old World primates The Old World primates possess multiple KIR genes that group into the four lineages I, II, III, and V. These are expressed on NK cells in all species examined. Rhesus macaques have 22 KIR genes that are differentially distributed through their populations so that different haplotypes may have from 4 to 15 KIR genes. This diversity is entirely due to 19 KIR genes in lineage II. Lineages I and V are each represented by a single gene, while lineage III has only a pseudogene. Correlating with this expansion of KIR genes and their products in the rhesus macaques, there has also been a corresponding expansion of their ligands, the MHC class I molecules. Thus there are four MHC-A and 19 MHC-B genes in this species. There is no equivalent of HLA-C in any Old World monkeys including the rhesus macaque. Thus in the Old World monkeys, diversification has focused on MHC-A and MHC-B and their receptor, lineage II KIR. Gibbons lack MHC-C and MHC-G and their NK cells express distinctly different KIRs [6]. Rhesus macaques also use genes located in their NKC complex to encode other NK receptors. Their NK cells express NKp80, NKG2A, and NKG2D, at similar levels as humans. NK cells in rhesus macaques are CD32 and can be divided into a major cytolytic population that expresses high levels of the Fc receptor CD16 and a minor population that expressed low levels of CD16. Macaque NK cells but not human NK cells express high levels of CD8α [43]. Rhesus macaques do not express MICA but they do possess many repeats of MICB [31].

24.6.13 New World primates The New World primates have their own unique KIR lineage called VI. Thus the spider monkey (Ateles) has five functional KIR genes with a pseudogene at the centromeric end and FCAR at the telomeric end. The KIR genes in the centromeric portion of the locus are in inverted orientation. In the owl monkey (Aotus) there is an additional KIR gene at each end of the locus. These owl monkey KIR proteins have four Ig-like domains rather than the usual three.

24.6.14 Prosimians The gray mouse lemur (Microcebus murinus) a prosimian, has a lymphocyte receptor locus containing three KIR genes. Two are pseudogenes and only one is functional. Unlike other primates, the most diverse NK cell receptors in this species are not KIRs, but killer cell lectin-like receptors (KLRs) encoded by genes located in the NK complex. Here, their KLR, CD94:NKG2A receptor is highly conserved and shows no polymorphism whatsoever. In the gray mouse lemur, however, there are three divergent CD94 genes whose products differ in amino acid sequence by 23%
24%. These KLR genes are also polymorphic. It also has four active MHC class II genes and four inactive class I genes within its MHC. Six active MHC class I genes have been moved to another chromosome. Lemurs also lack MHC-E the ligand of nonpolymorphic KLRs in higher primates [44].

24.7

B cells and immunoglobulins

24.7.1 Humans In primates, B cells with diverse antigen-binding receptors are produced by the bone marrow throughout life. For obvious reasons, it remains unclear to what extent human B cell diversity depends upon input from the commensal microbiota. It is apparent however, that the microbiota does play a significant role in the development of allergies and autoimmune diseases. They also influence class switching to IgD. The human IGH locus is located on chromosome 14 at band 14q32.33 at the telomeric end of its short arm. It is in a reverse orientation where the 50 end is telomeric and the 30 end is centromeric. It occupies 1.25 mb and contains

The primates: humans and their relatives Chapter | 24

IGH

5’

123-129

27

3’

9 M

IGK

IGL

76

D

G3

G1

A1

G2

G4

E

A2

387

FIGURE 24.6 The organization of the human immunoglobulin heavy and light chain loci.

5

73-74 7-11 V-J clusters

M

D

G

E

A

V

D

J

LC

170
176 genes (excluding orphons) (Fig. 24.6). Humans use gene recombination to generate most of their antibody diversity. Additional diversity is generated by base deletion and insertion and by somatic mutation. The IGH locus is one of the most complex and segmentally duplicated regions of the human genome. Thus 15 repetitive IGHV gene-containing segments showing high sequence similarity and ranging in size from 4 to 24 kb are found at multiple locations across the locus. One of these, about 10 kb in size and containing genes from the IGHV3 and IGHV4 subfamilies, is found in the IGH sequence 11 times. These duplications are a major cause of the many insertion/deletion variants found in the region. Even within the IGHC cluster, six types of multigene deletions, 150 kb in size, have been described in healthy humans. Most 50 IGHV genes are located very close to the 14q telomere while the IGHC genes are centromeric. Thirty-five IGH genes have been located in other chromosomal locations. These are designated as orphons since they cannot contribute to heavy chain diversity. Nine IGHV orphons and ten IGHD orphons have also been found on the short arm of chromosomes 15 and 16, respectively. IGHV orphons have also been located on the long arm of chromosome 16. One IGHC processed gene, IGHEP2 is found on the long arm of chromosome 9. (Both functional genes and ORFs must have full ORFs. In addition, functional genes must not have any described mutations in splice sites, recombination signals, or regulatory elements. Genes are classified as ORFs if they have any such mutations) [45].

24.7.2 Immunoglobulin heavy chains Humans have 11 IGHC genes of which nine are functional. They encode in order; IgM, IgD, IgG3, IgG1, IgA1, IgG2, IgG4, IgE, and IgA2. In addition, they have two pseudogenes, one an IGHE pseudogene (IGHEΨ), and one an IGHG pseudogene (IGHGψ).

24.7.3 IgM Human IgM is a pentamer of five immunoglobulin units linked by interchain disulfide bonds. As a result, it has ten antigen-binding sites. What these lack in affinity they make up in numbers. IgM accounts for up to 10% of the immunoglobulins in human blood.

24.7.4 IgD Despite the fact that it has been evolutionarily conserved, the functions of IgD in humans remain obscure. As B cells develop, they undergo alternative mRNA splicing so that they can express both IgM and IgD antigen receptors with identical specificities at the same time (Fig. 24.7) [46]. These B cells appear to respond especially well to polymeric

388

SECTION | 2 Mammalian orders

B cell receptor B cell signaling in association with IgM Responds to multimeric antigens Driven by antigens from the microbiota

Soluble immunoglobulin Basophil and mast cell binding and activation

IL-1, IL-4, BAFF Regulates the microbiota FIGURE 24.7 Some of the apparent functions of human IgD. Thus its production is stimulated by the microbiota and it appears to play a role in their regulation. It may also play a role in type 2 immune responses by binding to mast cells and prompting their production of type 2 cytokines.

repetitive antigens and the microbiota appear to play an important role in driving class switching to IgD. The role of IgD has so far defied explanation but it probably regulates B cell responses. IgM to IgD class switching has been described in the upper respiratory mucosa of humans. This generates IgD-producing plasma cells whose products bind respiratory tract bacteria. Thus in humans, IgD probably orchestrates a mucosal defense system at the interface between innate and adaptive immunity. IgD only constitutes about 0.25% of the total circulating immunoglobulins. This amounts to 3
40 μg/mL. It has a half-life of 2.8 days in humans. Its synthesis rate is at least 10 times lower than the other immunoglobulin classes [47]. Circulating IgD can bind to basophils and mast cells through the CD44 binding protein, galectin 9. This induces them to produce the Th2 cytokines, IL-1, IL-5, and IL-4 [48]. In other studies, cross-linked IgD induces human mononuclear cells to produce cathelicidins as well as IL-4, IL-13, IL-21, and B cell-activating factor [49]. As a result, this amplifies IgG1 and especially IgE production by B cells.

24.7.5 IgG As in other mammals, IgG is the most abundant immunoglobulin isotype and accounts for up to 20% of the plasma proteins in normal humans [50]. It consists of four subclasses named in order of decreasing abundance IgG1, IgG2, IgG3, and IgG4. These are more than 90% identical at the amino acid level but they each have unique biological properties including antigen binding, complement activation, effector cell triggering, half-lives, and placental transport. The four IgG subclasses while highly conserved, differ in their constant regions. These differences are primarily located in their hinge regions and upper CH2 domains. It is this region that is involved in binding to Fc receptors as well as to the first component of complement, C1q. As a result, these subclasses have different effector functions. It is also relevant to note that it is in this region that many species-related differences occur among other primates as well [51]. Human IgG3 has a hinge region of 62 amino acids that is not found in other primates [52]. Human IgG2 has a shortened hinge region. B cell responses to soluble protein antigens generally induce the production of IgG1, the most abundant of the IgG subclasses. On the other hand, the antibody response to bacterial capsular polysaccharides may be completely restricted to IgG2. IgG3 antibodies are highly effective proinflammatory antibodies. Viral infections generally stimulate IgG1 and IgG3 responses, but IgG3-dominated responses are rare. Allergens often stimulate an IgG4 response in addition, to IgE. This is especially prominent following chronic antigenic stimulation.

24.7.6 IgA IgA accounts for about 15% of the serum immunoglobulins in humans. As in other species, it is the predominant immunoglobulin class in external secretions. In serum, IgA is mainly monomeric but in secretions it is polymeric (mainly dimeric, secretory IgA) consisting of two or more subunits connected by a joining chain. Because of the large amount of mucosal lymphoid tissues, there is more IgA produced by the body than any other immunoglobulin class. Humans, chimpanzees, gorillas, and gibbons possess two IGHA constant region genes and as a result produce two IgA subclasses, IgA1 and IgA2. Other primates such as the orangutan and the Old World rhesus and cynomolgus macaques only have one. While they clearly arose through gene duplication, the two IgA subclasses are structurally and

The primates: humans and their relatives Chapter | 24

389

biologically different. The major differences lie in their hinge regions [53]. They also differ in their biological properties as a result of different binding and signaling properties. They also have different glycosylation profiles. IgA1 constitutes about 80% of the total serum IgA. It has an extended hinge region due to the insertion of a duplicated peptide sequence. Its hinge region contains multiple O-linked glycans and two N-linked glycosylation sites. IgA1 contains significantly more sialic acid than IgA2 As a result of this longer hinge region however, it is more susceptible to cleavage by proteases produced by Streptococci, Neisseria, and Haemophilus. It is primarily directed against protein antigens. Most lymphoid tissues contain a predominance of IgA1 secreting plasma cells. IgA2 lacks a disulfide bond in its CH1 domain that normally binds it to the light chain. IgA2 also has a shortened hinge region with four N-linked glycosylation sites. Thus the light chains in this isotype are noncovalently bound to the heavy chain [54]. IgA2 accounts for about 20% of human serum IgA but about 35% of the IgA in secretions on mucosal surfaces, and in tears. It is primarily directed against polysaccharides and lipopolysaccharides. Polysaccharide antigens tend to induce more IgA2 than do protein antigens. IgA2 also has proinflammatory effects on neutrophils and macrophages by stimulating their production of activating cytokines. whereas IgA1 does not. These functional differences appear to be due to their differences in glycosylation since the removal of its sialic acid increases the proinflammatory activity of IgG1 [55]. Two or three alleles of the IgA2 subclass are present in humans.

24.7.7 CD89 The functions of IgA are in large part mediated through its receptor, (FcαR1 or CD89). FcαR1 is an activating receptor expressed on diverse myeloid cells such as neutrophils, monocytes, macrophages, dendritic cells, and eosinophils. As a result, the signaling of IgA through this receptor can promote opsonization and phagocytosis, cytokine release, the neutrophil respiratory burst, and macrophage antibody-dependent cell-mediated cytotoxicity. It also stimulates the release of leukotriene B4 which serves as a neutrophil chemoattractant. These reactions generally occur at sites of IgA synthesis
the mucosal surfaces and mediate local inflammation [56]. In addition, to acting as an IgA receptor, CD89 can also serve as a receptor for C-reactive protein and as a pattern recognition receptor for bacteria such as Streptococcus pneumoniae and Escherichia coli [57]. The CD89 gene, unlike the other major Fc receptors, is located along with the NK cell receptors of the LRC on chromosome 19 in humans, orangutans, and macaques. It is also found on chromosome 1 in rats, chromosome 6 in pigs, chromosome 18 in cattle, and chromosome 10 in horses [58,59]. It appears to be totally absent in mice. CD89 appears to be more structurally related to the LRC-encoded receptors than to the other FcRs.

24.7.8 IgE Found in very small amounts in serum, most IgE is bound to tissue mast cells where it acts as an inflammatory signaling molecule and mediates acute allergic responses (Table 9.2). It may play an important role in immunity to some gastrointestinal parasites. There are three IGHE genes in the human and gorilla genomes, one active, one truncated, and one processed [60]. The chimpanzee has only two, one active and one processed.

24.7.9 IGHV The current immunoglobulin V region and allelic databases are likely still incomplete since we have only sampled a limited amount of the genetic diversity that occurs within the human species [45]. The human IGH locus is currently considered to contain 123
129 IGHV genes in total. The precise number differs between haplotypes. 82
88 of these IGHV genes belong to seven subfamilies but there are also 41 highly divergent pseudogenes. Seven unmapped IGHV genes have been described as insertion/deletion polymorphisms but have not yet been precisely located. The potential genomic IGH repertoire in humans is somewhat more limited since it is encoded by only 38
46 functional IGHV genes depending upon the specific haplotype. The IGH locus also contains 27 D gene segments belonging to seven subfamilies. Of these, 23 are functional. The IGH locus contains nine IGHJ segments of which six are functional. The human CDR3 region averages 13 amino acids in length but can range from 5 to 25.

24.7.10 Immunoglobulin Light chains 24.7.10.1 IGK About 70% of human immunoglobulins use kappa light chains. The human IGK locus is located on chromosome 2p11.2 and spans 1820 kb. It consists of 76 IGKV genes belonging to 7 subfamilies, 5 IGKJ segments, and a single

390

SECTION | 2 Mammalian orders

IGKC gene. The IGKV genes are organized into two clusters separated by 800 kb [61]. The most centromeric IGKV cluster spans 400 kb and contains 36 genes. The IGKV proximal cluster spans 600 kb and contains 40 genes. The potential functional IGK gene repertoire consists of 31
36 functional IGKV genes, the five IGKJ gene segments plus the single constant region gene. One rare IGK haplotype has been described that contains only the proximal V gene cluster. 25 IGKV orphons have been found dispersed among 5 different chromosomes.

24.7.10.2 IGL The human IGL locus is found on the long arm of chromosome 22q11.2. In addition, orphons have been found on chromosomes 8 and 22. The IGL locus spans 1050 kb. It is located 6 mb from the centromere [61]. It contains 73
74 IGLV genes occupying 900 kb, as well as 7
11 IGLJ, and 7
11 IGLC genes depending on the haplotype. Each C gene is preceded by a J gene. The 56
57 IGLV genes belong to 11 subfamilies plus there are 14 unclassified pseudogenes. Thus the total number of IGLV genes per haploid genome is 87
96 of which 37
43 are functional.

24.7.10.3 The Vλ
microRNA association Some IGLV gene segments also contain an miRNA gene called miR-650. This seems to be specific to primates. This gene is 96 nucleotides long and overlaps the IGLV leader exon (89 nucleotides). The untranslated region of the IGLV leader exon contains the mature miRNA sequence. Nine of these IGLV gene segments containing miR-650 are found in humans as both functional and pseudogenes and all belong to one specific phylogenetic V gene subfamily. They appear to be transcribed independently in different cell types [62].

24.7.11 Other primates While humans produce the four IgG subclasses described above, other primates produce alternative, species-specific IgG subclass mixtures [51]. Analysis of IGHC exons in Platyrrhines found that they possess all the major classes including IgD.

24.7.12 Great apes In the Hominids, not only do IGHG duplications occur but also duplications in IGHA and IGHE genes. Chimpanzees (P. troglodytes) possess three IGHG genes coding for IgG1, IgG2, and IgG3. The chimpanzee IgG2 molecule contains epitopes also found on both human IgG2 and IgG4, suggesting that the IGHG2 and IGHG4 genes diverged after the human-chimpanzee divergence. In addition, chimpanzees and bonobos but not gorillas, possess the human pseudogene IGHGψ but not the pseudogene for IgE. Gorillas, on the other hand, possess an IGHE pseudogene together with isolated IGHG pseudogenes. All the great apes, with the exception of the orangutan, make two IgA subclasses. All the hominids possess the genes for IgD. The orangutan has four IGHG genes and one pseudogene. However, these IgGs have a common origin with human IgG1 while there is no apparent relationship with the other human IgG subtypes. The IGHE gene in other primates is functional whereas the IGHE gene is a truncated pseudogene in the Orangutan [60].

24.7.13 Old World monkeys In the Old World monkeys, the Cercopithecoidea, exon 2 of IGHE is also duplicated. One copy is located within the IGHE locus while the other copy is an orphon. Some of these species may also have, duplicated IGHG genes. For example, rhesus macaques possess three IGHG genes coding for IgG1, IgG2, and IgG3. Baboons (Papio cynocephalus) have four IGHG genes, but their subclasses differ significantly from human IgG in their hinge region sequences. Light chain usage is evenly split between kappa and lambda chains.

24.7.14 New World monkeys The IGHD CH2 exon is absent from the New World monkeys, and they appear to have an extra IGHE exon that may be an orphon. Their IgG subclasses are highly conserved, but Aotus and Ateles express two IgG variants while Cebus and Alouatta express three [52].

The primates: humans and their relatives Chapter | 24

391

24.7.15 Prosimians The prosimian IGH locus encodes 70 exons encoding IgM, IgG, IgE, and IgA but not IgD. IGHD genes are absent in all prosimians tested [51].

24.7.16 Immunoglobulin D 24.7.16.1 Other primates The amino acid sequence of the chimpanzee IgD is 98.1% identical to that of human IgD. All the cysteines responsible for both the inter- and intrachain disulfide bonds are conserved. Likewise, the N-glycosylation motifs are conserved [63]. Both the human and chimpanzee N terminal portion of the hinge region are highly O-glycosylated and almost all are present in both species. IgD has also been sequenced in rhesus macaques, cynomolgus macaques, baboons, and a sooty mangabey. A single IGHD sequence was present in each species except that the mangabey has two that differ by two amino acid substitutions. The amino acid sequence identities between human and chimpanzee IgD are 72%
78% [63]. However, the identities between each of the Old World species are 96%
98%. Their IgD CH1 domain is two amino acids longer while their hinge is three amino acids shorter. Their N- and O- glycosylation sites are largely conserved. Out of interest, this study also examined dog IgD (Chapter 20). Dog IgD shares 39% amino acid identity with primate IgD. The sequence differences were greatest in the hinge region [63].

24.8

T cells and cell-mediated immunity

Gamma/delta T cells are the first T cells to appear in the fetal thymus. As a primate ages, however, the proportion of γ/δ T cells progressively drops. By the time adulthood is reached, the average proportion of γ/δ T cells in human peripheral blood is B4% of all CD31 T cells. Thus humans are a “γ/δ-low” species. A similar proportion of T cell receptor use is also found in the major lymphoid organs such as the spleen, thymus, tonsils, and lymph nodes. Unlike rodents, γ/δ T cells do not accumulate in human mucosal tissues or in the skin [64]. γ/δ and α/β T cells produce similar sets of cytokines especially the Th1 cytokines, IFN-γ, and TNF-α, and mount similar cytotoxic responses. However, the frequency of γ/δ T cells may climb significantly during some infections such as tuberculosis [64]. Humans employ two subsets of γ/δ T cells as determined by their δ chain usage. Vδ1 T cells predominate in the thymus and peripheral tissues and recognize various stress-related antigens [65]. Conversely, Vδ2 T cells form the majority of blood γ/δ T cells. They associate with the Vγ9 chain. These, Vγ9Vδ2 T cells use their TCR to recognize phosphoantigens. These phosphoantigens include hydroxymethylbutenyl pyrophosphate which is produced by many pathogens and isopentenyl pyrophosphate which is produced by many tumor cells. To be activated, the Vγ9Vδ2 T cells must also interact with cells producing butyrophilin-3A1 (Chapter 8). There is evidence that the three genes encoding Vγ9, Vδ2, and butyrophilin have coevolved in many species. In addition, to nonhuman primates, the alpaca also possesses functional homologs of all three genes [66,67]. Many other mammals possess nonfunctional γ/δ pseudogenes. T cells expressing Vγ9Vδ2 can account for up to 95% of all blood γ/δ T cells in humans. Between 10%
30% of these may also express Vδ1 paired with other Vγ chains [68]. A small percentage of γ/δ T cells are also CD81 [69]. Vδ2 T cells in humans are found at high frequency in the bloodstream and at inflammatory sites, whereas Vδ1 and Vδ3 T cells are found primarily in noninflamed mucosal tissues [70]. Both of the human γ/δ T-cell subsets are effectively cytotoxic and mediate target cell destruction by the release of perforins and granzymes [65]. Human γ/δ T cells can also act as antigen-presenting cells. Blood Vγ9Vδ2 T cells can respond to microbial signals and prime both CD41 and CD81 T cells. Activated γ/δ T cells may also phagocytose tumor cells. Human γ/δ T cells can be readily activated by signals transmitted through their γ/δ receptor in addition, to the activating receptors they share with NK cells such as NKG2D [69]. Specific activating ligands include the phosphoantigens for the Vγ9Vδ2 receptor, as well as MICA and MICB for NKG2D. The receptors on NK cells including both KIRs and LILRs, can also negatively regulate γ/δ T cell activation. These receptors possess an intracytoplasmic ITIM that turns off the T cell activation signals. Both humans and nonhuman primates also possess invariant NKT cells activated by the glycolipid antigen α-galactosylceramide. Unlike the rest of the primate MHC system, this invariant NKT cell receptor is strictly conserved across all the primate clades and the TCR β chains are nearly identical between humans and rhesus macaques. However, the NKT cells themselves may show diverse phenotypes depending upon their tissue location [71].

392

SECTION | 2 Mammalian orders

In addition, up to 10% of human blood T cells are mucosal-associated invariant T cells. These cells specifically recognize antigens generated by a precursor of riboflavin (vitamin B2). These antigens are present on most, but not all bacteria and are processed and presented by the nonpolymorphic class Ib MHC molecule, MR1. These cells are especially abundant in the female genital tract [72].

