Interleukin-21 in immunity and autoimmunity

Alexis Vogelzang

A thesis submitted for the degree of Doctor of Philosophy in the Faculty of Medicine, University of New South Wales

Mucosal Autoimmunity Unit, Garvan Institute of Medical Research Sydney, Australia

Awarded September 2010

1 ORIGINALITY STATEMENT

‘I hereby declare that this submission is my own work and to the best of my knowledge it contains no materials previously published or written by another person, or substantial proportions of material which have been accepted for the award of any other degree or diploma at UNSW or any other educational institution, except where due acknowledgement is made in the thesis. Any contribution made to the research by others, with whom I have worked at UNSW or elsewhere, is explicitly acknowledged in the thesis. I also declare that the intellectual content of this thesis is the product of my own work, except to the extent that assistance from others in the project's design and conception or in style, presentation and linguistic expression is acknowledged.’

Signed ……………………………………………...... Alexis Vogelzang

Date ……………………………………………......

2 COPYRIGHT STATEMENT

‘I hereby grant the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all proprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

Signed ……………………………………………......

Date ……………………………………………......

AUTHENTICITY STATEMENT

‘I certify that the Library deposit digital copy is a direct equivalent of the final officially approved version of my thesis. No emendation of content has occurred and if there are any minor variations in formatting, they are the result of the conversion to digital format.’

Signed ……………………………………………......

Date ……………………………………………......

3 ABSTRACT

T cell help to B cells is a fundamental property of adaptive immunity, yet only recently have many of the cellular and molecular mechanisms of T cell help emerged. T follicular helper (Tfh) cells are the CD4+ T helper cells that provide cognate help to B cells for high affinity antibody production in germinal centres (GC). This study has reveals a critical role of IL-21 in the upregulation of Tfh signature molecules. Expression of ICOS was found to be necessary for optimal production of IL-21; indicative of interplay between two Tfh expressed molecules. We also demonstrate that IL-21's costimulatory capacity for T helper differentiation operates at the level of the TCR through Vav1 signalling that controls T cell helper function and survival. Tfh cells express uniquely high levels of the IL-21 receptor relative to other T cell subsets, which reflects an IL-21- driven autocrine loop that is important for the generation and function of Tfh cells, which in turn were critical for supporting primary and secondary T dependent antibody responses. This study reveals a previously unappreciated role for Tfh cells in the formation of the GC through a CD4+ T cell intrinsic requirement for responsiveness to IL-21. IL-2 and IL-21 are crucial growth factors for distinct T helper subsets with opposing regulatory and effector functions, respectively. Mice made genetically deficient in IL-2 or its high affinity receptor chain (CD25) suffer from a fatal autoimmune disease characterized by ulcerative colitis and haemolytic anaemia. The observed autoimmunity and associated splenomegally are thought to be caused, in part, by a loss of regulation of effector T cells due to a deficit in IL-2- dependent Foxp3 regulatory T cells. Since IL-21 is known to facilitate the development of a number of autoimmune diseases, the possible contribution of IL-21 to the autoimmune pathology observed in Il2-/- mice was investigated in this study. Our findings demonstrate that IL-21:IL-21R signalling contributes to the destruction of tissues attributed to autoimmunity in Il2-/- mice and suggests that IL-21-producing T cells are significant targets of immune system regulation.

4 PUBLICATIONS, PRESENTATIONS AND PRIZES ARISING FROM

THIS THESIS

PUBLICATIONS Vogelzang, A., McGuire, H.M., Yu, D., Sprent, J., Mackay, C.R., and King, C. (2008). A fundamental role for interleukin-21 in the generation of T follicular helper cells. Immunity 29, 127- 137. Vogelzang, A., and King, C. (2008a). The modulatory capacity of interleukin-21 in the pathogenesis of autoimmune disease. Frontiers in Bioscience 13, 5304-5315. McGuire, H.M., Vogelzang, A., Hill, N., Flodstrom-Tullberg, M., Sprent, J., and King, C. (2009). Loss of parity between IL-2 and IL-21 in the NOD Idd3 locus. Proceedings of the National Academy of Sciences of the United States of America 106, 19438-19443.

ORAL PRESENTATIONS A.Vogelzang, H. McGuire, J. Sprent, C. King. Interleukin-21 in a model of ulcerative colitis. Australasian Society for Immunology, NSW branch meeting. Bowral, Australia (September 2009) A.Vogelzang, H. McGuire, J. Sprent, C. King. The role of Interleukin-21 in T Follicular Helper cell differentiation. Australasian Society for Immunology, 37th Annual Scientific Meeting. Canberra, Australia (Dec 2008) A.Vogelzang, H. McGuire, J. Sprent, C. King. A fundemental role for Interleukin-21 in the generation of T Follicular Helper Cells. The Federation of Immunological Societies of Asia- Oceania 4th Congress. Taipei, Taiwan (October 2008) A.Vogelzang, H. McGuire, C. Mackay, J. Sprent, C. King. The role of Interleukin-21 in T Follicular Helper cell differentiation. Australasian Society for Immunology, 36th Annual Scientific Meeting. Sydney, Australia (Dec 2007) A.Vogelzang, H. McGuire, C. Mackay, J. Sprent, C. King. The role of Interleukin-21 in T Follicular Helper cell differentiation. 17th St Vincents & Mater Health Sydney Research Symposium. Sydney, Australia (Sept 2007)

5 POSTER PRESENTATIONS A.Vogelzang, H. McGuire, J. Sprent, C. King. A fundamental role for Interleukin-21 in germinal centre antibody responses. Immunology 2009 96th Annual Meeting with the American Association of Immunologists. Seattle WA, USA (May 2009). A.Vogelzang, H. McGuire, C. Mackay, J. Sprent, C. King. The role of Interleukin-21 in T Follicular Helper cell differentiation. Frontiers in Immunological Memory Newport Beach CA, USA (Septemeber 2008).

PRIZES Postgraduate Research Student Support Scheme Award (UNSW) for travel to AAI Immunology 09 96th Annual Meeting with the American Association of Immunologists, Seattle, USA. 2009 Prestigious New Investigator Prize at the Australasian Society of Immunologists (ASI) Annual meeting, 2008. Australasian Society of Immunologists Travel Bursary of $3000 awarded to travel to the Federation of Immunological Societies of Asia-Oceania 4th Congress, 2008.

6 ACKNOWLEDGMENTS

Thank you to my supervisor Cecile King for your guidance and constant enthusiasm during this project. Thanks also to Helen McGuire who has helped on so many long days in the laboratory, becoming a great friend in the process. I’ve really enjoyed working closely with you both, I think we make a great team and I look forward to working with you in the future. Thank you to Kylie Webster, Rachael Kohler, Jaeho Cho and Heeok Kim, Marcel Batten and Jon Sprent for all your sage advice during lab meetings and meetings in labs. Thank you to Chris and Yovina for running a wonderful flow cytometry service, the many BTF staff who have helped me in the past four years, and the staff of the Garvan Institute which has been a great workplace. Thank you also to the members of the Brink, Silveira, Tangey, 2xMackay and Sprent labs for your generous provision of reagents and help during my PhD. Lastly, thanks to all my friends and family for providing me with support, sympathy, and big dinners while I’ve been a student, especially Blake and my parents who were extremely generous with all of the above throughout.

Experiments that were not the sole work of the author All experiments were performed by the author at the Garvan Institute, with the following exceptions; TaqMan ABI assays were carried out by Helen McGuire on samples prepared by the author; Paraffin embedding of histology samples was done by the St Vincents Hospital pathology unit, while sections and H&E staining of histology was performed by Alice Boulghourjian of the Garvan Institute Histology Unit. Assessment of pathology of the colon and pancreas sections was generously undertaken by Doctor Peter Earls of the St Vincents Hospital pathology unit.

7 TABLE OF CONTENTS

1 Introduction ...... 18 1.1 A brief review of the mammalian immune system...... 18 1.1.1 Primary lymphoid organs...... 18 1.1.2 T cell development...... 18 1.1.3 B cell development...... 19 1.1.4 Secondary Lymphoid organs ...... 19 1.2 Innate and adaptive immune responses...... 21 1.2.1 Innate immune responses...... 21 1.2.2 Adaptive immune responses ...... 22 1.2.3 Central T cell tolerance...... 22 1.2.4 Antigen presentation in the periphery ...... 23 1.2.5 T cell activation...... 24 1.2.6 Peripheral T cell tolerance ...... 25 1.2.7 CD8+ T cells ...... 26 1.2.8 CD4+ T helper cells ...... 26 1.2.9 T cell memory and homeostasis ...... 29 1.2.10 T helper subsets in allergy and autoimmunity ...... 30 1.2.11 B cell activation ...... 30 1.3 T follicular B helper cells ...... 32 1.3.1 TCR affinity and Tfh cell development ...... 33 1.3.2 Tfh cytokines...... 33 1.3.3 IL-21 and Tfh cells...... 34 1.3.4 Tfh cell positioning in the GC ...... 34 1.3.5 Tfh cell co-stimulatory molecules ...... 35 1.3.6 Transcriptional regulation of Tfh cells...... 36 1.3.7 Tfh memory...... 36 1.4 Interleukin-21...... 37 1.4.1 Cytokines...... 37 1.4.2 Interleukin-21...... 37 1.4.3 IL-21 signalling pathways...... 38 1.4.4 IL-21 and T cell priming...... 38 1.4.5 Th1 and Th2 cells...... 39 8 1.4.6 Th17 cells ...... 39 1.4.7 T regulatory cells...... 40 1.4.8 CD8+ T cells ...... 40 1.4.9 Opposing effects of IL-21 on B cells ...... 42 1.4.10 Transgenic IL-21 mouse models ...... 43 1.4.11 IL-21 deficiency and B cells...... 43 1.4.12 IgE production...... 43 1.4.13 NK T cells and NK cells...... 44 1.5 IL-21 in human autoimmune disease and lessons from animal models...... 44 1.5.1 Autoimmune Disease: breakdown of tolerance ...... 44 1.5.2 Systemic Lupus Erythematosus (SLE)...... 45 1.5.3 Rheumatoid Arthritis (RA) ...... 46 1.5.4 Multiple Sclerosis (MS) and Experimental Allergic Encephalitis (EAE)...... 46 1.5.5 Type-1 diabetes ...... 47 1.5.6 Gastro-intestinal inflammation and autoimmunity ...... 49 1.5.7 IL-21 in autoimmunity - concluding remarks ...... 49 1.6 Experimental objectives...... 51 2 Materials and methods ...... 52 2.1 Buffers ...... 52 2.2 Mice...... 53 2.3 Flow cytometry ...... 54 2.4 Intracellular staining ...... 56 2.5 Biotinylation of IL-21...... 56 2.6 BrdU Proliferation studies ...... 56 2.7 CFSE Proliferation studies...... 56 2.8 Bone Marrow Chimeras...... 56 2.9 Immunisations...... 57 2.10 Adoptive transfer studies...... 57 2.11 Primary cell sorting...... 57 2.12 Primary cell culture and proliferation studies ...... 58 2.13 T cell suppression assay ...... 58 2.14 Immunohistochemistry and immunofluorescence...... 58 2.15 Enzyme Linked Immunosorbance Assay (ELISA) ...... 59 2.16 Cytokine Bead Array ...... 59

9 2.17 Single cell sorting of GC B cells and Somatic Hypermutation assay...... 60 2.18 LPL and IEL isolation ...... 60 2.19 SDS Page and Immunoblotting...... 61 2.20 RNA analysis ...... 61 2.21 Polymerase chain reaction...... 62 2.22 Data analysis and Statistics ...... 63 3 An overview of IL-21 in the adaptive immune system ...... 64 3.1 Introduction ...... 64 3.2 Results ...... 66 3.2.1 IL-21 is redundant for immune development and homeostasis...... 66 3.2.2 IL-21 is produced by activated effector phenotype CD4+ T cells ...... 69 3.2.3 ICOS expression on CD4+ T cells modulates IL-21 production ...... 72 3.2.4 IL-21 costimulates the TCR to modulate activation markers...... 75 3.2.5 IL-21 acts at the level of the TCR to impact T helper cell fate ...... 78 3.3 Discussion ...... 81 4 A fundamental role for IL-21 in the generation of Tfh Cells ...... 83 4.1 Introduction ...... 83 4.2 Results ...... 85 4.2.1 IL-21 is necessary for GC formation in response to T dependent antigen...... 85 4.2.2 IL21 is an autocrine growth factor for Tfh...... 87 4.2.3 The GC defect in Il21-/- mice is present early in the response to SRBC...... 88 4.2.4 Tfh cells express the receptor for IL-21 ...... 90 4.2.5 Tfh have a specific requirement for co-stimulation through IL-21R ...... 90 4.2.6 Tfh cells are distinct from Th17 cells...... 92 4.2.7 The GC defect in IL-21 deficient mice is CD4+ T cell intrinsic...... 92 4.2.8 GC B cells require IL-21 acting on T cells ...... 96 4.2.9 Dose of IL-21 responsive CD4+ T cells correlates with GC formation...... 99 4.2.10 Il21r-/- T and B cells compete poorly with their WT counterparts ...... 100 4.2.11 Il21r-/- mice have poor GC responses to the T-dependent haptenated protein

NP13OVA...... 104 4.2.12 Il21r-/- OTII T cells restore GC B cells in response to NP-OVA but fail to reconstitute recall responses...... 106 4.2.13 IL-21 acting on CD4+ T cells is required for affinity maturation of IgG1 in response to T dependent antigen ...... 110

10 4.2.14 IL-21 is required for OTII Tfh expansion or survival ...... 112 4.2.15 IL21 aids early clonal expansion of antigen specific T cells...... 114 4.3 Discussion ...... 122 5 IL-21 in autoimmunity...... 126 5.1 Introduction ...... 126 5.2 Results ...... 128 5.2.1 Autoimmune Il2-/- mice produce high levels of IL-21...... 128 5.2.2 IL-21 drives morbidity in autoimmune Il2-/- mice...... 129 5.2.3 Histological analyses of colitis and pancreatitis in Il2-/- mice...... 132 5.2.4 Exacerbation of Treg defect in the absence of IL-21...... 136 5.2.5 Altered CD4+ T cell effector phenotype in Il2Il21r-/- mice ...... 137 5.2.6 IL-21 shapes the serum cytokine profile during autoimmunity ...... 140 5.2.7 IL-21 drives expansion of the autoimmune CD8+ T cell compartment ...... 143 5.2.8 IL-21 promotes T dependent antibody production in Il2-/- mice...... 146 5.3 Discussion ...... 151 6 General Discussion ...... 156 6.1 Research outcomes...... 156 6.2 Clinical Relevance ...... 162 7 References ...... 165

11 LIST OF FIGURES

Figure 1-1 Structure of murine spleen and lymph node ...... 20 Figure 1-2 Treg differentiation in the thymus and periphery...... 28 Figure 3-1 Il21-/- thymic T cell populations are equivalent to WT...... 66 Figure 3-2 Normal expression of T cell co-stimulation molecules by Il21-/- mice ...... 67 Figure 3-3 Il21-/- mice have a normal peripheral T cell phenotype ...... 68 Figure 3-4 Normal peripheral B cell phenotype of Il21-/- mice ...... 68 Figure 3-5 IL-21 production is elevated at mucosal immune sites ...... 69 Figure 3-6 IL-21 is produced by memory/effector phenotype CD4+ T cells...... 70 Figure 3-7 IL-21+ CD4+ T cells at mucosal sites have a recently activated phenotype...... 71 Figure 3-8 ICOS expression affects IL-21 mRNA levels in spleen...... 72 Figure 3-9 B cells enhance IL-21 production by CD4 T cells via ICOS:ICOSL ...... 73 Figure 3-10 Costimulation through ICOS and CD28 drives IL-21 expression ...... 74 Figure 3-11 Exogenous IL-21 recovers Il21-/- CD4+ T cell proliferation defect ...... 75 Figure 3-12 T helper phenotype is modulated by IL-21 co-stimulation ...... 76 Figure 3-13 Exogenous IL-21 can recover Il21-/- ICOS defect...... 77 Figure 3-14 CD28 expression is normal in Il21-/- mice...... 77 Figure 3-15 IL-21 compounds TCR signals to CD4+ T cells ...... 78 Figure 3-16 IL-21 alone can phosphorylate Vav1 ...... Error! Bookmark not defined. Figure 3-17 IL-21 acts downstream of proximal Zap70 activation...... 79 Figure 3-18 Tfh cells generated in response to T dependent antigen exhibit strong Vav1 phosphorylation...... 80 Figure 4-1 T dependent GC formation is reduced in Il21-/- mice...... 86 Figure 4-2 Reduced serum IgG1 production by Il21-/- mice in response to T dependent antigen...... 87 Figure 4-3 IL-21 deficiency effects Tfh cell generation following T dependent immunisation...... 88 Figure 4-4 Reduced generation of Tfh cells from Il21-/- CD4+ T cells...... 88 Figure 4-5 Timecourse of GC defect Il21-/- mice ...... 89 Figure 4-6 Both IL-21 and IL21R are expressed by Tfh cells ...... 90 Figure 4-7 Agonistic anti-CD28 treatment boosts Treg populations ...... 91 Figure 4-8 Agonistic anti-CD28 treatment cannot recover the Il21-/- Tfh cell defect...... 91 Figure 4-9 IL-21 producing Tfh cells are distinct from Th17 cells...... 92 Figure 4-10 Donor WT CD4+ T cells proliferate and assume a Tfh phenotype ...... 93 Figure 4-11 IL-21 responsive T cells boost GC B cells in Il21r-/- mice ...... 94 12 Figure 4-12 GC structures are restored to Il21r-/- mice by transfer WT CD4+ T cells...... 95 Figure 4-13 WT CD4+ T cells can restore generation of GC but not plasma cells from Il21r-/- B cells...... 97 Figure 4-14 WT CD4+ T Cells promote GC in Il21r-/- mice ...... 98 Figure 4-15 IL-21 acting on CD4+ T cells increases high and low affinity antibody titres...... 99 Figure 4-16 IL-21 responsive CD4+ T cells display increased Tfh cell markers...... 100 Figure 4-17 Mixed chimerism of lymphocytes in irradiated recipients...... 96 Figure 4-18 IL-21 is required for Tfh cell formation...... 101 Figure 4-19 Il21r-/- T and B cells compete poorly with WT cells in BM chimeras ...... 102 Figure 4-20 WT and Il21r-/- CD4+ T cells migrate to the B cell follicle in mixed chimeras ...... 103 Figure 4-21 Time course of GC B cell generation after NP OVA immunization...... 104

Figure 4-22 The T dependent GC response to NP13OVA is reduced in Il21r-/- mice...... 105 Figure 4-23 Il21r-/- GC B cells responses can be restored by donor OTII T cells ...... 107 Figure 4-24 IL-21 is not required for effective class switch to IgG1 isotype...... 108 Figure 4-25 Il21r-/- OTII T cells were unable to support secondary immune response...... 109 Figure 4-26 IL-21 responsive CD4+ T cells are important for the production of high affinity antibody from B cells...... 110 Figure 4-27 Il21r-/- are capable of SHM ...... 112 Figure 4-28 IL-21 supports Tfh cell numbers and the phenotype of donor OTII T cells...... 113 Figure 4-29 Il21r-/- T cells are excluded from the B cell follicle at day 2.5...... 115 Figure 4-30 Il21r-/- OTII T cells migrate poorly to the B cell follicle...... 116 Figure 4-31 IL-21 aids early clonal expansion of antigen specific OTII T cells...... 117 Figure 4-32 IL-21 is necessary for CD4+ T cells to form stable conjugates with B cells ...... 118 Figure 4-33 B cells and IL-21 conversely effect PD-1 expression on antigen specific T cells ...... 119 Figure 4-34 IL-21 maintains survival signals in antigen specific CD4+ T cells ...... 120 Figure 4-35 IL-21 sustains Bcl-xL in polyclonal Tfh cells ...... 121 Figure 5-1 IL-21 production is enhanced in autoimmune Il2-/- CD4+ T cells...... 128 Figure 5-2 Improved survival of Il2 Il21-/- mice compared to Il2-/- littermates ...... 129 Figure 5-3 Reduced wasting disease in Il2 Il21r-/- mice ...... 130 Figure 5-4 Enlarged secondary lymphoid organs in both Il2-/- strains ...... 131 Figure 5-5 Increased cell numbers is spleen but not mucosal sites in Il2 Il21r-/-mice...... 131 Figure 5-6 Crypt branching and mononuclear cell infiltrate in Il2-/- strains...... 133 Figure 5-7 Mucosal barrier integrity despite mononuclear infiltration of Il2-/- strains...... 134 Figure 5-8 Severe pancreatitis in Il2-/- mice is improved in the absence of IL-21 ...... 135

13 Figure 5-9 Cellular infiltrates at mucosal sites of aged Il2 ll21r-/- mice...... 135 Figure 5-10 Further decrease in T reg cells in Il2 Il21r-/- mice...... 136 Figure 5-11 Il-21 is not necessary for in vitro Treg suppression ...... 137 Figure 5-12 Expanded CD4+ T cells in Il2-/- strains with a mucosal homing phenotype both in the presence and absence of IL-21...... 138 Figure 5-13 Ameliorated Th17 T cell phenotype in the absence of IL-21 ...... 139 Figure 5-14 Limited influence of IL-21 on Th1 cytokines at mucosal sites...... 140 Figure 5-15 Altered serum cytokine profile in Il2-/- mice ...... 141 Figure 5-16 IL-21 does not effect IL-10 production by CD4+ T cells ...... 142 Figure 5-17 IL-21 drives CD8+ T cell expansion in Il2-/- mice...... 143 Figure 5-18 Il21r-/- hosts drive proliferation of adoptively transferred memory CD8+ T cells...... 144 Figure 5-19 High circulating IL-21 drives homeostatic proliferation of CD8+ T cells ...... 145 Figure 5-20 Both IL-2 and IL-21 are necessary for CD8+ T cell proliferation...... 146 Figure 5-21 B cell numbers in Il2-/- mice increase in the absence of IL-21 ...... 147 Figure 5-22 IL-21 drives GC B cell and plasma cell generation or survival in Il2-/- mice ...... 148 Figure 5-23 IL-21 enhances GC and plasma B cell proportions in Il2-/- mice ...... 148 Figure 5-24 IL-21 supports Tfh like cells in Il2-/- mice...... 149 Figure 5-25 IL-21 promotes T dependent antibody isotype production in Il2-/- mice...... 150 Figure 6-1 Schematic diagram showing that IL-21:IL21R interactions activate Vav1...... 158

14 LIST OF TABLES

Table 2-1 Antibody clones and concentrations…………………………….. 51 Table 2-2 Antibody clones and concentrations…………………………….. 53 Table 2-3 Primer sequences ……………………………………………………62 Table 4-1 SHM non-silent amino acid mutation analyses ………………….113

15 ABBREVIATIONS

AID Activation induced cytidine deaminase AICD Activation induced celld eath APC Allophycocyanin APC Antigen-presenting cell B6 C57BL/6 BALT Bronchial associated lymphoid tissue Bcl B cell lymphoma Bcl-xl B cell lymphoma extra large BCR B cell receptor BCS Bovine calf serum BrdU Bromodeoxyuridine Blimp-1 B lymphocyte induced maturation protein 1 BSA Bovine serum albumin CFSE Carboxyfluorescein succinimidyl ester EAE Experimental Allergic Encephalomyelitis EDTA Ethylenediamine-tetraacetate ELISA Enzyme-linked immunosorbent assay FITC Fluorescein isothiocyanate CCR Chemokine motif receptor CCL19 Chemokine motif ligand CXCL21 Chemokine CXC motif ligand CD Cluster of differentiation CFSE Carboxyfluorescein succinimidyl ester CTLA-4 Cytotoxic T lymphocyte antigen 4 CXCR5 CXC motif receptor 5 DC Dendritic cell DN Double negative thymocytes DNA Deoxyribose nucleic acid DP Double positive thymocytes ERK Extracellular signal related kinase FACS Flourescence activated cell sorting FDC Follicular dendritic cell FO Follicular GAPDH Glyceraldehyde 3 phosphate dehydrogenase GC Germinal centre ICAM-1 Inter cellular adhesion molecule 1 ICOS Inducible co-stimulation factor Ig Immunoglobulin IL Interleukin IFN Interferon- JAKs Janus family tyrosine kinases JNK Jun N-terminal kinase LFA-1 Lymphocyte function associated antigen 1 LPS Lipopolysaccharide mAb Monoclonal antibody MACS Magnetic activated cell sorting MALT Mucosal associated lymphoid tissue

16 MAPK Mitogen activated protein kinases MHC Major histocompatibility complex MS Multiple sclerosis MZ Marginal zone NFAT Nuclear factor of activated T cells NF- Nuclear factor- NK cells Natural Killer cells NKT cells Natural Killer T cells NOD Non-obese diabetic PAMP Pathogen associated molecular patterns PD-1 Programmed death-1 PD-L Programmed death ligand PBS Phosphate buffered saline PE R-phycoerythrin PCR Polymerase chain reaction PerCP Peridinin chlorophyll protein PLC1 Phospholipase C 1 PMA Phorbol myristate acetate PRR Pattern recognition receptors RA Rheumatoid arthritis Ras Rat sarcoma superfamilily RBC Red blood cells RORt Retinoic-acid-receptor-related orphan receptor-t RNA Ribonucleaic acid RT room temperature SAP SLAM associated protein SCID Severe combined immunodeficiency SLAM Signalling lymphocytes activation molecule SLE systemic lupus erythematosus SMH somatic hypermutation SNP single nucleotide polymorphism SP Single positive thymocytes SRBC sheep red blood cells STAT Signal transducers and activators of transcription T1D Type 1 diabetes TCR T cell receptor Tfh Follicular B helper T cells TGF- Transfering growth factor Tg Transgenic Th T helper TLR Toll like receptor TNF Tumour necrosis factor Treg regulatory T cells VDJ Variable, diversity and junction segments WT wild-type ZAP-70 -associated protein 70

17 1 Introduction

1.1 A brief review of the mammalian immune system

The mammalian immune system is a complex cellular network, which protects the body from invading pathogens. Immune protection derives from two arms of the immune response. Innate (natural) immunity can respond promptly to conserved molecules on pathogens. Secondly, the adaptive (acquired) immune cells expand in response to specific antigens and create long-lived protection via memory B and T cells. T cells that express the CD4 co-receptor wield immense influence over many diverse downstream arms of the immune system such as protective antibody production by B cells and the mobilisation of cytotoxic CD8+ T cells which can lyse invading pathogens or infected cells. These effects are carried out through cell-to-cell interactions and are also influenced by soluble factors such as cytokines. This thesis is focussed on the how a particular cytokine, IL-21, modulates CD4+ T cells to influence some of the adaptive immune outcomes that stem from their diverse functions within the immune system.

1.1.1 Primary lymphoid organs

Lymphoid organs are positioned around throughout the body and their specialised roles in immunity are largely based on their location. The bone marrow and the thymus are primary lymphoid organs where immature immune cells develop. Immune cell arise from hemopoietic precursors, which are derived from bone marrow stem cells.

1.1.2 T cell development

Thymocytes, or T cells develop in the thymus from bone precursors that migrate from the bone marrow. They begin as double negative for the CD4 and CD8 T cell co-receptors (DN CD4- CD8-), then become immature single positive cells (ISP CD4-CD8lo), double positive (DP CD4+CD8+) and finally become either single CD4 or CD8+ (SP CD4+ or CD8+). During this process the gene segments of the T cell receptor (TCR) are rearranged to form a unique specificity for a particular epitope that can signal in concert with either CD4 or CD8. Thymocytes also undergo both positive selection to ensure their TCR displays sufficient affinity for self major histocompatibility complexes (MHC) that display antigen, and negative selection which eliminates TCR rearrangements that can respond to self epitopes bound to MHC (Alam, 2003; Wiest, 1999).

18 Mature T cells exit from the thymus and circulate through the lymphatic system in search of antigen. CD4+ T cells regulate cellular immune responses against intra- and extra-cellular pathogens and humoral antibody responses against extra-cellular pathogens, whereas CD8+ T cells directly kill cells infected by intracellular pathogens.

1.1.3 B cell development

Early B cell development takes place in the bone marrow where germ-line heavy and light chain immunoglobulin genes are rearranged. Hemopoietic stem cells commit to the B cell lineage and become pro B cells. The variable (V), diversity (D) and junctional (J) segments of the immunoglobulin heavy chain locus are then rearranged by excision of the intervening DNA, resulting in a VDJ segment, which encodes a complete μ heavy chain. In successful recombinations, this heavy chain interacts with protein complexes to form the Pre-B cell receptor. A similar process of rearrangements follows at the light chain V and J locus, the completed light chain can then form an IgM B cell receptor (BCR) and the immature B cell is released into the periphery (Kouskoff, 2001). Immature B cells traffic to the spleen where they are classified as transitional phase B cells, which are sensitive to apoptosis upon BCR ligation. Transitional B cells can develop into two different B cells subsets. Marginal zone (MZ) B cells are found only in the splenic white pulp, where they can respond rapidly to blood borne antigens. Follicular (FO) B cells make up the majority of mature B cells and can survey the blood and lymph nodes in search of antigen (Allman, 2001; Kouskoff and Nemazee, 2001).

1.1.4 Secondary Lymphoid organs

The Secondary immune organs such as the lymph nodes, Mucosal and Bronchial associated lymphoid tissues (MALT and BALT) and Peyer’s patches are specialised structures that have evolved to efficiently conduct the interactions between rare antigen specific cells and local antigen. The organization of secondary lymphoid organs allows T and B lymphocytes to recognise their cognate antigen presented by antigen presenting cells.

19 Figure 1-1 Structure of murine spleen and lymph node

(a) The spleen. Lymphocytes enter the splenic white pulp via the MZ, and migration to the B and T cell zones is mediated by signalling through chemokine receptors. B cells are attracted to the B-cell follicles by CXCL13, whereas T cells are directed to the T-cell zone by responding to CCL19 and CCL21. Lymphocyte exit from the spleen is still unclear. (b) The lymph nodes. Lymphocytes enter through the blood and HEV and then migrate to the B-cell follicles or the T-cell zone much like in the spleen. Lymphocytes exit lymph nodes in efferent lymphatic vessels, and they then re-enter the bloodstream from the lymph. Figure reproduced from Mebius and Kraal 2005.

Both tissues contain T cell zones (also known as the periarteriolar lymphoid sheath, PALS) and B cell follicles where cells are guided via chemokine gradients of CCL19 and CCL21 or CXCL13, respectively. Lymphocytes enter lymph nodes through the afferent lymphatic and local blood vessels called high endothelial venules (HEV). Lymph nodes primarily filter antigen that drains from local lymphatics and blood, whereas the spleen mainly filters the blood for antigens and debris and their structures differ accordingly (Figure 1-1). The spleen is made up of the red pulp, which is mainly where old erythrocytes are broken down, and areas of white pulp that are dominated by lymphocytes. Within the white pulp, T cell zones surround central arterioles, which are in turn enclosed in large B cell follicles and a halo of MZ cells (Mebius, 2005).

20 1.2 Innate and adaptive immune responses

1.2.1 Innate immune responses

The cells of the innate immune system recognise antigens with a restricted panel of germ- line encoded receptors that detect evolutionarily conserved epitopes on pathogens or transformed cells, such as bacterial cell walls or virus derived double stranded RNA. The conserved molecules are called pathogen-associated molecule patterns (PAMPS) and they are recognised by the pattern- recognition receptors (PRR) on immune cells. Surface and internal PRR such as the Toll-like receptor TLR) family are expressed on many innate immune cells such including macrophages, neutrophils and dendritic cells (DC) (Janeway, 2002; Medzhitov, 2002). Interaction with PAMPs and their individual receptors generally leads to activation of specific adaptor protein and induction of signalling pathways that generate appropriate inflammatory transcription factor activation and gene transcription. For example, TLR engagement can lead to nuclear factor (NF)- activation and production of interferons, which inhibit viral reproduction (Kaisho, 2006). Innate cells form an essential first wave of immune defence against pathogens, as well as initiating the downstream adaptive immune response. Phagocytic cells such as macrophages, neutrophils and dendritic cells can ingest pathogens and proteolytically digest them in to protein components, which can be presented to lymphocytes; hence they are known as antigen presenting cells (APC). When APC are stimulated via PRR at the time of antigen uptake, they become activated and begin to express co-stimulation molecules that accompany lymphocyte activation by antigen itself. Innate immune cells such as mast cells and basophils have receptors which bind the Fc portion of soluble IgE. Cross-linking of these receptors leads to degranulation releasing pro- inflammatory factors such as histamine, prostaglandin and leukotrines into the local environment. Natural Killer T (NKT) cells survey tissues for expression of foreign lipids and glycoproteins presented by CD1d molecules. Natural killer (NK) cells can also discriminate between healthy cells and tumours or virally transformed cells, which may have altered surface expression of MHC I molecules as a result of infection or transformation. Both types of NK cells are classified as part of the innate immune system as they respond immediately with bursts of pro- inflammatory cytokines and chemokines such as INF, which attract and activate local leukocytes (Biron, 1999). Some T and B cell subsets respond quickly to antigen and express more restricted germ-line encoded receptors than the highly diverse T and B cell receptor repertoire, and are thus classed as innate immune cells. These include NKT cells, whose specificity is restricted to the 21 antigen in the context of the alternative MHC class-I superfamily molecule CD1d, and also T cells which respond to glycoproteins or nonclassical MHC molecules with their more restricted TCR chains and are generally found in the mucosa (Adams, 2005). Innate type B cells are also found in the immune system; MZ B cells can rearrange their BCR among common specific conditions, but are also uniquely sensitive to conserved mitogens such as LPS. Lastly, B1b B cells are largely observed in the peritoneal and pleural cavities and display germ-line BCRs that exhibit little variation. Both these B cells subsets are important for providing T independent antibody responses by quickly maturing into plasma cells in response to antigen, without the need for help from T cells (Bendelac, 2001; Martin, 2001).

1.2.2 Adaptive immune responses

PPR are limited in the range of epitopes they can recognise compared to the randomly generated specificities formed in lymphocytes by gene rearrangement of antigen receptors. The unique antigen receptors on T and B cells are capable of recognising a limitless array of antigens through recombination. Adaptive immune cells also create a memory response whereby re-infection with the same antigen leads to a more rapid and robust immune response. The shaping of this broad repertoire of naïve T and B cells in the periphery is subject to selective pressure applied by pathogen exposure in an individuals’ specific environment. Clonal selection by antigen drives the expansion of clones that have randomly generated a protective receptor. Cells bearing non- productive receptors die by neglect, thereby delivering the broadest range of useful affinities with the smallest metabolic cost (Billingham, 1953; Burnet, 1965).

1.2.3 Central T cell tolerance

Small numbers of undifferentiated CD4-CD8- (DN) lymphocyte precursors migrate from the bone marrow to the thymus where they begin to differentiate into T cells, rearranging their antigen receptor genes to generate diverse clones (Miller, 1961, 1965). Random recombination of the V, D and J segments of the Tcr loci results in clones which can potentially bind to both self and non-self, and so tolerance mechanisms are required to continually limit the responsiveness of auto reactive mature T cells into the periphery. This is particularly important as T cells can activate and recruit many other downstream effector cells such as B cells and macrophages. These small number of DN stem cell precursors generate large numbers of CD4+CD8+ double positive (DP) thymocytes. Productive rearrangements of the TCR chain that lead to binding to intrathymic peptides in the context of MHC molecules with relatively low affinity, deliver positive survival signals to immature T cells (Borgulya, 1992). This process occurs in the

22 thymic cortex, where a dense web of epithelial cells expressing self peptides is in close contact with DP thymocytes. The vast majority of DP thymocytes die within 3-4 dies as they fail to be rescued by this process of positive selection (Berg, 1989). Immature thymocytes also express both the CD4 and CD8 coreceptors, which bind to non polymorphic sites of MHC. Stronger singals derived from ligation of both the TCR and CD4 co-receptor on double positive cells will lead to commitment to the CD4 lineage of T cells, whereas shorter or weaker signals that utilise CD8 will instead commit to CD8 T cell maturation (Brugnera, 2000; Yasutomo, 2000). The TCR ligation that leads to positive selection of thymoctes is generally very weak and would probably not cause unwanted expansion of these clones to self antigens in the periphery, but a mechanism of negative selection also exists to filter out very strongly binding cells. Very strong TCR ligation with either co-receptor leads to negative selection of T cells by apoptosis, or diversion into other innate T cell lineages (Kappler, 1989; Kappler, 1987; Kisielow, 1988; Swat, 1991). This is different from mature T cells in the periphery, where strong TCR engagement leads to proliferation and differentiation, and it seems to be only the HSA+ single positive (SP) thymocytes that respond to strong TCR signals in this way. Diverse self tissue antigens from around the body are expressed in the thymus in small concentrations under the Aire promoter for the purpose of negative selection, mainly by medullary epithilial cells but also bone marrow derived cells (Anderson, 2002; Bonasio, 2006).

1.2.4 Antigen presentation in the periphery

The initiation of many aspects of an adaptive immune response stems from the recognition of antigen presented to CD4+ T cells in the context of MHC class II complexes expressed on the surface of APCs. Dendritic cells (DC) are specialised immune APCs, but macrophages and B cells can also express MHC class II for this purpose. Tissue resident immature DC constitutively phagocytose and process antigen, but are unable to activate naïve T cells due to a lack of co- stimulation molecules. When they are exposed to an infectious agent, danger signals trigger the maturation process via PRR or cytokines (Banchereau, 1998). DC then migrate to the T cell zones of secondary lymph nodes where they up regulate co-stimulation molecules and begin secreting pro-inflammatory cytokines that can influence the downstream behaviour of activated T cells (Morelli, 2001). Other cells such as macrophages and B cells are also capable of processing and presenting antigen (Itano, 2003) and B cells are also important for the subsequent expansion of T cell clones that have been primed by DC (Ron, 1987).

23 1.2.5 T cell activation

Emigrant naïve T cells from the thymus circulate through lymphoid organs and the blood surveying DC for activating MHC-peptide complexes. Mature T cells in the periphery co-express either CD8 or CD4 alongside the TCR, which dictates whether they can respond to antigen in the context of MHC class I or class II, respectively (Starr, 2003). Professional APC such as DC can present peptide loaded into the groove of the MHC class II molecule, along with key co-stimulatory molecules, which are necessary for complete activation of CD4+ T cells. Antigen presentation to CD4+ T cells is a key step monitoring immunity versus tolerance to self in the periphery, as without the ‘second signal’ delivered by co-stimulation, TCR:MHC class II engagement leads to anergy or apoptosis (Bretscher, 1999). The B7 family is a well characterised group of co-stimulatory molecules that is coexpressed with MHC class II:peptide complexes on activated APC. B7.1 and B7.2 on APC give activating signals via CD28 on the surface of T cells. ICOSL on B cells and ICOS on T cells form another co-stimulatory pair important for humoral responses. There are also inhibitory molecules in this family, which can temper the immune response including PD-L1 and PD-L2 on APC and CTLA-4 and PD-1 on T cells (Coyle, 2001; Greenwald, 2005; Tafuri, 2001). Many cytokines are produced by APC at the time of antigen presentation that can enhance or modulate the T cell response such as interleukin (IL)-1, IL-6, TNF, TGF and IL-12. Adhesion molecules are also necessary to form a synapse between T cells and APC through which intracellular signals are bilaterally initiated and cytokines are exchanged. These are mediated by molecules such as ICAM-1, LFA-1 and LFA-3 for T:DC interactions, while transmembrane members of the SLAM family have shown to be vital for stable T:B cell interactions in vivo (Greenwald, 2005; Qi, 2008). Differential expression of co-stimulatory molecules from various APC subsets leads to modulation of the resulting T cell responses. T cells can also influence antigen presentation by stimulating APC via CD40:CD40L interactions and release of cytokines. (Coyle and Gutierrez-Ramos, 2001). TCR engagement with its cognate peptide:MHC complex leads to signalling cascades which drive altered gene transcription, proliferation, cytokine secretion, anergy or apoptosis during tolerance induction. The TCR and its downstream biochemical signalling pathways must be extraordinarily sensitive to discriminate foreign peptide cognate antigen from the many thousands of self peptides displayed on the plasma membrane of the APC. Recent research has proposed that discrete lipid-microdomains facilitate the organisation and recruitment of discrete signalling components during TCR engagement with MHC:peptide complexes (Werlen, 2002), whereas an alternative model suggests preformed ‘microclusters’ of 24 TCR and CD28 associate to increase sensitivity to antigen (Yokosuka, 2009). TCR engagement activates Src and Syk/ZAP-70 family proteins via tyrosine phosphorylation, mobilisation of Ca2+ ionophores, and activation of mitogen-activated protein kinases (MAPKs) such as extracellular signal related kinase (ERK), jun N-terminal kinase (JNK) and p38 MAPK. The TCR activates the ERK cascade when it binds peptide:MHC complex, which activates the transcription factor NFAT leading to downstream IL-2 production. IL-2 is an important growth factor for T cells, which promotes entry into the cell cycle (Cantrell, 1983). The promoter region for IL-2 contains many binding sites for transcriptional elements that respond to T cell activation and local inflammation such as Oct1, AP-1 and NF- among others (Durand, 1988; Serfling, 1989; Ullman, 1991). Co- stimulation via B7 family ligands by their ligands on APC also leads to IL-2 production via activation of the JNK pathway (Coyle and Gutierrez-Ramos, 2001). Importantly, CD28 ligation also leads to upregulation of pro-survival molecules such as (B cell lymphoma) Bcl-xL thus protecting newly activated T cells from apoptosis, amplifies Ca2+ levels within the cell and leads to cytoskeletal changes which lower the activation threshold of effector T cells (Salazar-Fontana, 2003). These cellular signalling pathways can translate the external receptor engagement into modulation of effector T cell genes.

1.2.6 Peripheral T cell tolerance

In some cases, the low level expression of antigens in the thymus may not be sufficient to drive negative selection, and permit cells to mature which have specificities for highly expressed peripheral antigens. Alternatively, antigens that are only expressed in certain stages of maturation, such as puberty, could also fail to be negatively selected in the thymus, leading to inappropriate peripheral activation. Peripheral toleralance is maintained through several pathways. Firstly, immature T cells are limited in their circulation patterns to the lymphoid system and blood, where thay can remain ignorant of tissues antigens. Antigen presentation to CD4+ T cells is a key step monitoring immunity versus tolerance to self in the periphery, as without the ‘second signal’ delivered by co-stimulation, TCR:MHC class II engagement along with inhibitory co-receptors expressed by immature DCs leads to anergy or apoptosis (Bretscher, 1999). CTLA-4 for example, can increase the TCR stimulation threshold of T cells required for activation. Deletion of CTLA-4 leads to autoimmunity in mouse models, suggesting that is one important mechanism for preventing the activation of self reactive T cell clones in the periphery (Eggena, 2004; Hattori, 2005; Tivol, 1995). The large proliferation of antigen specific T cell clones during an immune response must be controlled in order to limit tissue damage and to maintain homeostatic T cell levels of memory and

25 naïve cells. Apoptosis derived from repeated stimulation of the TCR by antigen is called activation induced cell death (AICD), and is largely mediated by Fas-Fas ligand interactions in vivo (Brunner, 1995). Fas and its ligand are transcriptionally upregulated in the presence of local IL-2 generated by activated T cells and their eventual oligomerisation leads to activation of a caspase dependent catalytic pathway culminating in apoptosis (Refaeli, 1998). The importance of this self limiting aspect of T cell activation is apparent from the autoimmunity that arises in mice and human when the Fas pathway is interrupted (Fisher, 1995; Nagata, 1995). Lastly, dedicated regulatory CD4+ T cells (Tregs) maintain tolerance through a variety of cytokines and cell mediated mechanisms in the periphery (See 1.2.8.2 T regulatory cells)

1.2.7 CD8+ T cells

CD8+ T cells recognise peptide antigen in the context of MHC class I presentation. MHC class I is expressed almost universally by tissues and displays peptides derived from cellular products, generally from the cytosol. When a cell is infected, some pathogen derived peptides are generated by proteosomal degradation of proteins, then transported to the endoplasmic reticulum where they bind MHC class I molecules and are presented on the cell surface. In this way, CD8+ T cells can survey the cells for intracellular infections. CD8+ T cells are able to directly lyse infected cells through perforin release, and they also are capable of Fas mediated killing. Activated CD8+ T cells secrete a range of cytokines such as TNF and IFN, which also influence microbial defence (Wong, 2003).

1.2.8 CD4+ T helper cells

Effector CD4+ T cells and their signature cytokines orchestrate many components of the immune system during health and disease and much research has been carried out on their genesis and function. It is not clear whether APC shape the differentiation of naïve T cells in response to distinct pathogens, or whether a particular T helper outcome is governed by the sensitivity of multi- potent newly activated T cells to the cytokine microenvironment (Mosmann, 1986; Reinhardt, 2006). Following activation, CD4+ T cells acquire various effector functions such as cytokine secretion and adhesion molecule expression allowing migration to infection sites. Chromatin changes may also maintain differentiated T helper states and ensure progeny have a similar phenotype (Wilson, 2009).

26 1.2.8.1 Th1 and Th2 cells The first CD4+ T cell subsets to be described were T helper 1 and 2 (Th1 and Th2). They were segregated based on their mutually exclusive expression of IFN and IL-4, 5 and 13, respectively (Mosmann, 1986). It was later determined that Th1 cells arose following intracellular infections such as Leishmania major, whereas Th2 cells were specialised for protection against extracellular pathogens such as Helminths (Heinzel, 1989). Th1 cells develop when the transcription factor STAT1 is activated by cytokines such as IFN and IL-21 in the cytokine milieu during activation. STAT1 in combination with TCR signals switches on the T-bet and RUNX3 transcription factors which amplify the Th1 phenotype by creating IFN production and inhibiting Th2 transcription factors and gene products such as IL-4 (Djuretic, 2007). APC derived IL-12 also reinforces Th1 commitment through the STAT4 transcription factor, again enhancing IFN and creating a positive feedback loop (Wilson, 2009). Th1 cells activate macrophage and CD8+ T cells that can directly lyse target cells containing intracellular pathogens. Th2 cell differentiation is driven by the presence of IL-4 in the microenvironment at the time of TCR priming, which activates the STAT6 transcription factor. STAT6 and signals derived from transmembrane receptor Notch induce the transcription factors GATA3 and MAF which lead to transcription of the signature Th2 cytokines IL-4, IL-5 and IL-13 (Amsen, 2007). These Th2 transcription factors also concomitantly antagonise Th1 differentiation and IFN production (Ansel, 2006). Th2 cytokines and co-stimulation molecules help humoral immunity and also activated epithelial defence mechanisms that target parasites in the mucosa (Finkelman, 2004; Reinhardt, 2009; Takeda).

1.2.8.2 T regulatory cells Thymic selection leads to apoptosis of cells with TCRs that respond inappropriately to self-antigen expressed in the thymus. But occasionally T cells that retain specificity for a self-antigen will be released into the periphery. Immune responses to external factors that are too strong or too long lasting can also cause major damage to self-tissues and so mechanisms that enforce tolerance in the periphery are required. Tregs can inhibit immune response with a combination of anti-inflammatory cytokines, cell-to-cell contact and modulation of APC function (Fontenot, 2003; Sakaguchi, 2000; Sakaguchi, 1995). Tregs are generated both in the thymus (natural T regs) and in the periphery (inducible Treg) from CD4+T cells with self reactive TCR in response to tolerising signals (Figure 1-2) (Josefowicz, 2009; Sakaguchi, 1995). Tregs express the high affinity IL-2 receptor, CD25, on their surface and also characteristically express the forkhead box (Fox)p3 transcription factor, which 27 inhibits pro-inflammatory cytokine expression by this subset (Hori, 2003; Williams, 2007). Humans or mice with defects in the Foxp3 locus suffer from a fatal lymphoproliferative disease, which affects many tissues (Bennett, 2001; Brunkow, 2001, Khattri, 2003 #569). Natural T regs are generated in the thymus, where strong TCR affinity for self peptide:MHC that falls just below the threshold for negative selection is thought to correlate with Foxp3 expression (Hsieh, 2006). Peripheral naïve CD4+ T cells can also acquire Foxp3 expression when exposed to TGF- and high IL-2 levels (Chen, 2003; Elias, 2008). Recent research has shown the gut to be a particularly effective locale for generation of inducible Treg through combinations of tolerising APC, dietary derived retinoic acid and cytokines such as TGF- (Annacker, 2005; Benson, 2007; Coombes, 2007).

Figure 1-2 Treg differentiation in the thymus and periphery Natural Treg: (A) Most thymic SP CD4+ and CD8+ exit into the periphery, but some SP CD4+ T cells with a strong affinity for self-peptide:MHC on thymic epithelium or DC differentiate into Treg with the addition of CD28 and IL-2 signalling. Once in the periphery, these natural Foxp3+ Treg are dependent on exocrine IL-2 for survival. Inducible Treg: (B) Suboptimal TCR signalling combined with IL-2 and TGF- can lead to peripheral Foxp3 induction. The gut associated lymphoid tissues facilitate Treg differentiation via tolerising CD103+ DC, metabolites such as retanoic acid and TGF-. (Figure adapted from Josefowicz and Rudensky, 2009)

Tr1 cells can also suppress several types of autoimmune disease, although they do not express Foxp3 (Cottrez, 2004). They are thought to arise from antigenic stimulation in concert with high levels of the suppressive cytokine IL-10, which they also produce in order to dull local

28 immune responses. These cells are also said to utilise IL-21 and IL-27 in order to amplify IL-10 production (Pot, 2009; Spolski, 2009).

1.2.8.3 Th17 cells Th17 cells have been linked with clearance of bacterial and fungal infections through mobilisation of neutrophils and release of inflammatory chemokines (Ivanov, 2009; Ma, 2003; Pepper). This effector subset shares interesting similarities with Treg induction, as they are also generated in response to TGF- exposure, when combined with IL-6 rather than IL-2 (Manel, 2008). These cytokines upregulate the retinoic-acid-receptor-related orphan receptor-t (RORt), which can antagonise Treg development (Ivanov, 2007). Th17 cells produce IL-21, which can add to STAT3 activation driven by APC derived IL-23 and IL-6 to stabilise Th17 commitment (Korn, 2007a; Zhou, 2007). Th17 cells are often associated with mucosal surfaces and can sometime produce IL-22, another STAT3 signalling cytokine which targets epithelial cells with diverse effects such as cell survival and proliferation, promoting maintenance of epithelial barriers as well as induction of inflammatory proteins (Aujla, 2009).

1.2.8.4 T follicular Helper cells T follicular helper (Tfh) cells are a specialised CD4+ T helper subset that have the capacity to migrate from T cell zones into B cell follicles to provide help for high affinity antibody production within specialised structures called germinal centres (GC). They are characterised by high expression of surface markers CXCR5, ICOS and PD-1, the transcription factor BCL-6 as well as cytokines such as IL-21, IL-10, IL-4 and IFN. Tfh are integral in the studies described in this thesis and are reviewed at length in section 1.3 of the introduction.

1.2.9 T cell memory and homeostasis

The adaptive immune system has a unique capacity to form memory cells in response to a primary infection that can respond more robustly and quickly in the occurrence of secondary infection with the same pathogen. Generally, rapid clonal expansion of the appropriate T cell clones in response to local inflammatory cytokine production leads to the elimination of a pathogen, leaving an expanded T cell pool that is now redundant. 90-95% of these T cells die after the removal of pro-inflammatory cytokines and other signals, leaving only a small population that survive as memory cells due to competition for homeostatic cytokine signals (Surh, 2002). Memory T cells have a lower activation threshold to an identical antigen than naïve cells and are less dependent on co-stimulation in order to be reactivated (Cho, 1999; Rees, 1999). They have diverse

29 homing abilities and can remain in lymph nodes to aid antibody production or alternatively migrate to inflamed tissue sites. Memory T cells can be loosely partitioned into central memory cells which patrol lymphoid organs, and effector memory cells which survey tissues such as the gut for reinfection (Sallusto, 2009). The size of the T cell pool within an individual is very stable due to homeostatic balance between naïve, effector and memory T cells. The pool is maintained by a combination of pro and anti-apoptotic molecules, proliferation and differentiation. The common (c)-chain family of cytokines play a pivotal role in maintaining various T cell populations (Boise, 1995; Surh, 2005; Teague, 1997; Vella, 1997). The naïve T cell population is maintained by a combination of signals derived from the IL-7 receptor and continuing contact between the TCR and self-MHC, which sustain survival in interphase (Kishimoto, 1999). This is achieved by Bcl-2 and Mcl-1 expression preventing apoptosis via the mitochondrial pathway. Memory cells survive due to a mixture of signals from IL-7 and IL-15 (Surh, 2008; Tan, 2002). These support not just survival of memory T cells but also occasional cell division that promotes the longevity of a certain useful TCR specificity within the T cell repertoire. IL-2 drives proliferation of naïve T cells on exposure to antigen, and is also vital for Treg survival or metabolic fitness in the periphery (Fontenot, 2005; Pipkin). Differential receptor expression and transfer experiments suggest that IL-2 favours effector memory survival, while IL-15 sustains long-lived central memory (Cho, 2007).

1.2.10 T helper subsets in allergy and autoimmunity

The polarisation of T cell responses into the various T helper subsets has evolved in order to combat common mammalian pathogens, but is also relevant for the development of autoimmune and allergic diseases (Constant, 1997; Tao, 1997). Th1 cell activity has been identified as a driving force in inflammatory or autoimmune disorders including colitis (Hans, 2000a) and type 1 diabetes (TID) (Liblau, 1995). In contrast, Th2 T cells participate in allergic and atopic reactions such as asthma (Cohn, 2004). Th17 cells have been identified in autoimmune conditions such as experimental allergic encephalomyelitis (EAE) (Langrish, 2005) and Tfh cells in humoral immunity based diseases such as lupus (Bubier, 2009; Linterman, 2009; Vinuesa, 2009; Vinuesa, 2005b).