24.8.1 T cell antigen receptors 24.8.1.1 TRA/D The human TRA/D locus is located on the long arm of chromosome 14 very close to the centromere (Fig. 24.8). It contains 127 genes and occupies 1000 kb. There are 54 TRAV genes of which 45
47 are functional and they belong to 33
35 subfamilies; 61 TRAJ genes of which 50 are functional; and a single TRAC gene. It includes five TRAV/DV genes [73]. The TRD locus is embedded within the TRA locus. It contains 16 genes (Including the five TRAV/DV genes). There are eight TRDV genes of which at least seven are functional and they belong to 7
8 subfamilies; three TRDD genes, all are functional, four TRDJ genes of which all are functional; and a single TRDC gene.

24.8.1.2 TRB The human TRB locus is found on the long arm of chromosome 7 and is 620 kb in size. It contains 64
68 TRBV, 2 TRBD, 14 TRBJ, and 2 TRBC genes. The 40
48 functional TRBV genes belong to 21
23 subfamilies. There are 12
14 functional J genes, and both the D and C genes are functional. There are also six orphons located on chromosome 9. As in other mammals, with one exception (TRBV30), all the TRBV genes are located upstream of the two D-JC clusters. There are six TRBJ genes in cluster one and eight in cluster two. TRBV34 is in inverted orientation. The rhesus macaque TRB locus contains 77 TRBV, 2 TRBD, 14 TRBJ, and 2 TRBC genes.

24.8.1.3 TRG The human TRG locus is found on the short arm of chromosome 7. It spans 160 kb and contains only 19
22 genes per haploid genome. These include 12
15 TRGV genes of which four to six are functional. They belong to two subfamilies. There are five functional TRGJ genes and two functional TRGC genes. The TRGV genes are located upstream of the duplicated J-C clusters [73]. The first cluster contains three J genes plus the TRGC1 gene while the second contains two TRGJ genes plus the telomeric TRGC2 gene. As described above, TRGV9 is expressed in 80%
90% of human γ/δ T cells. TRGV10 and 11 are ORFs as a result of a splicing defect in their messenger sequence. Polymorphisms affecting the precise number of TRGV genes and the number of exons in the TRGC2 gene occur in different populations. Thus the TRGC1 gene has three exons and encodes a domain of 173 amino acids. The TRGC2 region has four or five exons as a result of duplication or triplication of the region that encodes exon 2 as well as a C domain of 189 or 205 amino acids [73].

TRA/D

TRB

5'

54 AV

8 DV

63-68

6

3 DD

12-15 3

V

61 AJ

3'

AC

8 C2

C1 TRG

4 DJ DC

2

D

J

C

FIGURE 24.8 The organization of the three human T cell antigen receptor gene loci. The red arrow denotes gene orientation.

The primates: humans and their relatives Chapter | 24

393

References [1] Finch CE. Evolution of the human lifespan and diseases of aging: roles of infection, inflammation and nutrition. Proc Natl Acad Sci USA 2010;107:1718
24. [2] Quintana-Murci L. Human immunology through the lens of evolutionary genetics. Cell 2019;177(3):184
99. [3] Mathieson I, Lazaridis I, Rohland N, Mallick S, et al. Genome-wide patterns of selection in 230 ancient Europeans. Nature 2015;528:499
505. [4] Quach H, Rotival M, Pothlichet J, Loh Y-HE, et al. Genetic adaptation and Neanderthal admixture shaped the immune system of human populations. Cell 2016;167:643
56. [5] Snyder-Mackler N, Sanz J, Kohn JN, Brinkworth JF, et al. Social status alters immune regulation and response to infection in macaques. Science 2016;354:1041
5. [6] Carter AM. Comparative studies of placentation and immunology of non-human primates suggest a scenario for the evolution of deep trophoblast invasion and an explanation for human pregnancy disorders. Reproduction 2011;141:391
6. [7] Clements T, Rice TF, Vamvakas G, Barnett S, et al. Update on transplacental transfer of IgG subclasses: impact of maternal and fetal factors. Front Immunol 2020. Available from: https://doi.org/10.3389/fimmu.2020.01920. [8] Palmeira P, Quinello C, Silviera-Lessa AL, Zago CA, Carniero-Sampaio M. IgG placental transfer in healthy and pathological pregnancies. Clin Dev Immunol 2012. Available from: https://doi.org/10.1155/2012/985646. [9] Mickleson KNP, Moriarty KM. Immunoglobulin levels in human colostrum and milk. J Pediatr Gastroent Nutr 1982;1:381
4. [10] Ozdemir SA, Ozer EA, Kose S, Ilhan O, et al. Reference values of serum IgG and IgM levels in preterm and term newborns. J Mat-Fet Neonat Med 2015. Available from: https://doi.org/10.3109/14767058.2015.1027680. [11] Hawkey CM. Mammalian hematology: Cellular components and blood coagulation of captive wild animals. London: Heinemann Medical Books; 1975. [12] Albert DL, Brodie SJ, Sasseville VG, Ringler DJ. Peripheral blood eosinophilia in owl monkeys is associated with increased eosinophilopoiesis yet depressed recruitment kinetics to the chemokine RANTES. Blood 1996;88(5):1718
24. [13] Hawash MBF, Sanz-Remon J, Grenier J-C, Kohn J, et al. Primate innate immune responses to bacterial and viral pathogens reveals an evolutionary trade-off between strength and specificity. Proc Natl Acad Sci USA 2021. Available from: https://doi.org/10.1073/pnas.2015855118. [14] Brinkworth JF, Van Etten K, Bhatt P, McClure K, et al. Functional comparison of human and non-human primate neutrophil responses. J Immunol 2019;73(1 Suppl):21. [15] Mestas J, Hughes CCW. Of mice and not men: differences between mouse and human immunology. J Immunol 2004;172:2731
8. [16] Gabay C, Kushner I. Acute-phase proteins and other systemic responses to inflammation. New Engl J Med 1999;340(6):448
54. [17] Palmer DB. The effect of age on thymic function. Front Immunol 2013;4. Available from: https://doi.org/10.3389/fimmu.2013.00316. [18] Steinmann GG, Klaus B, Mu¨ller-Hermelink HK. The involution of the aging human thymic epithelium is independent of puberty. A morphometric study. Scand J Immunol 1985;22(5):563
75. [19] Haynes BF, Hale LP. The human thymus: a chimeric organ composed of central and peripheral lymphoid components. Immunol Res 1998;18(2): 61
78. [20] Steininger BS. Human spleen microanatomy: why mice do not suffice. Immunology 2015;145:334
46. [21] Udroiu I. Evolution of sinusal and non-sinusal spleens of mammals. Hystrix It J Mamm 2006;17(2):99
116. [22] Perry M, Whyte A. Immunology of the tonsils. Immunol Today 1998;19(9):414
21. [23] Spencer J, Finn T, Isaacson PG. A comparative study of the gut-associated lymphoid tissue of primates and rodents. Virchows Arch 1986;51: 509
19. [24] Kararli TT. Comparison of the gastrointestinal anatomy, physiology, and biochemistry of humans and commonly used laboratory animals. Biopharm Drug Dispos 1995;16:351
80. [25] Veazey RS, Rosenzweig M, Shvetz DE, Pauley DR, et al. Characterization of gut-associated lymphoid tissue (GALT) of normal rhesus macaques. Clin Immunol Immunopathol 1997;82(3):230
42. [26] Fisher RE. The primate appendix: a reassessment. Anat Rec 2000;261:228
36. [27] Xie T, Rowen L, Aguado B, Ahern MA, et al. Analysis of the gene-dense major histocompatibility complex class III region and its comparison to mouse. Genome Res 2003;13(12):2621
36. [28] Yu CY, Yang Z, Blanchong CA, Miller W. The human and mouse MHC class III region: a parade of 21 genes at the centromeric segment. Immunol Today 2000;21(7):320
8. [29] Heijmans CMC, de Groot NG, Bontrop RE. Comparative genetics of the major histocompatibility complex in humans and nonhuman primates. Int J Immunogenet 2020;47:243
60. [30] Piontkivska H, Nei M. Birth-and death evolution in primate MHC class I genes: divergence time estimates. Mol Biol Evol 2003;20(4):601
9. [31] Fukami-Kobayashi K, Shiina T, Anzai T, Sano K, et al. Genomic evolution of the MHC class I region in primates. Proc Natl Acad Sci USA 2005;102(26):9230
4. [32] Knapp LA, Cadavid LF, Watkins DI. The MHC-E locus is the most well conserved of all known primate class I histocompatibility genes. J Immunol 1998;160:189
96. [33] Wu HL, Wiseman RW, Hughes CM, Webb GM, et al. The role of MHC-E in T cell immunity is conserved among humans, rhesus macaques and cynomolgus macaques. J Immunol 2018;200:49
60. [34] Wong ESW, Sanderson CE, Deakin JE, Whittington CM, et al. Identification of natural killer cell clusters in the platypus genome reveals an expansion of C-type lectin genes. Immunogenetics 2009;61:565
79.

394

SECTION | 2 Mammalian orders

[35] Hao L, Klein J, Nei M. Heterogeneous but conserved natural killer receptor gene complexes in four major orders of mammals. Proc Natl Acad Sci USA 2006;103:31292
3197. [36] Kumar S. Natural killer cell cytotoxicity and its regulation by inhibitory receptors. Immunology 2018;154:383
93. [37] Guethlein LA, Norman PA, Hilton HG, Parham P. Co-evolution of MHC class I and variable NK cell receptors in placental mammals. Immunol Rev 2015;267:259
82. [38] Parham P, Norman PJ, Abi-Rachad L, Guethlein LA. Human-specific evolution of killer cell immunoglobulin -like receptor recognition of major histocompatibility complex class I molecules. Philos Trans R Soc Lond B Biol 2012;367(1590):800
11. [39] Lanier LL. On guard
activating NK cell receptors. Nat Immunol 2001;2(1):23
7. [40] Yang Y, Bai H, Wu Y, Chen P, et al. Activating receptor KIR2DS2 bound to HLA-C1 reveals the novel recognition features of an activating receptor. Immunology 2022;165:341
54. [41] Rajalingam R, Parham P, Abi-Rached L. Domain shuffling has been the main mechanism forming new hominid killer cell Ig-like receptors. J Immunol 2004;172:356
69. [42] Storm L, Bruijnesteijn J, de Groot NG, Bontrop RE. The genomic organization of the LILR region remained largely conserved throughout primate evolution: implications for health and disease. Front Immunol 2021. Available from: https://doi.org/10.3389/fimmu.2021.716289. [43] Messaoudi I, Estep R, Robinson B, Wong SW. Nonhuman primate models of human immunology. Antiox Redox Signl 2011;14(2):261
73. [44] Averdam A, Petersen B, Rosner C, Neff J, et al. A novel system of polymorphic and diverse NK cell receptors in primates. PLoS One 2009. Available from: https://doi.org/10.1371/journal.pgen.1000688. [45] Watson CT, Breden F. The immunoglobulin heavy chain locus: genetic variation, missing data, and implications for human disease. Genes Immun 2012;13:363
73. [46] Gutzeit C, Chen K, Cerutti A. The enigmatic function of IgD: some answers at last. Eur J Immunol 2018;48(7):1101
13. [47] Vladutiu AO. Immunoglobulin D: properties, measurement and clinical relevance. Clin Diag Lab Immunol 2000;7(2):131
40. [48] Shan M, Carrillo J, Yeste A, Gutzeit C, et al. Secreted IgD amplifies humoral T helper 2 cell responses by binding basophils via galectin-9 and CD44. Immunity 2018;49:709
D724. [49] Zhai G-T, Wang H, Li J-X, Cao P-P, et al. IgD activated mast cells induce IgE synthesis in B cells in nasal polyps. J Allergy Clin Immunol 2018. Available from: https://doi.org/10.1016/j.jaci.2018.07.025. [50] Vidarsson G, Dekkers G, Rispens T. IgG subclasses and allotypes: from structure to effector functions. Front Immunol 2014. Available from: https://doi.org/10.3389/fimmu.2014.00520. [51] Olivieri DN, Gambon, Deza F. Immunoglobulin genes in primates. Mol Immunol 2018;101:353
63. [52] Yepes-Perez Y, Rodriguez-Obediente R, Camargo A, Diaz-Arevalo D, et al. Molecular characterization of parvorder Platyrrhini IgG subclasses. Mol Immunol 2021;139:23
31. [53] Woof JM, Kerr MA. IgA function
variations on a theme. Immunology 2004;113:175
7. [54] Tizard IR. Veterinary immunology. An introduction. 10th ed St Louis: Elsevier; 2017. [55] Steffen U, Koeleman CA, Sokolova MV, Bang H, et al. IgA subclasses have different effector functions associated with distinct glycosylation profiles. Nat Commun 2020. Available from: https://doi.org/10.1038/s41467-019-13992-8. [56] Wines BD, Ramsland PA, Trist HM, Gardam S, et al. Interaction of human, rat and mouse immunoglobulin A (IgA). With Staphylococcal superantigen-like 7 (SSL7) decoy protein and leukocyte IgA receptor. J Biol Chem 2011;286(38):33118
24. [57] De Tymowski C, Heming N, Correia MDT, Abbad L, et al. CD89 is a potent innate receptor for bacteria and mediates host protection from sepsis. Cell Rep 2019;27:762
75. [58] Akula S, Mohammadamin S, Hellman L. Fc receptors for immunoglobulins and their appearance during vertebrate evolution. PLoS One 2015. Available from: https://doi.org/10.1371/journal.pone.0124530. [59] Morton HC, Pleass RJ, Storset AK, Brandtzaeg, et al. Cloning and characterization of equine CD89 and identification of the CD89 gene in chimpanzees and rhesus macaques. Immunology 2005;115:74
84. [60] Ueda S, Takenaka O, Honjo T. A truncated immunoglobulin e pseudogene is found in gorilla and man but not in chimpanzee. Proc Natl Acad Sci USA 1895;82:3712
15. [61] Lefranc MP, ed. The global Immunogenetics web resource for immunoglobulins (IG), T cell receptors (TR), major histocompatibility (MH) and related proteins of the immune system. Montpelier France, 2021. Available from: http://imgt.org. [62] Das S. Evolutionary origin and genomic organization of micro RNA genes in immunoglobulin lambda variable region gene family. Mol Biol Evol 2009;26(5):1179
89. [63] Rogers KA, Richardson JP, Scinicariello F, Attanasio R. Molecular characterization of immunoglobulin D in mammals: Immunoglobulin constant delta genes in dogs, chimpanzees and four old world monkey species. Immunology 2006;118:88
100. [64] Chien Y-h, Meyer C, Bonneville M. γδ T cells: first line of defense and beyond. Ann Rev Immunol 2014. Available from: https://doi.org/ 10.1146/annurev-immunol-032713-120216. [65] Lawand M, Dechanet-Merville J, Dieu-Nosjean M-C. Key features of gamma-delta T cell subsets in human diseases and their immunotherapeutic implications. Front Immunol 2017. Available from: https://doi.org/10.3389/fimmu.2017.00761. [66] Karunakaran MM, Herrmann T. The Vg9Vd2 T cell antigen receptor and butyrophilin-3 A1: models of interaction, the possibility of coevolution, and the case of dendritic epidermal cells. Front Immunol 2014. Available from: https://doi.org/10.3389/fimmu.2014.00648. [67] Herrmann T, Fichtner AS, Karunakaran MM. An update on the molecular basis of phosphoantigen recognition by Vγ9Vδ2 T cells. Cells 2020. Available from: https://doi.org/10.3390/cells9061433.

The primates: humans and their relatives Chapter | 24

395

[68] Haas W, Pereira P, Tonegawa S. Gamma/delta T cells. Annu Rev Immunol 1993;11:637
85. [69] Pistoia V, Tumino N, Vacca P, Veneziani I, et al. Human gd T-cells: from surface receptors to the therapy of high risk leukemias. Front Immunol 2018. Available from: https://doi.org/10.3389/fimmu.2018.00984. [70] Holderness J, Hedges JF, Ramstead A, Jutila MA. Comparative biology of γδ T cell function in humans, mice and domestic animals. Annu Rev Anim Biosci 2013;1:99
124. [71] Yu KKQ, Wilburn DB, Hackney JA, Darrah PA, et al. Conservation of molecular and cellular phenotypes of invariant NKT cells between humans and non-human primates. Immunogenetics 2019;71:465
78. [72] Mondot S, Boudinot P, Lantz OMAIT. MR1, microbes and riboflavin: a paradigm for the co-evolution of invariant TCRs and restricting MHC1-like molecules? Immunogenetics 2016;68:537
48. [73] Antonacci R, Massari S, Linguiti G, Jambrenghi AC. Evolution of the T-cell receptor (TR) loci in the adaptive immune response: the tale of the TRG locus in mammals. Genes 2020. Available from: https://doi.org/10.3390/genes11060624.

This page intentionally left blank

Chapter 25

The Afrotheria: Elephants, manatees, and their relatives

Lesser hedgehog tenrec. Echinops telfairi

The eutherian mammals are currently considered to belong to four major clades, the Euarchontoglires, the Laurasiatheria, the Afrotheria, and the Xenarthra. However, the relationships between these clades have been a matter of considerable debate among systematologists. It is generally agreed that the Euarchontoglires and the Laurasiatherians are sister groups and together constitute the Boreoeutheria. However, the relationships between Afrotheria and Xenarthra are considered equivocal. The reason for this debate is, as noted in Chapter 1, that the divergence of these clades occurred over a relatively short period of around 90
100 mya. As a result, the timing of the branch points is difficult to discern from the fossil evidence. Traditionally, it has been considered that the Xenarthra diverged from the crown mammals first leaving the Afrotheria as a sister group of the Boreoeutheria. However recent molecular phylogeny studies have probably resolved these disputes by suggesting that the Afrotheria and Xenarthra are sister clades derived from a common branch that has been called the Atlantogenata [1]. The name Atlantogenata refers to the fact that South America, the home of most Xenarthra, separated from Africa when the Atlantic Ocean opened up through continental drift thus separating them from their African relatives, the Afrotheria. The split between the Atlantogenata and the Boreoeutheria is estimated to have occurred about 105 mya. The split between the Xenarthra and the Afrotheria is estimated to have occurred only two million years later B103 mya [1]. The superorder Afrotheria contains a very physically diverse group of mammals (Fig. 25.1). It consists of six orders: The elephants (Proboscidea); sea cows (Sirenia); hyraxes (Hyracoidea); the aardvark (Tubulidentata), elephant shrews (Macroscelidea), the golden moles (Chrysochlorida), and the tenrecs (Afrosoricida). While they bear a little superficial resemblance to each other, the molecular and genetic data support their monophyly. As a clade, the Afrotheria is also characterized by a failure of testicular descent as a result of the loss of functional forms of the RXFP3 gene (encoding a relaxin-family member found in sperm), and INSL3 (encoding an insulin-like hormone that encodes testicular descent). While not a direct function of the elephant immune system, it should be pointed out that Afrotherians such as elephants are remarkably resistant to the development of cancer. This appears to be the result of a pervasive duplication of tumor suppressor genes, especially TP53 [2]. Many, such as the elephants, also possess large numbers of olfactory receptors [3].

Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00010-1 © 2023 Elsevier Inc. All rights reserved.

397

398

SECTION | 2 Mammalian orders

Elephants 50 mya

Sirenians Hyrax

Afrotheria

Aardvark Elephant shrews

80 mya 100 mya

Golden moles Tenrecs

Sloths Xenarthra

Armadillos

FIGURE 25.1 The phylogeny of the Afrotheria. They are a sister taxon to the Xenarthra and diverged from them about 100 mya.

As in previous chapters, immunologic studies have tended to focus on economically important and charismatic species such as the elephants rather than the more inconspicuous members of the clade.

25.1

Elephants

The largest land mammals include two distinct species of African elephant, the Savannah elephant Loxodonta africanus and the Forest elephant Loxodonta cyclotis, as well as the Asian elephant (Elephas maximus). African and Asian elephants likely had a last common ancestor about 5
7 mya [4]. Their nearest living relatives are the Sirenians (Trichechus ssp) and the rock hyraxes (Procavia capensis).

25.1.1 Reproduction and lactation The placenta of both species of elephants resembles that of the dog or cat in overall structure. Thus it is an annular endotheliochorial placenta with a rudimentary marginal hematoma [5]. In dogs and cats, as described in Chapters 19 and 20, there is relatively little transfer of immunoglobulins across the placenta, so it might be anticipated that the situation would be the same in elephants—it is not (Fig. 25.2) [6]. By comparing antibody titers from mothers and their calves prior to suckling and before the ingestion of any colostrum, it has proved possible to determine the degree of colostral transfer. Based on antibody levels against tetanus toxoid and/or rabies it has been shown that the antibody titers in newborn elephant calves were equivalent to, or higher than that of their dams at birth. These antibody levels then doubled by 2
8 days after birth, presumably as a result of colostral immunoglobulin uptake. This situation persisted until at least three months of age. It appears therefore that the major route of antibody from mother to calf is transplacental. A similar, high level of passive transfer has been reported for antibodies against Mycobacterium tuberculosis in Asian elephants [7]. Antibodies of all three major classes, IgM, IgG, and IgA are present in elephant serum and milk.

25.1.2 Hematology The white blood cells of both African and Asian elephants include leukocytes with unique features. These include having heterophils instead of neutrophils. As a result, these cells have reddish granules when stained with one of the Romanowsky stains [8]. They also possess unique monocytes with bilobed or even trilobed nuclei. The lobes are connected by a very thin strand of chromatin that is often difficult to discern (Fig. 25.3). These monocytes possess cytoplasmic granules with peroxidase activity. This unique monocyte is present in addition, to monocytes of conventional morphology. It is sometimes difficult to differentiate from a lymphocyte. Elephant eosinophils, basophils, and lymphocytes are morphologically similar to those seen in other mammals. Basophils are rarely seen. As in other mammals, white cell counts decline with age as a result of a drop in the numbers of lymphocytes, monocytes, and basophils [9]. Elephants suffering from infections or other inflammatory diseases demonstrate significant changes in blood leukocyte morphology and numbers [10]. These include a heterophil left shift as a result of an influx of immature cells, and

The Afrotheria: Elephants, manatees, and their relatives Chapter | 25

399

FIGURE 25.2 Evidence for the maternal transfer of antibodies against rabies and tetanus in elephants. RVNA 5 rabies antiviral neutralizing antibody. From Nofs SA, Atmar RL, Keitel WA, Hanlon C, et al. Prenatal passive transfer of maternal immunity in Asian elephants (Elephas maximus). Vet Immunol Immunopathol 2013; 153;308
11. With permission.