1.2.11 B cell activation

B cells are vital for protection against pathogens due to their ability to secrete antibodies that can protect against invading pathogens in several ways: Antibody can bind to extracellular components required for adhesion or entry to host cells, thereby neutralising them. Antibodies can also protect against pathogens in intercellular spaces by binding and enhancing their phagocytic

30 uptake through Fc receptor (FcR) signalling on granulocytes and APC (Radaev, 2002). Lastly, certain antibody isotypes can opsonise pathogens by binding components of the complement system, causing cytolysis and release of pro-inflammatory proteins (Tedesco, 2008). Selection and expansion by antigen of diverse B cells clones derived from germ-line BCR rearrangement generates antibody secreting cells, plasma cells, and long-lived memory B cells with the ability to respond quickly to reinfection (Ribatti, 2009). Most, but not all protein antigens require concomitant help provided by CD4+ T cells at the time of BCR cross-linking for this activation to occur.

1.2.11.1 T dependent antibody production The interactions between B and T cells during T dependent immune responses are critical for the generation of high affinity memory cells and long lived plasma cells that can effectively neutralise invading pathogens. B cells internalise and proteolytically digest antigen into small peptide components, which are presented on their surface to CD4+ T helper cells in the context of MHC class II (Ron and Sprent, 1987). This is thought to occur at the border of the T and B cell zones. B cells that receive co-stimulation via CD40L on T cells at this site migrate deep into the B cell area where they form a primary follicle close to follicular dendritic cells (FDC), which serve as local depots of antigen due to retention of immune complexes on their surface. B cells dividing in newly formed GC move between sub structures called the light zone, which is filled with ‘centrocytes’ with lower proliferation rates, and the dark zone filled with dividing ‘centroblasts undergoing mutation of the BCR (Kelsoe, 1995). These two regions are compartmentalised based on sensitivity to the chemokines CXCR4 in the dark zone and CXCL13 in the light zone (Allen, 2007a). Tfh cells can also be found in both regions but are more common within the light zone, as are the FDC (Allen, 2007b). Tfh cells provide vital survival signals such as ICOS, CD40 and cytokines like IL-21 and IL-4, which is only available to the B cell clones that can compete for T cell help (Allen, 2007b). Additional diversity in addition to BCR rearrangement during development is generated in the GC, where B cells undergo a program of somatic hypermutation (SHM). SHM involves the introduction of a very high rate of point mutations into the V regions of the rearranged heavy- and light-chain genes that gives rise to mutated immunolglobulin BCRs, allowing for the generation of additional B cell diversity (Tarlinton, 2000). The multitude of BCR variants that arise from this process are sensitive to cell death, and must compete for survival signals from antigen on FDC and cognate signals from T cells within the GC (Allen, 2007a; Dal Porto, 1998; Garside, 1998).

31 Initial interactions between T and B cells and various co-stimulation molecules leads to immunoglobulin isotope switching as well as proliferation and differentiation (Stavnezer, 1996). T cell derived cytokines seem to be very important for this process as IL-4+ T-B conjugates purified from immunised mice were shown to contain increased class switching to IgG1, whereas IFN+ conjugates tended to express IgG2a (Reinhardt, 2009). In vitro studies have shown a hierarchy of certain cytokines that drive various isotype switching but the relevance is not known (Deenick, 2005). Some B cells with migrate into follicles and to participate in GC, but an alternative B cell fate is possible as a result of cognate B-T cell interactions at the T-B border. Some B cell clones migrate instead to extrafollicular sites and undergo a short burst of proliferation and differentiation into short-lived plasma cells. Extrafollicular derived antibody is generally low affinity as is occurs parallel to SHM and B cell selection in the germinal centre (Berek, 1991). The decision between GC and extrafollicular responses may be driven by affinity for antigen, with very high affinity cells moving quickly to extrafollicular development to make a first wave of low affinity antibody while slightly lower affinity cells are refined in the GC (Chan, 2009; Paus, 2006).

1.2.11.2 T independent B cell responses Some B cells can be activated in a T dependent manner by certain types of antigen such as polysaccharides from bacterial capsules of Toll-like receptor (TLR) ligands. These responses occur in extrafollicular areas, and like the similar reaction since early in T dependent responses, generally produce short lived plasma cells with a low affinity for antigen. MZ and B1b cells commonly participate in T independent responses rather than follicular B cells (Martin, 2000). T independent responses are not thought to result in a memory recall response and are thus considered part of the innate immune system, although long lasting immunity has been observed in some models (Alugupalli, 2008; Alugupalli, 2004).

1.3 T follicular B helper cells

The provision of help by T cells within the GC is essential for production of high affinity, long lasting humoural immunity. The specific CD4+ T cell subset responsible for B cell help, termed T follicular helper (Tfh) cells, were initially identified due to their unique ability to migrate into B cell follicles due to surface expression of the chemokines receptor CXCR5, and they were also shown to induce antibody production when cultured with B cells in vitro (Breitfeld, 2000; Schaerli, 2000). Tfh cells also express the B7 family molecule ICOS that stimulates antibody production by B cells, as well as programmed cell death (PD)-1, which is upregulated following

32 TCR engagement. Tfh cells also secrete cytokines such as IL-21, IL-10 and IL-4 which can influence B cell antibody production and class switching, as well as the follicular B cell chemoattractant CXCL13 (Haynes, 2007; Kim, 2004; Reinhardt, 2009). The provenance of the recently described Tfh cell subset and the exact sequence of events during their differentiation are subject to ongoing investigation. Tfh cells are derived from naïve T cells that are primed to antigen by the DC that populate the T cell zone like other T helper subsets. Upon TCR engagement, the progeny of the first IL-2 induced clonal expansion express the early activation marker CD69 and downregulate sphingosine 1 phosphate receptor 1 (S1P1R1), which sequesters these cells within the lymph node (Shiow, 2006). After 2-3 days some cells exit the lymph node as effector cells that can traffic to infection sites, and some retain CD69 expression and remain in the lymph node. These include potential Tfh cells, which express the chemokine receptor CXCR5, allowing movement towards the B cell border with the T cell zone (Haynes, 2007). T cells that form strong interactions with B cells may migrate into the primary follicle or GC where constant interactions with antigen specific B cells may cement the Tfh differentiation process (Ebert, 2004). Those CD4+ T cells that do not form strong conjugates with B cells may enter different T helper lineages or participate in extrafollicular B cells responses (Odegard, 2008; Reinhardt, 2009).

1.3.1 TCR affinity and Tfh cell development

As described earlier, CD4+ T cells with varying affinity for antigen will receive different biochemical signals upon ligation that modulate effector differentiation. Studies by Fazileau et al showed an interesting link between TCR affinity and Tfh cell differentiation whereby adoptively transferred Tg TCR with the highest affinity for the protein immunogen were more commonly recruited to a Tfh cell phenotype, defined as being CD62Llo and CXCR5+. This work suggests that to become a Tfh cell, naïve T cells require a strong signal through the TCR to commit to the lineage (Fazilleau, 2007a; Fazilleau, 2007b).

1.3.2 Tfh cytokines

Cytokines produced by T helper cells primed by antigen on DC in the T cell zone can induce antibody class switching by B cells in the follicle. It is not completely understood how the different cytokine profiles released by various types of T helper cells initiate B cell differentiation pathways that lead to a range of antibody isotypes appropriate for fighting a specific infection. Cross talk between B and T cells at the border region (Cunningham, 2002; Toellner, 1998), as well

33 as stable conjugates formed in B cell follicles with Tfh cells both contribute to acquisition of cytokine competency (Qi, 2008; Reinhardt, 2009). Although IL-4 and IFN and typically produced by Th2 and Th1 cells respectively in the periphery, they also influence antibody production in the secondary lymphoid organs. IL-4 is known to induce class switch to IgG1, or later to IgE, while IFN causes class switch to the IgG2a isotype (Abbas, 1990). Careful analysis of individual T-B cell conjugates purified from immunised or Leishmania major infected mice showed that this direct T cell help was directly associated with production of the different isotypes and also increased levels of activation induced deaminase (AID), which drives SHM (Reinhardt, 2009). IL-10 has shown to induce a range of IgG and IgA antibodies in cultured human B cells, but its relevance in B cell help in vivo is unknown (Briere, 1994; Fujieda, 1996; Malisan, 1996). Cytokines also contribute to the generation of the Tfh cell subset, which is particularly important for selection of high affinity clones in the GC. The STAT3 signalling cytokine IL-6 has been shown to enhance a B helper phenotype in CD4+ T cells (Eddahri, 2009).

1.3.3 IL-21 and Tfh cells

IL-21 is another STAT3 signalling cytokine with a putative role in the GC, where it is highly expressed by CD4+ T cells (Chtanova, 2004). When human tonsillar CXCR5+ cells are cultured with naïve human B cells, their capacity to induce antibody production is largely dependent on IL-21 (Bryant, 2007). IgG1 is the dominant isotype produced after immunisation with the protein antigen ovalbumin in alum, and is thought to be derived from GC. Interestingly IgG1 is much reduced in Il21r-/- mice suggesting a role for Il-21 in this reaction (Ozaki, 2002). IL-21 can influence B cell proliferation and differentiation, but it is unclear how Tfh cell production of IL-21 in the GC affects both T and B cells. The autocrine role for IL-21 acting on T cells to drive or sustain a Tfh cell phenotype as well as the relevance of IL-21 for GC B cells has been a particular focus of the research described in this thesis (Nurieva, 2007; Nurieva, 2008; Vogelzang, 2008), and has gained further credence when recent studies showed IL-21 could regulate expression of the Tfh cell associated transcription factor Bcl-6 (Nurieva, 2009)

1.3.4 Tfh cell positioning in the GC

The chemokine receptor profile of Tfh cells has been well characterised. Many studies have highlighted the expression of CXCR5 on T cells found in the B cell follicle, which allows it to migrate towards the CXCL13 rich GC (Haynes, 2007; Kim, 2004; Schaerli, 2000). CXCR5 is transiently upregulated by CD4+ T cells during cognate interactions with APC, in a manner

34 dependent on cell:cell contact at the costimulatory molecules CD28, OX40 and ICOS on T cells (Akiba, 2005; Ansel, 1999; Obermeier, 2003). As CXCR5 is an early response marker and is present on many T cells in an activated lymph node, for example in human tonsil more than half of CD4+ T cells express CXCR5 while only a small proportion are actually found in the follicle (Breitfeld, 2000) implying other factors influence Tfh cell migration. CXCR5 is also expressed on a small subset of human CD4+ in the blood, whose relationship to Tfh cells is not known (Breitfeld, 2000; Forster, 1994). CCR7 tethers naïve T cells in the T cell zone where its ligands CCL19 and CXCL21 are expressed, and must be downregulated for T cells to enter the B cell zone. Indeed even CD4+ T cells engineered to over express CXCR5 were unable to move into the follicle without removing the influence of CCR7 (Haynes, 2007). Studies employing intravital two-photon microscopy to trace cellular movements during GC reactions shows that there is intense competition among B cells for interactions with a small population of Tfh cells (Allen, 2007b). Recent research into the signalling lymphocytic activation molecule (SLAM) family proteins has shown a critical role for stable T-B cell interactions in Tfh cell generation. SLAM associated protein (SAP), an adapter molecule that binds to the signalling SLAM molecule, is expressed of CD4+ T cells and has been shown to enhance Th2 differentiation (Wu, 2001) and TCR signals (Cannons, 2004). SAP deficient T cells are unable to support GC formation, though other T helper capabilities are retained. This was found to be due to a critical role for SAP in forming stable conjugates with B cells, which enhance T cell recruitment to the nascent GC (Qi, 2008). These stable T-B conjugates have also been shown to present effective conduits for Tfh cell cytokines to modulate B cell responses such as SHM and class switch (Reinhardt, 2009).

1.3.5 Tfh cell co-stimulatory molecules

Bearing in mind the putative role for a strong TCR signal and stable interactions with B cells in the differentiation of the Tfh cell subset, it is no surprise that co-stimulation molecules that can modulate these signals have also shown to be important. Tfh cells have shown to express very high levels of co-stimulatory molecules such as ICOS and OX40L compared to other helper subsets, particularly those that activate antibody production by B cells (Breitfeld, 2000; Oxenius, 1996; Rasheed, 2006; Schaerli, 2000). ICOS is stimulated by its ligand on B cells, which induces cytokines such as IL-2, IL-4 and IL-10, some of which are known to help B cells (Hutloff, 1999; Lohning, 2003; Tafuri, 2001). ICOS is particularly important for GC reactions, which are absent in Icos-/- mice along with CXCR5+ T cells in the follicles (Bossaller, 2006). Conversely, the inhibitory co-stimulation molecule PD-1 is also found at high levels on Tfh cells, as it is on many activated T cells (Okazaki, 2006). This may arise as a result of the multiple

35 TCR ligations formed in the follicle, as PD-1 high cells are found after T-B interactions are initiated (Haynes, 2007). Certainly, PD-1 expression on CD8+ cells is associated with chronic antigen stimulation (Barber, 2006) and, in this regard, PD-1 may be important for ensuring the selective survival of Tfh cells with the highest affinity and most productive interactions with B cells as well as defining the lifespan of GC responses through limiting Tfh cell numbers.

1.3.6 Transcriptional regulation of Tfh cells

Lineage commitment to T helper subsets is acquired through the actions of specific transcription factors, which drive the upregulation of signature genes, while often suppressing alternative helper outcomes. High mRNA expression of the transcription factor Bcl-6 was first identified by microarray analysis of human tonsillar CXCR5+ Tfh cells compared to other CD4+ T helper and memory subsets (Chtanova, 2004). Bcl-6 activity has been shown to be important for germinal centre B cell commitment and suppression of the alternative plasma cell differentiation pathway dependent on the transcription factor Blimp-1 (Fukuda, 1995). However, the GC defect observed in Bcl-6 deficient mice also stems from actions on T cells. Enforced retroviral expression of Bcl-6 was shown to suppress Th1 and Th17 cell differentiation while enhancing the Tfh cell phenotype in a series of recent publications (Johnston, 2009; Nurieva, 2009; Yu, 2009).

1.3.7 Tfh memory

It is currently unclear whether long-lived memory cells arise from the Tfh subset. After their residence in the light zone of the GC, it is possible that they undergo apoptosis due to their high surface expression of PD-1 and Fas, which could make them metabolically sensitive to this outcome (Marinova, 2006), leaving memory B cells generated in the GC to perform the role of memory recall responses (Vieira, 1990). However, some studies have found that a population of sessile CD69+CXCR5+ CD4+ T cells remained in the B cell follicle associated with FDC and possible depots of antigen many months post immunization (Fazilleau, 2007a). Further research in this area is required to dissect whether Tfh cells perish as the GC responses are resolved, or whether they are locally retained in order to quickly produce cytokines again in the case of recurrent infection.

36 1.4 Interleukin-21

1.4.1 Cytokines

Cytokines are regulatory molecules that act on cells of the immune system, they constitute a network that coordinate the various stages of the immune response against pathogens, and the development of its hemopoietic constituents. They govern the initial activation of innate immune cells mediated by dendritic cells, monocytes, granulocytes and natural killer cells expressing pattern recognition receptors for conserved foreign molecules. Cytokines also drive activation of adaptive lymphoid immune cells with specific affinity for the pathogen at hand, the ability to form a circulating memory population, and the resolution of these destructive cellular and humoral forces in a timely manner that limits damage to self tissues. IL-21 is the most recently discovered member of the Type 1 cytokine family and signals via the c-chain component, its pleiotropic effects are being discovered at all levels of this diverse system (Leonard, 2005).

1.4.2 Interleukin-21 The receptor for IL-21 comprises the IL-21 receptor (IL-21R) chain and the common c- chain (Asao, 2001). The c-chain cytokine family shares this receptor component and includes IL-2, IL-4, IL-7, IL-9, IL-15, and IL-21 (Leonard and Spolski, 2005; Ozaki, 2000). The groups’ integral role in the immune response is evident in the severe immune defects observed when their signalling is disrupted. X-linked severe combined immunodeficiency arises from a mutation in the common c- chain and results in a failure to generate T cells, NK cells, or a functional B cell population (Leonard, 2000). The IL-21 receptor chain was discovered in 2000 as a putative Type 1 family receptor bearing close resemblance to the IL-2 receptor beta (Ozaki, 2000). High levels of transcript were found in lymphoid tissues from humans and mice (Chtanova, 2004; Parrish-Novak, 2000; Ueda, 2005). IL-21 was subsequently identified as the ligand, and found to display a four -helical bundle structure and a high structural homology with IL-2 and IL-15 (Parrish-Novak, 2000). These structural similarities combined with their proximity on chromosome 3 suggest these cytokines may have arisen from gene duplication. Further studies revealed IL-21 induced growth and proliferation of in vitro NK cells, CD4+ T and B lymphocytes in the presence of other c-chain cytokines (Parrish-Novak, 2000).

37 1.4.3 IL-21 signalling pathways IL-21 signalling is mediated via the common c-chain and associated Janus Kinases (JAKs) in concert with the IL-21R. The IL-21R cytoplasmic domain contains six tyrosine residues. One of these, Tyr 510, is phosphorylated and serves as a critical docking site for both transcription factors STAT1 and STAT3 (Zeng, 2007). Phosphorylation of JAK1 and JAK3 upon IL-21 stimulation activates sustained STAT3 and STAT1 signalling (Habib, 2002; Parrish-Novak, 2002). STAT3 is critically important to the actions of IL-21 since signalling responses to IL-21 are defective in T cells that lack expression of STAT3 (Zeng, 2007). IL-21R expression has been observed on B cells, T cells, NK cells and dendritic cells, where both stimulatory and inhibitory effects have been described among these subsets (Chtanova, 2004; Parrish-Novak, 2000; Strengell, 2006). The pluripotency of IL-21 within both arms of both the innate and adaptive immune response indicate an integral role in both the mobilisation of innate immune cells and the transition to antigen specific defence. IL-21 is produced by CD4+ T cells and by NKT cells upon mobilisation of intracellular calcium stores following T cell receptor (TCR) cross-linking - controlled by regulatory motifs such as Sp1 and the T cell transcription factor NFATc2 (Kim, 2005; Spolski, 2008; Wu, 2005). Two recent publications have showed that early IL-21 production by CD4+ T cells is dependent on STAT4 signals from APC derived IL-12, though these findings were not relevant in mice and have yet to be confirmed in vivo (Ma, 2009a; Schmitt, 2009). IL-21 is produced in response to immunization and during viral infection (Holm, 2006).

1.4.4 IL-21 and T cell priming

The functional capabilities of CD4+ T cells during the immune response are thought to vary in a manner dictated by co-stimulation events delivered at the time the T cells encounters its cognate antigen in the context of MHC II on antigen presenting cells (APC). This can require co- stimulatory molecules such as the B7 family molecules on APC, and cytokines in the local environment. CD4+ T cells are the main producers of IL-21, and the receptor for this cytokine is upregulated on these cells after activation. IL-21R is upregulated on T cells in the thymus at the double positive stage, although it may not be involved in thymic development or selection because mice genetically deficient in the IL-21 receptor (IL-21R-/- mice) have normal thymic development (Kasaian, 2002; Ozaki, 2004). Therefore, it is most likely that IL-21 begins to play an important role for T cells at a later stage, i.e. during proliferation, differentiation, and acquisition of effector functions in response to antigen. CD4+ T cells express several surface molecules that facilitate interactions with CD8+ T cells and B cells to guide the acquisition of cytotoxic function and antibody production, respectively. 38 The role of IL-21 in T helper cell differentiation remains controversial. This is largely due to the fact that IL-21 does not readily fit the established Th1/Th2 cytokine models. Some studies have demonstrated a functional contribution of IL-21 to the production of Th1 cytokines (Fina, 2007; Strengell, 2004; Strengell, 2002) and to the generation of Th2 responses (Frohlich, 2007; Mehta, 2005; Wurster, 2002). The T helper cell subsets that have been reported to produce IL-21 include Tfh, Th17 and Th2 cells (Chtanova, 2004; Korn, 2007a; Nurieva, 2007; Pankewycz, 1991; Pesce, 2006).

1.4.5 Th1 and Th2 cells

Mice deficient in IL-21R reveal that IL-21 responsiveness is critical for the differentiation of Th2 cells (Frohlich, 2007; Pesce, 2006). Infection of IL-21r-/- mice with Toxoplasma Gondii induces a strong induction of a Th1 response to intracellular pathogens characterised by interferon (IFN) production similar to wildtype mice (Ozaki, 2002). However, Th2 responses in IL-21R-/- mice appear to be compromised. IL-21R-/- CD4+ T cells migrate poorly to the site of infection and produce fewer Th2 cytokines in response to infection with the extracellular parasite Nippostrongylus Brasiliensis. These findings imply that IL-21 is necessary for potent Th2 responses but does not play a role in the polarisation towards Th1 responses (Frohlich, 2007). A predominant role for IL-21 in Th2 responses is also supported by studies demonstrating an inhibitory role of IL- 21 on the generation of Th1 cells in vitro. Exposure of naive Th cell precursors to IL-21 inhibited IFN production from developing Th1 cells through repression of eomesodermin expression (Suto, 2002).

1.4.6 Th17 cells

Th cells producing IL-17A (Th17) are a pro-inflammatory T helper subset, which attract neutrophils and other inflammatory cells to the site of immune response (Ivanov, 2009; Khader, 2009; Pepper). Th17 cells produce large amounts of IL-21, relative to Th1 and Th2 cells (Korn, 2007a). IL-21 has been shown to enhance production of IL-17 in vitro and to be able to substitute for IL-6 in Th17 differentiation, but it remains unclear whether these results reflect what happens in vivo (Korn, 2007a; Wei, 2007). EAE is a murine autoimmune model characterised by influx of Th17 into the central nervous system. While IL-21 appears necessary for a Th17 response in EAE in mice lacking the key differentiation factor IL-6, it seems unlikely that IL-21 is critical for Th17 generation either during infection or in an autoimmune setting where IL-6 is abundant. Kopf et al have shown that IL-21

39 signalling is redundant for several types of in vivo Th17 responses Il21r-/- mice are normally susceptible to EAE and to a similar experimental autoimmune response directed against the myosin protein in the heart tissue (Sonderegger, 2008). In summary, since the results in tissue culture experiments imply that IL-21 action may be more important during T cell activation only in the absence of other factors such as IL-6 or costimulatory molecules such as ICOSL (Korn, 2007b), it is vital to dissect the actions of IL-21 in in vivo settings.

1.4.7 T regulatory cells

As described earlier, T regulatory cells (Tregs) are characterized by high levels of expression of the chain of the IL-2 receptor (CD25) and the transcription factor Foxp3. Tregs are known to suppress the proliferation and function of other T cells, however, a definitive mechanism explaining the suppressive abilities of Tregs in vivo remains largely elusive. Nevertheless, studies demonstrating that a reduction in Treg number and function precipitate autoimmunity indicate a critical function of these cells in peripheral tolerance to self-antigens. IL-2 signals through STAT5a/b to up-regulate Foxp3 and, as such, is the primary growth factor for Tregs in the periphery (Yao, 2007). In contrast, IL-21 predominantly induces STAT3 phosphorylation and evidence suggests that IL-21 may have a negative impact on the function of Tregs (Bucher, 2009; Peluso, 2007). Recent studies have shown that IL-21 inhibits TGF-–driven differentiation of naïve Th cells into Foxp3+ Tregs (Fantini, 2007). However, it remains unclear whether IL-21 exerts its effect directly on Tregs or whether IL-21 renders T cells resistant to Treg-mediated suppression (Li, 2008; Peluso, 2007).

1.4.8 CD8+ T cells

Compared to the effects of IL-21 on T cells of the CD4+ T cell lineage, the actions of IL-21 on CD8+ T cells appear relatively straightforward. IL-21 promotes the proliferative response of several lymphocyte subsets. However, transgenic over-expression of murine IL-21 shows that IL-21 predominantly expands the memory CD8+ T cell subset (Allard, 2007). The homeostatic accumulation of memory phenotype CD8+ T cells in resting mice with excessive production of IL- 21 is likely to be explained by an effect on both their proliferation and their survival. Studies showing an increase in CD8+ T cell numbers following the administration of IL-21 and activation of the phosphatidylinositol-3 kinase (PI3K)-signalling cascade support a role for IL-21 on CD8+ T cell survival (Moroz, 2004; Ostiguy, 2007). In contrast, there is recent evidence for a pro-apoptotic effect of IL-21 on primate CD8+ T cells through the down-regulation of Bcl-2 (Barker, 2007). Whether these conflicting data reflect species-specific differences in responses to IL-21 or

40 differences in the actions of IL-21 on distinct CD8+ T cell subsets remains unclear. Thymic CD8+ T cell development and maintenance of T cell numbers in the periphery appears normal in mice made genetically deficient in the IL-21R, indicating that other factors can compensate for IL-21 effects on CD8+ T cell growth (Kasaian, 2002; Ozaki, 2002). The potent effect of IL-21 on CD8+ T cell growth was initially observed as accelerated outgrowth of CD8+CD4- cells in cultures of mouse thymocytes (Parrish-Novak, 2000). IL-21 co- stimulates T cell proliferation and numerous studies demonstrate that IL-21 enhances the proliferation, IFN production, and cytotoxic function of CD8+ effector T cells (Kasaian, 2002; Liu, 2007). In concert with the c-chain cytokines IL-15 and IL-7, IL-21 can promote proliferation and the acquisition of cytotoxic effector functions of CD8+ T cells both in vitro and in vivo (Liu, 2007; Zeng, 2005). The ability of IL-21 to synergise cytokine-mediated proliferation occurs in the absence of TCR signalling suggests that IL-21 may play a cooperative role in antigen-independent expansion. In contrast to other c-chain signalling cytokines such as IL-2 and IL-15, IL-21 does not drive CD8+ T cell proliferation in the absence of stimulation through the TCR or by other c-chain cytokines. Adding further complexity, the proliferative and functional effects of IL-21 appear to differ for naive and memory CD8+ T cells. As noted above, IL-21 can augment the IL-15 or IL-7 induced proliferation of CD8+ memory T cells but has less effect on antigen-dependent proliferation of memory cells (Liu, 2007). However, IL-21 effectively co-stimulates antigen-driven proliferation of naïve CD8+ T cells (Liu, 2007). Interestingly, IL-21 added to CD8+ effector T cells from HIV-infected patients upregulates perforin production in the absence of cell activation or proliferation, whereas IL-15-mediated upregulation of perforin occurs only in the presence of proliferation (White, 2007). Although beyond the scope of this review, it is important to note that the effects of IL-21 on cytotoxicity have functional consequences in vivo demonstrated by a potent anti-tumour ability in several models and against many tumour targets (Cretney, 2007; di Carlo, 2007; Ma, 2003; Spolski and Leonard, 2008; Yoon, 2008). Viral infections require the actions of cytotoxic CD8+ T cells to destroy infected cells and limit infection. Chronic viral infections can persist when constant antigen exposure can inhibit T cell activity and cause them to lose function, a condition called “exhaustion”. A series of reports in Science outlined a role for CD4+ T cell derived IL-21 in supporting these long term responses by preventing CD8+ T cell exhaustion thereby decreasing viral titre. Il-21 was shown to be redundant for acute responses to the same pathogen (Elsaesser, 2009; Frohlich, 2007; Yi, 2009). Taken together, these studies define IL-21 as a co-stimulator of T cell proliferation. The co-stimulation of CD8 T cells in vitro by IL-21 invokes a distinct modulation of cell surface molecules by sustaining the expression of co-stimulation molecules such as CD28 (Zeng, 2005),

41 which could increase CD8+ T cell responsiveness to the antigen presented by APC (Li, 2005). However, whether IL-21 operates via other co-stimulatory molecules or delivers its own co- stimulatory signal to T cells remains a relevant unanswered question.

1.4.9 Opposing effects of IL-21 on B cells

IL-21 applied to B cells purified from humans or mice yields unremarkable effects but can induce interestingly paradoxical results when combined with other stimuli. IL-21 can either deliver co-stimulation to B cells or induce B cell apoptosis depending on the activating signals that accompany it. These implications are important for autoimmunity: costimulation amplifies signals through the BCR to facilitate autoantibody production, whereas apoptosis is necessary for the removal of both the self-reactive clones that could target tissues for destruction and antibody- producing cells generated during infection (which could cause chronic formation of inflammatory immune complexes). IL-21R is expressed on both immature and mature B cells, and is upregulated further upon antigen binding or stimulation by TLR ligands (Strengell, 2002). IL-21 can also increase human B cell proliferation induced by ligation of CD40 in vitro (Ettinger, 2005; Good, 2006; Jin, 2004; Parrish-Novak, 2000). However, when paired with BCR stimulation, IL-21 reduces the response and inhibits proliferation of murine B cells induced by TLR ligands - namely LPS and CpG - by inducing apoptosis (Jin, 2004; Jin, 2006; Mehta, 2003). These different outcomes on B cells led to the hypothesis that while IL-21 could enhance antigen specific responses, it could also induce apoptosis for B cells that lack help provided by molecules such as CD40L, expressed on activated antigen specific CD4+ T cells, thus protecting against inappropriate B cell activation (Good, 2006). However, studies on purified murine splenic B cells indicated induction of apoptosis by IL-21 combined with both mitogens, such as LPS, and T dependent stimulation (Mehta, 2003). Adding IL-21 to cultures down-regulated the anti-apoptotic molecules Bcl-2 and Bcl-xL, suggesting the possibility of different composition in terms of maturity and activation of purified splenic B cells from mice compared to those circulating in humans (Mehta, 2003). Interesting new data has described a role for IL-21 in the small intestine, where it was found to inhibit TGF- induced IgG2b class switching thus skewing towards IgA production by mucosal B cells (Pistoia, 2009). In vitro data on IL-21’s role in isotype switching remains unclear with IgA, IgG1 and IgG3, said to result from exposure to IL-21 during culture of B cells in various studies (Avery, 2008; Pene, 2004)

42 1.4.10 Transgenic IL-21 mouse models

The evidence for both an activating and regulatory role of IL-21 for B cells is exemplified by the introduction of a plasmid inducing constitutive high levels of human IL-21 in mice. Mature B cell populations were reduced in number - which could be explained by apoptosis of resting cells or by a developmental defect. Conversely, there was an increased class-switch to the IgG1 antibody isotype, and upregulation of the germinal centre B cell transcription factor Bcl-6 (Ozaki, 2004). Excessive levels of IL-21 also pushed differentiation into plasma cells in vitro as well as in vivo via Blimp-1 upregulation, indicating that some of these effects could be attributed to an intrinsic B cell signalling response to IL-21 (Ozaki, 2004).

1.4.11 IL-21 deficiency and B cells

While an excess of IL-21 promotes plasma cell differentiation, IL-21R-/- mice have normal circulating mature B cell populations that proliferate normally to mitogens. Perhaps surprisingly - considering that IL-21 can induce dramatic B cell outcomes in vitro - it seems that IL-21 may not be essential for B cell development and function in vivo, and probably it has many overlapping functions with other cytokines such as IL-4 (Ozaki, 2004). One exception to the redundant roles of IL-21 on B cells is IgG1 production, which is heavily dependent on the ability of lymphocytes to respond to IL-21 in vivo (Ozaki, 2002). However, it remains to be seen if this is a B cell intrinsic effect or due to the effect of IL-21 on T cell help. IL-4 and IL-21 together are responsible for the production of high levels of switched antibodies. Mice made genetically deficient in both IL-4 and IL-21 display a severe pan-hypogammaglobulinaemia that is almost identical to X-linked SCID phenotype, implying that in combination - but not individually - IL-21 and IL-4 drive class switch in germinal centres (Leonard, 2000; Ozaki, 2002).

1.4.12 IgE production

Increased IgE has been described in Il21r-/- mice on immunization with experimental antigens and also after infection with the parasite Toxoplasma gondii (Alves, 2005). Conversely, injection of exogenous IL-21 during immunization can inhibit germline IgE heavy chain transcription induced by TLR mitogens in vivo. The mechanism of this inhibition though remains unknown (Leonard, 2000). Any attempts to modulate IL-21 for therapeutic reduction of harmful self-reactive IgG1 production must therefore acknowledge the risk of possible inverse effects on IgE secretion, which is the dominant isotype produced in allergy and atopy.

43 1.4.13 NK T cells and NK cells

NK T cells are a distinct subset of T cells that recognize lipid antigens presented by CD1d. Following activation by CD1d-restricted glycosphingolipid antigens, NK T cells are prolific producer of cytokines which can in turn influence the generation of immune responses. Early studies document an association between decreased numbers and function of NKT cells and the progression of autoimmune disease (Baan, 2007; Falcone, 1999; Laloux, 2001; Wang, 2001). In addition, treatment of type-1 diabetes-prone non-obese diabetic (Onoda) mice with - galactosylceramide, known to activate NKT cells, reduced the severity of autoimmune diabetes in a CD1-dependent manner (Wang, 2001). More recently, IL-21 was shown to enhance NKT cell survival and increase the proliferation of NKT cells in the response to -galactosylceramide, and in combination with IL-2 or IL-15 in vitro. In addition, NKT cells produced IL-21 and IL-21 was shown to increase granzyme B expression, suggesting autocrine stimulation of NKT cells by IL-21 (Coquet, 2007). However, whether IL-21 production by NKT cells negatively or positively impacts on autoimmune disease needs future studies. This thesis focuses on the role of IL-21 in the adaptive arm of the immune response, but IL- 21 has well-established effects on innate immune cells. The ability of IL-21 to modulate NK cell function is of interest to autoimmune disease pathogenesis since the interaction between IL-21 and NK cells has been reported to influence the course of EAE (Vollmer, 2005). IL-21 has been shown to stimulate the production of IFN-inducible genes important in innate immunity and to enhance the cytotoxic activity and IFN production in NK cells, but did not support their viability (Kasaian, 2002; Strengell, 2002). IL-21 also has potent antitumor activity, which has been attributed, in part, to its effects on NK cells (Spolski and Leonard, 2008). Expression of the effector molecules perforin and granzyme A and B was up-regulated in human NK cells by IL-21 at both mRNA and protein levels (Sakaguchi, 2000). In contrast, IL-21 blocked IL-15-induced expansion of both NK cells and memory-phenotype CD8+ T cells (Kasaian, 2002).

1.5 IL-21 in human autoimmune disease and lessons from animal models

1.5.1 Autoimmune Disease: breakdown of tolerance

Autoimmune diseases are a group of diseases where a breakdown in self-tolerance mechanisms at the level of T and/or B cells results in sustained immune responses against self- antigens (Chen, 2008; Forte, 2006). Autoimmune diseases occur in up to 3-8% of the population

44 (Zhou, 2007). The antigens targeted in different types of autoimmune diseases are diverse resulting in very different presentations of pathological signs and symptoms. Many autoimmune diseases involve a specific immune response to a self-antigen expressed solely in one organ, such as Graves disease where follicular cells of the thyroid are targeted, pernicious anaemia the parietal cells of the stomach and T1D the pancreatic cells are the targets of autoimmune destruction. However, there are also autoimmune diseases where the response is directed against antigens ubiquitously expressed throughout the body. A patient suffering from SLE can often develop pathology in their skin, heart, joints, lungs, blood vessels, liver, kidneys and nervous system as a result of deposits of immune complexes of antibody and nuclear antigens (Crowson, 2001). The causes of autoimmune diseases are multifactorial, consisting of genetic, environmental and stochastic components

1.5.2 Systemic Lupus Erythematosus (SLE)

A myriad of different genetic and environmental factors contribute to SLE. A genetic association between two common single nucleotide polymorphisms in the IL-21 gene has been recently established in SLE, and its known effects on plasma cell differentiation and T cell activation make IL-21 an interesting focus in SLE studies (Sawalha, 2008). Serum levels of IL-21 are high in patients with SLE and correlate with severity of the disease, although the clinical significance of these findings remains unknown (Wang, 2007). The identification of an autoimmune regulator gene, named roquin, and studies on the single nucleotide substitution hypomorphic sanroque mouse strain also implicated IL-21 in a lupus like disease (Vinuesa, 2005a). Without the action of the ubiquitin ligase roquin, these mice displayed dysregulated expression of the co-stimulator molecule ICOS, excessive generation of CD4+ Tfh cells and increased IL-21 production, leading to spontaneous formation of germinal centres producing auto-reactive antibodies (Vinuesa, 2005a). IL-21 production was not necessary for pathogenic germinal centre formation in the sanroque model, which suggests that the excessive IL-21 is a result of excessive ICOS ligation rather that a causative agent (Linterman, 2009). In a human study with implications for antibody mediated autoimmune diseases such as SLE, the application of CD40L and IL-21 to mimic co-stimulation by activated CD4+ T cells was found to increase the viability of purified tonsillar GC B cells. These cells are responsible for the generation of the high affinity, class switched antibody that can be harmful in chronic activation of B cells by self-antigen (Good, 2006). Elevated IL-21 production has been reported in a genetic model of SLE, the BXSB-Yaa mouse strain (Ozaki, 2004). Promising therapeutic studies indicate that neutralisation of circulating IL-21 can delay disease the progression of disease in BXSB-Yaa mice, with increased survival in

45 treatment groups and lower levels of serum anti DNA antibodies. Intriguingly, it appears that IL-21 neutralisation may have a duel effect - sequestration of excessive IL-21 can reduce inappropriate activation of self-reactive B cell clones and inhibit protective processes mediated by CD8+ suppressor cells that require IL-21 (Bubier, 2007). The combination of pro- and anti-inflammatory effects of IL-21 on various immune cells will likely challenge researchers aiming to develop studies with clinical benefits from manipulation on IL-21 signalling in SLE.

1.5.3 Rheumatoid Arthritis (RA)

Rheumatoid arthritis is a humoral autoimmune disease in which tissue damage of the joints is excited by the build up of large, pro-inflammatory immune complexes. Antibody complexes recruit a cellular infiltrate to the synovial lining, consisting of T and B cells as well as macrophages, neutrophils and fibroblasts. The initiation of this inappropriate response against unknown self- antigens in the joint are not as well understood as the downstream pathological cycle of cellular inflammation. IL-21 plays a critical role supporting antibody production and there have been several recent studies investigating its role in RA. Just as the Idd3 genetic locus containing the genes Il21 and Il2 has been linked to higher incidence of T1D (discussed below), single nucleotide polymorphisms (SNP)s in the region have also been confirmed in linkage studies with RA (Zhernakova, 2007), leading to the hypothesis that this region may confer a general genetic susceptibility to autoimmune diseases that share some common mechanisms in etiology. The analyses of inflamed synovial membranes by both histological and molecular biology methods reveals that IL-21 and its receptor are overexpressed in RA patients and, despite a somehow ambiguous role in the disease process, studies of therapeutic modulation of IL-21 in RA have been initiated (Jungel, 2004; Li, 2006). Treatment with IL-21R conjugated to an Fc fragment reduces clinical outcomes in models of RA in both rats and mice, perhaps by reducing IgG1 titres in these animals and the subsequent release of harmful cytokines from the cells encountering immune complexes in the joints (Young, 2007).

1.5.4 Multiple Sclerosis (MS) and Experimental Allergic Encephalitis (EAE)

MS is a disease of the central nervous system (CNS) that occurs when the immune system attacks the protective myelin coating around nerve cells with resulting inflammation disrupting nerve signals. A genetic association of IL-21 with MS has been proposed but remains to be established (Forte, 2006). Animal models of CNS inflammation, such as EAE are used to study MS 46 and offer direct support for a role of IL-21 in the disease process. EAE is induced by immunization of mice with myelin antigen in the presence of adjuvants. A recent study suggests that IL-21 acts predominantly on the priming of T cell responses to myelin antigen (Vollmer, 2005). The administration of IL-21 to mice before the induction of disease increases the severity of disease, this was characterized by increased numbers of inflammatory cells in the central nervous system (Vollmer, 2005). In contrast, IL-21 does not affect disease severity when administered after the disease has been initiated. A role for NK cells is suggested, since depletion of NK cells before disease induction abrogates the effect of exogenous IL-21 (Vollmer, 2005). Th17 cells have a critical role in EAE (Chen and O'Shea, 2008). Recent studies have linked IL-21 to the induction and expansion of the Th17 population and disease severity in EAE (Chen and O'Shea, 2008; Korn, 2007a; Nurieva, 2007). A potential interplay between Th17 cells and Tregs has been suggested which could also upset the balance of tolerance versus autoimmunity (Chen and O'Shea, 2008; Fantini, 2007; Korn, 2007a; Nurieva, 2007; Wei, 2007; Zhou, 2007). In this context, the enhanced autoimmune symptoms in mice injected with IL-21 before the initiation of EAE may be the result of increased numbers of Th17 cells and reduced numbers of Tregs. Conversely, the blockade of IL-21 before and after the induction of EAE has been shown to enhance the influx of inflammatory cells into the CNS (Piao, 2008). Intriguingly, when EAE was induced following adoptive transfer of proteolipid peptide (PLP(139-151))-reactive T cells, the blockade of IL-21 induced the proliferation of these self-reactive T cells and decreased the number and function of Tregs (Piao, 2008). These divergent results may reflect the suppressive effect of IL-21 on dendritic cell function or, alternatively, these models might exploit the agonistic versus antagonistic effects of IL-21 on T cell and NK cell expansion to IL-15, respectively (Brandt, 2003; Kasaian, 2002; Strengell, 2006). However, whether these conflicting data for IL-21 in EAE reflect differences in the models, strains of mice used or fundamental distinctions in the role of IL-21 during the course of CNS inflammation await further studies.

1.5.5 Type-1 diabetes

Diabetes is the name given to disorders in which the body has trouble regulating its blood glucose levels. Type 1 or insulin-dependent diabetes mellitus (T1D) in humans is a chronic autoimmune disease involving multiple genes that are under strong environmental influence and promote the destruction of the insulin-producing cells in the islets of Langerhans of the pancreas. In the non-obese mouse model of T1D, as in humans, the target of the autoimmune response is the insulin-producing cells in the pancreas, which are attacked and destroyed by activated T cells (Onoda). The chronic inflammation of the islets of Langerhans in the T1D-prone NOD mouse

47 begins at 4–6 weeks of age. The infiltrate consists of T cells, B cells, dendritic cells and macrophages, and begins as an apparently inconsequential inflammation around the islets (peri- insulitis) progressing to insulitis over the next 2-3 months and culminating in clinical T1D (Bach, 1997; Green, 1999; Ludewig, 1998). Experiments utilizing knockout mice and antibody depletion indicate that all of these cell types have a part in the T1D autoimmune disease process. Determining the factors and conditions that fuel the destructive transformation of pancreatic islet inflammation are vital questions for a better understanding of the pathogenesis of T1D and possible identification of better therapeutic targets. There are a number of genetic loci that are associated with susceptibility to autoimmune diabetes in humans and in the NOD mouse model. One locus that has garnered much interest lies on chromosome 4 in humans and chromosome 3 (Idd3) in mice has a major impact on insulitis and encodes the cytokines IL-2 and IL-21 (Denny, 1997; Todd, 1999). Reduced expression of the NOD IL-2 allele has been demonstrated and is thought to explain much of the effect of the Idd3 susceptibility locus (Yamanouchi, 2007). In contrast, the relative importance of IL-21 in the Idd3 locus effect remains controversial and studies are confounded by the fact that Il2 and Il21 are in strong linkage disequilibrium, such that a haplotype of IL-2 is inherited with the same haplotype of IL-21 (Ikegami, 2003). The NOD mouse has elevated levels of IL-21 but whether the effect of IL-21 on T1D pathogenesis is due to the IL-21 gene itself or is secondary to a defect in IL-2 production remains unclear (King, 2004). Genetic linkage studies support an association of the IL2/IL21 region on human chromosome 4q27 with T1D and this finding has been supported by a recent study in both T1D and RA patients (Todd, 2007; Zhernakova, 2007). In addition, polymorphic variants of IL-21 and its receptor have been associated with genetic susceptibility to T1D in an additive manner (Asano, 2007). Functional analyses are lacking in regard to the function of IL-21 in T1D. One likely target of IL-21- necessary for the development of diabetes in the NOD model - is the CD8+ T cell. CD8+ T cells proliferate to an increased extent in association with elevated IL-21 levels in the lymphoid organs of NOD mice (King, 2004). As discussed above, IL-21 also has a profound effect on B cell proliferation and antibody production and studies showing that a genetic defect in B cells prevents diabetes indicate that B cells play a part in T1D in the NOD mouse. The exact role of B cells in T1D pathogenesis remains unclear, however, studies point to their role as antigen-presenting cells (Silveira, 2004).

48 1.5.6 Gastro-intestinal inflammation and autoimmunity

Inflammatory bowel conditions such as Crohn’s disease (CD), ulcerative colitis (UC) and inflammatory bowel disease (IBD) occupy a hazy area of autoimmunity as studies on germ-free mice indicated that the target of inflammation is often an aberrant response to microflora which should be regarded as ‘self’ in the healthy individual. Self-tissue is destroyed in a collateral manner following the dysregulated response to either the microbiota itself or some other factor generated in the context of a colonised gut. These other factors may include aspects of the mucosal immune system itself, which fail to develop normally in germ-free mice. These diseases all involve an inflammatory cascade that affects both immune and non-immune cells in the mucosa, that combine to promote tissue damage in which cytokines play an essential role. The Idd3 locus containing Il2 and Il21 was found to have an association with celiac disease and also ulcerative colitis in recent genome wide association studies (Festen, 2009; Glas, 2009; Marquez, 2009). Mice deficient in IL-2 develop a diffuse lymphoproliferative disorder with clinical and histological similarities to UC in humans, but the participation of IL-21 in this disorder is not known (Sadlack, 1993). Increased IL-21 transcript has also been identified in biopsies from ulcerative colitis patients compared to healthy controls (Yamamoto-Furusho, 2009). As discussed previously IL-21 is highly expressed by the Th17 subset, which has been shown to participate in pathogenesis in murine models of IBD (Liu, 2009). In a chemically induced experimental model of colitis in Il21-/- had ameliorated disease due decreased recruitment of granulocytes to the mucosa, though the relevance of this link to human diseases mediated by adaptive immune cells remains unclear (Fina, 2008). IL-21 is also thought to be able to act on non-immune cells in a manner that increases pathogenisis. Human intestinal fibroblasts constitutively express the IL-21R, and its actions were found to include release of matric metalloproteinases in vitro, which can potentially degrade the mucosa (Monteleone, 2006).

1.5.7 IL-21 in autoimmunity - concluding remarks

Identifying the cytokines that participate in the development of autoaggressive T cell responses and the mechanisms underlying their involvement in the disease processes will enable the design of more specific immunotherapies for the treatment of autoimmune disease. IL-21 is an interesting candidate for immunosuppressive therapy because of its involvement in the generation of both cytotoxic and antibody responses against self-antigen. Blockade of the IL-21/IL-21R interactions prevents or limits the development of a range of diseases including psoriasis, T1D, lupus (Andersson, 2008; Caruso, 2009; Herber, 2007; Spolski, 2008; Young, 2007). These studies 49 demonstrate that therapeutic inhibition of IL-21 actions have beneficial anti-inflammatory effects in autoimmunity including the reduction of autoantibody production and the prevention of the differentiation of self antigen destructive T cells. However, a number of questions remain that are highly relevant to the pathogenesis of autoimmune disease, including the manner in which IL-21 influences T and B cell differentiation during immunization or infection, and whether IL-21 affects the migration of T cell subsets to lymphoid or tissue sites. The role of IL-21 in Th17 differentiation remains controversial and further analyses of the contribution of IL-21 to T helper cell differentiation will be beneficial to further our understanding of the role of IL-21 in autoimmune disease. Finally, determining whether is there a lymphoid microenvironment where IL-21 is predominantly produced would better define the context in which IL-21 exerts its effects on the immune system.

50 1.6 Experimental objectives

The broad aims of this study were to elucidate the physiological function of IL-21 during healthy immune responses and also during autoimmunity. At the time this began, very little was known about how IL-21 participated in humoral immune responses and Tfh differentiation in vivo, and this study has succeeded in contributing to the current pool of knowledge in this field. Additionally, I was interested in the phenotype of IL-21 expressing cells in vivo, and whether this might reflect varying roles in different tissues. Bearing in mind concurrent expression of IL-21 and IL-21R on T cells, an additional aim of this study was to decipher how autocrine IL-21 may act at the level of the TCR during priming. Another important aim was to discover whether IL-21 was necessary for isotype class switching or SHM and whether these roles might be due to actions of either or both T and B cells during GC reactions. Lastly, this thesis probes how IL-21 can influence dysregulated CD4+ help during autoimmunity, specifically in a murine model of ulcerative colitis. Il2-/- mice lack functional peripheral tolerance imposed by Treg cells leading to diffuse autoimmune disease, and I used this model to determine how IL-21 influences the various components of the adaptive immune system in the absence of regulation.

51 2 Materials and methods

2.1 Buffers

Buffer/Solution Components Suppliers

PBS (10X) 3.6% di-sodium hydrogen orthophosphate (Na2HPO4) Merck 0.2% KCl Ajax Finechem

0.24% KH2PO4 Merck 8% NaCl Ajax Finechem Carboxyfluorescein 0.1% foetal calf serum (FCS) Gibco succinimidyl ester 5uM CFSE Ebioscience (CFSE) buffer 1x phosphate buffered saline (PBS) Gibco ELISA buffer 1X PBS 0.1% Tween 20 ICN Biomedicals

ELISA coating 34.88 mM NaHCO3 Merck buffer 15 mM Na2CO3 Merck

3.08mM NaN3 Amersham

1mM MgCl2 Merck

Red blood cell 8.26g NH4Cl Merck (RBC) lysis solution 1g KHCO3 Merck 0.037g ethylenediamine-tetraacetate (EDTA) Gibco 1L dH20 Lymphocyte 1x RPM1 Gibco isolation media 10% bovine calf serum (BCS) Gibco

FACS buffer 0.1% NaN3 Amersham 0.5% bovine serum albumin (BSA) Gibco 1X PBS Gibco MACS buffer 2mM EDTA Gibco 3% BCS Gibco 1X PBS MACS running 2mM EDTA Gibco buffer 0.5% BSA Gibco 1X PBS MACS rinsing 2mM EDTA Gibco buffer 1X PBS cell culture media 10% Foetal Calf Serum (FCS) Gibco 50 Units/ml penicillin G sodium Gibco 50μg/ml streptomycin sulphate Gibco 2mM L-glutamine Gibco 50μM 2-mercapto-ethanol (2ME) Sigma 1X RPMI 1640 Gibco Standard lysis 50mM tris buffer pH 7.4 buffer 250mM NaCL 5mM EDTA 5mM NaF

1mM Na3VO4 52 1% Nonidet P40

0.02% NaN3 1% protease inhibitor cocktail (P8340 SIGMA) 1mM phenylmethanesulfonylfluoride Western running 24mM tris buffer Gibco buffer 192mM glycine Sigma 0.1% (w/v) SDS Merck Western transfer 12 mM tris buffer Gibco buffer 96 mM glycine Sigma 0.1% (w/v) SDS Merck 20% (v/v) methanol 10x Biotinylation 8.274g Sodium hydrogen carbonate Merck buffer 0.159g Sodium Carbonate Merck 100ml dH20 IEL stripping buffer 1mM EDTA Gibco 1mM DTT SIGMA 5% FCS Gibco 50 Units/ml Penicillin G Sodium Gibco 50μg/ml Streptomycin Sulphate Gibco 1xPBS Gibco DNA isolation buffer 670 mM Tris pH8.8 Gibco 166mM Ammonium sulfate Amersham 65mM Magnesium chloride Amersham 10% 2ME Gibco 5% Triton X-100 Sigma 100μg/mL Proteinase K Promega

2.2 Mice

The Il21-/- mice in this study were created through an NIH initiative with Lexicon and Deltagen, on a mixed C57BL/6 and 129 background and bred to B6 N10. Il21r-/- mice were obtained from Dr Warren Leonard (NIH) via Dr Mark Smyth (Melbourne) at B6 N6 and backcrossed to N7 for experimental use. μMT, IL-7tg, Thy1.1+ OTII and Il21r-/- Thy1.1+ OTII mice were bred in house, all on the C57BL/6 background. Icos-/- and Il2-/- mice were purchased from Jackson Laboratories (ME, USA) and crossed onto Il21r-/- in house. Ly5.1 congenic mice were purchased from the Animal Research Centre in Perth, Australia. Animals were housed under specific pathogen-free conditions and handled in accordance with the Australian code of practice for the care and use of animals for scientific purposes. Age matched littermate mice used for experimental purposes were between 7 and 14 weeks of age.

53 2.3 Flow cytometry

Spleen and lymph nodes were homogenized using 70μm cells strainers in lymphocyte isolation buffer. RBC were removed from spleens using 2ml RBC lysis buffer for 1 minute on ice before washing in Lymphocyte isolation buffer. 50μl of a single cell suspension at 2x107 cells/ml from spleen and lymph nodes were stained in FACS buffer containing pre-titred antibodies in 96 well V-bottomed microtitre plates (Nunc, Roskilde, Denmark) at concentrations shown in table 2-1. To reduce non-specific binding cells were pre-treated with anti-CD16 for 20 minutes (2.4G2 made in house). Cells were acquired using Canto cytometer (BD Biosciences, CA) and analysed using Flowjo (Treestar, CA). Biotin conjugated IL-21 was made in house as described previously (King C, Immunity 2004). Doublets were excluded by forward scatter height and width, expect when analyzing T-B. T-B conjugates were defined as CD4+ and CD19+ double positive doublet cells.

Table 2-2 Antibody clones and concentrations Uses other than flow cytometry are indicated.