FIGURE 25.3 Examples of the unique bi- and tri-lobed monocytes found in the blood of an Asian elephant. Courtesy Andrea Lee/Houston Zoo.

400

SECTION | 2 Mammalian orders

TABLE 25.1 The white blood cells of selected Afrotheria. Asian Elephant [11]

African elephants [12]

Florida manatee [13]

Rock hyrax [14]

Total WBC. 3 103 /μL

8
26

7.8
10.8

8.4
9.1

11
21

Neutrophils (%)

15
60

23
68

26
70

75

Lymphocytes (%)

44

23
47

26
70

16

Monocytes (%)

2
10

2
8

4

1
3

Eosinophils (%)

0
11

,1

,1

4

Basophils

0
1.5

,1

,1

,1

the presence of band neutrophils, accompanied by dysgranulopoiesis, hypersegmentation, large reactive lymphocytes, plasma cells, and vacuolated monocytes. It is clear from the hematology data from many species, including elephants, that cell counts, and relative numbers are highly variable within wild, free-ranging populations. This is probably a result of recurrent minor infections and individual parasite burdens (Table 25.1). Rock hyraxes (Procavia capensis) also possess the unique lobulated monocytes seen in elephants [14]. Up to 6% of the total lymphocyte population in the hyrax are medium to large granular cells. Of their monocytes, 1%
3% have the characteristic bilobed nuclei connected by a thin filament and a light blue cytoplasm.

25.1.3 Innate immunity Toll-like receptors have been studied in two species of manatees, Trichechus inunguis, the Amazonian manatee, and Trichechus manatus, the West Indian manatee. Both TLR4 and TLR8 genes show variability in these species. Thus the Amazonian manatee has seven SNPs in TLR4 while TLR8 is much less variable. Only one TLR4 site has been subject to positive selection and none in TLR8. These differences may well reflect the greater immunologic challenges encountered by a freshwater species like the Amazonian manatee rather than the coastal West Indian manatee [15]. This has been noted previously in Chapter 16 where river dolphins, both in South America and China, show much more major histocompatibility complex (MHC) diversity than do their marine relatives, perhaps for the same reason.

25.1.4 Cytokines Elephants are the epitome of K strategists. They are very long-lived, and they spend a long time carefully and considerately raising their young. As the largest land mammals, they have a large mass of tissue to maintain a healthy state. Clearly, the type of immune system that sustains a mouse will not suffice for an elephant. To understand their increased resistance to disease, the genome of both African and Asian elephants has been examined for the presence of conserved regulatory elements [16]. It has been found that there are 862 such elements in the Asian elephant genome and 1017 elements in the genome of the African savannah elephant. Investigations of their reported functions found that many of these regulatory pathways are involved in the control of innate immunity. For example, in both species, there is significant enrichment in the elements associated with the “TNF-mediated signaling pathway” and “positive regulation of TNF production.” In addition, African elephant genomes are enriched in elements involved in the “negative regulation of TNF production.” The innate immune pathways of the Asian elephant are enriched relative to those in the African elephant, and it has been speculated that this may be a consequence of encounters with diverse pathogens as they migrated out of Africa [16]. It is also of interest to note that Asian elephant calves are more susceptible than African elephant calves to the cytokine storm triggered by elephant endotheliotropic herpesvirus. Analysis of the regulatory elements of the Asian elephant genome also shows enrichment in gene ontology terms that include “interleukin-1β production,” “interleukin-18 production,” and “neutrophil activation involved in an immune

The Afrotheria: Elephants, manatees, and their relatives Chapter | 25

401

response.” Uncontrolled activation of all of these pathways may contribute to severe inflammatory disease such as occurs in elephant endotheliotropic herpesvirus infection.

25.1.5 Acute-phase responses The acute-phase responses have been followed in Asian Elephants suffering from disease caused by endotheliotropic herpesvirus-1. Serum amyloid A (SAA) levels increased threefold during periods of viremia [17]. However, there were no significant differences in their haptoglobin levels. Similar increases in SAA have been detected in an African elephant with bronchopneumonia [18]. Routine vaccination of a single Asian elephant against tetanus and rabies stimulated an increase in serum SAA, TNF-α, IFN-γ, IL-2, IL-6, and IL-10 [18].

25.1.6 Adaptive immunity 25.1.6.1 Lymphoid organs The elephant thymus is located in its normal position within the anterior mediastinum [8]. Likewise, the histologic structure of the manatee thymus is typical of those in mammals [19]. Elephants have both conventional lymph nodes and hemolymph nodes with no unusual morphologic features and are structurally similar to those seen in other mammals such as cattle or dogs [8,20]. Thus the inguinal and cervical nodes consist of several segments with an outer cortex and inner medulla. Afferent lymphatics open into a subcapsular sinus. The medulla is made up of medullary cords and sinuses [21]. This is very different from the structure of porcine lymph nodes. The spleen of the Asian elephant is a flat rectangular organ located on the left side of the abdominal cavity. It is covered by a dense thick connective tissue capsule consisting of two layers. One of connective tissue and one consisting of bundles of smooth muscle cells. The trabeculae are thick and also contain large numbers of smooth muscle cells. The red pulp forms the major component of the spleen parenchyma and to that extent, the elephant spleen may be considered a storage organ. The penicillary arteries consisted of groups of 5
7 vessels adjacent to the white pulp [22]. Both African and Asian elephants possess palatine tonsils located on each side of the pharynx [8]. Their epithelium is invaginated into numerous crypts containing subepithelial follicles and large numbers of intraepithelial lymphocytes. Intestinal mucosal lymphoid tissues consist of Peyer’s patches in the small intestine and small lymphoid nodules scattered along the cecum and colon [8].

25.1.7 The major histocompatibility complex The overall organization of the MHC of Loxodonta africana is currently unclear since the class II region does not appear to be linked to the class III region [23]. In the extended class I region, within the κ, λ, and the β blocks there are four, one, and eight functional genes respectively. This region contains 31 class I genes of which 19 appear to be functional and the remaining 12 are pseudogenes (Fig. 25.4). The MHC of the Lesser hedgehog tenrec (Echinops telfairi) is arranged in conventional order. However, the tenrec class I region appears to be truncated since it is lacking a κ block. The class I region contains eight identified genes of which two appear to be pseudogenes [23]. Studies comparing the class II DQA gene products of the major elephant species have identified six alleles in 40 African savannah elephants and three alleles in three Asian elephants [24]. There is moderate polymorphism and allelic diversity and similar selection patterns that are consistent with balancing selection at the antigen-binding site. There is also clear evidence of trans-species allelism that has been maintained for at least six million years.

DR

II

FIGURE 25.4 The organization of the elephant major histocompatibility complex. It appears to have a similar structure in both African and Asian elephants.

DP DQ DRA

II

III

I

402

SECTION | 2 Mammalian orders

The loss of MHC diversity in small populations results in a decline in the number of antigens that can be recognized by the immune system. Thus in small declining populations or populations going through a bottleneck, a loss of MHC diversity is a characteristic feature. This was the case in Wooly mammoths (Mammuthus primigenius) who went through a well-documented bottleneck about 10,000 years ago [25]. (Mammoths went extinct B 4000 years ago). Mammoths and African elephants had a last common ancestor about 6.6
8.8 mya. Studies of ancient mammoth DNA show that all their DQA alleles were either identical to the African savannah and Asian elephants or differed in only one or two positions. Interestingly, when standardized for sample numbers it appears that wooly mammoths had more DQA alleles than do modern elephants! However, it is also clear that they suffered decreased heterozygosity resulting in a decline in the number of alleles in their DQA locus in the isolated population on Wrangel Island in Russia as a result of inbreeding and genetic drift [25].

25.1.8 The natural killer cell receptor complex African savannah elephants and Lesser hedgehog tenrecs (E. telfairi) are among the mammalian species that can generate type 1 NKT cells [26]. These cells, as discussed earlier, are T cells that possess invariant antigen receptors that bind lipid antigens presented by the non-classical class I molecule CD1. Searches for CD1d transcripts have detected its presence in both elephants and tenrecs. These NKT cells are implicated in some forms of antimicrobial activity, specifically defense against bacterial pathogens expressing glycolipid antigens. The elephant leukocyte receptor complex (LRC) is arranged into three clusters separated by conserved groups of non-LRC genes. The African elephant genome contains 83 exons encoding LRC-like immunoglobulin domains [27]. Twenty-five of these exons may be pseudogenes with stop codons, frameshifts, or disrupted splice sites. Analysis indicates that the elephant has at least 17 functional genes within the LRC. One is an ortholog of human A1BG. Elephants also possess orthologs of human FCAR, NCR1, TARM1, OSCAR, AIBG, and IGSF1 in addition, to members of the LIR, LILR, and LAIR families. The elephant LILR family is contracted relative to humans [27].

25.1.9 B cells and immunoglobulins Elephants produce IgM, IgG, and IgA [28]. Antibodies of all three classes are present in serum and milk and are generated in response to injected antigens such as tetanus toxoid. Elephants can probably produce IgE (Fig. 25.5). FIGURE 25.5 The organization of the elephant [29] and manatee [34] IGH gene locus. The two differ significantly in the number of immunoglobulin subclasses. This is well recognized as a species-specific feature among the mammals.

The Afrotheria: Elephants, manatees, and their relatives Chapter | 25

403

The immunoglobulin genes of the African elephant (L. africana) have been analyzed based on the published elephant genome [29]. The IGH locus is 2974 kb in length. In this study, a single IGHM and eight different IGHG genes were identified. Probably as a result of gaps in the genome, IGHA and IGHE genes were not identified. However, elephant IgA has been isolated by physicochemical methods. With respect to IgA, the African and Asian elephant nucleotide sequences are 99% identical.

25.1.9.1 IGHM The single elephant IGHM gene contains four CH and two transmembrane exons. Sequence comparisons with other mammalian IgM molecules have indicated that the amino acids required for folding of these domains, cysteine, and tryptophan, are present and highly conserved. The heavy chain of 465 amino acid residues is 99% identical between the three elephant species. The elephant IgM constant region shows 63.8% identity to the human IgM constant region and 50.8% to the echidna IgM [29].

25.1.9.2 IGHD While IgD has not been detected in the elephant, a search of the sequences 10 kb downstream from IGHM has detected a gene fragment homologous to the Cδ3 domain in other mammals [29].

25.1.9.3 IGHG Further downstream, eight IGHG loci have been detected on one scaffold while a ninth locus IGHG9, has been detected on another [29]. This may be an additional subclass or an allelic variant. This large number of IGHG loci is consistent with a previous serum fractionation study that had detected five IgG subclasses in elephant serum [30]. Although it is not known if all of these IGHG loci are functional, this number is unusually large and very different from the single IGHG locus of the manatee. The genome study also found that CH1 from IGHG1 and IGHG2 were missing. The CH3 exon of the elephant IGHG3 is pseudogenized as a result of a premature stop codon and a frameshift mutation caused by the insertion of two additional nucleotides. Phylogenetic analysis indicated that the elephant IGHG loci form their unique cluster. Further analysis of the IGH locus demonstrated the presence of switch regions upstream of the IGHM locus as well as six of the IGHG loci (G1, G4, G5, G6, G7, and G8). The switch region of IGHG2, G3, and G9 could not be identified, probably as a result of gaps in the currently available sequence [29].

25.1.9.4 IGHV A total of 112 IGHV gene segments have been identified in the elephant and of these, 51 appear to be functional. The remaining 61 IGHV genes contain premature stop codons or frameshifts and are considered to be pseudogenes. Guo et al. also detected 17 partial segments of about 200 bp in length that appear to be truncated IGHV segments [29]. Phylogenetic analysis of these IGHV genes identified seven subfamilies. These were shown to be homologous to their corresponding human V genes. The elephant IGHV4 is the largest subfamily with 72 members. They fall into all three mammalian clans, I, (16) II (13), and III (16). Again, the presence of such a large number of clan III gene subfamilies is in marked contrast to the manatee that has none. It has been suggested based on horse, cattle, and sheep sequences that clan II is favored by herbivores. This large number of gene subfamilies is also a feature of the mouse, human, and horse IGH loci. Six germline IGHJ segments have been detected. IGHJ1 is a pseudogene since a tryptophan has been replaced by a stop codon. Eighty-seven IGHD gene segments have also been detected. This is likely a minimum estimate based on the existence of gaps in the sequence [29]. Their potential coding regions are 10
37 bp long and they are flanked by the usual heptamers and nonamers. Sequence analysis suggested that these D segments fall into seven families with greater than 70% sequence identity [29].

25.1.9.5 Light chains The elephant immunoglobulin kappa chain locus consists of 153 IGKV segments. Of these, 53 are apparently functional and 100 are pseudogenes. At least 148 of these IGKV segments can be assigned to eight families. Six of the families are homologous to the six human Vκ families. On one scaffold, 18 of the IGKV genes were in the reverse transcriptional orientation. The elephant IGL locus contains only 12 IGLV genes belonging to six families. Three are pseudogenes. All are arranged in the same transcriptional direction. At the 3’ end of the locus are three IGLC genes. Both IGLC2 and

404

SECTION | 2 Mammalian orders

IGLC3 are preceded by an IGLJ gene segment. The IGLJ segment preceding IGLC1 is likely missing because of a sequence gap. The three IGLC genes have about 90% sequence identity [29].

25.2

MANATEES

Within the order Afrotheria, the extant Sirenia consists of a single species of dugong and three species of manatee. All are exclusively herbivorous, and all appear to be closely related to elephants. They inhabit warm waters with variable salinity. Thus the Florida (West Indian) (T. manatus) and the African manatees (Trichechus senegalensis) can live in fresh, brackish, or saltwater. The Amazon manatee (T. inunguis) is a strictly freshwater species.

25.2.1 Hematology White cell counts in Florida manatees average 5.98 (2.77
13.5) 3 103/μL in free-ranging animals [31]. Of these heterophils account for about 39%, lymphocytes for 48%, monocytes for 8.5%, eosinophils for 3%, and basophils for less than 1%. Captive animals tend to have slightly higher heterophil counts. Total leukocyte, heterophil, and eosinophil counts tend to be lower in adults than in younger animals. Their heterophils have a pale blue cytoplasm with round, oval, or rod-shaped pink to red granules [31].

25.2.2 Lymphoid organs The Florida manatee is a hindgut fermenter with an enlarged cecum. The small intestine contains numerous Peyer’s patches up to one cm in diameter located on the anti-mesenteric side of the intestinal wall [32]. They are more numerous in the distal intestine. There is a diffuse lymphocytic infiltration with some nodules in the lamina propria of the large intestine. The manatee lacks Paneth cells in its duodenum, unlike other large herbivores.

25.2.3 The major histocompatibility complex The Florida manatee MHC class I molecules are similar in structure and antigen-binding ability to other mammals (Fig. 25.4) (Fig 7.4). The manatee class II locus shares overall synteny with other mammals and contains the main classical class II genes. However, most DR loci are translocated from the canonical location past the extended class II region. Studies of the genomes of related taxa suggest that this translocation is shared with the other Afrotheria [33]. Other presumptive chromosome rearrangements in the Afrotheria include the deletion of the DQ loci in Chrysochloris asiatica (the Cape golden mole) and the deletion of DP in E. telfairi (Lesser hedgehog tenrec). Thus the translocation of the manatee DR region has split the DR sequences into four subregions. There is a non-translocated class II subregion and three additional translocated subregions. The manatee has presumably functional DRA and DRB genes in the nontranslocated and second translocated (Tr2) regions. Manatee DRB also has a gap in its sequence but is believed to be functional. The first translocated region (Tr1) is only found in the Paenungulata (L. africana and T. manatus). All the Afrotherians appear to have lost a functional DRB from the non-translocated region whereas the tenrec, E. edwardii appears to lack all DR loci in that region. The other Afrotherians appear to maintain only a single region containing a functional DRB. The Paenungulata, aardvark, Orycteropus afer, and the Afrosoricida use the Tr2 region whereas the elephant shrew (Elephantulus edwardii) uses Tr3. The manatee has a single DQA gene and four DQB loci although only one of these is functional. The number of DQB loci varies among taxa. The only Afrotherian known to have multiple functional DRB genes is L. africana. DQ loci have not been detected in the genome of C. asiatica or E. telfairi. The manatee sequences cluster with those of the elephant although the manatee DQB pseudogene clustered with the P. alecto (Flying fox) pseudogene suggesting that this gene may have been present in the ancestor of the eutherians. The manatee, uniquely among the marine mammals has functional paired DPA and DPB genes. It has three DPA and four DPB loci but only one DPA and one DPB are functional. The aardvark possesses four in-tandem duplications of the DP loci whereas E. telfairi has lost all its DP loci [33]. It appears that the tenrec has one of the simplest mammalian MHC class II regions with only three DRA genes and only a single DRB gene [33].

The Afrotheria: Elephants, manatees, and their relatives Chapter | 25

405

25.2.4 B cells and immunoglobulins 25.2.4.1 IGH The Florida manatee (Trichechus manatus) is one of the few Afrotherians to have its immunoglobulin genes sequenced [34] (Fig. 25.5). The IGH locus spans 3765 kb. The locus codes for each of the five heavy chain constant regions, M, D, G, E, and A, but only contain a single subclass for each isotype. The IGHM gene is the most 5’ of the constant genes. It consists of four constant exons plus two transmembrane exons. The IgM heavy chain is 445 amino acids long and contains all the conserved cysteines needed for inter- and intrachain bonding. It also contains seven possible N-linked glycosylation sites, one of which in the C4 domain, is unique to manatees and elephants. There is a single 273 bp IgD constant domain fragment located about 10 kb downstream from the last IGHM exon. This is similar to that in the elephant as described above. It appears to correspond to the CH3 domain of IgD. It is prematurely terminated and likely a pseudogene. The manatee genome encodes a single IGHG constant domain gene. It consists of three exons in the order CH1, hinge, CH2, and CH3 so that the secreted protein is 330 amino acids long. It also contains two transmembrane region exons. The manatee IgG is structurally similar to human IgG2. There are no sequence gaps between IGHG and IGHE suggesting that the manatee has only a single IgG subclass. The manatee IGHE gene encodes a peptide chain with four constant domains and two transmembrane domains. The expressed heavy chain contains 426 amino acids. It differs from the IgE of other mammals in the number and location of its glycosylation sites. The IGHA gene contains three CH exons with a seven amino acid hinge region encoded by the 5’ end of the CH2 exon, as well as a single transmembrane exon. The expressed IgA heavy chain is 339 amino acids in length. It contains the cysteines required for binding the J chain as well as secretory component (pIgR) and FcαR1. It possesses features of both IgA1 and IgA2 in humans.

25.2.4.2 IGHV One hundred and thirty-nine IGHV gene segments have been identified in the manatee IGH locus. Of these, 118 are pseudogenes as a result of either premature stop codons or frameshifts, and eight have been designated ORFs as a result of missing splice sites, missing motifs, truncations, insertions, or no detectable RSS [35]. Thus the manatee appears to have among the fewest functional IGHV gene segments and the largest number of IGHV pseudogenes yet found in a mammal. The functional V genes are arranged in seven subfamilies based on 70% homology. When compared to human and mouse sequences, three IGHV segments cluster with clan I, and ten cluster with clan II. None of the functional IGHV segments belong to clan III. There are only three other mammals that have been shown to lack clan III V genes. These are cows, sheep, and horses. All these species also have relatively few IGHV genes (Cows have 13, sheep 10, and horses have 14). This loss of clan III likely occurred independently in the manatee. There are six functional IGHJ gene segments and one ORF IGHJ segment. They range from 45 to 66 bp in length. Between the IGHV segments and the IGHJ segments are 48 functional IGHD segments. These range from 10 to 41 bp in length. They are classified into 10 subfamilies. Thus compared to most mammals, manatees have relatively few diversity segments. This is significantly fewer than the elephant, mice, or humans, but more than the opossum, or cow. The number of D segments is dwarfed by the 87 D segments in the African elephant. These all suggest that the greatest sequence diversity in the manatee immunoglobulins is located in the heavy chain V domain CD3 region [35].

25.2.5 T cells and cell-mediated immunity T cells use heterodimeric antigen receptors to bind and recognize antigens presented by MHC class I or class II molecules. These receptors use paired chains, either alpha, and beta or gamma and delta. In effect, the β and δ chains have D sequences in their V domains and serve as heavy chains while the α and γ chains lack D sequences and so serve as light chains. Each receptor chain is therefore encoded either by four genes, -V-D-J-C- or just by three -V-J-C-. Three T cell receptor loci encode the genes for the four peptide chains since the TRD genes are nested within the TRA locus. Both TRA and TRD loci contain a collective set of TRV genes that can rearrange with either TRA or TRD genes or with both. The TRB and TRG genes are encoded at separate loci. Depending upon the species, the TRG loci may be divided into one or two separate loci. The details of the elephant are unclear but their TCR α/β receptors appear to be encoded by a large number of genes [36].

406

SECTION | 2 Mammalian orders

VHD

MANATEE TCR

TRA/D

TRB

5'

107 A/DV

3 DV

70

6

DD DJ DC

3'

AC

4 C1

TRG

50 AJ

C2

14 7 J 5C not assembled

V

D

J

C

FIGURE 25.6 The organization of the manatee T cell receptor gene loci. Note the the unusual presence of an immunoglobulin G pseudogene, VHD within the TRA/D locus.