Antibody/Reagent Clone Label Dilution 47 DATK32 PE 1:200 B220 RA3-6B2 PE 1:200 PerCP-C5.5 1:200 APC 1:200 APC-Cy7 1:200 Bcl-2 3F11 FITC 1:50 A19-3 (isotype control) 1:50 Bcl-xL 54H6 purified 1:400 isotype control made in house 1:400 BrdU B44 FITC 1:100 CCR7 4B12 APC 1:30 CCR9 eBioCW-1.2 FITC 1:150 CD3 145-2C11 FITC, CD3 1:200 eBio500A2 Biotin (histology) 1:200 eBio500A2 FITC (histology) 1:200 CD4 RM4-5 APC 1:400 Alexa750 1:300 PB 1:300 GK1.5 FITC (histology) 1:200 CD5 53-7.3 PE 1:200 CD8 53-6.7 PerCP-C5.5 1:200 CD11b M1/70 biotin 1:400 CD11c N418 APC CD19 Ebio103 Pacific Blue 1:200 CD21/35 7G6 FITC 1:800 54 eBio8D9 PE 1:300 Biotin (histology) 1:200 CD23 B3B4 FITC 1:400 PE 1:100 PE-Cy7 1:200 CD25 PC61 FITC 1:200 B3B4 PeCy7 1:400 CD27 LG.3A10 PE 1:200 CD38 90 FITC 1:200 CD44 IM7 FITC, APC 1:200 CD62L MEL-14 FITC, PE 1:200 CD69 H1.2F3 PerCP 1:200 CD80 B220 biotin 1:100 CD86 GL1 PE 1:200 CD122 PO3.1 PE, 1:100 CTLA-4 UC10-4F10-11 PE 1:200 CXCR5 2G8 biotin 1:100 Fas Jo2 biotin 1:100 Foxp3 FJK-165 Alexa647 1:200 GL7 GL7 FITC 1:200 Granzyme B GRB04 PE 1:200 ICOS C398-47 PE 1:1000 IgD 11-26c.2a FITC 1:200 Alexa647 (histology) 1:200 IgG1 A85-1 PE 1:200 Biotin (histology) 1:200 IgM II/41 APC 1:100 R6-60.2 biotin 1:100 IL-10 JES5.16E4 PE 1:100 IL-17A TC11-18H10 FITC 1:100 IL-21 Polyclonal Bioitin 1:80 BAF594 R&D IFN XMG1.2 FITC 1:200 Ly5.1 A20 Biotin (histology) 1:200 PeCy7 1:500 PD-1 J43 FITC, PE 1:100 PD-L1 MIH5 PE 1:100 PD-L2 TY25 Biotin 1:100 PNA SIGMA Biotin (histology) 1:400 Streptavidin - PerCP-C5.5 1:500 APC 1:1000 Cy3 (histology) 1:200 Syndecam-1 281-2 PE 1:200 Thy1.1 HIS51 PeCy7 1:300 Alexa750 1:500 TNF MP6-XT22 PE 1:200

55 2.4 Intracellular staining

Extracellular molecules were stained as described previously. Nuclear foxp3 was detected using the ebioscience intracellular staining kit according to the manufacturer’s instructions. Intracellular molecules and were detected using the BD sciences intracellular staining kit according to the manufacturer’s instructions. Cytokines were detected either directly ex-vivo or after 4 hours stimulation at 37°C in cell culture media with Phorbol 12-myristate 13 acetate (PMA 5ng/ml, BIOMOL USA), ionomycin (1μg/ml, Invitrogen) and GolgiStop (1:1000, BD Biosciences, CA),

2.5 Biotinylation of IL-21

The cell surface receptor for IL-21 was detected with the use of biotinylated IL-21. rmIL-21 (100μg) (R & D Systems, Minneapolis, MN) was biotinylated at a 12:1 molecular weight ratio with biotin-DNP (9.5μg in DMSO, Invitrogen) in 1x biotinylation buffer. The mixture was incubated at RT for 40 minutes, mixing every 10 minutes, and then desalted using a 15ml ultracentrifugation device (Amicon, Millipore). Biotinylated IL-21 was used at 3μg/ml, unbiotinylated IL-21 was used as a negative control.

2.6 BrdU Proliferation studies

1mg of BrdU was injected i.p in 200μl of PBS and organs were collected overnight, 10 hours later. Dividing cells were detected by intracellular flow cytometry for incorporation of BrdU using the BD sciences intracellular staining kit according to the manufacturer’s instructions, including cytoperm plus and incubation with DNAse to reveal nuclear antigens.

2.7 CFSE Proliferation studies

Cells were washed twice to remove excess fetal calf serum (FCS) and resuspended at 5x107 per ml for CFSE staining in CFSE buffer containing 5μM CFSE. Cells were incubated at 37°C for 10 minutes then washed twice with ice-cold lymphocyte isolation media before being prepared for cell culture or adoptive transfer.

2.8 Bone Marrow Chimeras

Cohorts of B6 mice were lethally irradiated by a 137Cs source (B6: two doses of 0.45Gy 4 hours apart) and reconstituted the following day by i.v. injection with 10x106 BM cells isolated 56 from femurs and tibiae obtained by flushing bone with lymphocyte isolation media in sterile conditions. Mature T cells were depleted from bone marrow by incubation for 20 minutes on ice with a T cell specific anti-Thy1.2 antibody (clone Jij), generated from ascites preparations (1:200, produced in house). This was followed by the addition of LowTox Guinea Pig complement at 1:10 in lymphocyte isolation media for 40 minutes at 37°C (1:20, Cedar Lane, Canada), and two washes in lymphocyte isolation media. A 1:1 ratio of ly5.1 congenic WT or μMT:IL21r-/- BM cells were found to result in ~50% of lymphocytes derived from each donor. After irradiation, mice were maintained on antibiotic water containing cotrimoxazole (Roche) for 10 days. Mice were used for immunisation protocols as described above at eight to twelve weeks post-reconstitution.

2.9 Immunisations

Mice were immunised i.p with 2x108 sheep red blood cells (IMVS, Australia) or 100μg of

NP13-OVA absorbed to alum and spleens analysed at various timepoints. An equal volume of alum

(Imject) was added dropwise to NP13-OVA at 1mg/mL then vortexed for 30 minutes. 100μg of anti- CD28 agonistic antibody (37-51, ebioscience), or an armenian hamster control IgG antibody (Clone HTK888, Biolegend) in PBS was given prior to immunisation in some experiments. In other experiments 25μg of anti-PD-1 blocking antibody (J43, ebioscience) was injected i.v at the time of immunisation.

2.10 Adoptive transfer studies

For transfer experiments 2 x 107 B220+, CD4+ Ly5.1 congenic cells (purified at day 6 post sheep red blood cells (SRBC) immunisation for primed CD4+ T cells), or 3 x 104 OTII CD4+ T cells were given i.v at the time of immunisation.

2.11 Primary cell sorting

CD4 T cells were stained in MACs buffer with the CD4 biotin MACSbeads negative selection kit according to the manufacturer’s instructions and collected using automacs (both from Miltenyi Biotec, Auburn, USA) giving cell purity of 90-97% for use in cell culture or adoptive transfer. In brief, lymphocytes were resuspended at 1x108 cells/ml and were then incubated with 10μl CD4 negative MicroBeads (Miltenyi Biotec) per 100μl of labelled splenocytes for 20min at 4ºC. These cells were washed once then incubated as above with anti-biotin secondary beads. Labelled lymphocytes were washed twice in MACS buffer and T cells were isolated using the

57 “deplete sensitive” program (to maximize purity) on an autoMACS magnetic Separator (Miltenyi Biotec).

2.12 Primary cell culture and proliferation studies

2x105 lymphocytes or purified T cells were cultured in 96-well flat-bottomed microtitre plates (Nunc) in 200μl cell culture medium. Some assays contained plate bound or soluble anti-CD3 at 2μg/ml (145.2C11, Ebioscience) and soluble anti-CD28 antibody (37-51, Ebioscience) at 1μg/ml for 48 hours before analysis of CFSE dilution or surface markers at 48 hours. For some assays the TCRs were cross-linked without co-stimulation using plate bound anti-CD3. 100μl of 5μg/ml anti- CD3 in PBS was bound to wells at 37°C for 2 hours, followed by 2 washes with 150μl fresh PBS prior to addition of cells. Various soluble molecules were added in some in vitro experiments including, anti-ICOS 5μg/ml (7E.17G9 ebioscience) , CTLA-4Ig 5μg/ml (BD biosciences), IL-21 50ng/mL (R&D systems).

2.13 T cell suppression assay

CD4+ CD25+ T regs or CD4+ CD25- T responder populations were sorted from lymphocyte preparations isolated as for flow cytometry in sterile conditions in cell culture media to high purity using a FACS Aria. T responders were stained with CFSE as described earlier. APCs were obtained by incubation of RBC depleted splenocytes on 15cm cell culture plates at 37°C for 2 hours. Non-adherent T and B lymphocytes were washed away using 3x 20ml volumes of ice cold PBS. The remaining plate bound APCs were collected using a cell scraper (Nunc), resuspended in cell culture media and irradiated by a 137Cs source (2Gy). 8x104 irradiated APC were added to each well along with 0.5μg /ml soluble anti CD3 (145.2C11). 1x105 T responders were added in cell culture media with a 1:1 – 1:16 ratio of T responders to T regs. Cells were then analysed after 72 hours by flow cytometry and the proportion of divided CFSE+ CD4+ T responders calculated by gating on diluted CFSE peaks.

2.14 Immunohistochemistry and immunofluorescence

5μm frozen OCT (Tissue Tek, Australia) spleen sections cut using a cryostat (Leica, Wetzlar, Germany) then fixed in ice-cold acetone for 7 minutes, dried for one hour before re- hydration in PBS for 5 minutes. Primary biotin or fluorochrome conjugated antibodies (Table 2-1)

58 were incubated in 100μl at RT for 2 hours followed by amplification with Streptavidin-Cy3 (Jackson ImmunoResearch) for 1 hour. Slides were washed in PBS and then 90% glycerol was used as a mounting agent. Various issues were also embedded in paraffin by the St Vincent’s pathology department following overnight immersion in 10%formalin in PBS. Hematoxylin and Eosin stained 4μn sections were cut and stained by the Garvan Institute histology facility. Sections were analysed using a Leica DM RBE TCS confocal microscope or Leica light microscope (Leica Microsystems, Wetzlar, Germany). The images were processed using the Leica acquisition and analysis software ImageJ (Freeware NIH Bethesda, USA) or Adobe Photoshop, version 7 (San José, CA).

2.15 Enzyme Linked Immunosorbance Assay (ELISA)

Blood was allowed to clot at RT for one hour before centrifugation at 0.3 RCF for 10 minutes to separate serum for analysis of cytokines or Ig. Serum immunoglobulin was captured by coating plates overnight with anti mouse Ig(H+L) (2μg/ml, Southern Biotech), OVA3BSA or

OVA30BSA (10μg/ml, Molecular Solutions) in ELISA coating buffer. The plate was washed then blocked with 4% milk powder (Coles, Australia) diluted in ELISA buffer for 1 hour at 37°C. Plates were washed, then serum samples were incubated for 2 hours at 37°C at 1:200 in ELISA buffer for antibody detection or neat for cytokine analysis along with 8-15 1:2 dilutions. After washing, analytes were detected using Alkaline phosphatase conjugated anti-mouse IgG, IgG1, IgG2b, IgG2c, IgM, IgA (1:2000, BD Biosciences) at 37°C for 1 hour. All standards used for each isoptype were purchased from Southern Biotech and were used at a top dilution of 1μg/ml. Plates were given 5 final washes before detection of bound AP enzyme with 4-Nitrophenyl phosphate disodium salt hexahydrate at 1mg/ml (SIGMA). The reaction was stopped using 2M NaOH. The titre of NP specific IgG1 was calculated as Log2 of the last dilution factor where the OD was 3x that of background. Serum immunoglobulin was captured using anti mouse Ig(H+L) (2ug/ml, Southern Biotech). IgE was analysed using the BD Biosciences kit according to the manufacturers instructions. IL-21 was assayed from cell culture supernatant using ELISA, according to the manufacturers instructions (R&D Systems).

2.16 Cytokine Bead Array

Analysis of serum from WT, Il21r-/-, Il2-/- and Il2Il21r-/- was carried out using a Flowcytomix mouse Th1/Th2 multiplex including an IL-22 simplex assay (Bender MedSystems). MFI was used to compare binding of the beads to serum cytokine levels, each cytokine was measured using discrete band of fluorescence intensity on the FL1 and FL2 channels using flow 59 cytometry as above. Serum samples were divided into mice from 6-11 weeks of age, and 11-28 weeks of age.

2.17 Single cell sorting of germinal centre B cells and Somatic Hypermutation assay

GL7+fas+CD19+ B cells were collected by flow cytometry activated sorting at day 14 post NP-OVA immunisation and OTII transfer as described above. Single cells were collected directly into 10ul of Taq buffer (Invitrogen), 0.5 mg/ml protease K (Promega), 0.1mM EDTA, 0.1% Tween- 20 in skirted PCR plates (ThermoScientific, Australia) and frozen at –80C. Plates were spun at 2100rpm for 5 minutes and DNA digested by heating to 56C for 40 minutes, followed by 98C for 8 minutes. The dominant VH186.2 segment was amplified as described earlier (Reinhardt et al, 2008). In brief, the sequence containing the VH186.2 intron was amplified using nested primers, and 2.5 μ l of the first round product was used in the second PCR. First round; 5'- ACACAGGACCTCACCATG-3' and 5'TCACAAGAGTCCGATAGACC-3' (with 35 cycles of 95 °C for 10 s, 60 °C for 30 s and 72 °C for 1.5 min), and second round, 5'- GGGTGACAATGACATCCA-3' and 5'-GAGGAGACTGTGAGAGTGGTGCC-3' (with 32 cycles of 95 °C for 10 s, 66 °C for 30 s and 72 °C for 30 s) using Taq DNA polymerase (Invitrogen). Excess primers and nucleotides were removed from the PCR product using ExoSAP-IT (USB, OH) according to the manufacturer’s instructions before sequencing at the Garvan institute in house facility.

2.18 LPL and IEL isolation

Colons were extracted from mice and placed in a petri dish containing pre-warmed PBS. Peyer’s patches were removed and the colon was cut longitudinally to allow faeces to be washed away. Samples were cut into small pieces with scissors and transferred into 50ml falcon tubes. Samples were then washed by using a 25ml pipette to bubble PBS through tissues for 10 seconds. This process was repeated until media cleared. To isolate intraepithelial lymphocytes (IEL), the tissues were incubated in 20ml of IEL stripping buffer for 20 minutes at 37°C while shaking. Tissues were allowed to settle and the supernatant decanted through a cell strainer, then washed twice in lymphocyte isolation media and suspended in 8ml of 40% Percoll (GE Healthcare). 3 ml of 70% Percol was underlayed using a glass pipette and the sample was centrifuged at 600g for 20 minutes at RT. The IEL were then removed from the resulting interface, washed twice in lymphocyte isolation media by centrifuging at 300g for 5 minutes at 4°C and used immediately for

60 flow cytometry analysis. To isolate the lamina propria lymphocytes (LPL), the tissue remaining after treatment with stripping buffer was washed twice as above and resuspended in 5 ml of 5mg/ml collagenase D (Roche) and 0.05% DNAse (Promega) in lymphocyte isolation media. Tissues were incubated in the enzyme solution for 15 minutes at 37°C then another 10mL were added and incubated for another 15 minutes. Tissues were removed and washed twice as above, then passed through a 70μm cell strainer. These cells were run on a Percoll gradient as above then used immediately for analysis.

2.19 SDS Page and Immunoblotting

1x106 purified CD4 were stimulated in vitro in plates coated with 20μg/ml agonistic anti- CD3 and anti CD28 (Ebioscience) for various timepoints, with or without rmIL-21 (Peprotech, NJ) at 50ng/ml, or PD-L2-Fc (R & D systems) 10μg/ml. Cells were washed and lysed in standard lysis buffer (Table 2-1) and run by SDS polyacrylamide-gel electrophoresis using a 4-12% gradient gel and 0.45μm or 0.2μm pore size nitrocellulose membranes (Invitrogen). Membranes were first blocked using 5% BSA in TBS, probed with anti Vav1, anti phospho-Vav1 Tyr 174, (from SantaCruz, CA) and anti B-actin (Sigma Aldrich, USA) or anti ZAP-70, anti ZAP-70 Tyr 319 and Tyr 2202/204 (Cell signalling). Antibody binding was detected using goat anti-rabbit or anti-mouse IgG-HRP (DAKO, Denmark). Detection antibodies were diluted in 2% BSA in TBS with the addition 0.01% Tween 20 (SIGMA). Membranes were developed by incubation with enhanced chemiluminescence substrate (Perkin-Elmer, MA) for one minute before immediate exposure to X- Ray sensitive film (Fuji).

2.20 RNA analysis

Pooled lymphocytes from 3 mice were stimulated 4 hours in triplicate and RNA isolated using TRIzol reagent (Invitrogen) for RNA isolation. Cells were pelleted by centrifugation then resuspended in 300μl TRIzol reagent (Invitrogen) with 5ng nucleic acid carrier GenElute linear polyacrylamide (LPA, Sigma) with immediate vortexing for 10s. 60μl of chloroform (Sigma) was added to each TRIzol lysate followed by vortexing for 30s. Samples were pelleted at 12,000 x g for 15min at 4°C before removal of the upper aqueous phase to a new tube. After the addition of 180μl of isopropanol, RNA was precipitated at -20°C for 24-48h. RNA was subsequently pelleted at 12,000 x g for 30min at 4°C before removal of supernatant and washing once with 500μl of 75% ethanol. RNA pellets were dried and resuspended in nuclease free water (Sigma) and quantitation performed on a 2100 Bioanalyser (Agilent Technologies, CA, USA) using a Eukaryote Total RNA Nano assay (Agilent Technologies) or a NanoDrop 1000 (Wilmington, DE, USA). cDNA was 61 prepared using Superscript II reverse transcriptase (Invitrogen) and oligo-dT primers. We determined the relative abundance of cDNAs in triplicate by qRT-PCR analysis using the ABI Prism 7700 Sequence Detection System (Applied Biosystems). Real-time PCR primers for IL-21, and GAPDH were obtained from Applied Biosystems. The RT-PCR intronic probe used to detect IL-21 was Fam-TGCTCACAATTTACAGCCTCACAGTTTAGTCATTGT-Tamra. Fluorescence signals were measured over 40 PCR cycles and the cycle (Ct) at which signals crossed a threshold set within the logarithmic phase was recorded. For each assay, standard curves were generated to identify positive signals on the linear part of the curve. Values for IL-21 were normalized to GAPDH expression in each sample. Fold modulation of mRNA was calculated by employing a comparative Ct method; Relative abundance of genes = 2({Delta}Ct), where {Delta}Ct is the difference between the Ct of target and the artithmetic mean of Cts of GAPDH.

2.21 Polymerase chain reaction

2mm of mouse tail was incubated overnight at 65°C in 200ul DNA isolation buffer with 1.13ul of Proteinase K (10mg/ml, Promega) and 0.5μl of the resulting digest was used as a template in PCR reactions. All primers and probes used for PCR are listed in Table 2-2 from 5’ to the 3’ end. Mice were screened for presence/absence of genes and transgenes by PCR analysis. PCR was conducted using 0.5μl of DNA template, 0.24μM forward and reverse primer (Sigma-Aldrich), 50μM each dNTP (Promega, Madison, WI, USA), 0.625 U Taq polymerase (Promega) and 1X Green GoTaq Reaction Buffer (Promega) in a 25μl reaction volume. All reactions were performed on an iCycler Thermal Cycler (Biorad, CA, USA). PCR products were visualized by electrophoresis on a 1-2% agarose (Sigma) gel containing 10μg/ml ethidium bromide (Sigma- Aldrich).

62 Table 2-3 Primer sequences F indicates the forward primer, whereas R indicates the reverse primers for each pair. Neo indicates primers that bind within the neomycin resistance cassette used to detect mutant knock out gene mice, whereas WT annotates primers that were used to detect the native gene sequence. Primer Sequence 5’ to 3’ Annealing T IL-21 F neo GCAGCGCATCGCCTTCTATC 65 IL-21 R neo ACCATTCTACTGACTTGTTAGACTC 65 IL-21 F WT GGAGACTCAGTTCTGGTGG 65 IL-21 R WT GGAGCTGATAGAAGTTCAGG 65 IL-21R F neo ATCGCCTTCTATCGCCTTCTTGACG 60 IL-21R F WT GACTCTTGGCCTGCAGTTCTGACG 60 IL-21R R common CCAAAGAGCTCCAGTAAACAG 60 IL-2 F neo TCGAATTCGCCAATGACAAGACGCT 64 IL-2 F WT CTAGGCCACAGAATTGAAAGATCT 64 IL-2 R common GTAGGTGGAAATTCTAGCATCATCC 64 OTII F GCTGCTGCACAGACCTACT 59 OTII R CAGCTCACCTAACACGAGGA 59 THY1.1 F GTGCTCTCAGGCACCCTC 65 THY1.1 R CCGCCACACTTGACCAGT 65 THY1.2 F GCGACTACTTTTGTGAGAGCTTCA 65 THY1.2 R CGCCACTTGACCAGC 65

2.22 Data analysis and Statistics

Data were analysed using Prism software (Graphpad software, CA) to calculate un-paired, two-way Student’s t test, with an F test to compare variances. Analysis of more that two groups was performed using one-way ANOVA followed by Bonferroni’s post test to compare groups.

63 3 An overview of IL-21 in the adaptive immune system

3.1 Introduction

The cytokine interleukin-21 (IL-21) signals through the c-chain in concert with a unique co-receptor chain (IL21R) which is widely expressed by many immune cells including T and B lymphocytes, NK cells and DCs. Conversely, IL-21 protein production is largely restricted to CD4+ T cells and NKT cells in response to TCR crosslinking. IL-21 signals through the IL-21R to induce phosphorylation of JAK1 and JAK3, which, in turn, activates sustained STAT3 signalling and also STAT1 and STAT5 transcription factors though it is not known if these transcription factors alone orchestrate IL-21’s downstream effects (Coquet, 2007; Spolski and Leonard, 2008). IL-21R expression by a broad range of immune cells may reflect its pleiotropic roles, which include both stimulatory and inhibitory effects reported among these cell subsets. The role of IL-21 in T helper cell differentiation remains controversial. This is largely due to the fact that IL-21 does not readily fit the established Th1/Th2 cytokine models. Some studies have demonstrated a functional contribution of IL-21 to the production of Th1 cytokines (Fina, 2007; Strengell, 2004; Strengell, 2002) and to the generation of Th2 responses (Frohlich, 2007; Mehta, 2005; Wurster, 2002). Recent studies have highlighted a controversial new role for IL-21 as a growth factor for newly characterised CD4+ T cell subsets (King, 2004). The T helper cell subsets that have been reported to produce IL-21 include Tfh, Th17 and Th2 cells (Chtanova, 2004; Korn, 2007a; Nurieva, 2007; Pesce, 2006). The highest levels of IL-21 mRNA transcript have been found in a subset of CD4+ T cells which is thought to provide help to B cells producing antigen specific immunoglobulin, termed Tfh (Chtanova, 2004; Rasheed, 2006). This non-polarised lineage of T cells expresses chemokine receptors such as CXCR5, which pilot Tfh cell navigation to the B cell follicles where high levels of IL-21 and co-stimulatory molecules can effectively influence high affinity antigen specific antibody responses by germinal centre B cells (King, 2008). Tfh express significantly greater amounts of IL-21 than other Th subsets (Chtanova, 2004) - a finding suggesting an important role of IL-21 in humoral responses. The requirement for IL-21 by B cells in vivo has been studied in mice using two different methods. Firstly, introduction of a plasmid expressing constitutive high expression of IL-21 into mice showed IL-21 could both activate and regulate B cells. Reduced mature B cell populations were observed which could be explained either by apoptosis of resting cells, or a developmental defect. Conversely, there was also an increased class-switch to the IgG1 antibody isotype, and upregulation of the germinal centre B cell transcription factor Bcl-6. Constitutive excessive IL-21 64 propelled B cells to develop into plasma cells in vivo via Blimp-1 upregulation (Asao, 2001). Secondly, Il21r-/- mice were created, and shown to have normal circulating mature B cell populations (Kasaian, 2002; Ozaki, 2002), implying that there may not be a linear association between nil and excessive IL-21 production. Perhaps surprisingly, considering IL-21 can induce such dramatic B cell outcomes in vitro, it seems that IL-21 is not essential for B cell development and function and probably has many overlapping functions with other co-stimulation molecules. The studies described in this chapter were designed to examine what role IL-21 may play in the development and homeostasis of the adaptive immune system. Il21r-/- mice were first created by Kasaian et al. in order to study how IL-21 affected cells of the innate immune system, particularly NK cell development (Kasaian, 2002). In neither this work, nor consequent studies by Ozaki et al describing serum antibody defects in these mice, was general phenotyping of these mice published, but they were reported to be identical to WT in terms of lymphocyte development and numbers (Ozaki, 2002). Both Il21r-/- and mice where the Il-21 gene has been deleted (Il21-/-) were integral in our studies, so we performed a range of phenotyping of cell numbers and cell surface marker expression and confirmed that both these strains of mice did indeed have normal mature T and B cell numbers, as well as normal granulocyte and monocyte populations. We were particularly interested in further characterising the phenotype of CD4 T cells that produce IL-21 and its role in both CD4 T cell activation and proliferation, as well as exploring alternative downstream routes of activation that may lie outside the JAK:STAT system. In addition, we studied adhesion and cell trafficking molecules that were expressed in concert with IL-21 by CD4+ T cells in order to locate naturally occurring IL-21-producing CD4 T cells in secondary lymphoid organs.

65 3.2 Results

3.2.1 IL-21 is redundant for immune development and homeostasis

This thesis is particularly concerned with how IL-21 may modulate mature T cells responses in the periphery. We purchased IL-21 deficient (Il21-/-) mice (Lexicon/NIH initiative (UCR)), and performed general phenotyping of the immune system to confirm that IL-21 has no obvious role in lymphoid development and validate these mice as a model for study of mature lymphocytes. We confirmed that Il21-/- mice displayed normal thymic development in terms of numbers and proportions of double negative, double positive and single positive thymocytes (Figure 3-1). Co- stimulation via CD5 alongside TCR ligation of double positive thymocytes has been shown to enhance CD4 single positive cell development due to its ability to recruit inhibitory phosphatases to the TCR signallosome (Azzam, 2001; Tarakhovsky, 1995a; Wong, 2001). As such it is an interesting target for investigation as it can reflect the strength of TCR signals received in the thymus. In particular, CD5 ligation was found to be important for Bcl-2 upregulation, and thus survival of DP thymocytes and therefore we examined expression of this key marker of thymic output in Il21-/-. IL-21 had no role in modulation of CD5 levels on thymocytes as surface expression was comparable between WT and Il21-/- mice (Figure 3-1).

A

B

Figure 3-1 Il21-/- thymic T cell populations are equivalent to WT (A) Representative dot plots showing CD4 and CD8 expression on CD3+ cells in the thymus. (B) Histograms showing double negative (DN), double positive (DP), CD4+ single positive (CD4 SP) and CD8+ single positive (CD8 SP) thymocytes from WT (unbroken line) and Il21-/- (dashed line). Representative of 2 experiments in which n=4 mice at 8 weeks.

66 Antigen presenting cells (APC) can shape the behaviour of mature T cells by modulation of co-stimulatory cell surface molecules and cytokines at the time of antigen presentation. IL-21 has previously been reported to suppress human monocyte derived dendritic cell activation in vitro via SOCS signalling (Strengell, 2006), and thus an Il21-/- deficient strain could be expected to display increased APC activation. To rule out any disparate activation by APC in the absence of IL-21, Il21-/- mice were immunised with a T dependent antigen, SRBC, and APC populations were analysed 2 days later. We found no difference in CD11c+ dendritic cells (DC) or CD11cloB220+ plasmacytoid DC numbers at this timepoint, nor any difference in expression of pro-inflammatory (CD80, CD86) or anti-inflammatory (Programmed cell death ligands PD-L1, PD-L2) co-stimulatory molecules (Figure 3-2A and B).