25.2.5.1 TRA/D locus The manatee TRA/D locus spans over 2.01 mb—much larger than the human (1 mb) or the mouse (1.65 mb) but smaller than cattle (2.4 mb). It has a typical mammalian structure with the TRDD, TRDJ, and TRDC genes and an inverted TRDV segment nestled between the pool of shared TRA/DV segments and the TRAJ gene segments and TRAC (Fig. 25.6). The TRAD locus contains 112 TRDV segments of which 60 are functional while the remainder are pseudogenes or ORFs. As in other mammals, some of these TRV segments are expressed with either TRA or TRD or with both. Thus of the 50 functional V loci, the TRA/D locus contains 41 TRAV, 6 TRA/DV, and 3 TRDV. An interesting feature of the TRAD locus in amphibians and monotremes in the presence of a gene encoding an immunoglobulin heavy chain V gene segment located near the TRD-specific gene segments. It is expressed with TRDC and is called VHD. The VHD gene is not present in marsupials or eutherians. In the manatee however, a pseudogene has been detected downstream of the TRDV3 pseudogenes that appear to be a highly frameshifted gene that resembles the immunoglobulin heavy chain gene segments and the platypus (O. anatinus) VHD gene. The platypus gene is 38% homologous to manatee TRA/DV, whereas its similarity to manatee IGHV is 51%. The functional V segments in the TRA/DV locus are classified into 32 subfamilies based upon 75% nucleotide identity. The inclusion of pseudogenes increases the number of these subfamilies to 47. When compared to the shared V genes in humans, several synteny blocks are apparent. These range in size from two to 14 TRA/DV gene segments. In general, the manatee is similar to other mammals in its overall number of TRA/DV segments, but the number of pseudogenes (55%) is high compared to other species. The Florida manatee is the only species known to have only a single functional TRDD gene segment of 14 bp. They also have 50 TRAJ segments, two of which are pseudogenes. These range from 18 to 22 amino acids in size. They also have a single functional TRDJ segment that is present in all expressed TRD sequences.

25.2.5.2 TRB locus The manatee TRB locus spans 433 kb and is thus smaller than in humans and mice. It contains 70 TRBV segments of which 40 are functional and the remainder are either pseudogenes or ORFs. They are divided into 19 families based on 75% nucleotide sequence identity. Compared to other mammals this is an average number of V gene segments. Its sequence is similar to other mammals with two -D-J-C- clusters and an inverted TRBV gene downstream of the -D-JC- clusters. The two D-J-C clusters have each one TRBD segment that shares 69% nucleotide identity. TRBD1 is 13 nucleotides long while TRBD2 is 15 nucleotides long. The two TRBJ segments are also split between the -D-J-C- clusters with six in cluster 1 and four in cluster two [35].

25.2.5.3 TRG locus The structure of the manatee TRG locus is largely undefined as a result of the low coverage of this region. There appear to be two functional TRGV loci. Sixteen TRGV segments have been identified, of which 14 are functional. Only one of

The Afrotheria: Elephants, manatees, and their relatives Chapter | 25

407

these V gene segments has more than 75% identity with any human subfamily. The significance of the above-average number of TRGV segments is unclear since it is as yet unknown whether manatees are a γ/δ-high or -low species.

25.3

Mammalian life-spans

Living mammals have a vast range of natural life-spans ranging from slightly over one year in the case of forest shrews (Mysorex varius) to over 200 years in the bowhead whale (Balaena mysticetus). In general, larger species tend to live longer than smaller ones as a result of greater fitness and a decreased susceptibility to predation as well as their specific environmental situation. Thus the female bowhead whale has an estimated maximum life span of 211 years and a mass of over 100 tons. Likewise, the African elephant (L. africana) weighs more than six tons and lives for up to 65 years. Some animals fail to obey this mass-longevity rule and although very small, live an inordinately long time. Thus Brandt’s bat (Myotis brantii) weighs 5
20 g and may live for up to 40 years. A rodent, the naked mole-rat (Heterocephalus glaber), lives on average for more than 28 years whereas similarly sized rodents such as the mouse, live for a tenth of that time (Chapter 23). It should be noted that both the bat and the mole-rat are very effective at avoiding predation by their ability to fly or burrow. Life-span, however, is usually proportional to body size [37]. The relationship between maximum lifespan (MSLP) and body mass (M) is MSLP 5 10.2M0.22. Thus the naked mole-rat (H. glaber) lives for approximately six times longer than predicted based on its body size [37]. If they are to live a long time, mammals must be able to attenuate the aging process while developing increased resistance to many different diseases. Thus for example, the naked mole-rat has unique coding changes in its hyaluronan synthase 2 gene and so secretes high molecular weight hyaluronan that may contribute to cancer resistance (Box 25.1). Obviously, some mammals live abnormally long life-spans such as humans (aided by modern medicine) while others such as the mouse, lead unusually short life spans compared to this formula. In short-lived r strategists such as mice, the immune system has to keep them alive for only a few months or years while they breed. The situation in humans is different. They have much longer life-spans than the rest of the great apes. Whereas the normal human lifespan around 1900 was about 40 years in North America, it has now doubled. This is largely due to the prevention of infectious diseases by drugs and vaccines as well as the availability of food. Infections were the most important cause of mortality in hunter-gatherers, early farmers, and city dwellers as well as in wild chimpanzees. However, this increase in longevity has also revealed interesting aspects of the human immune system that are not encountered in other short-lived BOX 25.1 Peto’s paradox. In 1977 a British scientist, Richard Peto, made the argument that large animals have more cells. When each of these cells divides it carries a risk of a mutation developing in its daughter cells. Some of these mutations will result in a cell becoming cancerous. Thus large, long-lived mammals should accumulate more mutations and as a result, have a much greater risk of developing cancer than small mammals. But this is not the case! Many large long-lived animals rarely develop cancer—hence Peto’s paradox. Elephants are long lived mammals in which cancer is rare. As described above, they happen to have 20 copies of the tumor suppressor gene TP53 as well as additional copies of the Leukemia Inhibitory Factor gene (LIF). TP53 detects even minor DNA damage and responds by triggering apoptosis in affected cells. Thus cancerous cells are destroyed immediately they develop. LIF is also activated by TP53 and as a result also triggers tumor cell death. As described in Chapter 23, the Naked mole-rat (Heterocephalus glaberi) is also relatively resistant to cancer and has an unusually long life span for a mouse-sized animal. The cells of this creature are unusually sensitive to contact inhibition. This is due to the activation of the cGAS-STING pathway. Overcrowded cells get thinned out by apoptosis. These mole-rats also produce large quantities of high molecular weight hyaluranan that has been claimed to have antitumor properties. The capybara (H. hydrochaeris), the world’s largest rodent, also appears to have evolved enhanced T cell-mediated antitumor cytotoxicity that effectively reduces its cancer risks despite its large size. The longest-living mammal is probably the bowhead whale (Balaena mysticetus) that may live for over 200 years. Studies on its genome and transcriptome have shown specific mutations in some genes associated with cancer and aging as well as duplications of genes associated with DNA repair, cell cycle, and aging. In addition, the bowhead genome contains large numbers of long noncoding DNAs that are potentially tumor suppressive. Keane M, et al. Insights into the evolution of longevity from the bowhead whale genome. Cell Rep 2015;10:112
122. Jiang J-J, Kong Q-P. Comparative analysis of long noncoding RNAs in long-lived mammalsprovides insights into natural cancer resistance. RNA Biol 2020;17(11):1657
1665

408

SECTION | 2 Mammalian orders

mammals. Elderly humans suffer from an eventual decline in their humoral immune responses. This decline is a result of alterations in B cell behavior and in the BCR repertoire. In fact, the BCR repertoire becomes increasingly specialized but less plastic over many years [38]. Much of the space in the B cell repertoire becomes increasingly occupied with IgD1 B cells. Additionally, many of the BCR genes develop premature stop codons which suggest that B cells may no longer be able to discriminate between functional and nonfunctional antigen receptors.

25.4

The r/K trade-off

As an animal population exploits a new habitat and food and other resources are abundant, it will grow at a maximum rate r until such time until the carrying capacity of the habitat K, is reached. At that time, growth will level off and reach equilibrium. When examined from a broad perspective, mammals may evolve to exploit their environment in two distinctly different ways. Some species seek to maximize their growth rate r. In other words, they will multiply rapidly, and the population will grow as long as resources permit. This strategy will work well, especially in less crowded or unbounded environmental niches. These r-selected species multiply rapidly, and produce many offspring but as a result, may have high mortality and a short lifespan once the carrying capacity K is reached. r-strategy works well when invading new habitats and in rapidly changing circumstances. These r-strategists tend to have high fecundity with an early onset of maturity and a short generation time. Moreover, because of their short lifespan, their immune systems do not have to provide for long-term survival. Extreme examples of this are seen in some small mammals. For example, in the elephant shrew (E. edwardsii) the plains viscacha (Lagostomus maximus, a rodent), the dunnart (Sminthopsis spp, a marsupial), and the opossum (O. virginianus) almost all the cohort of ovarian follicles are fertilized at a single mating. These may amount to 50
100 embryos. As a result, only those blastocysts that can find a space in the uterus will survive. If they survive, only those neonates that can find a free teat will suckle and survive and this is rarely more than twelve. Thus selection in these species occurs soon after fertilization. An alternative life-history strategy is to seek to exploit the carrying capacity of the environment while at the same time investing heavily in fewer offspring. This is called a K-strategy. A K-strategy works best in very stable and closed environments. The investment in relatively few offspring and their associated care maximizes their individual chances of surviving to sexual maturity and producing offspring. These K-selected offspring will place different demands on the immune system. It will have to ensure parental survival to a much greater age. K-selected species tend to be in equilibrium with their environment and perform well in exploiting and even dominating stable environments. K-selected species have a larger body size, a long life expectancy, and the production of fewer offspring over a longer period of time. Their offspring also tend to be altricial and this too can have immunological implications. Species classically considered to be K-strategists include humans, whales, and elephants. In practice, however, the r/K dichotomy should be considered a continuous spectrum. Some species may also change behaviors in response to environmental pressures and resource availability—humans and rabbits certainly do!

25.5

Body mass and immunity

A large body mass provides an animal with several ecological advantages. The most important of these is the relatively reduced energy input needed for maintenance and a greatly enlarged gastrointestinal tract from which to obtain that energy. Thus the largest land mammals such as the elephants, hippos, and bovids are all herbivores capable of obtaining their energy from plentiful but nutritionally poor plant sources. Relatively few studies have focused on the relationship between immune function and body size. One theoretical study has suggested that the minimum size of a functional set of B cells is 107. This is the size of the smallest known immune system—in the tadpole [39]. It has been suggested that this functional unit, termed the “protecton,” is the basic unit of the B cell system. Animals with more than 107 cells simply add more protectons as needed. Thus a mouse may express the equivalent of 10 protectons. The number of such units likely depends in large part on space limitations. Larger mammals simply possess more protectons [39]. Comparative studies on the lymph nodes of small species such as the rat have also suggested that they are made up of basic units called “deep cortex complexes” [40]. When the nodes of mice, rats, guinea pigs, hamsters, rabbits, dogs, sheep, and humans are compared it is clear that the size of these basic units is essentially the same in each species studied. (Larger units, for example, might not be functionally optimal and so slow or impair lymphocyte migration through the node). In effect, in larger mammals, there is an increase in the number of small units rather than an increase in the size of individual units. This supports the protecton concept described above. Given the complexity of the immune system, one might expect that the scaling relationships between body mass and specific immune traits may differ. Three theories have been applied to these relationships. Thus the “protecton” theory

The Afrotheria: Elephants, manatees, and their relatives Chapter | 25

409

is described above [39]. suggests that lymphocyte concentrations should be proportional to body mass. Likewise, theory suggests that the number of lymphocytes depends upon delivery to the site of invasion and also predicts that lymphocyte numbers would scale with body mass. A third suggestion is that lymphocyte numbers may scale with metabolic activity rather than size since cellular turnover and differentiation are driven by metabolic processes. As discussed above, mass is only one factor that will affect these numbers. Life-history traits are also important. For example, r strategists that live fast and die young may invest minimally in immunity. Social system complexity and diet will also affect exposure to pathogens and hence phylogeny also affects immune variation. Studies on the neutrophil and lymphocyte counts in over more than 250 mammal species with a range of over 250,000-fold differences in body mass have revealed significant relationships [41]. Thus phylogeny explains .70% of the variation in both lymphocyte and neutrophil numbers and body mass. However, neutrophils scale hypermetrically with body mass. Thus it is estimated that an African elephant circulates 13.3 million times the neutrophils of the house mouse although their body masses differ by only 250,000-fold. Mass explains just 9% of the neutrophil variation. Lymphocytes in contrast scale just shy of isometrically. Body mass explains just 3% of lymphocyte variation. In other words, lymphocyte numbers are relatively proportional to body mass, declining just slightly with increased size [41].

References [1] Hallstro¨m BM, Kullberg M, Nilsson MA, Janke A. Phylogenomic analysis provide evidence that Xenarthra and Afrotheria are sister groups. Mol Biol Evol 2007;24(9):2059
68. [2] Vazquez JM, Lynch VJ. Pervasive duplication of tumor suppressors in Afrotherians during the evolution of large bodies and reduced cancer risk. eLife 2021;. Available from: https://doi.org/10.7554/eLife.65041. [3] Yazhini A, Srinivasan N, Sandhya S. Signatures of conserved and unique molecular features in Afrotheria. Sci Rep 2021;. Available from: https://doi.org/10.1038/s41598-020-79559-6. [4] Reddy PC, Sinha I, Kelkar A, Habib F, et al. Comparative sequence analysis of genome and transcriptome reveal novel transcripts and variants in the Asian elephant Elephas maximus. J Biosci 2015;40(5):891
907. [5] Cooper RA, Connell RS, Wellings SR. Placenta of the Indian elephant, Elephas indicus. Science 1964;146:410
12. [6] Nofs SA, Atmar RL, Keitel WA, Hanlon C, et al. Prenatal passive transfer of maternal immunity in Asian elephants (Elephas maximus). Vet Immunol Immunopathol 2013;153:308
11. [7] McGee J, Wiedner E, Isaza R. Prenatal passive transfer of Mycobacterium tuberculosis antibodies in Asian elephant (Elephas maximus) calves. J Zoo Wildl Med 2014;45(4):955
7. [8] Fowler ME, Mikota SK. Biology, medicine, and surgery of elephants. Blackwell, Oxford; 2006. [9] Reichert S, Berger V, Franco dos Santos DJ, Lahdenpera¨ M, et al. Age related variation of health markers in Asian elephants. Exp gerontol 2022;. Available from: https://doi.org/10.1016/j.exger.2021.111629. [10] Stacy NI, Isaza R, Wiedner E. First report of changes in leukocyte morphology in response to inflammatory conditions in Asian and African elephants (Elephas maximus and Loxodonta africana). PlosOne 2017;. Available from: https://doi.org/10.1371/journal.pone.0185277. [11] Silva ID, Kuruwita VY. Hematology, plasma, and serum biochemistry values for free-ranging elephants (Elephas maximus ceylonicus) in Sri Lanka. J Zoo Wildl Med 1993;24(4):434
9. [12] Hawkey CM. Comparative mammalian hematology. Cellular components and blood coagulation of captive wild animals. London: Heinemann Medical books; 1975. [13] Kiehl AR, Schiller CA. A study of manatee leukocytes using peroxidase stain. Vet Clin Pathol 1994;23(2):50
3. [14] Aroch I, King R, Baneth G. Hematology and serum biochemistry values of trapped, healthy, free ranging rock hyraxes (Procavia capensis) and their association with age, sex, and gestational status. Vet Clin Pathol 2007;36(1):40
8. [15] Maia de Oliveiera T, Burlamaqui TCT, Alves de Sa AL, Breaux B, et al. TLR4 and TLR8 variability in Amazonian and West Indian manatee species from Brazil. Genet Mol Biol 2021;. Available from: https://doi.org/10.1590/1678-4685-GMB-2019-0252. [16] Tollis M, Ferris E, Campbell MS, Harris VK, et al. Elephant genomes reveal accelerated evolution in mechanisms underlying disease defenses. Mol Biol Evol 2021;. Available from: https://doi.org/10.1093/molbev/msab127. [17] Stanton JJ, Cray C, Rodriguez M, Arhart KL, et al. Acute phase protein expression during elephant endotheliotrophic herpesvirus-1 viremia in Asian elephants (Elephas maximus). J Zoo Wildl Med 2013;44(3):605
12. [18] Edwards K, Miller MA, Siegal-Willott J, Brown JL. Serum health biomarkers in African and Asian elephants. Value ranges and clinical values indicative of the immune response. Animals 2020;. Available from: https://doi.org/10.3390/ani10101756. [19] Goldbach K, Lewis P, Samuelson D. Fine structure of the young thymus in the Florida manatee (Trichechus manatus latirostris). IAAAM 2010;. [20] Cave AJE, Aumonier FJ. Elephant and rhinoceros lymph node histology. J Roy Micro Soc 1962;80:209
14. [21] Aijima H, Hara H, Harada T, Hoshi H. Histological findings of the lymph nodes in a case of an African elephant. J Nihon Univ Med Ass 2018;77(1):19
22. [22] Rajani CV, Indu VR, Patki HS, Surjith KP, Pradeep M. Morphology of spleen of Asian elephant (Elephas maximus indicus). Ind J Vet Anat 2021;33(1):24
6.

410

SECTION | 2 Mammalian orders

[23] Abduriyim S, Zou D-H, Zhao H. Origin and evolution of the major histocompatibility complex class I region in eutherian mammals. Ecol Evol 2019;9:7861
74. [24] Archie EA, Henry T, Maldonado JE, Moss CJ, et al. Major histocompatibility complex variation and evolution at a single expressed DQA locus in two genera of elephants. Immunogenetics. 2010;62:856 -100. [25] Pecnerova´ P, Diez-de-Molino D, Vartanyan S, Dalen L. Changes in variation at the MHC DQA locus during the final demise of the wooly mammoth. Sci Rep 2016;. Available from: https://doi.org/10.1038/srep25274. [26] Looringh van Beeck FA, Reinink P, Hermsen R, Zajonc DM, et al. Functional CD1d and/or NKT cell invariant chain transcript in horse, pig, African elephant, and guineapig but not in ruminants. Mol Immunol 2009;46:1424
31. [27] Guselnikov SV, Taranin AV. Unravelling the LRC evolution in mammals:IGSF1 and A1BG provide the keys. Genome Biol Evol 2019;11 (6):1602
17. [28] Humphries AF, Tan J, Peng R, Benton SM, et al. Generation and characterization of antibodies against Asian elephant (Elephas maximus) IgG, IgM and IgA. PLoS One 2015;. Available from: https://doi.org/10.1372/journal.pone.0116318. [29] Guo Y, Bao Y, Wang H, Hu X, et al. A preliminary analysis of the immunoglobulin genes in the African elephant (Loxodonta africana). PlosOne 2011;. Available from: https://doi.org/10.1371/journal.pone.0016889. [30] Kelly PJ, Carter SD, Azwai SM, Cadman HF. Isolation and characterization of immunoglobulin G and IgG subclasses of the African elephant (Loxodonta africana). Comp Immunol Microbiol Inf Dis 1998;21:65
73. [31] Harvey JW, Harr KE, Murphy D, Walsh MT, et al. Hematology of healthy Florida manatees (Trichechus manatus). Vet Clin Pathol 2009;38 (2):183
93. [32] Reynolds III JE, Rommel SA. Structure and function of the gastrointestinal tract of the Florida manatee, Trichechus manatus latirostris. Anat Rec 1996;245:539
58. [33] Alves de Sa´ AL, Breaux B, Cesar T, Burlamaqui T, et al. The marine mammal class II major histocompatibility complex organization. Front Immunol 2019;. Available from: https://doi.org/10.3389/fummu.2019.00696. [34] Breaux B, Deiss T, Chen PL, Cruz-Schneider MP, et al. The Florida manatee (Trichechus manatus latirostris) immunoglobulin heavy chain suggests the importance of clan III variable segments in repertoire diversity. Dev Comp Pathol 2017;72:57
68. [35] Breaux B, Hunter ME, Cruz-Schneider MP, Sena L, et al. The Florida manatee (Trichechus manatus). T cell receptor loci exhibit subfamily V synteny and chain-specific evolution. Dev Comp Immunol 2018;85:71
85. [36] Olivieri D, von Haeften B, Sanchez-Espinel C, Gambon-Deza F. The immunologic V-gene repertoire in mammals. bioRxiv 2014;. Available from: https://doi.org/10.1101/0022667. [37] Hulbert AJ, Pamplona R, Buffenstein R, Buttemer WA. Life and death: metabolic rate, membrane composition, and life span of animals. Physiol Rev 2007;87(4):1175
213. [38] De Bourcy CFA, Lopez Angel CJ, Vollmers C, Dekker CL, et al. Phylogenetic analysis of the human antibody repertoire reveals quantitative signatures of immune senescence and aging. Proc Natl Acad Sci USA 2017;114(5):1105
10. [39] Langman RE, Cohn M. The E-T (Elephant-tadpole) paradox necessitates the concept of a unit of B-cell function. Protecton Mol Immunol 1987;24(7):675
97. [40] Be´lisle C, Sainte-Marie G. Topography of the deep cortex of the lymph nodes of various mammalian species. Anat Rec 1981;201:553
61. [41] Downs CJ, Dochtermann NA, Ball R, Klasing KC, Martin LB. The effects of body mass on immune cell concentrations in mammals. Amer Naturalist 2020;. Available from: https://doi.org/10.1086/706235.

Chapter 26

Four other orders: the Xenarthra, the Scandentia, the Eulipotyphla, and the Pholidota

A Pichi (Dwarf armadillo). Zaedyus pichiy.