(A) Representative flow cytometry dot plots showing equivalent B220+CD11c- B cell, CD11b+B220- Dendritic cells and B220+CD11blo plasmacytoid dendritic cells in spleens of WT and Il21-/-. (B) Histograms showing expression of co-stimulatory molecules (CD80 and CD86) and co-inhibitory molecules (PD-L1 and PD-L2) on splenic antigen presenting cell populations 2 days after immunisation i.p with SRBC. Data represents 2 experiments in which n=3 mice at 8 weeks of age.

Figure 3-2 Normal expression of T cell co-stimulation molecules by Il21-/- mice

67 We found that peripheral CD4 and CD8 single positive T cell numbers were also normal in Il21-/- mice, complementing unpublished observations from Il21r-/- mice by Kaisaian et al and Osaki et al (Figure 3-3A and B) (Kasaian, 2002; Ozaki, 2002). Likewise, mature follicular CD23+CD21+ B cells, CD23+CD21- marginal zone (MZ) and immature CD23-CD21- cell populations in the spleen and peripheral lymph nodes were found in equivalent numbers in WT and Il21-/- mice (Figure 3-4 and data not shown).

Figure 3-3 Il21-/- mice have a normal peripheral T cell phenotype Representative dot plots (A) and quantification (B) of the equivalent numbers of memory CD62L- CD44hi and resting and naïve CD44lo CD4+ T cells (A left) and memory CD44hi CD122+ and resting CD44lo CD8+ T cells (A right) in a CD3+ gate in the spleen of WT and Il21-/-. Representative of 3 experiments in which n=10 mice at 8-12 weeks of age. P>0.05 by Student’s T test between WT and Il21-/- in all organs.

Figure 3-4 Normal peripheral B cell phenotype of Il21-/- mice Representative dot plots showing (A) CD23+CD21 follicular, CD21+CD23- MZ and CD23-CD21- immature B cells in WT and Il21-/-. (B) Further characterisation of immature B cell gate from (A) showing CD21-IgMhi T2 transitional and CD21-IgMlo T1 transitional B cells. Data is representative of 2 experiments in which n=3-4 mice at 8 weeks of age.

68 3.2.2 IL-21 is produced by activated effector phenotype CD4+ T cells

IL-21-producing CD4+ T cells were detected by intracellular immunostaining in order to see if particular CD4+ T cell effector or memory surface phenotypes correlated with protein expression. IL-21 production was higher at peripheral mucosal sites such as the mesenteric lymph nodes (MLN), iguinal lymph nodes (ILN) and peyers’ patches compared to the spleen (Figure 3-5A). The specificity of IL-21 detection was controlled against IL-21-deficient CD4+ T cells (Figure 3-5B).

Figure 3-5 IL-21 production is elevated at mucosal immune sites (A) Quantification and (B) representative MLN dot plots of IL-21 expression measured by flow cytometry as a percentage of CD3+ CD4+ T cells. Data represent individual mice between 9-12 weeks pooled from 6 different experiments, each in a different colour, plus the mean +/- SEM.

IL-21 producing CD4+ T cells in resting B6 have all the hallmarks of a CD4+ T helper effector memory population with high expression of CD44 and CD27 (Figure 3-6). 80-90% of IL- 21+ CD4+ T cells had a memory phenotype (CD44hi) and included cells that were intermediate for CD44 expression, whereas no IL-21 was produced by CD44lo naïve cells. (Figure 3-6 and Figure 3-7). CD27 has been described as a CD8+ T cell marker, which can distinguish long-lived effector memory type cells from short-lived central memory cells (Dolfi, 2008; Hendriks, 2000) and recent publications suggest a similar expression pattern of CD27 on memory CD4+ T cells (Pepper, 2009). IL-21 expression is associated with a potentially long-lived CD27+ phenotype in the spleen and lymph nodes, whereas the peyers patch phenotype appears to consist of a mixture of short and long- lived memory cells (Figure 3-6). This may reflect the ongoing immune activity in peyer’s patches from proximity to commensal microbiota in SPF mice. 69 Many IL-21+ CD4+ T cells also expressed the T helper Inducible Co-stimulation molecule (ICOS), which interacts with its ligand on B cells and suggests T-B interactions may be important for IL-21 induction (Figure 3-6). Very little IL-21 was co-expressed alongside the Treg marker CD25 implying it is not associated with regulatory function of CD4+ T cells (Figure 3-6).

Figure 3-6 IL-21 is produced by memory/effector phenotype CD4+ T cells Quantification of cell surface markers co-expressed with IL-21 in vivo measured by flow cytometry gated on IL-21+ CD4+ CD3+ T cells. Data are pooled from 2-3 experiments, points represent individual mice plus the mean -/+ SEM.

Effector and memory T cells can also be classified based upon their ability to traffic through peripheral tissues, a property controlled by the expression of tissue-specific adhesion and chemoattractant receptors (Kunkel, 2003). CD62L denotes CD4+ T cells which are able to respond to its ligand, PSGL-1, allowing entry into the lymph node, and is generally high on naïve CD4+ T cells and downregulated on memory T cells that can migrate to tissues to execute effector functions (Bevilacqua, 1991; Lasky, 1992). IL-21+ CD4+ T cells displayed a mixed phenotype, with approximately 60% remaining CD62L high and being retained in the lymph node. Bearing in mind these cells were all CD44 high or intermediate (Figure 3-7), this may reflect a T helper phenotype associated with past T-B cell interactions within lymph nodes or spleen (Fazilleau, 2007a). This 70 was not the case in the peyers’ patches where the dominant CD62L low status implies that these IL- 21+ cells are activated/memory phenotype cells that are destined to enter the circulation. CD69 is the earliest molecule upregulated after T cell activation (Ziegler, 1993), and high co-expression in the peyers’ patches suggested that these IL-21+ CD4+ T cells contained a distinct recently activated, CD62L low circulating subset as opposed to cells from other tissues (Figure 3-7).

Figure 3-7 IL-21+ CD4+ T cells at mucosal sites have a recently activated phenotype Quantification of migration and cell adhesion molecules co-expressed with IL-21 in vivo measured by flow cytometry gated on IL-21+ CD4+ CD3+ T cells. Data are pooled from 2-3 experiments, points represent individual mice plus the mean -/+ SEM.

CD4+ T cells activated in intestinal lymph nodes selectively acquire responsiveness to the ligand for the G protein-coupled chemokine receptor 9 (CCR9) (Papadakis, 2000), the intestinal and thymus-expressed chemokine (TECK), and also express high levels of alpha 4 beta 7 (47) integrin (Campbell, 2002). Co-expression of IL-21 with CCR9 would suggest that they originated from the small intestinal mucosa and/or have the potential to home to tissues that express CCL25 (TECK), including small intestine (Calabrese, 2009). Despite data showing IL-21 levels were highest in the peyers’ patches, it seems that as only 10-20% of mucosal homing adhesion markers

71 coincided with IL-21 expression in resting B6, the origin of T cell priming of these cells was largely within lymphoid tissues outside of the mucosa (Figure 3-7).

3.2.3 ICOS expression on CD4+ T cells modulates IL-21 production

IL-21 transcript is highly expressed by the Tfh subset that delivers help to B cells within GC (Chtanova, 2004; Korn, 2007a). The interactions between TFH cells and B cells within the GC are governed by signals through the TCR and BCR antigen receptors in conjunction with MHC class II and several co-stimulatory molecules and cytokines. ICOS and its ligand on B cells have been described as key modulators of T and B cells responses at the time of B cell antigen presentation to T cells. Indeed, mice made genetically deficient in ICOS (Icos-/-) do not generate Tfh cells and exhibit defective GC formation in response to T dependent antigen (Akiba, 2005; Bossaller, 2006; Tafuri, 2001). Tfh cells are characterised by their expression of both ICOS and IL- 21 but the relationship between these two molecules remains unclear (Chtanova, 2004; Hutloff, 1999). As we had seen co-expression of IL-21 and ICOS previously (Figure 3-6), we determined whether ICOS expression had an impact on IL-21 production by comparing IL-21 levels in splenocytes from Icos-/- and WT mice, following in vitro stimulation with anti-CD3 antibody. Splenocytes from Icos-/- mice exhibited markedly reduced IL-21 mRNA expression when cells were analysed directly ex vivo and following stimulation with anti-CD3 in vitro (Figure 3-8A) and when Il21 mRNA was normalized for background expression of Il21 mRNA present ex vivo in each genotype (Figure 3-8B).

Figure 3-8 ICOS expression affects IL-21 mRNA levels in spleen (A) IL-21 mRNA levels from WT and Icos-/- splenocytes ex-vivo (0 hours) and after 2 hours stimulation with soluble anti-CD3 measured by Real Time PCR. IL-21 mRNA expression is presented as fold modulation compared to WT ex-vivo levels. (B) IL-21 mRNA from (A) presented as delta change from ex-vivo levels from each genotype. Data is shown as the mean +/- SEM and individual values from 2 experiments in which n=4-11

72 These data revealed an integral role for ICOS in IL-21 production, but it remained unclear whether the splenocytes analysed from Icos-/- mice contained fewer activated IL-21+ CD4+ T helper cells or whether co-stimulation via ICOS:ICOSL interactions was necessary for IL-21 production. To differentiate between these two possibilities, we partitioned the naïve and memory CD4+ T cell subsets based on CD44 expression from Icos-/- and WT mice and measured IL-21 mRNA 4 hours after stimulation with anti-CD3 and anti-CD28 in vitro in order to compensate for any differences in memory phenotype numbers in Icos-/- mice. The naïve (CD44lo) T cells from Icos-/- mice produced similar amounts of IL-21 mRNA relative to naïve CD4+ T cells from WT mice (Figure 3-9A). Similarly, the memory (CD44hi) populations also exhibited no significant difference in their IL-21 mRNA levels after stimulation (Figure 3-9A). However, the provision of WT ICOSL bearing B cells to the cultures boosted the IL-21 expression levels in WT CD4+ T cells but had no effect on Icos-/- CD4+ T cells (Figure 3-9B), indicating that optimal IL-21 expression by CD4 T cells requires co-stimulation via ICOSL expressed by B cells at the time of antigen presentation.

Figure 3-9 B cells enhance IL-21 production by CD4 T cells via ICOS:ICOSL (A) IL-21 mRNA from either naïve (CD44lo) CD4+ T cells or memory (CD44hi) CD4+ T cells cultured with soluble CD3 and CD28 mAbs (both at 2 μg/ml). (B) IL-21 mRNA expression in memory (CD44hi) CD4+ T cells cultured with CD3 mAb and WT B cells. Data are presented as fold modulation compared to ex-vivo WT naïve cell expression. Data is shown as the mean +/- SEM and individual values from 2 experiments in which n=4-11.

73 The importance of ICOS for IL-21 production by WT T cells was recapitulated by blocking the interaction of splenic T and B cells in vitro with a neutralising antibody, resulting in reduced IL- 21 mRNA induction (Figure 3-10A). In addition, CD28-B7 interactions were also found to be necessary for maximal IL-21 mRNA expression as blockade of these interactions in vitro again reduced IL-21 mRNA levels (Figure 3-10B). Finally, we confirmed the decreased IL-21 mRNA from Icos-/- CD4+ T cells, and from cells with anti-ICOS blockade, corresponded with a decreased production of IL-21 protein (Figure 3-10C). Taken together, these findings demonstrated that T helper cells utilize B cells via ICOS:ICOS-L interactions that quantitatively contribute to IL-21 production.

Figure 3-10 Costimulation through ICOS and CD28 drives IL-21 expression IL-21 mRNA from splenocytes stimulated with soluble CD3 mAb in the presence or absence of (A) ICOS mAb (5μg/ml) or (C) IL-21 production measured by ELISA in supernatants from duplicate cultures of splenocytes stimulated for 4 days with CD3 mAb. Data is shown as the mean +/- SEM as well as individual points from cultures derived from individual mice (n=2), representative of 2 experiments.

74 3.2.4 IL-21 costimulates the TCR to modulate activation markers

A specific requirement for IL-21 in CD4+ T cell activation was evident when we performed comparative analyses of the proliferation of Il21-/- and WT CD4+ T cells in response to CD3 and CD28 monoclonal antibodies. A considerably reduced fraction of Il21-/- CD4+ T cells were observed in the divided populations and this was found to be a direct effect of IL-21 deficiency because the addition of exogenous rmIL-21 recovered the proliferation of Il21-/- cells (Figure 3-11).

Figure 3-11 Exogenous IL-21 recovers Il21-/- CD4+ T cell proliferation defect Histograms show division of purified CD4+ T cells measured by dilution of CFSE, analysed by flow cytometry. Cells were cultured for 48 hours with soluble anti-CD3 with and without the presence of soluble anti-CD28 and exogenous IL-21. The rightmost peak represents undivided cells. Data is representative of 4 experiments in which n=8-9

Since most costimulatory signals are additive to the TCR, we argued that cross-linking the TCR or providing ample CD28 monoclonal antibodies (mAb) might overcome the need for IL-21. Therefore, we examined the effect of IL-21 on the expression of the signature CD4+ T cells activation molecules with an emphasis on the provision of T cell help to B cells. We investigated ICOS, CXCR5 as well as CCR7 and CTLA-4 on purified CD4+ T cells from Il21-/- and WT mice following activation with CD3 mAb and increasing concentrations of CD28 mAb in vitro. 75 IL-21 deficient CD4+ T cells stimulated with CD3 mAb and low dose CD28 mAb failed to upregulate either ICOS or CTLA-4 in vitro (Figure 3-12 and Figure 3-13). However, the expression of CTLA-4 on Il21-/- CD4+ and WT CD4+ T cells could not be distinguished at the highest dose of CD28 mAb nor in the presence of cross-linked CD3 mAb (Figure 3-12) meaning IL-21 may be important for optimal activation of polyclonal CD4 T cells that receive a weak TCR signal. Similarly, the activation-induced down-regulation of CCR7 was impaired in IL-21 deficient CD4+ T cells but this defect was overcome by increasing the strength of signal through the TCR (Figure 3-12).

Figure 3-12 T helper phenotype is modulated by IL-21 co-stimulation Representative histogram overlays showing flow cytometric analysis of CD4+ T cells stimulated for 48 hours with soluble CD3 mAb (2 μg/ml) in the presence of increasing doses of CD28 mAb, or plate bound CD3 mAb. Overlays show modulation of CTLA-4, CCR7 and ICOS from WT ex-vivo (filled) and cultured WT (unbroken line) and Il21-/- (dashed line). Data are representative of 3 experiments in which n=7.

By contrast, the levels of ICOS remained relatively reduced on Il21-/- CD4+ T cells (Figure 3-13), implying a dependence on signals via IL21R in addition to the TCR. Indeed, ICOS levels were boosted following the addition of IL-21 to CD3 mAb stimulated Il21-/- CD4+ T cells, confirming that IL-21 upregulated ICOS expression (Figure 3-13). Purified CD4+ T cells from both groups failed to upregulate CXCR5 under these in vitro conditions (data not shown). Collectively, these findings indicated that the defect in activation of CD4+ T cells in the absence of IL-21 could be partially explained by decreased costimulation of the TCR.

76 Figure 3-13 Exogenous IL-21 can recover Il21-/- ICOS defect Representative dot plot showing ICOS expression on CFSE stained CD4+ T cells from WT or Il21- /- after 48 hours culture with CD3 mAb (2μg/ml) and CD28 mAb (0.1μg/ml) in the presence or absence of exogenous IL-21 (50ng/ml). Data are representative of 2 experiments in which n=4.

Il21-/- cells cultured in vitro exhibited reduced modulation of molecules associated with CD4+ T cell activation after 48 hours, notably CTLA-4 which is an inducible inhibitory B7 family member that is upregulated following CD28 ligation (Sansom, 2006). In order to rule out any intrinsic defects in sensitivity to co-stimulation through CD28 such as low expression levels, we compared the surface phenotype of WT and Il21-/-, and found they had identical CD28 surface expression. This was the case both ex vivo and after 4 days stimulation where similar high inducible levels were achieved by both genotypes (Figure 3-14). CD28 expression was also identical on CD4+ T cells from the MLN (data not shown).

Figure 3-14 CD28 expression is normal in Il21-/- mice Representative histograms showing CD28 expression on splenic CD4+ T cells both ex vivo and after 4 days stimulation with soluble CD3 and CD28 in vitro, WT (unbroken line), Il21-/- (dashed line), ex-vivo WT CD4 levels (filled). Datais representative of 2 experiment in which n=3.

77 3.2.5 IL-21 acts at the level of the TCR signallosome to impact T helper cell fate

Since IL-21 potentiated the TCR mediated activation of CD4+ T cells in these in vitro cultures, we determined whether signalling through the TCR was affected by IL-21. Vav1 is activated by tyrosine phosphorylation following TCR stimulation and is also important for the PI3K-dependent pathways that lie downstream of the IL-21R. We focused our attention on Vav1 for two reasons; first, Vav1 is an integral component of the TCR signallosome and, secondly, Vav1 deficient CD4+ T cells are defective in their T cell helper function (Gulbranson-Judge, 1999; Villalba, 2001).

Figure 3-15 IL-21 compounds TCR signals to CD4+ T cells Immunoblots probed for tyrosine phosphorylation of Vav1, total Vav1 and -actin after addition of exogenous IL-21, cross-linked CD3 mAb or CD28 mAb as indicated in purified WT and Il21r-/- CD4+ T cells after 10 minutes of anti-CD3 monoclonal antibody stimulation in the presence or absence of exogenous IL-21. Densitometry is shown of relative tyrosine phosphorylation of Vav1 compared to unstimulated control samples. Data are representative blots from 3 experiments where cells were pooled from 2-3 mice.

Crosslinking of the TCR with anti-CD3 antibody led to rapid phosphorylation of Vavl in CD4+ T cells purified from both Il21r-/- and WT mice. Some basal phosphorylation was also detected, which is probably due to residual activation derived from in vivo activation of CD4+ T cells within the purified cohort which contained both memory and naive cells. The addition of IL-21 to WT cells greatly potentiated TCR-induced phosphorylation of Vav1 with no such effect on Il21r- /- cells (Figure 3-15). The lack of Vav1 phosphorylation observed by adding IL-21 to Il21r-/- CD4+ T cells confirmed the specificity of the effect (Figure 3-15).

78 Immunoblots probed for tyrosine phosphorylation of Vav1, total Vav1 and -actin after addition for 10 minutes of exogenous IL-21, cross-linked CD3 mAb or CD28 mAb to purified WT CD4+ T cells purified from 2-3 mice. Data are representative blots from 3 experiments.

Figure 3-16 IL-21 alone can phosphorylate Vav1

In contrast, experiments performed in parallel demonstrated that IL-21 alone did not phosphorylate Zap70, indicating that IL-21 signalling intersects with the TCR downstream of Zap70 and importantly does not simulate a TCR signal, but rather augments antigen stimulation (Figure 3-17). These findings support the notion that IL-21 co-stimulates TCR signals to promote T helper cell activation and differentiation.

Figure 3-17 IL-21 acts downstream of proximal Zap70 activation Immunoblots probed for tyrosine phosphorylation of Zap70 and -actin after addition of exogenous IL-21, cross-linked CD3 mAb or CD28 mAb to purified WT CD4+ T cells pooled from 2-3 mice. Data are representative blots from 2 experiments.

IL-21 expression may modify or enhance TCR signals, which effect eventual T helper outcomes. In order to see whether IL-21’s ability to co-stimulate through Vav1 phosphorylation may be occurring in the Tfh subset that has been described to produce particularly high levels of IL- 21, we performed Vav1 immunoblots on various purified T cell subsets following immunisation with a T dependent antigen, SRBC. We found that among activated, CD44hi ICOS+ CD4+ T cells from the spleen, Vav1 phosphorylation was particularly apparent among CXCR5+ expressing cells capable of migration into the B cell follicle to provide help, implying that IL-21 may be important for this subset’s generation or function (Figure 3-18). Apart from Tfh cells (Nurieva, 2007;

79 Vogelzang, 2008), a number of other T helper subsets have been reported to be dependent upon IL- 21 for their generation, such as Th17 and Th2 cells (Korn, 2007a; Nurieva, 2008). Co-stimulation from IL-21 via Vav1 may not be unique to Tfh, as the ICOS- CD44hi fraction also contained strong phosphorylation, showing that other discrete cell activation pathways such as Th1 or Th2 activation may share this requirement.

Figure 3-18 Tfh cells generated in response to T dependent antigen exhibit strong Vav1 phosphorylation Immunoblots probed for tyrosine phosphorylation of Vav1, total Vav1 and -actin on flow cytometry sorted CD4 T cell subsets isolated 6 days after immunisation with SRBC. Each sample contains pooled cells from 5 mice. Data is representative of 2 experiments.

Our findings suggested that IL-21 acts to drive the differentiation of CD4+ T cells, supporting the phenotypic changes in the expression of molecules that define helper T cells such as ICOS and CCR7. IL-21 costimulates the TCR-induced proliferation of CD4+ T cells (Parrish- Novak, 2000), indicating that IL-21 may mediate its effects on T helper generation by potentiating signals through the TCR.

80 3.3 Discussion

As the Il21-/- strain used here are previously unstudied, we performed a range of phenotyping of lymphoid and myeloid cells in order to confirm that IL-21 does not play an important role in development of the immune system. Our findings agree with previous reports regarding normal lymphogenesis in the Il21r-/- strain (Kasaian, 2002; Ozaki, 2002). Having established that mature Il21-/- and Il21r-/- CD4+ T cells do not exhibit gross defects in development and homeostasis, we went on to characterise the CD4+ T cells that produce IL-21 in a healthy immune system. IL-21 could be detected at low levels within the CD4+ T cell compartment by flow cytometry, but was increased substantially in the peyers’ patches of the small intestine, which may reflect ongoing immune activity towards commensal microbiota in the small intestine. Certainly the IL-21+ cells in the peyers’ patches displayed a discrete phenotype from their counterparts in the spleen and lymph nodes. IL-21 expression in the spleen and lymph nodes correlated with a memory phenotype that was CD44hi, and did not express the recent activation markers CD69 or CD25. Approximately 60% of IL-21 producing cells in these organs retained high expression of CD62L suggesting that they acquire the ability to produce IL-21 while they are retained within lymph nodes (Fazilleau, 2007a). In contrast, CD44hi IL-21 cells in the peyers’ patches were mostly CD62L- and could thus exit from secondary lymphoid organs into the circulation. In addition, more IL-21+ CD4+ T cells were observed in the peyers’ patches than other lymphoid organs that expressed CD69, indicating a recently activated phenotype. IL-21 appears to be produced by a variety of memory phenotype cells. The memory T cell marker CD27 (Hendriks, 2000; Pepper, 2009), for example, was differentially expressed. The expression of the long-lived central memory marker CD27 was more frequent amongst IL-21+ cells in lymph nodes, whereas the peyers’ patches showed more heterogenous CD27 expression, with both short and long lived effector and central memory phenotype cells. IL-21 is therefore present in the resting mouse in two distinct populations, memory phenotype cells retained in the spleen and lymph nodes, and a mobile mix of recently activated and memory cells found in the peyers’ patches, which drain mucosal surfaces. However, relatively few IL-21+ CD4+ T cells also expressed mucosal homing markers such as CCR9 or 47. We found that IL-21+ CD4+ T cells from all sources also commonly express ICOS, which interacts with its ligand (ICOSL) on B cells during immune responses. B cells have been thought to be the predominant recipient of IL-21’s effects, driving class switch and antibody production. This study demonstrates that ligation of ICOS is required for optimal IL-21 production, which is

81 produced in abundance by Tfh cells (Bryant, 2007; Chtanova, 2004). Our findings reveal an IL-21- driven autocrine loop whereby IL-21 can influence T cell modulation of B helper molecules and associated chemokines receptor such as ICOS and CCR7. As CXCR5 is not upregulated under in vitro priming conditions in our hands, the following chapter assesses upregulation of CXCR5 in vivo to determine whether this chemokines receptor, which is critical for T helper function, is also dependent on costimulation with IL-21. Engagement of lymphocyte antigen receptors sets in motion the tyrosine phosphorylation of numerous proteins that are essential for cellular activation and differentiation. Vav1 has been shown to be necessary for T cell help during the GC response (Gulbranson-Judge, 1999) and this finding is supported by diminished IL-4 production and Th2-cell responses in Vav1-/- T cells (Tanaka, 2005). Our data demonstrated that IL-21, alone or in conjunction with CD3 mAb, induced the tyrosine phosphorylation of Vav1. This costimulation intersects with TCR signalling downstream of highly proximal events such as ZAP-70 phosphorylation, but further research is required to pinpoint where the signalling cascades combine. We were able to detect a small amount of Vav1 phosphorylation of both genotypes directly ex-vivo without TCR stimulation, and it is not clear whether this reflects residual in vivo activation of memory cells, or low level signalling in naive cells, both of with were included in the cell preparations. This could be addressed by repeating these western blots on separate populations of purified memory and naïve cells in order to identify its source. These findings are a good indication that IL-21 increases the strength of the TCR signal and stronger TCR signals are thought to favour Tfh cell differentiation (Fazilleau, 2009). Our data propose an intriguing relationship between costimulation of the TCR by IL-21 and activation of Vav1 during the generation of CD4+ helper cells that may intersect via the P13K kinase pathway. Moreover, the functional coupling of Vav1 and IL-21 reveals a molecular mechanism explaining a fundamental role of IL-21 in T helper cell differentiation.

82 4 A fundamental role for IL-21 in the generation of Tfh Cells

4.1 Introduction

Germinal centres (GC) are specialized structures that develop within B cell follicles of secondary lymphoid tissues, such as lymph nodes, spleen, tonsils, and the Peyer’s patches of mucosal-associated lymphoid tissues. The GC is the principle site for processes such as somatic hypermutation, class switch recombination, and selection of high-affinity B cells (Kelsoe, 1995; Liu, 1996; MacLennan, 1994). GC CD4+ T cells provide direct help to antigen-specific naive B cells, which promotes the differentiation of antigen-selected high-affinity GC B cells into memory B cells or plasma cells, thereby ensuring long-term humoral immunity (Banchereau, 1994; Garside, 1998; Odendahl, 2005). Live imaging of the GC reveals that T-B cell interactions in the GC are weighted toward B cell competition for a small number of CD4+ T cells (Allen, 2007b; Hauser, 2007; Schwickert, 2007). Linked recognition of antigen and B cell competition for T cell help provides a stringent control that ensures self-tolerance during the production of high affinity antibodies of varying specificities (Mitchison, 2004). Despite the long held appreciation of a fundamental role for CD4+ T cells in the GC reaction, it is only recently that components of the cellular and molecular mechanisms for T cell help have emerged. T cell help for B cells has been considered the domain of Th2 and Th1 cells, which direct B cells to undergo Ig isotype switching to IgE and IgG2a, respectively. More recently, a subset of CD4+ T helper cells termed Tfh were observed in the GC and displayed potent help for antibody responses (Breitfeld, 2000; Schaerli, 2000; Walker, 1999). At least six different follicular T cell subsets have been identified, on the basis of surface markers and/or anatomical location (Vinuesa, 2005b) including Tfh memory cells that reside in the lymph node and regulate the memory B cell response for long-term immunity (Fazilleau, 2007a). Expression of the chemokine receptor CXCR5 is transiently upregulated when T cells interact with peptide-MHC Class II and costimulatory molecules on antigen-presenting cells and continued CXCR5 expression following priming may reflect qualitative or quantitative aspects of this stimulation (Ansel, 1999; Sallusto, 1999). Characteristically, Tfh cells retain intense expression of CXCR5, which directs these cells towards CXCL13-rich areas within GC (Breitfeld, 2000; Schaerli, 2000; Walker, 1999). High amounts of costimulatory molecules, such as ICOS and CD40L on Tfh cells reflect both the sustained multisignal pathways necessary for their generation

83 as well as their function to provide cognate help to B cells (Bossaller, 2006; Coyle, 2000; Dong, 2001; Hutloff, 1999; Tafuri, 2001). Tfh cells also express a number of costimulatory molecules that have the capacity to restrain their interaction with B cells and APCs, including CTLA-4 and PD-1, which may reflect their discriminating role in the GC as a limiting factor enhancing antibody output (Chtanova, 2004; Haynes, 2007). B helper cell function is not limited to cell surface molecules and Tfh cells also express a number of cytokines that facilitate antibody production, including IL-4 and IL-10 (Haynes, 2007). More recent studies have revealed that elevated expression of IL-21 and its receptor (IL-21R) distinguish Tfh cells from other Th subsets (Chtanova, 2004; Rasheed, 2006). The almost exclusive production of this cytokine by CD4+ T cells makes it a candidate for mediating T helper function during antibody production. Analyses of the behaviour of immune cells in response to IL-21 in vitro and studies of mice deficient in IL-21 or its receptor support this notion, and predominant actions on B cells in models of transgenic overexpression also point to a role in B cell help (Spolski, 2008a). In the previous chapter, IL-21 was shown to synergise with TCR derived signals in order to upregulate T helper molecules in vitro such as ICOS, CTLA-4 and CCR7, suggesting a novel autocrine role for IL-21 in acquisition of helper phenotypes. The research outlined in this chapter investigates the requirement for IL-21 during GC formation in response to T dependent antigen. It goes on to dissect IL-21 actions on B and T cells during the immune response.

84 4.2 Results

4.2.1 IL-21 is necessary for GC formation in response to T dependent antigen

IL-21 has previously been shown to be important for the generation of IgG1 antibody responses to T dependent antigen (Ozaki, 2002). However, it is not known whether IL-21 imparts its effects on B cells or T cells to generate these responses. Using mice made genetically deficient in IL-21 (Il21-/-), we analysed the generation of GC response following immunization with SRBC, which provides a strong antigen stimulus for T dependent IgG1 antibody production in the B cell follicle. The GC reaction can be detected in sections of lymphoid tissue by staining with the lectin PNA, which stains certain non-sialylated glycans on the surface of activated lymphocytes. Il21-/- mice exhibited a marked reduction in PNA+ GC compared with WT mice 7 days after immunization with SRBC (Figure 4-1A). The fraction of PNA+ B220+ B cells was significantly reduced in the absence of IL-21 (Figure 4-1B and C), and very little Brdu incorporation could be observed in Il21-/- B cells, particularly among PNA+ cells, indicating a reduced proliferative capacity (Figure 4-1D).

85 Figure 4-1 T dependent GC formation is reduced in Il21-/- mice (A) Representative light microscopy of haematoxylin stained spleen sections, showing PNA+ (brown) germinal centres on day 7 after SRBC immunisation. (B) Representative dot plot from flow cytometric analysis of IgD- PNA+ germinal centre B cells, gated on total B220+ cells. (C) PNA+ germinal centre B cells as a percent of total B220+ cells. Data is shown as values from individual mice +/- SEM. (D) Representative dot plot of flow cytometric analysis of BrdU+ B cells in the spleen. Data are representative or combined individual values from 2 experiments in which n=5-8.

In accordance with this reduced germinal centre compartment, SRBC immunisaed IL-21 deficient mice exhibited a marked defect in IgG1 production (Figure 4-2), supporting previous observations in Il21r-/- mice (Ozaki, 2002).

86 Figure 4-2 Reduced serum IgG1 production by Il21-/- mice in response to T dependent antigen (A) IgG1+ germinal centre B cells as a percent of total B220+ cells on day 7 after SRBC immunisation. (B) Quantification of serum IgG1 concentrations measured by ELISA. Data is shown as values from individual mice +/- SEM from a representative experiment of 3 performed.

4.2.2 IL21 is an autocrine growth factor for Tfh

As mentioned earlier, Tfh cells express the costimulatory molecule ICOS and the chemokine receptor CXCR5, which correlates with their ability to travel to the B cell follicle and provide cognate help to B cells, respectively (Rasheed, 2006). Microarray data obtained from Tfh cell populations using these markers found that human Tfh cells also express IL-21 and IL-21R (Chtanova, 2004), and we speculated that this co-expression might reflect an autocrine loop. We analysed the contribution of IL-21 to Tfh cell differentiation in Il21-/- mice following immunization with T dependent antigen. Analyses of the phenotype of GC CD4+ T cells 7 days after immunization with SRBC revealed a defect in the generation of Tfh cells, defined as CXCR5+ ICOS+ CD4+ T cells, in the absence of IL-21 (Figure 4-3 and Figure 4-4). Il21-/- CD4+ T cells also showed reduced levels of another Tfh cell marker indicative of recent antigen exposure, PD-1 (data not shown). Il21+/- heterozygous mice gave more varied results, suggesting that the dose of IL-21 could be a factor in Tfh cell generation (Figure 4-4). Interestingly, despite an earlier observation of a defect in vitro, immunization caused ICOS expression to increase on CXCR5- CD4+ T cells in the presence or absence of IL-21 suggesting redundancy with other factors in vivo. By contrast, maximal expression of CXCR5 was not achieved on CD4+ T cells in the absence of IL-21 (Figure 4-4). Therefore, IL- 21 was not necessary for ICOS expression per se, but was necessary for the maximal expression of CXCR5, a key marker of Tfh cells.

87 Figure 4-3 IL-21 deficiency effects Tfh cell generation following T dependent immunisation. Representative dot plots showing flow cytometric analysis of Tfh cells defined as ICOS+ CXCR5+ gated on total CD3+CD4+ T cells before and 7 days after SRBC immunization. Representative of 4 separate experiments in which n=3-4.

Figure 4-4 Reduced generation of Tfh cells from Il21-/- CD4+ T cells. (A) Tfh cells shown as a percent of total CD4+ T cells 6 days after immunisation with SRBC. (B) CXCR5 expression on Tfh cells as above, shown as mfi. Data are shown as mean +/- SEM including values from individual mice from (A) 2 pooled or (B) a representative experiments in which n=3-7.

4.2.3 The Il21-/- GC defect is present early in the SRBC response

CD4+ T cells are necessary in the GC for the production of high affinity antibody responses due to their high expression of B cell co-stimulatory molecules (Lipsky, 1997; Monson, 2001; Qi, 2008). GC reactions can be induced in a T dependent manner in some extraordinary situations by strong cross-linking the BCR of a large number of transgenic (Tg) antigen specific B cell precursors by polysaccharide based antigens such as NP-Ficoll. These GC are unable to progress past a massive apoptosis event between day 4 and 5 at the point where centrocyte selection is normally 88 carried out by CD4+ T cells (de Vinuesa, 2000; Toellner, 1998). We decided to examine the kinetics of SRBC immunisation by measuring GC B cells, and Tfh cells at day 4, 6 and 10 after immunisation and found that the defective formation of these GC cell types in Il21-/- mice could be observed at all timepoints examined, but was most apparent at later timepoints due to the progression of the WT response (Figure 4-5). The small numbers of GC B cells and memory B cells in Il21-/- mice changed very little after day four, suggesting that an absence of CD4+ T cell help in the follicle was arresting progression of the GC reaction at this stage. As mentioned earlier, CXCR5 is upregulated on T cells following Ag stimulation but sustained expression is unique to the Tfh subset. Il21-/- CD4+ T cells expressed similar levels of CXCR5 early in the response at day 4 (Figure 4-5 E and F) but the defect in Tfh cell numbers became clear at day 6 when WT T cells had developed into a distinct ICOS+CXCR5 high subset.

Figure 4-5 Timecourse of GC defect Il21-/- mice Quantification of splenic GL7+CD38-CD19+ GC B cells and CD38+GL7+CD19+ memory B cells shown as a percent of total B cells (A and B) or total numbers (C and D) at time points indicated after SRBC immunisation. CXCR5+ICOS+PD-1+ Tfh cells represented as (E) a percent of total CD3+CD4+ T cells or (F) total cell numbers. T test p value calculations on day 6 below 0.01 are represented by asterisks. Data are representative of the mean +/- SEM of individual mice from two combined experiments in which n=6-7.

89 4.2.4 Tfh cells express the receptor for IL-21

The dependence of Tfh cells on IL-21 suggested that they utilize IL-21 in an autocrine manner for their growth and/or survival, confirming earlier studies showing that high mRNA transcript levels of both cytokine and receptor distinguishes Tfh cells from other human T helper subsets (Chtanova, 2004). In order to confirm protein expression, we measured the expression of the IL-21R by flow cytometry using biotinylated ligand on CD4+ T cells following immunization and observed prominent expression of IL-21R on CXCR5+ Tfh cells but not on either CXCR5- memory (CD44hi) CD4+ T cells or naïve (CD44lo) CD4+ T cells. We also measured IL-21 production by Tfh cells by staining for intracellular IL-21 using Il21-/- CD4+ T cells as a negative control. We found that IL-21 was present at the highest levels in the CXCR5+ ICOS+ Tfh cell population compared to other activated and resting CD4 T cells directly ex-vivo after immunisation (Figure 4-6).

Figure 4-6 Both IL-21 and IL21R are expressed by Tfh cells (A) Histogram overlay of IL-21R expression on representative CD44lo naïve CD4+ T cells, CXCR5- CD44hi memory CD4+ T cells and CXCR5+ CD44hi Tfh cell subsets. WT spleen at day 7 after immunisation with SRBC. (B) Histogram overlay of intracellular IL-21 expression by ICOS+CXCR+ Tfh cells, ICOS+CXCR5+ T helper cells and resting CD44lo ICOS-CXCR5- CD4+ T cells after immunisation as above. Il21-/- CD4+ T cells in the filled histogram show non-specific background staining. Data are representative of combined individual values from 3 experiments in which n=6-9.

4.2.5 Tfh have a specific requirement for co-stimulation through IL-21R in vivo

A specific requirement for IL-21 in CD4+ T cell activation or survival was evident when we performed comparative analyses of the proliferation of Il21-/- and WT CD4+ T cells in response to CD3 and CD28 mAbs in vitro as described earlier. We next determined whether increasing signals through the TCR might restore the defect in Tfh cell generation in Il21-/- mice in vivo. To this end, we administered a single dose of CD28 mAb to Il21-/- and WT mice one day prior to

90 immunization with SRBC and monitored Tfh cell generation by FACs. CD25hi CD4+ Treg cell numbers in the spleen and lymph nodes increased in response to this strong artificial co-stimulation, confirming that IL-21 deficient CD4+ T cells were responsive to signals delivered through CD28 (Figure 4-7). However, while the Tregs were able to expand, the anti-CD28 antibody potentiated the generation of Tfh cells in WT mice but had little or no effect on Tfh cells in the Il21-/- mice (Figure 4-8) proving that Tfh cells were uniquely dependent upon IL-21.

Figure 4-7 Agonistic anti-CD28 treatment boosts Treg populations Flow cytometry showing Foxp3+CD25+ Treg cells as a percent of gated CD4+ T cells after treatment with 100μg anti-CD28 or isotype control prior to immunisation with SRBC. Spleens were analysed at day 7. Representative plots from 3 experiments in which n=4-8.

Representative dot plots of CD4+ T cells showing the percent of Tfh cells in immunized WT and Il21-/- mice after administration of 100μg anti-CD28 or isotype control prior to immunisation with SRBC. Spleens were analysed at day 7. Representative plots from 3 experiments in which n=4-8.

Figure 4-8 Agonistic anti-CD28 treatment cannot recover the Il21-/- Tfh cell defect

91 4.2.6 Tfh cells are distinct from Th17 cells

The newly described IL-17-producing Th17 subset has been reported to require IL-21 for their generation (Korn, 2007a; Nurieva, 2007; Wei, 2007; Zhou, 2007). As mentioned earlier, Tfh cells can be distinguished from other T helper cell subsets, including Th17 cells, by their expression of CXCR5 (Acosta-Rodriguez, 2007) yet they share a number of attributes including high expression of IL-21, ICOS, CTLA-4 and Fas, as well as provision of help to B cells for antibody production (Annunziato, 2007). Furthermore, an expanded population of IL-17-producing Th17 cells has recently been demonstrated to drive spontaneous germinal centre formation in the BXD2 strain of mice that develop erosive arthritis (Hsu, 2008).

Figure 4-9 IL-21 producing Tfh cells are distinct from Th17 cells (A) Representative dot plot showing intracellular immunostaining for IL-17 in CXCR5-CD4+ T cells on day 7 of immunization with SRBC. (B) Representative dot plots showing IL-17 in CXCR5- CD4+ T cells from WT host and OTII donor cells on day 8 of immunization with NP-OVA. These data are representative of 2 experiments (n=4-9).

To directly address whether the Tfh cells that developed following T dependent immunisation were phenotypically distinct from Th17 cells we performed intracellular immunostaining for IL-17, which revealed that the IL-21+ CXCR5+ Tfh cells generated by immunization of WT mice with SRBC (Figure 4-9A) and the CXCR5+ OTII cells observed in an alternate protocol where NP-OVA immunized hosts were given 30 000 adoptively transferred OVA specific Tg CD4 T cells (Figure 4-9B) did not co-express IL-17. Taken together, these data demonstrate that Tfh cells and IL-17-producing Th17 cells were distinct subsets in our models.

4.2.7 The GC defect in IL-21 deficient mice is CD4+ T cell intrinsic

The IL-21R is expressed intensely on both Tfh cells and B cells and our data reveals a central role for IL-21 in the generation of Tfh cells and T dependent antibody production. Previous compelling data support a role for IL-21 in B cell proliferation and antibody production from B

92 cells (Ettinger, 2005; Good, 2006; Jin and Malek, 2006; Ozaki, 2004; Ozaki, 2002). However, it remained unclear whether the lack of responsiveness to IL-21 by T cells or B cells, or both subsets, explained the defect in GC formation and antibody production observed.

Figure 4-10 Donor WT CD4+ T cells proliferate and assume a Tfh phenotype (A) Representative dot plots showing donor WT Ly5.1+ CD4+ T cells as a percent of the total lymphocyte gate 7 days after immunization of host Il21r-/- mice with SRBC. (B) Histogram overlay showing IL21R expression on CD4+ T cells from host Il21r-/- (filled), WT (unbroken line), and donor WT Ly5.1+ (dashed line) 7 days after SRBC immunization. (C) Representative dot plot showing flow cytometric analysis of Tfh cells, gated on total CD4+ T cells from WT, host Il21r-/- and donor WT Ly5.1+. Data are representative of 3 experiments (n=4-9).

To determine the relative role of T cell responsiveness to IL-21 in GC formation and IgG1 production we transferred 2 x107 IL-21R sufficient WT Ly 5.1 CD4+ T cells (Figure 4-10) into Ly 5.2 IL-21R-deficient recipients and immunized the recipient mice with SRBC. The transferred CD4+ T cells (Figure 4-10) were found to fully differentiate into Tfh cells with high expression of IL-21R, CXCR5 and ICOS, which was presumably fuelled by the excess available IL-21 in the Il21r-/- mice (Figure 4-10). However, the transfer ofWT CD4+ T cells into unimmunised Il21r-/- mice did not produce Tfh cells (data not shown), indicating that the effects of IL-21 required antigen for Tfh cell generation. The transferred IL-21 responsive WT CD4+ T cells were particularly effective since just 10% of this splenic population could precipitate B cell activation and recover approximately 40% of the WT levels of IgG1 and GC B cells (Figure 4-11A and B).

93 However, whilst the amount of IgG1 in the sera of recipient mice was double that of Il21r-/- sera, it remained lower than WT (Figure 4-11C). By contrast, the transfer of 2 x 107 IL-21R- sufficient WT B cells had no effect on IgG1 production. Similarly, the transfer of both CD4+ T and B cells increased cell numbers in the host, but did not increase IgG1 levels above that observed with CD4+ T cells alone (Figure 4-11C). Histological evaluation confirmed these findings by demonstrating PNA+ cells and a small number of Ly5.1 CD4+ T cells entering the B cell follicles of Il21r-/- hosts (Figure 4-12A and B).

Figure 4-11 IL-21 responsive T cells boost GC B cells in Il21r-/- mice (A) Representative dot plot showing flow cytometric analysis of IgD- PNA+ germinal centre B cells gated on total B220+ cells on day 7 of SRBC immunization. (B) Total IgG1+ IgD- PNA+ germinal centre B cell numbers in spleen after SRBC immunization. C) Serum IgG1 levels measured by ELISA on day 7 of immunization with SRBC. Data are shown as the mean +/- SEM. Data are representative of 3 experiments (n=4-9).

94 Figure 4-12 GC structures are restored to Il21r-/- mice by transfer WT CD4+ T cells Representative confocal images of spleen sections on day 7 of SRBC immunization showing (A) PNA expression (red) within IgD+ (green) B cell follicles and B) GL7+ (green) germinal centres with infiltrating donor Ly5.1+ (red) CD4+ T cells. Data are representative of 3 experiments (n=4-9).

These data demonstrated that the defect in GC formation and IgG1 production was CD4+ T cell intrinsic but said little about the antigen specificity of the Tfh cells. We used CD4+ T cells purified at day 7 from primed WT mice to SRBC, which afforded a small improvement in the amount of IgG1 produced (Figure 4-11B and C)). It was not clear whether the transferred B cells were efficiently trafficking to the spleen or whether they were as resilient as CD4+ T cell to the transfer process as they were infrequently detected after transfer. These approaches were also limited by the inability to distinguish antigen specific T cells from the polyclonal cohort.

95 4.2.8 GC B cells require IL-21 acting on T cells

In order to further understand the influence of IL-21 on T and B cells during the immune response, we created bone marrow (BM) chimeras. We combined Il21r-/- BM in equal proportions with μMT-/- BM, a mutation that delivers a defect early in B cell development and so can thus only reconstitute the T cell compartment of these chimeras with WT cells. 12 weeks later these mice were immunised with 100μg NP-OVA in alum, and the relative contribution of both WT and Il21r- /- BM to the immune response was analysed.

The NP13OVA immunisation system has several advantages for dissecting the unique contributions of the T and B cell compartments to GC reactions. The nitrophenyl (NP) hapten can cross-link the BCR of a large cohort of naïve B cells in B6 mice, whereupon covalently conjugated OVA protein is processed and several key epitopes presented to T cells in the context of MHC class II. It is also possible to measure NP specific antibody produced by coating ELISA plates with the immunogen (Griffiths, 1984; Jacob, 1993). In order to discover whether IL-21 lends a competitive advantage to lymphocytes during GC reactions we made BM chimeras consisting of a 50:50 ratio of WT and Il21r-/- lymphocytes. Irradiated B6 mice were reconstituted with 1x107 BM cells from mice expressing the congenic marker Ly5.1 to allow detection of WT cells, combined with equal proportions of BM from Ly5.2+ Il21r-/- mice. Our reconstitution protocol resulted in an approximate ratio of 45%:55% WT:Il21r-/- lymphocytes, measured by flow cytometry in the blood directly prior to immunisation (Figure 4-13).

The percentage of Ly5.1+ WT or Il21r-/- Ly5.2 among total lymphocytes in C57BL/6 recipients measured by flow cytometry in blood prior to immunisation. The values represent individual mice pooled from 2 experiments (n=14), the mean and the standard deviation are indicated above each group.

Figure 4-13 Mixed chimerism of lymphocytes in irradiated recipients. Despite an approximately 10% contamination of radio-resistant T cells remaining from the endogenous WT compartment of the B6 hosts, reconstitution with Il21r-/- bone marrow alone recapitulated earlier observed defects in both Tfh cell and GC B cell numbers compared to recipients of WT BM cells. These experiments were also repeated once in RAG-/- recipients where no host lymphocytes are present, with similar results proving donor cells were responsible rather

96 than radio resistant cells (data not shown). We saw an approximate doubling in the generation of Il21r-/- IgG1 expressing GC B cells when we introduced a 50% WT T cell compartment that was derived from μMT-/- and WT donors, confirming that IL-21 acting on T cells alone could drive the downstream GC B cell response to a T dependent antigen (Figure 4-14A). In contrast to this improvement in the GC phenotype with the addition of WT T cells, there was no change in the absence of plasma cells, indicating a strong requirement for IL-21 for commitment to this lineage (Figure 4-14B). This analysis was carried out at day 10 following immunisation which is after the collapse of early extra-follicular responses and therefore reflects a post GC plasma cell defect.

Figure 4-14 WT CD4+ T cells can restore generation of GC but not plasma cells from Il21r-/- B cells Bone marrow chimeras 12 weeks after reconstitution were immunised with NP-OVA and spleens analysed at day 10 for generation of (A) IgG1+ GL7+fas+ GC B cells or (B) B220lo syndecam-1+ plasma cells. Data is representative of 2 similar experiments, and shows the mean +/- SEM in which n=6-8 per group for each experiment.

Correct placement of recovered Il21r-/- GC B cells was confirmed by histological analysis, where it was possible to identify GC as IgD low regions that stained brightly for IgG1 in both the WT and Il21r-/-:μMT-/- groups but were absent when neither T nor B cells could respond to IL-21 (Figure 4-15). Again, one notable difference between these groups was the absence of brightly IgG1+ plasma cells in the bridging regions outside the follicle in Il21r-/-:μMT-/- mice (Figure 4-15), confirming an absolute requirement for IL-21 in the development of plasma cells in response to T dependent antigen and supporting earlier studies (Ettinger, 2005). A small number of IgG1 expressing B cells in Il21r-/- chimera could be detected though they were found sporadically throughout the B cell follicle rather than in IgD low GC regions (Figure 4-15).

97 Figure 4-15 WT CD4+ T Cells promote GC in Il21r-/- mice Bone marrow chimeras from WT, Il21r-/- and mixed Il21r-/- and μMT-/- donors 12 weeks after reconstitution were immunised with NP-OVA and spleens were analysed on day 10. Representative images from 3 transverse histological sections of 3 spleens from each group. Data is representative of 2 similar experiments, in which n=6-8 per group for each experiment.

To test the functional output of these GC, affinity maturation of antigen specific antibody was assessed by measuring the amount of IgG1 produced that could bind to NP hapten when coupled at very high (NP23) or low (NP3) levels with BSA. Only very high affinity antibodies are capable of binding robustly to NP3BSA, which can be detected using ELISA. Using this method we found that μMT-/- derived WT T cells could increase the output of high affinity antibody by Il21r-/- B cells, confirming that these GC B cells were capable of hyper-somatic mutation and high affinity clones had been selected in GC by the restored Tfh cell compartment (Figure 4-16). However, the amount of high affinity antibody was not completely restored to the level of the WT chimeras, which may be explained by an absolute requirement for IL-21 in the commitment to the plasma cell lineage after selection of high affinity clones in the GC. It is an intrinsic flaw in the BM chimera methodology used here that it is impossible to identify whether this partial recovery reflected a dose response from the number IL-21 responsive T cells (100% in WT, 50% in Il21r-/-:μMT-/- or mixed chimeras and lastly, 0% in Il21r-/-) or a partial requirement for IL-21 acting on B cells.

98 Figure 4-16 IL-21 acting on CD4+ T cells increases high and low affinity antibody titres

Quantification of serum titres of low (NP23BSA binding) and high affinity (NP3BSA binding) NP specific antibody, measured by ELISA at day 10 after immunisation with NP-OVA. Data is representative of 2 similar experiments, and shows the mean +/- SEM in which n=7-8 per group for each experiment. *p < 0.05.

4.2.9 Dose of IL-21 responsive CD4+ T cells correlates with GC formation

The Tfh cell phenotypes of WT or Il21r-/- reconstituted BM chimeras reflected the previously observed endogenous defects, with significantly reduced levels of CXCR5, ICOS and PD-1 expressed by mice reconstituted with Il21r-/- CD4+ T cells (Figure 4-17A and Figure 4-18A). There was an improvement in Tfh cell marker expression on the entire combined CD4+ T cell cohort in both groups of mixed genotype chimeras, due to the addition of WT T cells which can successfully generate a Tfh cell phenotype in response to antigen and IL-21, and possibly the improved GC B cell phenotype present in these groups (Figure 4-17). This increase in Tfh cell phenotype correlated well with the increase in GC B cells and antibody production described earlier (Figure 4-14 and Figure 4-16).

99 Figure 4-17 IL-21 responsive CD4+ T cells display increased Tfh cell markers Bone marrow chimeras from WT, Il21r-/- and mixed Il21r-/- and μMT-/- donors were immunised 12 weeks after reconstitution and spleens were analysed at day 10. (A) Quantification of Tfh cells as a percentage of total splenic CD4+ T cells and as a absolute numbers. (B) Mean fluorescence intensity (mfi) of Tfh cell markers CXCR5, ICOS and PD-1 on ICOS+CXCR5+CD4+ T cells. Data is representative of 2 similar experiments, and shows the mean +/- SEM in which n=7-8 per group for each experiment.

4.2.10 Il21r-/- T and B cells compete poorly with their WT counterparts

As described earlier, BM chimera from either WT or Il21r-/- donors resulted in normal or defective GC and Tfh cell responses respectively. Interestingly, when WT and Il21r-/- lymphocytes were combined in a mixed BM chimera, CD4+ T cells derived from either genotype seem to generate a similar, low proportion of ICOS+ CXCR5+ Tfh cells, though Il21r-/- CD4 +T cells generated a significantly lower proportion of Tfh cells when paired within each host (Figure 4-18A and B). This group displayed an interesting half-way phenotype between WT and abnormal Il21r-/- responses, perhaps partly affected by a limiting number of ovalbumin specific T cell precursors in the mixed chimeras (Figure 4-18C). The significant increase in the generation of ICOS+CXCR5+ Tfh cells from Il21r-/- BM alone or combined with WT BM in a mixed chimera was unexpected. This finding implies that either other soluble factors derived from the successful generation of GC in these mice, or an increased ability of WT GC B cells to “pull” T cells into the GC, as has been observed in a similar system by live imaging using 2-photon microscopy (Allen, 2007b), may

100 compensate for the absence of IL-21:IL-21R signalling to a small extent in this model. These experiments were also carried out in rag-/- hosts with the same response, ruling out a role for IL-21 acting on radio-resistant cells in creating the above phenotype (data not shown).