As is obvious throughout this book, there are enormous differences in the amount of information available regarding the immune systems of different mammals. Several mammalian orders have been largely ignored by immunologists and the limited information regarding their immune systems is described in this chapter. For example, I have been unable to find any information on the immune systems of the flying lemurs
the Colugos, despite being relatively common in Southeast Asia, and the sole living members of the order Dermoptera. Molecular evidence suggests that the Dermoptera are a sister group to the primates [1]. These however are not the only order of ignored mammals. Others include the Xenarthra, the South American mammals that encompass the sloths, armadillos, and anteaters; the Scandentia, the treeshrews, a small squirrel-like order that also appear to be closely related to the primates; the true shrews, the Eulipotyphla; and the highly endangered pangolins of the order Pholidota.

26.1

Xenarthra: sloths, armadillos, and anteaters

The mammalian order Xenarthra is one of the four major clades of eutherian mammals. As discussed in the previous chapter, there is an ongoing dispute regarding both the timing and order in which they and the Afrotheria diverged from the crown mammals. The relationships between Afrotheria and Xenarthra thus remain debatable. This is because these divergences occurred over a relatively short time span so the branching order is difficult to discern from the fossil evidence (Fig. 26.1). Recent molecular genetic studies have gone far to resolving these disputes by suggesting that the Afrotheria and Xenarthra are sister clades derived from a common branch called the Atlantogenata [2]. The name Atlantogenata refers to the fact that South America, the home of most Xenarthra, separated from Africa when the Atlantic Ocean opened up through continental drift thus separating them from the African mammals, the Comparative Mammalian Immunology. DOI: https://doi.org/10.1016/B978-0-323-95219-4.00023-X © 2023 Elsevier Inc. All rights reserved.

411

412

SECTION | 2 Mammalian orders

Boreoeutheria Cingulata 100 mya

Armadillos Xenarthra Sloths

Atlantogenata

Pilosa Anteaters Afrotheria

FIGURE 26.1 The phylogeny of the Xenarthra.

Afrotheria. The split between the Atlantogenata and the Boreoeutheria is estimated to have occurred about 105 mya. The split between the Xenarthra and the Afrotheria is estimated to have occurred only two million years later B103 mya [2]. The Xenarthra are restricted to Central and South America. There are 31 living species in the suborder. They include the sloths, the armadillos, and the anteaters. As with many orders, there are many more diverse fossil Xenarthra such as the giant ground sloths and glyptodonts, indicating that at one time they were the dominant land mammals in South America. They have several characteristic features not present in other mammals. For example, they have unique skeletal features including their intervertebral joints (Xenarthra means “strange joints”). They are also characterized by unique teeth and a very low metabolic rate. As shown in Fig. 4.1, the body temperature of the nine-banded armadillo (D. novemcinctus) ranges from 30 C to 35 C while that of the seven-banded armadillo (D. hybridus) ranges from 29.5 C to 32 C [3]. The nine-banded armadillo and Hoffmann’s two-toed sloth (Choloepus hoffmanni) have both had their complete genomes sequenced. It is, therefore, reasonable to expect much more information on their immunomes in the future.

26.1.1 Reproduction and lactation Female Xenarthra show no clear structural distinction between their uterus and vagina. Males have abdominal testicles, a trait also possessed by Afrotheria. Given the very early date of their divergence from the other eutherian mammals, the placentation of the Xenarthra has been investigated to determine the evolutionary processes involved in the acceptance of the semi-allogeneic graft that is the fetus and the prevention of its immunologic rejection. However, these studies have revealed considerable diversity in placental structure among the members of the order [4]. Thus the placenta of the nine-banded armadillo is hemochorial, but its structure is considerably different from that found in other eutherians. The armadillo endometrium contains preformed blood sinuses located under its luminal surface. When the embryonic trophoblast penetrates the uterine epithelium, it sends finger-like projections into these blood sinuses. It was long thought, therefore, that the placental tissue from the developing fetus protected itself from any maternal immune response by developing inside a preformed uterine blood sinus and thus had minimal contact with the uterine connective tissue. It was argued that how the armadillo placenta develops with minimal destruction of maternal tissues and little direct contact between the trophoblast and uterine stroma helped explain maternal-fetal tolerance. Reevaluation of the structure has confirmed that the chorioallantoic villae are separated from direct contact with the endometrial stroma by a vascular endothelial cell layer. However, this is not an intact layer [4]. Armadillo trophoblast cells can come into contact with the endometrial stroma by partially replacing the endothelial cells of the sinus. The fetal tissue is therefore not separated from the maternal connective tissue. It is also clear that large numbers of leukocytes are present in structures called placental fibrinoids located at the interface between the intervillous space and the endometrium. These fibrinoids are acellular structures, probably fibrin aggregates, and it is suggested that they may contribute to the immune tolerance of the fetus, perhaps by acting as an immunosorbent “sponge.” Thus, they may serve to trap leukocytes and so limit inflammation and promote tolerance. However, not all Xenarthra have a hemochorial placenta like the armadillo. Sloths have lobulated, labyrinthine, and endothelochorial placentas. Examination of the placentas of the giant anteater (Myrmecophaga tridactyla), and the lesser anteater (Tamandua tetradactyla) show a hemochorial placenta but there are no surviving remnants of maternal endothelium [5]. Different areas of the placenta show different structures. The villous and trabecular areas are complex

TABLE 26.1 Blood leukocyte numbers in selected Xenarthra. Giant anteater. M. tridactyla [6] (Mean of 13)

Collared anteater T tetradactyla [6] (Mean of 13)

Maned sloth. B. torquatus [7]

Hoffmans two-toed sloth C. hoffmanni [7]

Nine-banded armadillo D. novemcinctus -Wild [8]

Nine-banded Armadillo D. novemcinctus -Captive [8]

WBC 3 103/μL

11.87

8.07

10.5

15

8.8

8.8

Neutrophils

72.6

48.15

32

29

57

33

Lymphocytes

18.77

44.15

58

67

25

52

Monocytes

1.69

2.0

4.4

1

12

0.7

Eosinophils

6.92

5.69

5

2

5

4

Basophils

,1

,1

,1

1

1

0.7

414

SECTION | 2 Mammalian orders

and intermingled. The trabeculae consisted of proliferative trophoblast cells and connective tissue attached to the decidua. The villae consisted of fetal capillaries and hypertrophied mesenchymal cells. Given these differences between armadillos and anteaters, it is clear that there is no consistent placental pattern in this order although it is believed that the armadillo placenta probably represents the ancestral pattern. Armadillos are also unique in that they bear monozygotic quadruplets.

26.1.2 Hematology As with much of the immune system of the Xenarthra, there is limited data available on their normal blood cell values, and when reported, the number of individuals sampled is often very low. However, in a sample of thirteen giant anteaters and thirteen collared anteaters, both mean white cell counts and their differential counts were highly variable [6] Hematology data has also been obtained for several sloth species [7,8] (Table 26.1). The figures for female and juvenile sloths are very similar. No unusual leukocyte morphology has been reported. In the nine-banded armadillo (D. novemcinctus) lymphocytes predominate among the circulating white blood cells from captive animals while neutrophils predominate in wild armadillos [9].

26.1.3 Lymphoid organs 26.1.3.1 Sloths The thymus of sloths is a compact lobulated mass located within the anterior mediastinum. It presents as large lobules with poorly separated cortical and medullary zones and conspicuous Hassall’s corpuscles. The lymph nodes of adult two-toed sloths (C. hoffmanni) are small and numerous. Thus, the mesenteric lymph nodes are under four mm in length, discrete, and “abundant.” Hemal lymph codes have not been observed in the sloth [10].

26.1.3.2 Armadillos The thymus is very small in newborn armadillos but may be detected histologically in the ventral caudal region of the neck and sternum by 1 week of age it is identifiable and composed of numerous pink lymphoid lobules, surrounded by adipose tissue [9]. Its structure is similar to that of other mammals. Interestingly, in adult armadillos, the involuted thymus contains numerous mast cells in the periphery of the lobules. Hassall’s corpuscles are distributed through the medulla. By 18 months of age, the thymus has involuted to such an extent that 40% is replaced by fat, and by 4 years of age, 80% has been replaced by fat. The armadillo’s ribbon-like spleen has a thin capsule containing smooth muscle and collagen There is obvious hematopoiesis occurring in the red pulp of adult animals. The ellipsoids are very large measuring 40 3 100 microns. Poorly developed sinuses are present thus it is classified as a non-sinusal spleen. Lymph nodes have thin capsules and are of conventional structure. The lymph node medullas of some animals contain unusually large granulated histiocytes. The thymus-dependent areas of the lymph nodes contain abundant lymphocytes, presumably T cells. Armadillo tonsils and Peyer’s patches contain numerous lymphoid nodules and active germinal centers. Armadillos do not possess a vermiform appendix [9].

26.1.4 The Major histocompatibility complex The MHC class I genes of M. tridactyla, the giant anteater have been characterized. Its MHC is arranged in conventional mammalian order. The class I region contains 12 genes of which 10 are functional [11]. The spatial distribution of the MHC class II DRB exon 2 gene diversity across multiple biomes has been investigated in the lesser anteater (T. tetradactyla) [12]. They show significant diversity with 60 alleles being found in 65 individuals. As expected, some of these alleles are widespread whereas others are restricted to specific biomes. The highest allelic diversity in DRB was found in anteaters living in rainforest biomes, especially the Amazon rainforests. It was speculated that this increased diversity may be a response to greater threats from pathogens in that environment.

26.1.5 Natural killer cells While few studies have been performed on NK cell functions in the Xenarthra, the leukocyte receptor complex (LRC) of D. novemcinctus has been characterized [13]. Thus, the armadillo LRC contains 115 exons with Ig-like domains. It possesses orthologs of human FCAR, GPVI, NCR1, TARM1, VSTM1, OSCAR, AIBG, and IGSF1 in addition to members

Four other orders: the Xenarthra, the Scandentia, the Eulipotyphla, and the Pholidota Chapter | 26

415

of the LIR, LILR, and LAIR families. The LILR family is expanded compared to humans. The LRC also contains a gene for a protein adjacent to GPVI with A1 and A2 domain subtypes that is absent in humans and mice [13]. The LRC genes are scattered across multiple scaffolds but it appears to be arranged in a manner similar to that in humans and elephants.

26.1.6 B cell responses Fractionation of armadillo serum has confirmed the presence of IgM, IgG, and IgA [14]. The genomes of the armadillo (D. novemcinctus) and the sloth (C. hoffmanni) have been sequenced. These show that there are a total of 415 V genes in the armadillo and 124 in the sloth immunoglobulin and T cell receptor loci. As in other mammals, these are distributed among seven loci, four encoding the TCR and three for the immunoglobulins. Thus, in D. novemcinctus there are 109 IGHV genes, 115 IGKV and 69 IGLV genes. The high number of IGHV genes is probably a result of a recent expansion of the IGHV and IGKV loci [15]. In the IGH locus of C. hoffmanni there are 25 IGHV, 11 IGKV, and 40 IGLV genes. This large difference in the number of V genes between the armadillo and sloth likely suggests that they have evolved under very different evolutionary pressures. It is tempting to suggest that this may reflect the terrestrial lifestyle of the armadillo vs the more isolated arboreal lifestyle of the sloth.

26.1.7 T cell responses In the TCR loci of D. novemcinctus there are 47 TRAV, 46 TRBV, but only 17 TRGV and 12 TRDV genes. Likewise in C. hoffmanni, there are 25 TRAV, 15 TRBV, but only 5 TRGV and 3 TRDV genes. It is reasonable to suggest that both armadillo and sloth are “γ/δ-low” species [15].

26.2

Scandentia. The tree-shrews

The tree-shrews are widely distributed in the forests of Southeast Asia. They are small, rat or squirrel-sized animals with a short reproductive and life-cycle. While relatively common, it has been difficult to determine their precise phylogeny. They consist of twenty species placed in four genera. At the present time, they constitute the order Scandentia. Most belong to the genus Tupaia. They do not all live in trees, and they are not true shrews. Tree-shrews were originally classified as insectivores but because of their unusually large brains as well as new molecular data based on genome-wide sequencing, they are now considered to be most closely related to the primates (Fig. 26.2). If so, they likely split off from the primate clade between 83 and 91 mya [16,17]. The rodent clade diverged from the primate clade somewhat earlier around 96 mya [17]. Tree-shrews are noticeably more similar to humans than to rodents despite being rat or squirrel size and shape (Fig. 26.3). Because of their relatively close relationships to humans and other primates, the tree-shrews have recently been subjected to considerable genetic analysis.

26.2.1 Innate immunity Tree-shrews lack the NADPH oxidase (NOX1) gene that may affect their responses to infection and even the composition of their gut microbiota [17]. Tree-shrews have also lost their DDX58 gene [17]. This gene encodes the cytoplasmic RNA receptor RIG-1. RIG-1 triggers the transduction cascade that involves the signaling pathway mediated by the mitochondrial antiviral signaling protein resulting in NF-κB activation and is usually required for interferon induction and antiviral innate immunity. However, the tree-shrew compensates for this loss by substituting it with a related gene, MDA5 [18]. Along with the loss of RIG-1, MDA5 has undergone strong positive selection. As a result, tree-shrews can still activate their type 1 interferon pathway. Glires Lagomorphs and rodents Scandentia Tree-shrews Dermoptera Primates FIGURE 26.2 The phylogeny of the Scandentia and Dermoptera.

Colugos

416

SECTION | 2 Mammalian orders

FIGURE 26.3 A northern tree-shrew, Tupaia belangeri. Courtesy R. Tizard.

Thirteen TLR genes of the Chinese tree-shrew (Tupaia belangeri chinensis) have been characterized and identified These are orthologs of mammalian TLR1-TLR9 and TLR11-TLR13. TLR10 is a pseudogene. Both TLR8 and TLR9 have been subjected to significant positive selection. Overall, these TLRs have been evolutionarily conserved [19].

26.2.2 Major histocompatibility complex The MHC of tree-shrews has the same organization and gene syntenic order as do humans. Tree-shrews possess at least four class I genes that are homologous to those in humans and one MIC gene. However, the shrew MHC class I genes cluster away from humans as would be expected by paralogous amplification. Interestingly one shrew class I gene is closely related to a human HLA-A gene. Its functionality has still to be determined [20]. Their peptide structure, as in other mammals, includes a leader peptide, α1, α2, and α3 domains, a transmembrane domain, and a cytoplasmic domain. Twenty-one class I alleles have been reported in this species and it is possible that they belong to two different loci [21]. The tree-shrew MHC class II region is homologous to all the human class II genes including HLA-DP, HLA-DQ, and HLA-DR as well as the non-classical class II genes HLA-DM and HLA-DO. There are at least four different DRB loci. All are polymorphic but only one is transcribed [21] Twelve different DRB exon 2 alleles have been identified in fifteen shrews with one to four alleles detected in each individual. Exon 2 has been subjected to significant positive selection [22]. Subsequent studies have found up to eight different DRB exon 2 alleles per individual [21]. The MHC class III region in tree-shrews is most conserved with gene syntenic alignment. But in contrast to humans that have two C4 genes, tree-shrews have only one.

26.2.3 Natural killer cell receptors T. belangeri chinensis has had its genome sequenced and analyzed [17]. One example of a pathway shared between tree-shrews and other primates including humans is the NKG2D ligand interaction pathway The ligands of this NK cell receptor are stress proteins. Tree-shrews use the same pathway as humans. Thus, their ligands include the MHC-Irelated chain (MIC) as well as the ULBP (UL-16-binding protein) family. They have six members of the NKG2D family as do humans [17].

26.2.4 B cell responses The T. belangeri chinensis IGH locus is located on chromosome 7. Its genome encodes only four, single copy, IGHC genes,
M, G, E, and A. They have 60%
70% homology with human immunoglobulins. It does not possess a functional IGHD gene [23]. One major difference however is that tree-shrews have undergone a remarkable expansion in their lambda V gene family with 67 copies in tree-shrews as compared to 36 in the human genome.

Four other orders: the Xenarthra, the Scandentia, the Eulipotyphla, and the Pholidota Chapter | 26

26.3

417

Eulipotyphla. The shrews

Like the Scandentia, the members of the order Eulipotyphla were originally lumped together within the order Insectivora. Molecular studies have however shown that it is an order that emerged around 90 mya. This order contains a diverse mixture of small mammals including the hedgehogs, solenodons, moles, and the true shrews [24]. Shrews, like the small rodents, are consummate r- strategists. Thus, European shrews (Sorex araneus) have a short life span and invest heavily in reproduction. They have a high metabolic rate and very little adipose tissue. Females can produce up to three litters each with about seven offspring. Both males and females die shortly after breeding with an average life-span of 14
16 months. This raises the question, how much can shrews afford to invest in immune defenses? [25] Perhaps rapid and extreme proliferation is a better investment in the short term.

26.3.1 Hematology The Asian house shrew (Suncus murinus) has a total white cell count of 7.2 3 103/μL. These consist of 47% lymphocytes, 42% neutrophils, 5% eosinophils 5% monocytes and 0.3% basophils [26].

26.3.2 Lymphoid organs Bray et al. have examined the lymphoid tissues of wild-caught European shrews in detail. In general, lymphatic tissue functions appears normal for most of the animal’s lifespan but evidence of immune senescence eventually develops [25]. Despite the fact that lymphoid follicles are smaller in adults than in subadults, adult shrews are able to respond effectively to their natural parasites. Thymic tissue from shrews of all age groups consists of layers of lymphocytes arranged around blood vessels and covered by a thin fibrous connective tissue capsule. These thymuses show no evidence of significant age-related involution. Likewise, their bone marrow shows moderate to high activity regardless of the age of the shrew. The spleen of adult shrews shows less activity in the white pulp but has higher cellularity in their red pulp when compared to the spleens of subadult animals. The white pulp tends to be confined to the center of the spleen. T cells form the periarteriolar lymphoid sheaths. About half the cells in the white pulp are phagocytic cells, - macrophages, monocytes, and neutrophils. The white pulp of subadults contains exclusively secondary follicles. Mature adults have fewer and smaller follicles. The spleen is clearly a hematopoietic organ and both in the spleen and bone marrow the degree of hematopoiesis appears to be the same regardless of age. Shrew lymph nodes are structurally similar to those in other species. The medulla contains large numbers of plasma cells, especially in adults [25].

26.3.3 The pancreas of Aselli The pancreas of Aselli is a large pale ovoid lymphoid organ located at the root of the mesentery in at least four species of shrews (S. araneus, S. minutus, Neomys fodiens, and Crocidura russula) (Fig. 26.4). Given their small size, its average weight of 52 mg accounts for 0.76% of the shrew’s body mass
about the same proportion as a kidney! The organ shrinks with age so that in adults it is half to a quarter of its size in juveniles. On histology, its structure is similar to that of a lymph node [25]. Thus, it has a subcapsular sinus. Beneath the sinus is a cortex full of secondary follicles with germinal centers. Beneath this is a T cell-rich paracortex. At the center of the organ is a medulla with few macrophages but large numbers of plasma cells. In adults, the whole organ appears simply to be an accumulation of plasma cells surrounded by a fragmentary cortex and paracortex. The function of the “pancreas of Aselli” has been a matter of some dispute. Thus, it may simply be a very large mesenteric lymph node or a set of aggregated mesenteric lymph nodes or it may have other functions. For example, Tsiperson has suggested that it is a primary lymphoid organ producing B cells and thus a shrew equivalent of the avian Bursa of Fabricius as well as a site for the terminal differentiation and storage of plasma cells [27]. On the other hand, Bray et al. consider it to be simply a large, specialized lymph node [25]. They also noted that shrew lymph nodes tended to contain greater numbers of plasma cells than other mammals. However, it should also be pointed out that, as described in the rodent chapter, wild animals tend to be under fairly intense immune pressure as a result of a burden of both ecto- and endoparasites and this may be reflected in the size and structure of their secondary lymphoid organs. This aggregation of mesenteric lymph nodes is not unique to the Eulipotyphla. A similar structure has also been described in marine mammals including dolphins, belugas, and blue whales. They have also been noted in the opossum

418

SECTION | 2 Mammalian orders

FIGURE 26.4 Imprint from the pancreas of Aselli from a shrew (S. araneus). General pattern of the imprint showing mainly lymphocytes and plasmacytes. Numerous blast forms and maturing cells are present ( 3 400) [27]. From Tsiperson VP. Pancreas of Aselli in some species of the shrews (Sorex araneus & Neomys fodiens) as an analogue of the Bursa of Fabricius in birds. Cell Biol Int 1997;21(6):359
365. With permission.

[28]. While first described and named by the Italian anatomist, Gasparo Asellius, (1581
1626), it is more appropriately simply called a mesenteric lymphoid mass [28].

26.3.4 Vaginal tonsils In the Asian house shrew (S. murinus), the anus, urethra, and/or the vagina open into a large, shared cavity, the ostium urogenitoanalis. A pair of tonsil-like structures are present near the entrance to the ostium
the anal tonsils, while another, somewhat smaller pair are located bilaterally at the vaginal entrance into the ostium. The anal tonsils are consistently present, but the vaginal tonsils are only present in about two-thirds of females examined. All are ellipsoid in shape and are covered by epithelial folds separated by grooves. The epithelial surface in both locations contains M cells. The anal tonsils have deep mucosal crypts. Surrounding the crypts are numerous lymphoid follicles with germinal centers and outer lymph nodules including germinal centers, and they possess a fibrous capsule [29]. The boundaries between the germinal centers and interfollicular areas are unclear. B cells are present in large numbers in the parenchyma and are scattered through the epithelium. There are few T cells (10%) scattered throughout the vaginal tonsils [30].

26.3.5 Hedgehogs Limited studies on European hedgehogs (Erinaceus europaeus) have demonstrated that immunoglobulin A is present in large amounts in its saliva, bile, and intestinal lamina propria. They also have, as expected, IgM and two subclasses of IgG [31]. Studies on the MHC of E. europaeus and a related species E. concolor, have been used to determine population diversity. Among 84 individuals screened, only two DQA alleles were identified in each species. However, ten DQB alleles were found in E. concolor and six in E. europaeus [32].

26.4

Pholidota. The Pangolins

Pangolins belong to their own order, the Pholidota. The Pholidota were long considered to be a sister taxon to the Xenarthra. However molecular evidence now indicates that their closest relatives are the Carnivora. The split between the carnivores and the pangolins is estimated to have occurred around 79
87 mya. There are eight pangolin species, four African and four Asian, that diverged B 23 mya [33]. They are all highly endangered and under severe threat since their scales are believed to be of medicinal value. Pangolins have scales covering most of their body, no teeth (they eat ants exclusively using their muscular tongues), poor vision, and an acute olfactory system [34]. Complete genome assemblies have been made of both the Malayan (Manis javanica) and Chinese (M. pentadactyla) pangolins. As expected, genes encoding tooth enamel production and

Four other orders: the Xenarthra, the Scandentia, the Eulipotyphla, and the Pholidota Chapter | 26

419

TABLE 26.2 Blood leukocyte numbers in two species of Asian pangolins. Malayan pangolin M. javanica [39]

Chinese pangolin. M. pentadactyla -Wild [38]

Chinese pangolin. M. pentadactyla -Captive [38]

WBC 3 103/μL

6.25

2.34
8.58

2.34
8.58

Neutrophils

53

30
61

18
85

Lymphocytes

31

26
64

14
72

Monocytes

5

2
5

1
7

Eosinophils

9

1
11

3
8

Basophils

0.4

,1

,1

encoding a lens protein are pseudogenized. In addition, another gene that is pseudogenized in both species is that encoding interferon-epsilon (IFN-ε). Insertions and point mutations generate a premature stop codon and result in loss of function. Further investigations have demonstrated that all the African pangolin species also lack IFN-ε implying that the pseudogenization occurred before the African and Asian pangolin divergence. It has also been shown that pangolins have a greatly reduced number of other interferon genes. Of the ten genes in the interferon family, only three are present in the Malayan pangolin and two in the Chinese pangolin. The heat shock protein family has also contracted significantly in these species, perhaps increasing the sensitivity of pangolins to stress. Pangolin genes under strong positive selective pressure include those involved in several immune system pathways such as hematopoietic cell lineage, cytosolic DNA sensing pathway, complement and coagulation cascades, cytokinecytokine receptor interactions, and phagosomal pathways [34]. In addition to the loss of the IFN-ε gene described above, pangolins also lack certain viral sensor genes. Thus, a study of RNA sensor genes in three species of pangolins indicated that they had fully functional DDX58 (the gene encoding RIG-1), as well as the three toll-like receptor genes TLR3, 7, and 8. However, they lack IFIH1, the gene encoding MDA5, a RIG-1-like sensor of double-stranded RNA. Its gene has been inactivated by a series of mutations in the pangolin genome. In addition, the gene encoding the Z-DNA-binding protein (ZDP-1) which senses the presence of both Z-DNA and Z-RNA has also been lost. As a result, the antiviral innate immune response in pangolins is very different from that in other mammals. This may provide an advantage by reducing inflammation-induced damage. It is also possibly relevant to the apparent ability of pangolins to serve as healthy carriers of certain coronaviruses [35]. Pangolins are also among those mammals that have lost the gene encoding TLR5, the PRR for bacterial flagellin [36].

26.4.1 Hematology Limited hematology has been performed on M. pentadactyla [37]. The differential white cell count depends upon the conditions under which these pangolins are kept [38] (Table 26.2). Similar leukocyte numbers have also been recorded in Sunda pangolins (M. javanica) [40]. Differential white cell counts for African ground pangolins (Smutsia temminckii), are similar to the Asian species [39]. A large number of circulating eosinophils in some individuals suggests that they may have had a significant parasite burden.

26.4.2 Major histocompatibility complex The Sunda pangolin (M. javanica) has only four functional MHC class I genes and a single pseudogene. This is a relatively small number compared to other mammals. Two recombinant sequences have been detected in its class I genes [11].

References [1] Mason VC, Li G, Minx P, Schmitz J, et al. Genomic analysis reveals hidden biodiversity within colugos, the sister group to primates. Sci Adv 2016;. Available from: https://doi.org/10.1126/sciadv.1600633.

420

SECTION | 2 Mammalian orders

[2] Hallstro¨m BM, Kullberg M, Nilsson MA, Janke A. Phylogenomic analysis provide evidence that Xenarthra and Afrotheria are sister groups. Mol Biol Evol 2007;24(9):2059
68. [3] Cuba-Caparo´ A. Some hematologic and temperature determinations in the 7-banded armadillo (Dasypus hybridus). Lab Anim Sci 1976;26 (3):450
5. [4] Chavan AR, Wagner GP. The fetal-maternal interface of the nine-banded armadillo: endothelial cells of maternal sinus are partially replaced by trophoblast. Zool Lett 2016;. Available from: https://doi.org/10.1186/s40851-016-0048-1. [5] Mess AM, Favaron PO, Pfarrer C, Osmann C, et al. Placentation of the anteaters Myrmecophaga tridactyla and Tamandua tetradactyla (Eutheria Xenarthra)/. Repro Biol Endocrinol 2012;10:102
8. [6] Sanches TC, Miranda FR, Oliviera AS, Matushima ER. Hematology values of captive giant anteaters (Myrmecophaga tridactyla) and collared anteaters (Tamandua tetradactyla). Pesq Vet Bras 2013;33(4):557
60. [7] Catenacci LS, Nascimento A, Muniz-Neta ES, Cassano CR, et al. First record of hematological values in free-living and captive maned sloths (Bradypus torquatus) Xenarthra, (Bradypodidae). J Zoo Wildl Med 2017;48(2):312
18. [8] Wallace C, Oppenheim YC. Hematology and serum chemistry profiles of captive Hoffmann’s two-toed sloths (Choloepus hoffmanni). J Zoo Wildl Med 1996;27(3):339
45. [9] Purtilo DT, Walsh GP, Storrs EE, Gannon C. The immune system of the nine banded armadillo (Dasypus novemcinctus Linn). Anat Rec 1975;181:725
34. [10] Wislocki GB. Observations on the gross and microscopic anatomy of the sloths (Bradypus griseus griseus) Gray and Chololepus hoffmanni Peters. J Morphology Physiol 1928;46:318
97. [11] Abduriyim S, Zou D-H, Zhao H. Origin and evolution of the major histocompatibility complex class I region in eutherian mammals. Ecol Evol 2019;9:7861
74. [12] Clozato CL, Mazzoni CJ, Moraes-Barros N, Morgante JS, Sommer S. Spatial pattern of adaptive and neutral genetic diversity across different biomes in the lesser anteater (Tamandua tetradactyla). Ecol Evol 2015;5(21):4932
48. [13] Guselnikov SV, Taranin AV. Unraveling the LRC evolution in mammals: IGSF1 and A1BG provide the keys. Genome Biol Evol 2019;11 (6):1586
601. [14] Sasiain MC, Carosella ED, Balina LM, Brezavsek DM, Bachmann AE. A study of cellular and humoral immunity in three species of armadillos. Part I. Int J Lepr Other Mycobact Dis 1977;45(4):323 236. [15] Olivieri D, von Haeften B, Sanchez-Espinel C, Gambon-Deza F. The immunologic V-gene repertoire in mammals. bioRxiv 2014;. Available from: https://doi.org/10.1101/0022667. [16] Roberts TE, Lanier HC, Sargis EJ, Olson LE. Molecular phylogeny of treeshrews (Mammalia: Scandentia) and the timescale of diversification in Southeast Asia. Mol Phylogen Evol 2011;60:358
72. [17] Fan Y, Huang Z-Y, Cao C-C, Chen C-S, et al. Genome of the Chinese tree-shrew. Nat Commun 2013;. Available from: https://doi.org/10.1038/ ncomms2416. [18] Xu L, Yu D, Fan Y, Peng L, et al. Loss of RIG-1 leads to a functional replacement with MDA5 in the Chinese tree-shrew. Proc Natl Acad Sci USA 2016;113(39):10950
5. [19] Yu D, Wu Y, Xu L, Fan Y, et al. Identification and characterization of toll-like receptors (TLRs) in the Chinese tree-shrew (Tupaia belangeri chinensis). Dev Comp Immunol 2016;60:127
38. [20] Zhang A-H, Dai Z-X, Zhang G-H, Han J-B, et al. Molecular characterization, balancing selection and genomic organization of the tree-shrew (Tupaia belangeri) MHC class I gene. Gene 2013;522:147
55. [21] Oppelt C, Wutzler R, von Holst D. Characterization of MHC class II DRB genes in the northern tree-shrew (Tupaia belangeri). Immunogenetics. 2010;62:613
22. [22] Huang Z-Q, Sun X-M, Dai J-J, Gu M-L, et al. Sequence diversity of the MHC II DRB gene in Chinese tree-shrews (Tupaia belangeri chinensis). Biochem Syst Ecol 2015;61:417
23. [23] Zhang J, Xiao H, Bi Y, Long Q, et al. Characteristics of the tree-shrew immune system. Mol Immunol 2020;127:175
85. [24] Douady CJ, Cjatelier PI, Madsen E, de Jong WW, et al. Molecular phylogenetic evidence confirming the Eulipotyphla concept and support for hedgehogs as the sister group to shrews. Mol Phylogen Evol 2002;25:200
9. [25] Bray DP, Bennett M, Stockley P, Hurst JL, Kipar A. Composition and function of haemolymphatic tissues in the European common shrew. PLoS One 2008;. Available from: https://doi.org/10.1371/journal.pone/0003413. [26] Rahman MK, Islam S, Rahman M, Ferdous J, et al. Hematological and biochemical reference values for Asian house shrews (Suncus murinus) in Bangladesh. Vet World 2019;12(9):1514
18. [27] Tsiperson VP. Pancreas of Aselli in some species of the shrews (Sorex araneus & Neomys fodiens) as an analogue of the bursa of Fabricius in birds. Cell Biol Intern 1997;21(6):359
65. [28] Romano TA, Felten SY, Olschowka JA, Felten DL. A microscopic investigation of the lymphoid organs of the Beluga, Delphinapterus leucas. J Morphol 1993;215:261
87. [29] Kubo K, Isomura G. Presence and structure of anal and vaginal tonsils in the laboratory shrew (Suncus murinus). Ann Anat 1996;178:251
8. [30] Sakai K, Imada H, Shinzato M, Nishii K, et al. Peculiar tonsil-like structure near vagina of the laboratory shrew, Suncus murinus. Okajimas Folia Anat Jpn 2013;89(4):105
12. [31] Vaerman JP, Heremans JF. IgA and other immunoglobulins from the European hedgehog. J Immunol 1971;107(1):201
11.

Four other orders: the Xenarthra, the Scandentia, the Eulipotyphla, and the Pholidota Chapter | 26

421

[32] Berggren KT, Ellegren H, Hewitt GM, Seddon JM. Understanding the phylogeographic patterns of European hedgehogs, Erinaceus concolor and E. europaeus using the MHC. Heredity 2005;95:84
90. [33] Gaubert P, Antunes A, Meng H, Miao L, et al. The complete phylogeny of pangolins: Scaling up resources for the molecular tracing of the most trafficked mammals on earth. J Heredity 2018;. Available from: https://doi.org/10.1093/jhered/esx097. [34] Choo SW, Rayko M, Tan TK, Hari R, et al. Pangolin genomes and the evolution of mammalian scales and immunity. Genome Res 2016;26:1312
22. [35] Fischer H, Tschachler E, Eckhart L. Pangolins lack IFIH1/MDA5, a cytoplasmic RNA sensor that initiates innate immune defense upon coronavirus infection. Front Immunol 2020;. Available from: https://doi.org/10.3389/fimmu.2020.00939. [36] Sharma V, Walther F, Hecker N, Stuckas H, Hiller M. Convergent loss of TLR5 suggest altered extracellular flagellin detection in four mammalian lineages. Mol Biol Evol 2020;. Available from: https://doi.org/10.1093/molbev/msaa058. [37] Khatri-Chhetri R, Sun C-M, Wu H-Y, Pei KJ-C. Reference intervals for hematology, serum biochemistry, and basic clinical findings in freeranging Chinese pangolins (Manis pentadactyla) from Taiwan. Vet Clin Pathol 2015;44(3):380
90. [38] Heath ME. Hematological parameters of four Chinese pangolins (Mantis pentadactyla). Zoo Biol 1986;5:387
90. [39] Hooijberg EH, Lourens K, Meyer LCR. Reference intervals for selected hematology and clinical chemistry measurands in Temminck’s pangolin (Smutsia temminckii). Front Vet Sci 2021;. Available from: https://doi.org/10.3389/fvets.2021.654529. [40] Ahmad AA, Samsuddin S, Oh SJ, Martinez-Perez P, Rasedee A. Hematological and serum biochemical parameters of rescued Sunda pangolins (Manis javanica) in Singapore. J Vet Med Sci 2018;. Available from: https://doi.org/10.1292/jvms-0585.

This page intentionally left blank

Index Note: Page numbers followed by “b,” “f,” and “t” refer to boxes, figures, and tables, respectively.

A A proliferation inducing ligand (APRIL), 350
351 2,5 A endonuclease (2,5 OA), 285 Aardvark (Orycteropus afer), 397
398 Accessory spleens, 158 Actinobacillus suis, 78
79 Actinobacteria, 58 Activation-induced cytidine deaminase, 127, 129 Acute phase responses, 222, 401 Bats, 286 Cats, 301 Cetaceans, 239 Dogs, 313f Elephants, 401 Humans, 378 Lagomorphs, 344 Pigs, 222 Primates, 378 Rodents, 359 Adaptive immunity, 27, 101, 117, 300, 358 Suppression of, 23 Addax antelope (Addax nasomaculatus), 272 Adipose tissue, 43
44 Aeromonas hydrophila, 46 African elephant (Loxodonta africana), 332, 398
404 Acute phase responses, 401 B cells, 402 Colostrum, 398 Hematology, 398
399, 400t Immunoglobulins, 402
403 Innate immunity, 400 Leukocytes, 399f, 400 Lymphoid organs, 401 Major histocompatibility complex, 401
402 Natural Killer cell receptor complex, 402 Passive immunoglobulin transfer, 398, 399f Phylogeny, 397
398, 398f Reproduction and lactation, 398 African wild dog (Lycaon pictus), 318 Afrotheria, 8, 397
407 Afrosoricida, 397 Elephants, 398
407 Evolution, 397
398 Mammoth, wooly (Mammuthus primigenius), 402 Manatees (Trichechus spp), 404
407 Tenrec (Echinops telfairi), 404

Aggregated lymphoid nodules, camel, 208
209 Agouti (Dasyprocta sp), 370 AIM2 receptors, 77 Alarmins, 78 Allografts, fetal, 22 Alpaca (Lama pacos), 206 Alpha block (MHC), 91
92 Alpha-1-acid glycoprotein, 301, 313
314 Alpha-defensins, 70
71, 256 Alpha-galactosylceramide, 338 Altricial young, 342 Ameridelphia, 189, 192
194 Amniotes, 3 Amphibians, 118 Anal tonsils, 166 Dolphins, 242 Shrews, 418 Whales, 242 Anaplasma marginale, 270 Anapsids, 3 Anole, 118 Anteaters, 411
415 Anthrax, 298 Antibacterial peptides, 32, 70
71, 173, 190
192, 222, 331
332, 358
359 Antilocaprae, 251 Apocrine glands, 30 Appendix Humans, 380 Rabbits, 350
351 Armadillos, 44, 411, 415 Articular bone, 4
5 Artiodactyls, 20, 205, 241 Aryl hydrocarbon receptor, 65
66 Atlantogenata, 317, 411
412 Aurochs (Bos primigenus), 252, 270 Australidelphia, 189
190, 195
196

B B cells, 117
132, 149, 220, 335
337 Antigen receptors, 103, 117
122, 230 Bats, 290
292 Camels, 211
215 Cats, 304
305 Cattle, 264
267 Cetaceans, 244
246 Dogs, 318
320 Elephants, 402
403

Epigenetics, 131 Goats, 273
274 Horses, 335
337 Humans, 386
390 Lagomorphs, 346
351 Marsupials, 196
198 Microbiota and, 62
63 Monotremes, 177
180 Pigs, 227
231, 231f Primates, 386
391 Ruminants, 264
267, 273
274 Rodents, 364
366 Sheep, 273
274 Signal transduction, 121 Xenarthra, 415 B cell activating factor (BAFF), 351 Baboons. See Primates Bacille Calmette Guerin (BCG), 263 Bacillus, 33, 60
61, 63, 130, 350
351 Bacteria, 48
49, 60, 69, 332 Antigens, 77, 350
351, 367 DNA, 77 Lipopolysaccharides, 77, 367 Permeability increasing protein, 71 Bacteroides, 61 Bactrian camels (Camelus bactrianus), 206
209 Bandicoots, 145 Basophils, 84, 173, 191t, 221, 299t, 331t, 358, 400t, 413t Bats, 381
392 B cells, 290 Black flying fox (Pteropus alecto), 283
285, 288
292 Body temperature, 286
287 Classification, 281, 282f Egyptian fruit bats (Rousettus aegyptiacus), 281
282, 285
286, 289, 290f Flight and immunity, 286 Hematology, 283 Hibernation, 286
287 Immune reconstitution inflammatory syndrome, 287 Immunoglobulins, 290
292 Inflammation, 283
284 Innate immunity, 283
287 Interferons, 284
286 Lymphoid organs, 287
288 Major histocompatibility complex, 288
289 MicroRNA, 286

423

424

Index

Bats (Continued) Natural killer complex, 289
290 Phylogeny, 281, 282f Reproduction and lactation, 282 T cells, 292
293 Virus reservoirs, 282 White nose syndrome, 287 Bears Brown (Ursus arctos), 323 Climate change and immunity, 323 Colostrum, 322 Hibernation, 323 Major histocompatibility complex, 323
324 Microbiota, 322
323 Polar (Ursus maritimus), 323 Sloth bear (Melursus ursinus), 323 Beluga (Delphinapterus leucas), 240
241, 244 Beta block (MHC), 91
92 Beta-2-microglobulin, 34, 380
381 Bifidobacterium longum, 32 Black Flying fox (Pteropus alecto), 282
289 Blocking antibodies, 26 Blood leukocytes, 83, 231, 283 Lymphocytes, 35, 252, 254, 271
272, 330 Monocytes, 312
313 Body mass and immunity, 42, 408
409 Body odors, 97
98 Body temperatures, 42, 42f, 46, 286 Bone marrow, 154, 193, 349 Bonobos (Pan paniscus), 386 Boreoeutheria, 8, 397 Bottlenose dolphin (Turciops truncatus), 84, 145, 161, 238
239 Bovines, 251
271 B cells, 264 Colostrum, 252
254 Dendritic cells, 264 Hematology, 254
255, 254t Immunoglobulins, 264
267 Innate immunity, 255
257 Lymphoid organs, 257
260 Major histocompatibility complex (BoLA), 260
262 Milk, 253 Natural killer complex, 263 Phylogeny, 251, 252f Reproduction and lactation, 252
253 T cells, 112, 267
270 Ultralong V regions, 265
266 Virus diarrhea virus, 35, 254 Workshop cluster 1 antigens, 270
271 Bowhead whale (Balaena mysticetus), 407, 407b Bronchus-associated lymphoid tissue, 164
165 Camels, 208 Cats, 302 Cattle, 164f, 259 Horses, 333 Monotremes, 175 Pigs, 224 Rodents, 359 Brown adipose tissue, 43
45, 44f Brushtail possum (Trichosurus vulpecula), 186
187

Buffalo, water (Bulbalis bulbalis), 275 Butyrophilins, 30, 114b, 217

C C-reactive protein, 79, 286, 301, 313 C-type lectins, 76 Calprotectin, 71 Camels Alpacas (Vicugna pacos), 206 B cells, 211 Bactrian (Camelus bactrianus), 206, 209 Colostrum, 206 Complement, 207 Dromedary (Camelus dromedarius), 206, 209 Heavy chain only antibodies, 213
214 Hematology, 206, 207t Immunoglobulins, 211
215 Innate immunity, 207 Leukocyte receptor complex, 211 Llamas (Lama glama), 206 Lymphoid organs, 207
209 Aggregated lymphoid nodules, 209, 209f Major histocompatibility complex, 209
211 Milk, 206 Natural killer complex, 211 Phylogeny, 205, 205f Reproduction and lactation, 206 Somatic hypermutation, 217 T cells, 215, 217 VHH domains, 213
214 Wild (Camelus ferus), 206 Workshop cluster-1, 215 Caniforms, 311
312 B cells, 318 Bears, 322
324 Breeds, 311 Colostrum, 312 Dendritic cells, 314
315 Domestication, 311 Ferrets, 322 Hematology, 312
313 Blood leukocytes, 312f Eosinophils, 313 Immunoglobulins, 318
320 Innate immunity, 313
315 Lymphoid organs, 315
316, 324 Major histocompatibility complex (DLA), 316
318, 323
324 Mustelids, 322 Natural killer cells, 314 Phylogeny, 311, 312f Pinnipeds, 324
325 Raccoons, 322
324 Reproduction and lactation, 311
312 T cells. Receptors, 320
322 Ursids, 322
323 Canine distemper virus, 282 Capybaras (Hydrochaeris hydrochaerus), 370
371 Carboniferous period, 3
4 Carcinoembryonic antigen-related cell adhesion molecule, 284, 290

Carnivora, evolution, 54, 119, 297
298 Carnivory, disease risks, 54, 297
299 evolution, 297
301 gastrointestinal defenses, 55f hematology, 299
301 innate immunity, 301 prey selection, 297
301 Carrion eating, 307
308 Caruncles, 17 Caseins, 31 Caspases, 75 Catarrhines, 375 Cathelicidins, 173, 188 Cats. See Feliformes Cattle. See Bovines CD molecules CD1 system, 338, 367 CD3 complex, 104
105, 105b, 105f, 174 CD4, 106 CD8, 106 CD16, 142 CD36, 76 CD56, 137 CD79, 121 CD89, 389 CD94, 22 CD121, 81 CD163, 222, 271 Cecal appendix, 166, 350
351 Cecocolic lymphoid patches, 194 Cellulose, dietary, 53
54 Cenozoic era, 5, 7, 9 Ceruloplasmin, 313
314 Cetaceans B cells, 244
246 Beluga (Delphinapterus leucas), 244 Bottlenose dolphin (T. truncatus), 84, 145, 161, 238
239 Bowhead whale (Balaena mysticetus) Life-span, 407, 407b Chinese river dolphin (Lipotes vexillifer), 243 Harbor porpoise (Phocoena phocoena), 240f Hematology, 238
239 Immunoglobulins, 244
246 Innate immunity, 239 Cytokines, 239 Lymphoid organs, 239
242 Laryngeal gland, 241, 242f Tonsils, anal, 242 Oropharyngeal, 241, 242f Major histocompatibility complex, 242
244 Natural killer receptor complex, 244 Orca (Orcinus orca), 238 Phylogeny, 237
238, 238f Placenta, 238 Pressure effects, 246 Reproduction and lactation, 238 Sperm whale (Physetus macrocephalus), 241 Spleens, 240 T cells, 246 Antigen Receptors, 246
247 Yangtse river dolphins (Neophocaena phocaenoides), 244

Index

Cetartiodactyla, 8 Cheetah (Acinonyx jubatus), 300 Chemokines, 81
82, 257, 330 Chimpanzee (Pan troglodytes), 89
90, 382, 390 Chiropterans. See Bats Chorioallantoic placenta, 16
17 Cingulata, 8 Clans, immunoglobulin variable region, 124
127, 124f Climate change, 323 Clostridia, 58 Collectins, 79, 256 Colostrum, 20, 34
36, 220 composition, 34
35, 36f, 253 immunoglobulins in, 34, 253 lymphocytes in, 35
36, 254 production, 34 Colugos, 411 Commensals, 41, 56
57, 350
351 Complementary determining regions, 103
104, 119
120, 265, 335 Ultralong, 265
266 Complement system, 71 Conglutinin, 256 Continental drift, 6, 397, 411 Coronavirus, 281
282 Cotyledons, 17 CR1-related gene/ protein Y (Crry), 23 Cretaceous-Paleogene event (K-Pg), 8
9 Cretaceous-Tertiary event, 8
9 Cryptopatches, 166 Ctenohystrica, 355, 356f Cyclic CMP-GMP synthase (CGAS), 77, 284
285, 371 Cyclooxygenase 2, 45 Cynodonts, 4 Cytokines, 23, 25
26, 80
82, 192, 207, 222, 239, 256
257, 302, 332 Cytotoxic T lymphocyte antigen 4 (CTLA-4), 64

D Damage-associated molecular patterns (DAMPs), 69
70, 70f, 78 Dectin, 76t, 79 Deep cortex complexes, 408 Defensins, 70
71, 83, 173, 192, 256, 331, 344 Dendritic cells, 23, 47, 84
85, 136, 302, 333, 335 Cats, 302, 302f Dogs, 314
315 Horses, 335 Pigs, 225
226 Ruminants, 224 Dermcidin, 188 Dermoptera, 411 Devil facial tumor disease, 200b Diapsids, 3 Didelphidae, 185
186 Dinosaurs, 6
7 fungal infections, 50 Dipodoidea, 357

Diprotodonts, 192 Dogs. See Caniforms Do¨hle bodies, 173 Dolphins. See Cetaceans Donkeys. See Perissodactyls Double stranded RNA, 72 Dromedary. See Camels Duck-billed platypus (Ornithorhynchus anatinus), 49
50, 159 Dysbiosis, intestinal, 165

E Ebolavirus, 281
282 Echidna (Tachyglossus aculeatus), 158, 171
172, 175 Ectothermy, 46 Eggs, monotreme, 15, 72 Egyptian Rousette bat (Rousettus aegyptiacus), 281
282, 285
286, 289, 290f Elephants. See African elephant (Loxodonta africana) Elephant shrews (Macroscelididae), 397
398 Endothermy, 41
49 Benefits, 45 Costs, 46
47 Effects on immunity, 45
49 Evolution, 41
42 Fevers, 45
49 Fungal infections, 49
50 Mechanisms, 43
45 Enterococcus fecalis, 172 Enterocytes, 61
62 Environmental microbiota, 66 Eocene epoch, 9 Eosinophils, 84, 221, 239, 324, 342, 358 Epigenetics, B cell, 131 Equine. See horses Equinins, 331 Escherichia coli, 58, 331, 389 Euarchontoglires, 8, 397 Eugenin, 188 Eulipotyphla, 417
418 Asian house shrew (Suncus murinus), 417 European shrew (Sorex araneus), 417 Hematology, 417 Lymphoid organs, 417 Pancreas of Aselli, 417
418 Reproduction, 417 Vaginal tonsils, 418 European hamster (Cricetus cricetus), 371 European hedgehog (Erinaceus europaeus), 332 European rabbit. See Rabbits Eutherian mammals, 6
9 Family tree, 6
9, 9f Fetoembryonic defense system, 25 Placenta, 15
19 Superorders, 8 Explosive model, 7

F Fas-ligand, 142

425

Fc receptors, 131, 198, 366 -like molecules, 131
132, 179
180 neonatal (FcRn), 34, 253 Feliformes. See Cats B cells, 304 Cheetahs (Acinonyx jubatus), 300 Colostrum, 299 Domestication, 297 Feline infectious peritonitis, 301 Hematology, 299
300 Blood leukocytes, 299t, 300f Hyaenas, 307
308 Immunoglobulins, 304
305, 305f Innate immunity, 301 Dendritic cells, 302f Lions (Panthera leo), 300 Leopards (Panthera pardus), 300 Lymphoid organs, 302
303 Major histocompatibility complex, 303
304 Natural killer cells, 301 Passive immunity, 299 Phylogeny, 297 Reproduction and lactation, 299 T cells, 305
306 receptors, 306
307 Feline infectious peritonitis, 301 Feline leukemia virus, 304 Fenestra, skull, 3, 4f Ferret (Mustela putorius faro), 322 Alpha-fetoprotein, 25 Fetus, Immune tolerance to, 20
21 Fever B cell functions, 48 Bacterial diseases, 48
49 Behavioral, 46 Costs, 46
47 Fungal diseases, 49 Innate immunity, 47 Mechanisms, 45 T cell functions, 48 Viral diseases, 49 Ficolins, 79 Flying foxes. See Chiropterans Fo¨a-Kurloff cells, 370 Foregut fermenters, 58, 251 Fractalkine, 81 Fungal diseases and dinosaurs, 49
50

G Galactophore, 30 Galectins, 25, 79 Gamma/delta T cells, 65, 111
114, 232, 268
269 bovine, 112 “high” species, 111
113 “low” species, 113
114 Pigs, 113 Rabbits, 113 Sheep, 112 Gastric lymphoid tissues, 224 Gastrointestinal tract Foregut fermenters, 58, 251 Hindgut fermenters, 58
59, 329

426

Index

Gastrointestinal tract (Continued) Microbiota, 58
61 GATA-binding protein-3, 135
136 Gene conversion, 130 Gene duplication, 105b Genital lymphoid ring, 260 Geomyces destructans, 49
50 Gerbils (Rhombomys opimus), 369 Germinal centers, 158 Giant panda (Ailuropoda melanoleuca), 322
323 Gibbons (Hylobatidae), 377, 386, 388 Glires, 341, 355 Goats (Capra hircus). See also Ruminants Domestication, 271 Immunoglobulins, 273
274 Innate immunity, 272 Leukocytes, 254t Lymphoid organs, 272 Major histocompatibility complex, 273 T cell antigen receptors, 275 Workshop cluster 1, 275 Golden moles (Chrysochloridae), 397
398 Gompertz Law, 371 Gondwana, 5 Gorillas, 375, 377, 386 Immunoglobulins, 390 Major histocompatibility complex, 382
383 Granulocyte-macrophage colony stimulating factor, 112, 269 Granulysins, 71, 143 Grasslands, 251, 329 Gray eosinophils, 313 Gray mouse lemur (Microcebus murinus), 386 Great apes. See Primates Guanacos (Lama guanaco), 206 Guinea pigs (Cavia porcellus), 19, 369
370 Hematology, 369
370 Immunoglobulins, 370 Major histocompatibility complex, 370 Gut-associated lymphoid tissues (GALT), 175, 194, 350
351 Gut-mammary axis, 33, 63

Cetaceans, 238
239 Dogs, 312
313, 312f Elephants, 398, 399t Eulipotyphla, 417 Goats, 254t Horses, 330, 331t Humans, 377 Lagomorphs, 342 Manatees, 404 Marsupials, 190 Monotremes, 173 Pigs, 221, 221f Primates, 377
378 Rodents, 357
358, 369 Ruminants, 254
255, 254t Sheep, 254t Xenarthra, 414 Hemochorial placentas, 19, 282 Hemolymph nodes, 162 Hendravirus, 281
282 Herbivore microbiota, 53
54 Herd behavior, 11, 255 Herpesvirus, endotheliotrophic, elephant, 304 Hibernation Bats, 286
287 Bears, 323 Microbiota and, 66 Squirrels, 66 High endothelial venules, 156, 223
224 High mobility group band protein-1 (HMGB1), 45, 78, 344 Hindgut fermenters, 58
59 Homeothermy, evolution of, 41
42 Horses. See Perissodactyls Horseshoe bat (Rhinolophus hildebrandtii), 287
288 Humans. See Primates Hyaluronan, 371 Hyenas, 307
308 Hyraxes, 397
398 Hystricomorphs, 355
370

I H H-2 system, 89
90 Hair, evolution of, 5 Hamsters. See Rodentia Hantavirus, 355 Haptoglobin, 79
80, 301, 313
314 Harbor porpoises (Phocoena phocoena), 239 Hares. See Lagomorphs Hassall’s corpuscles, 150, 302 Heat shock protein (HSP) 90, 48 Heavy-chain only antibodies, 213
214 Hedgehogs (Erinaceus ssp), 418 Hemal nodes, 208 Cattle, sheep, goats, 258 Hematology Bats, 283 Camels, 206, 207t Cats, 299
300, 299t, 300f Cattle, 254
255

Ilium, Peter’s Patches, 153 Immune reconstitution inflammatory syndrome, 287 Immune system, development, 59
60 microbiota and, 59
60 Immunity, microbiota regulation by, 60, 65 Immunoconglutinins, 256 Immunoglobulins Amphibians, 118 Colostrum, 34, 253 Constant regions, 120
121 Diversity, 121
122, 126
132 Evolution, 122 Fish, 117 Gene rearrangement, 127
129 Heavy chains, 119, 123t Hinge regions, 121 Light chains, 119 Milk, 36, 36f Placental transfer, 19
20, 20f

Reptiles, 118 V regions, 124
126 Immunoglobulin A Bats, 291 Camels, 212 Cats, 304 Cetaceans, 245 Dogs, 320 Goats, 274 Horses, 336 Humans, 388
389 Lagomorphs, 347
348 Manatees, 405 Marsupials, 197 Monotremes, 178 Pigs, 229 Primates, 388
389, 391 Rodents, 365f Ruminants, 264f Secretory, 32
33 Sheep, 273 Subclasses, 347 Xenarthra, 415 IgA Receptors, 191, 198, 366 Immunoglobulin D Bats, 291 Cetaceans, 245 Dogs, 319 Elephants, 403 Goats, 274 Horses, 336 Humans, 387
388 Lagomorphs, 346 Manatees, 405 Microbiota and, 63 Monotremes, 178 Pigs, 229 Primates, 391 Rodents, 365, 365f Ruminants, 265 Sheep, 273 Structure, 125 Immunoglobulin diversity, 126
132 Base deletion and insertion, 127
129 Epigenetics, 131 Gene conversion, 130 Gene rearrangement, 127 Receptor editing, 129 Somatic hypermutation, 129
130 Immunoglobulin E Bats, 291 Camels, 212 Cats, 305 Cetaceans, 245f Dogs, 319
320 Goats, 274 Horses, 336 Humans, 389 Lagomorphs, 346 Marsupials, 197 Monotremes, 178 Pigs, 228 Primates, 390 Rodents, 365f

Index

Ruminants, 264 Serum levels, 126t Sheep, 273 Structure, 125
126, 126t Immunoglobulin G Bats, 291 Camels, 212 Cats, 305 Cattle, 265
267 Cetaceans, 245
246 Dogs, 319 Elephants, 403 Horses, 336 Humans, 388 Lagomorphs, 346 Manatees, 405 Marsupials, 197 Monotremes, 178 Pigs, 228 Primates, 390 Receptors, 34, 34f, 131, 198, 253, 366 Rodents, 364 Ruminants, 265 Transplacental transfer, 19
20, 34 Xenarthra, 415 Immunoglobulin kappa chains, 121
122, 349, 365, 389
390, 403 Immunoglobulin lambda chains, 121
122, 129, 365, 390, 403
404 Immunoglobulin M Bats, 291 Camels, 211
212 Cats, 305 Dogs, 318 Elephants, 403 Goats, 274 Horses, 336 Humans, 387 Lagomorphs, 346 Manatees, 405 Marsupials, 197 Monotremes, 177 Pigs, 228 Primates, 390 Ruminants, 264 Sheep, 273 Xenarthra, 415 Immunoglobulin O, 178 Immunoglobulin T, 117 Immunoglobulin Y, 179b Immunoreceptor, tyrosine based activation motifs (ITAM), 131, 143 Immunoreceptor tyrosine-based inhibitory motifs (ITIM), 131, 139
140, 335, 363 Immunosuppression, fetal, 22
23 Indole dioxygenase, 23, 25 Inflammasomes, 75, 80, 301 Inflammation, 78, 283, 291 Innate immunity, 32, 69, 71, 101, 173, 190, 192, 207 Bats, 283
286 Camels, 207

Cats, 301
302 Cetaceans, 239 Dogs, 313
314 Elephants, 400 Horses, 331
332 Humans, 378 Lagomorphs, 342
344 Marsupials, 190
192 Microbiota and, 54
56 Monotremes, 173 Pigs, 221
222 Rodents, 358
359 Ruminants, 255
257 Sheep, 272 Innate lymphoid cells, 135
136 Insulin-like growth factor, 34
35 Interferons, 21
22, 82, 192, 256
257, 284
285 Bats, 284
286 Pathways, 284
286 stimulated genes, 82, 286 Types, 82, 284 alpha, 256, 285
286 beta, 75, 222, 257, 285
286 chi, 257 delta, 222, 286 epsilon, 257, 286, 418
419 gamma, 82, 143, 362 kappa, 257 lambda, 82, 286 omega, 222, 256
257, 286 tau, 257 Interleukin-1, 24, 45, 75, 80
81, 222, 239, 400
401 Receptor antagonist (IL-1RA), 81, 239 IL-2, 64, 192, 207, 343 IL-4, 192, 207, 222, 343 IL-6, 45, 64, 75, 81, 192, 302, 343 IL-7, 115, 192 IL-8, 24, 81, 302 IL-9, 192 IL-10, 24, 64, 207, 343 IL-12, 192, 207 IL-13, 192, 207 IL-15, 115 IL-15L, 222 IL-17, 23
24, 45, 102f, 114 IL-18, 400
401 IL-22, 24 IL-23, 24
25 IL-26, 64, 332 IL-31, 192 IL-33, 45, 81, 192, 344 IL-37, 81, 222 IL-38, 81 Intestine, microbiota, 59, 130 Nematodes, 356 Intraepithelial lymphocytes, 59, 306, 380

J JAK/STAT pathway, 32 Jackrabbit (Lepus californicus), 343 Jurassic period, 4

427

K K strategy, 282, 408 K-Pg event, 6
8 Kangaroos, 156
157 Kappa block (MHC), 91
92 Kappa light chains, 121
122, 320, 349, 365, 389
390, 403 Killer cell immunoglobulin-like receptors (KIR), 140, 176
177, 363, 386 Functions, 139
140 Genes, 139, 262, 335 Ligands, 384 Structures, 139, 139f Koalas (Phascolarctos cinerus), 185, 187
188 Kowari (Dasyuroides byrnie), 194 Kurloff cells, 370

L Lactation, 29
38 Adaptive immunity, 33 Evolution, 29
30 Marsupials, 29
30, 29f Microbiota and, 32
33 Monotremes, 29
30, 29f Origins, 29
30 Lactoferrin, 32 Lactoperoxidase, 32 Lagomorphs B cells, 346 repertoire, 349 Colostrum, 342 Cottontail (Silvilagus floridianus), 342 Hares (Lepus spp), 341
342 Hematology, 342 Immunoglobulins, 346
349 Immunoglobulin A, 347
348 Innate immunity, 342
344 Necroptosis, 344 Jackrabbit (Lepus californicus), 343 Lymphoid organs, 344
345 Appendix, 350
351 Major histocompatibility complex (RLA), 345
346 Natural killer cells, 344, 346 Picas (Ochotonidae), 341 Phylogeny, 341
342 Rabbit (Oryctolagus cuniculus), 341
352 Reproduction and lactation, 342 T cells, antigen receptors, 351
352 Lambda light chains, 121
122, 129, 220, 330, 337, 349, 365, 390, 403
404 Laurasia, 5, 7 Laurasiatheria, 397 Lemurs, 375, 386 Leopards (P. pardus), 300 Leopard cat (Prionailurus bengalensis), 307 Leukocytes, 83
85 Bats, 283 Camels, 207, 207t Cats, 299
300, 299t Cetaceans, 238
239 Dogs, 312, 312f Elephants, 398
399, 400t

428

Index

Leukocytes (Continued) Horses, 330, 331t Lagomorphs, 342, 343f Marsupials, 190, 191t Monotremes, 173 Pigs, 221, 221f Primates, 377 Rodents, 358, 369
370 Ruminants, 254, 254t Xenarthra, 413t Leukocyte-associated, immunoglobulin-like receptors, 138, 140, 180, 364 Leukocyte immunoglobulin-like receptors (LILRs), 24, 138, 140, 180 Leukocyte receptor complex, 138
140, 196, 211, 262
263, 314, 384, 414
415 Life-history patterns, 10 Life-spans, 407 Lipid antigens, 367 Lipopolysaccharides, 71, 350
351 Llamas. See Camels Long-fuse model, 7 Ly49, 138f, 140
142, 140f, 329, 363 Lymph nodes Bats, 287 Camels, 208 Cetaceans, 240
241 Elephants, 401 Functions, 159
160 Marsupials, 193
194, 194f Pigs, 223
224, 223f Ruminants, 257 Species differences, 161 Structure, 158
159, 159f Xenarthra, 414 Lymphocytes, circulation, 160
161 colostral, 35
36, 254 sources, 149 Lymphoglandular complexes, 165, 225, 316 Lymphoid nodules, 174, 174f, 242 Lymphoid organs, 149
166 Bats, 287
288 Camels, 207
209, 209f Cats, 302
303 Cetaceans, 239
242, 240f Dogs, 315
316 Elephants, 401 Horses, 333 Lagomorphs, 344
345 Marsupials, 192
194 Monotremes, 174
175 Pigs, 223
225, 223f Primates, 379
380, 380f Rodents, 359 Ruminants, 257
260, 258f, 259f Xenarthra, 414 Lymphoid stem cells, 149 Lymphoid tissue inducer cells (LTi), 136, 149 Lymphotactin, 81, 112 Lysozyme, 32, 188

M Macaques. See Primates

Macrophages, 23, 26, 84, 155, 164
165, 174, 342 Pulmonary intravascular, 84, 225f, 302 Macropods, 189 Macropredators, 11 Major acute phase protein, 222 Major histocompatibility complex Bats, 288
289, 288f Bears, 323
324 Camels, 209
211, 210f Cats, 303
304, 303f Cetaceans, 242
244, 243f Class Ia molecules, 90
93, 92f Evolution, 91
92 Polymorphism, 93 Structure, 90
91, 91f Class Ib molecules, 22, 93
94 Class II molecules, 94
95 Evolution, 95 Gene arrangement, 94
95 Polymorphism, 95 Structure, 94 Class III molecules, 95
96 Disease and, 96
97 Dogs (DLA), 316, 318 Elephants, 401
402, 401f Goats, 273, 273f Great apes, 385
386 Horses (ELA), 333
334, 334f Humans (HLA), 380
384, 381f Lagomorphs, 345
346, 345f Manatees, 404 Marsupials, 194
196, 195f Mice (H-2), 360
362, 360f Monotremes, 175
176, 175f, 176f Odorant receptors, 97
98 Opossum, 194
195 Pigs (SLA), 226
227, 226f Pinnipeds, 324 Primates, 380
386 Rats (RT-1), 368 Ruminants (BoLA), 260
262, 261f Sheep, 272
273 Species variations, 96, 96f Structure, 89
90 Tree-shrews, 416 Trophoblast, 22 Xenarthra, 414 Mammals, classification, 8
9 Evolution, 3
9 Evolutionary models, 7 Life-spans, 407 Origins, 4
8 Phylogeny, 5
9 Mammary gland, evolution, 29
30 Mammoth, wooly (Mammuthus primigenius), 402 Manatees, 404
407 Marsupials Ameridelphia, 189 Australidelphia, 189
195 B cells, 196
198 Bandicoots, 200 Devil Facial tumor disease, 200b

Dunnart, 193 Hematology, 190 Immunoglobulins, 196
198 Innate immunity, 190
192 Kangaroos, 156
157, 186 Koalas (Phascolarctos cinerus), 185, 187
188 Lactation, 188
189 Ameridelphia, 189 Australidelphia, 189
190 Leukocyte receptor complex, 196 Leukocytes, 191t Lymphoid organs, 192
194, 193t, 194t Major histocompatibility complex, 194
196, 195t Australidelphia, 195
196 Opossum, 194
195 Natural killer Complex, 196 Phylogeny, 185
186, 186f Placenta, 186 Pouch Protection, 187
188, 188f Reproduction and lactation, 186
187, 187f T cells, 198 Receptors, 198
200 Tammar wallaby, 187
189, 190f, 200 Tasmanian devil (Sarcophilus harrisii), 200b Mass die-offs, 10
11 Mass extinction events, 4
5, 7, 9 Mast cells, 85 Maternal-fetal tolerance, 20
25 Matrotrophy, 15
16 Meadow voles (Microtus sp), 166 Megabats, 292 Memory T cells, 114
115 Mesozoic era, 5 Metatheria, 6 MHC, class I chain-related genes (MIC), 196, 263, 416 Microbats, 286
287 Microbiota B cells and, 59
60, 62 Behaviors and, 65
66 Carnivores, 54 Defensive functions, 59 Dysbiosis, 65 Evolution, 9
10 Gastrointestinal tract, 58
59 Genitourinary system, 58 Gut-mammary axis, 63 Herbivores, 54 Hibernation and, 66 Immune system development, 59
60 Regulation, 60
61, 60f, 63f Relationship, 54
56 Immunoglobulin A and, 62
63 Immunoglobulin D and, 63 Lactation, 32
33 Nutritional functions, 59 Odors and, 65
66 Respiratory tract, 57
58 Skin, 56
57, 57f T cells and, 63
65 Microchimerism, 26
27 MicroRNAs, 31, 131, 286

Index

Milk, 29, 36, 38 Composition, 342 Development, 31 Fat content, 31, 238 Functions, 31
32 Immunoglobulins, 34, 36
38, 36f, 206 Marsupial, 188
189 Monotreme, 172 Microbiota and, 32
33 Oligosaccharides, 172 Secretion, 253 Minke whale (Balaenoptera acutorostrata), 241
242, 245 Miocene epoch, 9 Dry period, 251, 329, 341 Missing-self strategy, 137 Mitochondrial proton leak, 44 Mitochondrial antiviral signaling adaptor (MAVS), 255 Mitochondrial target of rapamycin (mTORC), 44
45 Mole rats (Heterocephalus glaberi), 166, 371
372 Mongolian gerbil (Meriones unguiclatus), 369 Monkeys. See Primates Monocytes, elephant, 398 Monotremes, 4, 6, 15
16, 30, 172, 179 B cells, 177 Cell-mediated immunity, 180 Eggs, 172 Fungal diseases in, 49 Hematology, 173, 173t Immunoglobulins, 177
179 Immunoglobulin O, 178 Receptors, 179 Innate immunity, 173 Lactation, 30, 172, 172f Lymphoid organs, 174
175 Milk, 30 Placenta, 15
16 Phylogeny, 6, 171 Major histocompatibility complex, 175
176 Natural killer cell receptors, 176
177 Reproduction, 171
172, 172f T cells, 180 Antigen Receptors, 180
182 Mu chain, 182 Venom, 173 Mouse. See Rodentia MR1 system, 145
146 Mucosal associated invariant T cells, 146, 352 Mucosal associated lymphoid tissues Bats, 287
288 Camels, 208
209, 209f Cats, 302
303 Cetaceans, 241
242 Dogs, 315
316 Elephants, 401 Horses, 333 Lagomorphs, 345 Marsupials, 194 Monotremes, 175 Pigs, 224
225 Pinnipeds, 324

Primates, 379
380 Rodents, 359 Ruminants, 258
260, 260f Sheep, 272 Structure, 89
90 Xenarthra, 414 Muroid rodents, 357
359 Mustelids, 322 Mx genes, 239, 358 Mycobacteria, 74, 143, 163, 270, 398 Myeloid-derived suppressor cells, 23 Myeloid differentiation factor-2 (MD-2), 77

N Naked mole rats (Heterocephalus glaberi), 371, 407, 407b Nanoantibodies, 214 Natural cytotoxicity receptors, 138, 141
142 Natural killer (NK) cells and receptor complex, 136
137 Bats, 289 Camels, 211 Cats, 304 Cetaceans, 244 Dogs, 314 Elephants, 402 Effector mechanisms, 143 Functions, 139
140 Genes, 138f Horses, 332 Lagomorphs, 344, 346 Lysin, 143 Marsupials, 196 Monotremes, 176
177 Pigs, 227 Primates, 384
385, 385f Receptors, 137, 138f, 139, 139f Rodents, 363
364, 368 Ruminants, 262
264 T cells. See NKT cells Trained immunity, 143
144 Uterine, 23
24 Xenarthra, 414
415 Natural killer complex, 140
142, 140f, 196, 211, 314, 384 NKG2 ligands, 141 Receptors, 141 Necroptosis, 344 Negative selection, 151 Neonatal immunoglobulin receptor (FcRn), 34, 253 Neutrophil extracellular traps, 239 Neutrophils. See Leukocytes NKT cells, 142
146, 233
234, 338, 352, 363, 367 Nine-banded armadillo (Dasypus novemcinctus), 42, 412, 414
415 Nipahvirus, 281
282 Nocturnal bottleneck, 5 Northern fur seal (Callorhinus ursinus), 324 Nuclear factor-kappa B, NF-κB, 75 Nucleotide-binding oligomerization domainlike receptors (NOD), 76, 76t, 301

429

O Odd-toed ungulates. See Perissodactyls Odontoceti, 238 Olfactory receptors, 98 Oligocene epoch, 311 Opossums. See Marsupials Orangutans (Pongo pygmaeus), 382 Osteoclast-associated receptor (OSCAR), 140, 196 Otic notch, 3 Owl monkeys (Aotus trivirgatus), 84, 377

P Pacas, 19, 370 Paired immunoglobulin activating receptors (PIRs), 140, 364 Pancreas of Aselli, 161, 417
418 Paneth cells, 70
71 Pangaea, 5 Pangolins, 418
419 Pasteurella multocida, 10
11 Pathogen-associated molecular patterns (PAMPs), 69
70, 143, 173 Pattern recognition receptors, 69
70, 72, 72f, 283, 286, 358 Peccaries (Tayassuidae), 219 Pecora, 251 Pelycosaurs, 4 Pentraxins, 79 Peptidoglycan recognition proteins, 77 Periderm, 186
187 Perissodactyls B cells, 335 CD1, role of, 338 Colostrum, 330 Dendritic cells, 335 Donkeys (Equus asinus), 330 Hematology, 330
331, 331f blood leukocytes, 331t Immunoglobulins, 335
337 Innate immunity, 331
332 Interleukin-26, 332 Lymphoid organs, 333 Major histocompatibility complex (ELA), 333
334 Natural killer cells, 332, 335 Natural killer T cells, 338 Passive immunity, 330 Phylogeny, 329 Reproduction and lactation, 330 Rhinoceros, tonsils, 333 T cells, 337 antigen receptors, 337
338 Zebra, 329 Permian/Triassic mass extinction event, 5 Peroxynitrite, 23 Peto’s paradox, 370 Peyer’s patches, 152
154, 194, 209, 221, 287
288, 303, 316, 333, 379
380 Group 1 species, 152
153 Group 2 species, 153
154 Structure, 152
154 Pholidota. See Pangolins

430

Index

Phosphoryl choline, 364 Photosynthesis, 251 Picas. See Lagomorphs Pigs B cells, 227
228 Receptor development, 230
231, 231f Breeds, 219 Colostrum, 220 Dendritic cells, 225
226 Duroc, 228 Hematology, 221 Blood leukocytes, 221f Immunoglobulins, 227
230 Innate immunity, 221
222 Landrace, 228 Lymphocyte circulation, 224, 224f Lymphoid organs, 223
225 Lymph nodes, 223
224 Natural killer T cells, 233
234 Major histocompatibility complex (SLA), 226
227 Natural killer cell complex, 227 Phylogeny, 219 Reproduction and lactation, 220 T cells, 232 Antigen Receptors, 232
233 WC1 cells, 231, 231f Pilot whales (Globicephala macrorhynchus), 238
239 Pinnipeds, 324
325. See also Caniforms Colostrum, 324 Hawaiian monk seal (Monachus schauinslandi), 324 Leopard seal (Hydryrga leptonyx), 324 Lymphoid organs, 324 Major histocompatibility complex, 324 Natural killer cells, 325 Northern elephant seal (Mirounga augustirostris), 324 Northern fur seal (Callorhinus ursinus), 324 Southern elephant seal (Mirounga leonina), 324 Weddell seal (Leptonychotes weddelli), 324 Placenta, classification, 17
19 cotyledonary, 17 endotheliochorial, 18, 282, 299, 312, 398, 412 epitheliochorial, 17, 207, 220, 238, 252, 330 evolution, 15 fibrinoids, 412 hematomas, 18 hemochorial, 282, 342, 357, 377, 412 immunoglobulin transfer, 19
20 marsupial, 16 monotreme, 15
16 NK cells in, 23
24 Structure, 18f, 21
22 synepitheliochorial, 17 T cells in, 24
25 yolk sac, 16, 16f zonary, 17 Placentomes, 17 Platypus. See Monotremes Platyrrhines. See Primates

Polymeric immunoglobulin receptors (pIgR), 179
180, 336 Porcine respiratory and reproductive system virus, 226 Possums (Phalangeriformes), 185, 186f, 190 Pouch young, marsupial, 186
188, 190 Pressure adaptation, cetacean, 246
247 Primates Acute-phase response, 378f B cells, 386
387 Baboons (Papio ssp), 375, 377, 379, 382 Catarrhines, 375, 385 Chimpanzee (Pan troglodytes), 375, 377, 382
384, 388
389, 391 Colostrum, 377 Common marmoset (Callithrix jaccus), 380, 383
384, 386 Cynomolgus macaque (Maccaca fasicularis), 384, 388, 391 Gorillas (Gorilla gorilla), 375, 377, 382
383, 386, 388
389 Hematology, 377 Hominoidea, 375 Immunoglobulins Great apes, 390 Humans, 386
390 Immunoglobulin D, 387
388 New-world monkeys, 390 Old-world monkeys, 390 Prosimians, 391 Infectious disease history, 376
377 Innate immunity, 378 Lemur, gray mouse (Microcebus murinus), 386 Lymphoid organs, 379
380 Major histocompatibility complex Great apes, 382 Humans (HLA), 380
382 Old-world monkeys, 382 New-world monkeys, 383 Non-classical MHC genes, 383
384 Natural killer cell receptors, 384
386, 385f Orangutan (Pongo pygmaeus), 382
383, 385, 388
389 Owl monkeys (Aotus trivirgatus), 377, 386, 390 Passive immunity, 377 Phylogeny, 375, 376f Platyrrhines, 375, 390 Prosimians, 386, 391 Reproduction and lactation, 377 Rhesus macaque (Maccaca mulatta), 27, 91, 380, 382, 386 Social status, 377 Sooty mangabey (Cercocebus atys), 391 Strepsirrhines, 375 T cell antigen receptors, 391
392 Protecton theory, 408 Prototheria, 6, 171 Pseudogymnoascus destructans, 287 Pulmonary intravascular macrophages, 302 PYHIN proteins, 285 Pyroptosis, 80, 301

Q Quadrate bone, 4
5 Quokka (Setonyx brachyurus), 187, 189

R r/K trade-off, 408 Rabbits. See Lagomorphs Rats. See Rodentia Receptor editing, 129 Recombination activating gene (RAG), 117, 126
127 Recombination signal sequences (RSS), 122, 126
127, 365 Red kangaroo (Macropus rufus), 186 Red necked wallaby (Macropus rufogriseus), 196 Red-Queen concept, 55, 92 Red squirrel (Sciurus vulgaris), 362 Regakine-1, 257 Regulatory T cells. See Treg cells Reproduction and lactation Bats, 282 Camels, 206 Cats, 299 Cetaceans, 238 Dogs, 311
312 Elephants, 398 Horses, 330 Lagomorphs, 342 Marsupials, 186
190 Monotremes, 171
172 Pigs, 220
221 Primates, 377 Rodents, 357 Ruminants, 252
254 Xenarthra, 412
413 Respiratory tract microbiota, 57
58 Retinoic acid, 65 Retinoic acid-related orphan receptor -α, 135
136 Retroviruses, endogenous, 21 Rhesus macaque (Macacca mulatta). See Primates Rhinoceros, 329, 333 RNA interference, 72 Rodentia B cells, 364 Capybara (Hydrochoerus hydrochaeris), 370
371 Great gerbils (Rhombomys opimus), 369 Guinea pigs (Cavia porcellus), 369
370 Immunoglobulins, 370 Kurloff cells, 370 Hamsters (Mesocricetus ssp), 371 Hematology, 357 Blood leukocytes, 370 Immunoglobulins Fc receptors, 366 Guinea pigs, 370 Mouse, 364
366 Prairie voles, 369 Rat, 368
369 Innate immunity, 358

Index

Lymphoid organs, 359 Major histocompatibility complex (H-2) Mice, 359
362 Rats, 368 Mole rats (Heterocephalus glaberi, Spalax ehrenbergi) Cancer resistance, 371 Life-span, 371 Microbiota, 372 Myomorpha, 357 Mx genes, 358 Natural killer cell receptors, 363
364 Natural killer T cells, 367 Phylogeny, 355, 356f Prairie voles (Microtus ochrogaster), 369 Rats (Rattus norvegicus) Major histocompatibility complex RT1, 89
90, 368 Reproduction and lactation, 357 T cells receptors, 366
367 Thirteen-lined ground squirrel (Icidomys tridecemlineatus), 343 Thy-1 glycoprotein, 367
368 Wild versus laboratory rodents, 356
357, 357f Ruminants Bovines, 251
271 B cells, 264 Colostrum, 252
254 Dendritic cells, 264 Goats (Capra hircus), 271, 273, 275 Hematology, 254
255, 254t Immunoglobulins, 264
267 Innate immunity, 255
257 Lymphoid organs, 257
260 Major histocompatibility complex (BoLA), 260
262 Milk, 253 Natural killer complex, 263 Phylogeny, 251, 252f Reproduction and lactation, 252
253 Sheep (Ovis aires), 271
275 T cells, 112, 267
270 Virus diarrhea virus, 35, 254 Water buffalo (Bulbalis bulbalis), 275 Yak (Bos grunniens), 275 Workshop cluster 1 antigens, 270
271 Ruminococcus, 58
59

S Sacculus rotundus, 153
154 Saiga antelope (Saiga tatarica), 10
11 Sauropsids, 3, 4f Scandentia, 415
416 B cell responses, 416 Chinese tree shrew (Tupaia belangeri), 416f Innate Immunity, 415
416 Major histocompatibility complex, 416 Natural killer cell receptors, 416 Phylogeny, 415f Toll-like receptors, 416 Scavenger receptor, cysteine-rich family proteins, 270

Seals. See Pinnipeds Sea lions. See Pinnipeds Sea otter (Enhydra lutris), 322 Secondary lymphoid organs, 154, 159
160, 163 Secretory IgA, 32
33 Segmented filamentous bacteria, 57
58 Sentinel cells, 83 Serum amyloid A (SAA), 64
65, 222, 286, 300
301, 401 Serum amyloid P, 79, 286 Sheep (Ovis aires). See also Ruminants B cells, 273 Domestication, 271 Immunoglobulins, 273 Innate immunity, 272 Lymphoid organs, 272 Major histocompatibility complex, 272
273, 273f Natural killer cell receptors, 273 Reproduction and lactation, 271
272 T cells, 274 Antigen receptors, 274
275 Shell membrane, 16, 186 Short-beaked echidna (Tachyglossus aculeatus), 171
172, 177 Short-fuse model, 7 Shrews. See Eulipotyphla Sickness behavior and predation, 11, 78, 298 Signal transducing components (TCR), 104
106, 121 Simpson, George Gaylord, 8 Sirenia. See Afrotheria Skin microbiota, 56
57 Sloths, 411
415 Snowshoe hare (Silvilagus sp.), 348 Social rank and immunity, 307, 377 Somatic hypermutation Immunoglobulins, 129
130 T cell receptors, 217 Sosuga virus, 281
282 Sperm, 27 Spleen, 154
158, 156f Accessory, 158, 241f Bats, 287 Camels, 208 Cats, 302 Cetaceans, 240 Defensive, 157 Dogs, 315 Elephants, 401 Functions, 156
157 Horses, 333 Lagomorphs, 345 Marsupials, 193, 193f Monotremes, 174 Pigs, 223 Primates, 379 Red pulp, 155 Rodents, 359 Ruminants, 257 Storage, 157 Types, 156
157 White pulp, 155
156

431

Xenarthra, 414 Steller’s sea cows (Hydrodamalis gigas), 44 Stimulator of interferon genes (STING), 77, 284
285, 371 Streptococci, 79, 163, 331, 389 Striped dolphins (Stenella coeruleoalba), 161, 240 Striped hyaenas (Hyaena hyaena), 308 Sugar glider (Petaurus breviceps), 166 Suiformes. See Pigs Superantigens, 350
351 Surfactant proteins, 256 Synapsids, 4
5 Syncytin-1, 21 Syrian hamster (Mesocricetus aureus), 371

T T-bet, 335 T cells, classification, 102f Evolution, 101
102 Habitat links, 111 Phenotypes, 102f Receptors Alpha, 109 Antigen-binding, 106 Beta, 109 Butyrophilins and, 114 Delta, 109 Diversity, 107, 110t Functions, 63
64, 106 Gamma, 109 Gene duplication, 105b Gene structure, 107
111 Mu, 109
111, 199
200 Signal transduction, 104
106, 105t Somatic mutation, 107 Structure, 102
106, 104t Helper, 102 Helper 17 Cells (Th17), 64
65 Memory T cells, 114
115 Microbiota and, 63
65 Regulatory (Treg), 64 Microbiota and, 63f Species differences γ/δ high species, 111
113 γ/δ low species, 113
114 Tammar wallaby (Notamacropus eugenii), 16, 95, 175
176, 186
187, 189 Tapirs, 329 Tasmanian devil (Sarcophilus harrisii), 185, 187 Tenrec, hedgehog (Echinops telfairi), 43, 397, 401 Tertiary lymphoid organs, 166 Th17 cells, 23, 64
65 Therapsids, 4
5 Thermogenin, 43
44 Thy1, 367
368 Thymulin, 151 Thymus Bats, 287 Camels, 207 Cats, 302

432

Index

Thymus (Continued) Cetaceans, 239, 240f Dogs, 315 Elephants, 401 Function, 151 Hormones, 151 Horses, 333 Involution, 151
152 Lagomorphs, 344 Marsupials, 192
193 Monotremes, 174 Pigs, 223 Primates, 379 Ruminants, 257, 258f Species differences, 152 Structure, 150
151, 150f Xenarthra, 414 TNF-related apoptosis-inducing ligand (TRAIL), 23 Tolerance, maternal-fetal, 21
23, 21f Toll-like receptors, 72
76 Bats, 283 Camels, 207 Cats, 301 Cell surface, 74 Elephants, 400 Endosomal, 74
75 Horses, 331 Lagomorphs, 343 Ligands, 73t Manatees, 400 Pigs, 221 Primates, 378 Receptor signaling, 75
76, 344 Rodents, 358 Ruminants, 255 Scandentia, 416 Sheep, 272 Tonsils, 163, 163t Anal, 242 Camels, 208 Cats, 303 Cetaceans, 241
242, 242f Dogs, 315 Elephants, 401 Horses, 333 Lagomorphs, 345

Marsupials, 194 Monotremes, 175 Pigs, 224 Primates, 379, 380f Rhinoceros, 333 Rodents, 359 Ruminants, 258
259 Vaginal tonsils, shrew, 418 Torpor, 46 Trained immunity, 143
144 Transforming growth factor beta, 61 Transient receptor potential cation channels, 45 Tree shrews, 415
416 Treg cells, 63f, 64 Treponema pallidum, 49 Triassic mass extinction event, 4 Trichosurin, 188 Trophoblast, 19, 377 Trophectoderm, 220 Tuberculosis, 43b Tumor necrosis factor-alpha, 75, 80, 95, 143, 160, 318, 362 Tylopoda. See Camels

U Uncoupling protein-1, 43
44 Uterus, NK cells in, 22, 24, 139 UT gene family (MHC), 176

V Vaginal tonsils, 418 Van der Waals forces, 106 Variable regions, immunoglobulins T cell receptors, 106
107 Ultralong V regions, cattle, 265
266 Vascular endothelial growth factor, 23 Venom, platypus, 173 Vervet monkeys (Chlorocebus aethiops), 46
47, 380 VHH gene segments, 214 Vicugna (Vicugna vicugna), 206 Vitamin A, 65 Vitellogenins, 15 Viviparity, evolution of, 15
19 Vγ9Vδ2 T cells, 217

W Waldeyer’s ring, 379 Wallaroos (Osphranter robustus), 190 Water buffalo (Bulbalis bulbalis), 275 Weddell seal (Leptonychotes weddellii), 324 Whales. See Cetaceans White beaked dolphin (Lagenorhynchos obliquidens), 246 White Nose syndrome, 287 White pulp, spleen, 155
156 White rhinoceros (Ceratotherium simum), 337 Workshop cluster 1 (WC1), 231, 270
271

X Xanthine oxidoreductase, 30 Xenarthra Armadillos, 411
412, 413t, 414
415 B cell responses, 415 Evolution, 411 Giant anteater (Myrmecophaga tridactyla), 414 Hematology, 414 Blood leukocytes, 413t Immunoglobulins, 415 Lesser anteater (Tamandua tetradactyla), 412 Lymphoid organs Armadillos, 414 Sloths, 414 Major histocompatibility complex, 414 Natural killer cells, 414
415 Phylogeny, 412f Reproduction and lactation, 412 Sloth, two toed (Choloepus hoffmanni), 412 T cell Receptors, 415

Y Yaks (Bos grunniens), 275 Yangochiroptera, 281 Yinpterochiroptera, 281, 288
289 Yolk sac, 15
16