Figure 4-18 IL-21 is required for Tfh cell formation Bone marrow chimeras from WT, Il21r-/- and mixed Il21r-/- and μMT-/- donors were immunised 12 weeks after reconstitution and spleens were analysed at day 10. (A) Representative dot plots showing Tfh cells as a percentage of CD4 T cells from BM chimeras of Ly5.1+WT or Ly5.2+Il21r- /- alone, or combined in one B6 irradiated host. (B) Paired values within individual hosts showing either WT or Il21r-/- Tfh cells as a percentage of CD4+ T cells of matched genotype. (C) CXCR5+ICOS+ Tfh cells as percentage of CD4+ T cells from BM chimeras as in (A). Data is representative of 2 similar experiments in which n=7-8 per group, and shows the mean +/- SEM.

GC and plasma B cells were present, but those participating in the response were overwhelming derived from Ly5.1 expressing WT BM (Figure 4-19A) despite their slightly lower (45%) percentage among lymphocytes prior to immunisation, indicating a strong selective advantage for IL-21 responsive B cells in the GC reaction. The robust GC population in Il21r-/- :μMT-/- chimeras ruled out an absolute requirement for this cytokine by B cells participating in T dependent immune responses.

101 Recent studies by Reinhardt et al in 2008 showed that flow cytometry could be used to detect stable interactions of T and B cells, named “T-B conjugates”. These conjugates were shown to be the site of direct delivery of T cell help to B cells as they contained increased class switched, somatically mutated B cells at later stages of the immune response (Reinhardt, 2009). In our model, very few CD4+ B220+ T-B cell conjugates contained Il21r-/- Ly5.2+ cells, confirming an overwhelming dominance of IL-21 responsive T cells delivering B cell help in mixed BM chimeras (Figure 4-19B). It was also interesting to note the very high expression of PD-1 on T cells bound strongly to B cells of both genotypes, reflecting the strong antigen stimulus these T cells receive during these interactions. The small numbers of Il21r-/- T cells found in these conjugates implied that IL-21 may deliver a co-stimulus that counters co-inhibitory signals delivered by PD-1 (Figure 4-19C)

Figure 4-19 Il21r-/- T and B cells compete poorly with WT cells in BM chimeras Quantification of the proportion of WT Ly5.1+ or Il21r-/- Ly5.2+ derived (A) fas+ GL7+ GC cells or Syndecam-1+ B220lo plasma cells from mixed BM chimera. (B) The proportion of cells present in CD4+ B220+ T-B conjugates derived from each genotype in mixed BM chimera. (C) Representative dot plot and histogram showing the total CD4+ T cell gate in a mixed chimera with Ly5.1+WT (unbroken line) or Ly5.2+ Il21r-/- (dashed line) CD4+ T cells and the corresponding PD-1 expression on these conjugates. Data is representative of 2 similar experiments, and shows the mean +/- SEM in which n=7-8.

102 We also assessed histological sections for participation of IL-21 responsive and ignorant T cells in the GC response. In the mixed chimeras it was possible to detect both green Il21r-/- and yellow Ly5.1 WT in the GC, which was itself very strongly stained for Ly5.1 WT B cells, reflecting the competitive advantage derived from IL-21 (Figure 4-20).

Figure 4-20 WT and Il21r-/- CD4+ T cells migrate to the B cell follicle in mixed chimeras Lymphocytes from mixed Ly5.2+ Il21r-/- and Ly5.1+ WT chimeras were analysed for positioning in the spleen at day 10 following NP-OVA immunisation. Shown are GC (GC), B cell follicles (F) and T cell zones (T). Data are representative of 2 similar experiments, and shown are representative images from 3 transverse histological sections of 2 spleens from each group of 7-8 mice.

Ovalbumin that has been processed by antigen presenting cells presents three discrete peptide sequences via MHC class II to CD4+ T cells. The number of naïve ovalbumin specific precursors in B6 mice is low, for example approximately 20 CD4+ cells are responsive to the OVA 323-339 peptide bound to MHC are present in an individual naïve mouse (Moon, 2007). BM chimeras are derived from only 1x107 BM cells, which may confound the comparison between single genotype and mixed chimeras containing 50% of WT T cells, which may not provide a sufficient number of precursors. We therefore decided to use adoptive transfer of a fixed number of antigen specific CD4 precursors to examine whether T cell precursor frequency in mixed BM chimeras could be limiting the development of a normal Tfh cell population.

103 4.2.11 Il21r-/- mice have poor GC responses to the T-dependent

haptenated protein NP13OVA

The NP13OVA immunisation system has several advantages for dissecting the unique contributions of the T and B cell compartments to GC reactions. The nitrophenyl (NP) hapten can cross-link the BCR of a large cohort of naïve B cells in B6 mice, whereupon covalently conjugated OVA protein is processed and several key epitopes presented to T cells in the context of MHC class II (Griffiths, 1984; Jacob, 1993). Adoptively transferred OTII transgenic CD4+ T cells can then recognise OVA peptides and provide cognate help to NP specific B cells. While, endogenous CD4+ T cells could also provide this help, the response is largely dominated by transgenic T cells if they are present. By introducing a small number of TCR transgenic OTII T cells with or without the IL- 21R at the time of immunisation, we hoped to be able to dissect the importance of IL-21 for the provision of antigen specific T cell help in the GC. One drawback in the use of BM chimeras for dissecting the importance of the IL-21R expressed on T and B cells is that we had difficulty distinguishing whether the observed effects reflected the varying dose of IL-21 responsive T cells or the requirement for IL-21 by B cells. The dominance of the antigen specific WT or Il21r-/- OTII T cells in this model and the use of a homogenous host population of Il21r-/- B cells, should normalise the number of CD4+ T cell precursors responding to NP-OVA and make it easier to detect how and when IL-21 is required by these cells.

Figure 4-21 Time course of GC B cell generation after NP OVA immunization.

Quantitation of the number and proportion of CD38lo fas+ germinal center B cells at

various timepoints after immunisation with NP13OVA. One Il21r-/- group received 30 000 WT OTII cells at the time of immunisation. Representative of 2 experiments in which n=2-4, day 4 which represents values from only 1 experiment where n=3. *A difference of p<0.05 from the WT group values calculated by Student’s T test. No significant difference was found between the WT and Il21r-/- + OTII groups at any timepoint.

104 μ We began by immunising WT and IL21r-/- mice with 100 g of NP13OVA absorbed to 100μl of alum and found that the frequency and proportion of GC B cells in Il21r-/- mice was reduced, significantly at day 7 which seemed to be the peak of the primary response (Figure 4-21A and B). The difference in GC size was also apparent by histological staining of PNA, similar to the SRBC model (Figure 4-22A, C and D). We subsequently adoptively transferred 3x104 antigen specific WT OTII T cells into both groups at the time of immunisation, to see if this could rescue the GC defect in Il21r-/- mice more effectively than we had seen previously by transferring a large polyclonal CD4+ T cell population during SRBC immunisation. Indeed, this small number of IL-21 responsive OTII T cells was able to restore GC formation by endogenous Il21r-/- B cells to WT frequency, as assessed by flow cytometry enumeration (Figure 4-21) and histological measurement of PNA expression (Figure 4-22A, C and D). This finding was presumably due to increased antigen specific precursor frequency in the transgenic donor population. Similar to the SRBC model, IL- 21’s actions on CD4+ T cells were critical for the generation of a GC response of appropriate magnitude to the T dependent antigen NP13OVA.

Figure 4-22 The T dependent GC response to NP13OVA is reduced in Il21r-/- mice 3x104 WT OTII CD4 T cells were transferred i.v into recipient WT and Il21r-/- mice at the time of i.p μ immunisation with 100 g NP13OVA. Spleens were subsequently examined by histology on day 7. In four consecutive 5μm sections, StAv-Cy3 was used to detect biotinylated PNA (A, C and D) or anti- CD35 (B-E) to assess the GC and follicular dendritic cell networks of WT and Il21r-/- splenocytes respectively. Graphs represent the mean number of PNA+ GC foci (C) or relative area of PNA+ or CD35+ staining (D and E, respectively) +/- SEM of >6 fields per section at 10x magnification using the ImageJ image analysis software. *p value of <0.05 using ANOVA with Dunnett’s post test to compare individual groups. These data are representative of 2 experiments (n=4-9).

105 FDC are important for displaying antigen to B cells in the follicle, forming the basis for early B cell proliferation and GC formation after antigen exposure. Previous reports have found some effects of IL-21 on myeloid derived cells and therefore to rule out a defect in FDC that could be downstream of the observed poor GC formation we also analysed the amount of CD35+ staining in consecutive splenic sections (Strengell, 2006). We found no differences in these structures between WT and Il21r-/- mice (Figure 4-22B and E), implying that the defect observed in Il21r-/- mice lies further down the path of B cell activation where T cell selection of centrocytes becomes important for GC B cell expansion and survival.

4.2.12 Il21r-/- OTII T cells restore GC B cells in response to NP- OVA but fail to reconstitute recall responses.

In order to discover what aspects of T cell help are reliant on IL-21:IL21R interactions, TCR transgenic Il21r-/- OTII T cells bearing the Thy1.1 congenic marker were transferred at the time of immunisation as described above. Unexpectedly, both WT and Il21r-/- OTII T cells were capable of restoring a healthy GC B cell reaction to Il21r-/- recipients (Figure 4-23). GC B cells arising in Il21r-/- recipients of Il21r-/- OTII T cells were phenotypically indistinguishable from those produced by transfer of their WT counterparts, but IL-21 deficiency seemed not to support memory formation, as a recall response showed no re-entry into the GC pathway at day 5 after a booster immunisation (Figure 4-23). Thus, IL-21’s action on CD4+ T cells was important for generated B cell memory, which may be one of the important roles for Tfh cells in the B cell follicle.

106 Figure 4-23 Il21r-/- GC B cells responses can be restored by donor OTII T cells 3x104 WT or IL21r-/- OTII T cells were delivered i.v at the time of i.p immunisation with 100μg

NP13OVA absorbed to alum. Spleens were analysed at day 7, or at day 26, five days after secondary immunisation. (A) Representative dot plots showing CD38lofas+ GC B cell formation gated on total CD19+ B cells. (B) Quantification of the percent and total numbers of GC B cells in both the primary and secondary responses described above. P values were calculated by one-way ANOVA compared to WT values *p<0.05, **p<0.01, ***p<0.001. Data is representative of 3 experiments in which n=2-5 per group for each experiment.

It was not clear from our flow cytometry analysis whether the GC B cells restored by Il21r- /- OTII T cells in Il21r-/- hosts were capable of producing class switched IgG1 antibody in response to NP-OVA. Splenic histology sections collected at day 7 after immunisation with NP-OVA showed that Il21r-/- mice contained very few IgG1+ B cells, and also very few FITC+ T cells in the B cells follicles compared to sections from WT spleens (Figure 4-24). Though some IgG1+ B cells were detected in Il21r-/- mice, they appeared sporadically, rather than in GC where large regions of 107 B cells have downregulated IgD expression in the WT, indicating that some GC independent class switching was certainly still occurring. Transfer of WT OTII T cells in to Il21r-/- mice restored lo large IgD GC regions, containing many class switched B cells expressing IgG1. Interestingly, transferring antigen specific Il21r-/- OTII T cells could also restore GC structures and IgG1 expression, but compared to WT mice and Il21r-/- recipients of WT OTII T cells, there were still very few T cells infiltrating the follicle (Figure 4-24). The paucity of antigen specific Il21r-/- T cells in the B cell follicle, taken together with unimpeded isotype switch suggested that Il21r-/- OTII T cells, perhaps present at the border of the T cells zone and B cell follicle, were able to provide the stimulus for class switching that preceded T cell migration to the follicle. Although class switch and initial GC expansion seemed unaffected, we were interested in whether this generation of GC that seemed independent of large scale Tfh cell migration into the B cell follicle was defective in other respects, such as memory B cell recall responses to a secondary immunisation.

Figure 4-24 IL-21 is not required for effective class switch to IgG1 isotype Representative confocal images of WT and Il21r-/- spleen sections 7 days after immunisation with NP-OVA in alum. Immunostaining for IgD (blue), CD3 (green) and IgG1 (red). Data is representative of 3 experiments in which n=2-5 per group.

108 When spleen sections from the same groups were examined 5 days after a secondary immunisation of NP-OVA in PBS at day 21, we found Il21r-/- endogenous or Il21r-/- OTII T cells were completely unable to support a memory recall response. WT or Il21r-/- mice that had received WT OTII T cells showed large IgDlo GC and brightly staining IgG1+ plasma cells in the bridging channels of the follicles (Figure 4-25). This finding demonstrated that, in contrast to the dependence of plasma cell generation on IL-21 in the primary response, IL-21 responsiveness by B cells was not necessary for the generation of plasma cells during a secondary response. Large numbers of Tfh cells infiltrating the follicle were also evident in these groups, which could be detected in the locale of endogenous B cells re-entering the GC pathway during the secondary response (Figure 4-25). Thus, the action of IL-21 on CD4+ T cells was vital for supporting a memory B cell and plasma cell output from the GC during a secondary response to T dependent antigen, which may have reflected the increased presence of Tfh cells in the GC of mice receiving WT as opposed to Il21r-/- OTII T donor cells.

Figure 4-25 Il21r-/- OTII T cells were unable to support secondary immune response Representative confocal images of WT and Il21r-/- spleen sections after immunisation as above on day 26, 5 days after secondary immunisation. Immunostaining for IgD (blue), CD3 (green) and IgG1 (red). Data is representative of 3 experiments in which n=2-5 per group each experiment.

109 4.2.13 IL-21 acting on CD4+ T cells is required for affinity maturation of IgG1 in response to T dependent antigen

Histology from the previously described OTII transfer experiments showed that IL-21 was not required for class switching to IgG1 in response to T dependent antigen or for primary GC formation. However, IL-21 independent GC did not produce memory B cells or plasma cells. We next examined whether these GC were capable of producing high affinity antibody, which is produced after rounds of SMH and selection in the GC during competition for antigen and the cognate help provided by Tfh cells. WT and Il21r-/- mice were immunised with NP-OVA as described earlier, with some groups receiving either IL-21R competent or IL-21R deficient OTII T cells. Serum was collected every 7 days after primary immunization and 5 days after the booster immunisation, which was delivered at day 21. To test the functional output of the ensuing GC, affinity maturation was again assessed by measuring the amount of antibodies produced that could bind to NP hapten when coupled at very high (NP23) or low (NP3) levels with BSA. We found that WT OTII T cells but not Il21r-/- OTII T cells were able to stimulate endogenous Il21r-/- B cells to begin producing high affinity antibody, showing that IL-21 is important for effective T cell help and hence, productive GC formation (Figure 4-26). In this system, there was no requirement for B cells to respond to IL- 21 in order to make large amounts of high affinity antibody in response to immunisation.

Figure 4-26 IL-21 responsive CD4+ T cells are important for the production of high affinity antibody from B cells WT, Il21r-/- and Il21r-/- mice with or without transferred OTII of each genotype were immunised with 100μg NP-OVA in alum, then given a boost at day 21 in PBS. Serum was collected weekly and at 5 days after the secondary boost, and then assessed by ELISA for binding at either low or high affinity to either NP23BSA or NP3BSA, respectively. Data is presented as the mean of 3 combined experiments where n=2-5 per group.

110 Immunisation with the NP hapten results in the clonal expansion of B cells that are characterised by usage of a particular common variable heavy chain element, VH186.2 (McHeyzer- Williams, 1993; Nie, 1997). To assess whether both WT and Il21r-/- B cells could support SHM, we carried out analysis of SHM by quantifying the frequency of amino acid mutations in the complementarity determining regions (CDR) 1 and 2, and framework regions in the sequence of the VH186.2 heavy chain fragment. We purified and sequenced the VH186.2 region from single GL7+Fas+ GC B cells at day 14 of the NP-OVA response in WT, Il21r-/- mice and Il21r-/- mice reconstituted with either WT or Il21r-/- OTII T cells. All groups overwhelmingly utilised the VH186.2 heavy chain rather than other related noncannonical clones (Table 4-1 and data not shown). Interestingly, despite the differences in GC size and frequency between the groups, as assessed by histology (Figure 4-24 and Figure 4-25), all groups showed a similar higher rate of mutation in the CDR regions as opposed to framework. These findings demonstrate that IL-21 was not required for the process of SHM (Figure 4-27A and table 4-1). Importantly, in the presence of Il21r-/- CD4+ T cell help there was an increase in the proportion of clones that utilised an L>W mutation at amino acid 33, which confers 10 times stronger binding to NP (Bothwell, 1981; Cumano, 1985; Kelsoe, 1996; Weiss, 1992)(Figure 4-27 and Table 4-1). This may reflect the arrest of the immune response at the selection phase, before the progression to B cell memory and plasma cell formation and exit to the bone marrow. WT Il21r-/- Il21r-/- Il21r-/- + WT OTII + Il21r-/-OTII VH186.2 6.7 5.3 6.3 5.2 Mutations

frequency 4.3% 3.2% 4.2% 4.2% framework 6% 4% 6% 4% CDR1 18% 19% 16% 22% CDR2 7% 9% 8% 9%

Codon 33 L>W 22% 47% 26% 70% VH186.2 clones 9192317 Noncannonical 01 0 0 clones Table 4-1 SHM non-silent amino acid mutation analyses

111 WT Il21r-/- Il21r-/- +OTII Il21r-/- + Il21r-/-+OTII

Figure 4-27 Il21r-/- are capable of SHM (A) The mean percent of clones bearing amino acid changes from the germline VH186.2 sequence in various indicated regions. (B) The mean number of clones that bore an L>W mutation at position 33, conferring 10x higher affinity for NP. Data represent the mean of all clones analysed (see table 4.1). Data is representative of 2 experiments where n=3.

These data suggest that the defect in serum antibody levels in Il21r-/- mice is caused by an inability to amplify, differentiate and sustain B cell clones that have undergone SHM. Improved survival and expansion of somatically mutated Il21r-/- B cell clones can therefore be achieved by IL-21 acting on B cells or, as shown here, by WT Tfh cells that are capable of rescuing Il21r-/- B cells at this juncture.

4.2.14 IL-21 is required for OTII Tfh expansion or survival

The model of OTII T cell transfer and NP-OVA immunisation confirmed an important role for IL-21 acting on CD4+ T cells in ensuring a functional output of memory B cells and plasma cells from the GC. We examined the differences between WT and Il21r-/- donor OTII T cells during primary and secondary responses to their cognate antigen in order to understand how IL-21 may be supporting the Tfh cell response, and found a large difference in cell numbers at both timepoints (Figure 4-28A). WT OTII T cells expanded to ten fold greater levels in Il21r-/- rather than WT hosts, presumably reflecting a lack of competition in the former, and also perhaps higher level of circulating IL-21 in the system as it unable to be utilised by endogenous cells (Figure 4-28A). The reduced numbers of T cells available for provision of antigen specific help when Il21r- /- donor T cells were introduced was exacerbated by slightly lower expression of the chemokine receptor CXCR5 (Figure 4-28B), which might explain the reduced number of Il21r-/- OTII T cells observed in the B cell follicle (Figure 4-24 and Figure 4-25). Unexpectedly, levels of the Tfh cell signature co-inhibitory molecule PD-1 were higher on Il21r-/- donor T cells. PD-1 is present at high levels on T cells that have recently been exposed to antigen, and is thought to inhibit cell viability

112 when antigen is withdrawn (Figure 4-28B). ICOS was not affected by IL-21 in the primary immune response, which may indicate that it is upregulated upstream of IL-21’s effects on T cells, perhaps during interactions with the APC in the T cell zone or with B cells at the T-B border. Surviving donor Il21r-/- cells at 26 days after a booster immunisation had slightly reduced ICOS (Figure 4-28B). It seems that IL-21 is important for the maintenance of the numbers of ICOS expressing Tfh cells, but not for early ICOS upregulation on CD4+ T cells as it was only slightly reduced on transferred Il21r-/- OTII cells, and still higher than the non-responding WT host CD4 population shown in the shaded histograms (Figure 4-28B).

Figure 4-28 IL-21 supports Tfh cell numbers and the phenotype of donor OTII T cells WT and Il21r-/- OTII T cells were transferred into WT and Il21r-/- recipients and immunised with NP-OVA as described previously. (A) Quantification of the number of donor OTII T cells in the spleen at day 7 and day 26 of the primary and secondary response, respectively. Data are the mean and SEM of 3 combined experiments with 2-5 animals per group. (B) Representative dot plots showing splenic Tfh cell markers on host polyclonal WT (filled), WT OTII (unbroken line) or Il21r-/- OTII (dashed) donor CD4+ T cells at day 7 (primary) and day 26 (secondary) after immunisation The shaded histograms indicate the values on the WT polyclonal T cell population. Data is representative of 3 experiments in which n=2-5 per group for each experiment.

113 Taken together, these observations of the role of IL-21 responsiveness in antigen specific CD4+ T cells at the height of primary and secondary responses to NP-OVA showed that Il21r-/- OTII T cells display a defect in expansion or survival, and also an inability to migrate effectively to the B cell follicle. We speculated that these defects may stem from an inability to form strong interactions with B cells early in the immune response, and hence a defect in delivery of B cells help through co-stimulatory molecules such as ICOS and CD40L. IL-21 might therefore provide important survival and differentiation signals at a very early stage of the immune response when T cells upregulate CXCR5 and move towards the T cell border and subsequently into primary B cell follicles.

4.2.15 IL21 aids early clonal expansion of antigen specific T cells

In order to study the importance for IL-21 for early proliferation and positioning of antigen specific T cells at day 2.5 when T-B cell interactions are beginning, we immunised WT and Il21r-/- hosts with NP-OVA and transferred CFSE stained WT and Il21r-/- OTII T cells on day 0. CFSE+ donor cells could be easily detected by histology and could be visualised as yellow cells due to a co- stain with anti-CD3 on the red channel during confocal microscopy. Firstly, even at this very early timepoint, it was clear that Il21r-/- hosts contained fewer T cells from either donor or host origins within the B cell follicle, with endogenous red CD3 cells tending to gather around the edges of the B cell follicles or remaining in the T cell zone or red pulp (Figure 4-29). Yellow antigen specific donor cells also showed a need for IL-21 signalling for correct positioning in the B cell follicle, indeed Il21r-/- OTII T cells showed a particular disadvantage in WT hosts, presumably due to competition with many viable host WT T cells for interactions with B cells.

114 Figure 4-29 Il21r-/- T cells are excluded from the B cell follicle at day 2.5 5x105 purified CFSE+ WT and Il21r-/- OTII CD4+ T cells were transferred into hosts of each genotype and immunised as above. Representative images from day 2.5 of 3 transverse histological sections of 3 spleens from each group showing positioning of donor OTII T cells in the spleen. Data are representative of 2 experiments in which n=3.

115 We quantified the positioning of WT and Il21r-/- OTII T cells by examining consecutive transverse sections from 3 mice from each group and scoring each for whether T cells were in the red pulp, T cell zone or B cell follicle. Each genotype was represented evenly in the red pulp, whereas IL-21 responsive cells tended to be more frequent in the T cell zone, and were significantly increased in B cell follicle (Figure 4-30), showing that IL-21 has an important role for early migration of antigen-specific CD4 T cells during an immune response.

Figure 4-30 Il21r-/- OTII T cells migrate poorly to the B cell follicle 5x105 purified CFSE+ WT or Il21r-/- OTII T cells were transferred into hosts of each genotype indicated, immunised with NP-OVA and analysed on day 2.5. Quantification of the average number of T cells in >3 fields from >2 transverse histological sections scored for position within the spleen. Data are representative of 2 experiments in which n=3

In order to ascertain whether the observed differences in early T cell migration of WT and Il21r-/- OTII T cells were driven by early defects in chemokine regulation and proliferation, we examined these same donor OTII T cells by flow cytometry. IL-21 clearly played a role in the initial burst of clonal expansion in the T cell zone in response to antigen, as even at day 2.5 there was a marked reduction in the numbers of Il21r-/- OTII T cells that had proliferated (Figure 4-31A and B). In contrast, the number of peaks of proliferation, as indicated by CFSE dilution, were similar between the two groups (Figure 4-31A). IL-21 can produced promptly following TCR stimulation in vitro, where it can be detected at 24 and 48 hours by ELISA (data not shown)(McGuire, 2009). The absence of IL-21 leads to a significant defect in the numbers of proliferating Il21r-/- OTII T cells present as early as two days post antigen exposure in vivo .

116 In contrast to later timepoints, Tfh cell markers CXCR5, ICOS and CCR7 were modulated independently of IL-21 at day 2.5 in this system (Figure 4-31C). It is important to note that the Tfh cell markers CXCR5, PD-1 and ICOS have been described as early activation markers and are upregulated by all CD4+ T cells after cognate TCR interactions with MHC II-antigen complex, whereas they are retained at later stages only on the Tfh subset.

Figure 4-31 IL-21 aids early clonal expansion of antigen specific OTII T cells 5x105 purified CFSE+ WT or Il21r-/- OTII T cells were transferred into hosts of each genotype, immunised with NP-OVA and analysed at day 2.5. (A) Representative histograms showing cell proliferation measured by CFSE dilution and (B) total numbers of OTII T cells in the spleen. (C) Representative flow cytometry dot plots of Tfh cell markers on OTII donor T cells. Data represents 2 experiments in which n=3.

117 Low expression of chemokine receptors that would drive migration to the follicle, and high expression of PD-1 in the absence of IL-21 could indicate that these T cells were not forming strong synapses with B cells in this setting. Therefore, we looked for T:B conjugates containing endogenous WT and Il21r-/- B cells and OTII T cells and found that although a large proportion of Il21r-/- OTII T cells were bound to B cells, WT OTII T cells were more commonly bound to B cells of both genotypes than their Il21r-/- counterparts (Figure 4-32), though this trend was not found to be significant with the number of mice in the study. Also, there was no role for IL-21 acting on B cells in determining the number of these conjugates, as their numbers correlated with the genotype of the T cell donor, rather than the genotype of the host B cells (Figure 4-32).

Figure 4-32 IL-21 is necessary for CD4+ T cells to form stable conjugates with B cells CFSE+ OTII T cells were adoptively transferred at the time of immunisation with NP-OVA, and spleens examined at day 2.5. (A) Representative dot plots gated on donor OTII T cells showing cells bound to B220+ cells as doublets and (B) quantification of the total number of T:B conjugates in the spleens. Data is representative of 2 experiments in which n=3, no significant difference was found between WT and Il21r-/- by Student’s T test.

118 As well as a larger proportion of the relatively few remaining Il21r-/- OTII T cells detected bound to B cells, PD-1 was significantly increased on Il21r-/- OTII T cells (Figure 4-33A and B). PD-1 is a molecule from the B7 family that is upregulated in response to antigen exposure and ligation can lead to cell death through inhibition of cell metabolism pathways unless it is countered with strong survival signals from other co-stimulatory molecules and cytokines (Okazaki and Honjo, 2006). PD-1 expression was found to be higher on B220 bound, rather than free single cell OTII cells, which may indicate that Tfh cells are susceptible to activation induced cell death due to constant TCR triggering. IL-21 seemed to be able to ameliorate this pathway somewhat as PD-1 expression was higher again on Il21r-/- rather than WT OTII T cells (Figure 4-33A). It is possible that IL-21 may deliver a survival signal alongside TCR stimulation by B cells at the T:B border, and indeed throughout the continuing bombardment of antigen applied to Tfh cells in the B cell follicle where many centrocytes compete for signals from limiting infiltrating T cells.

Figure 4-33 B cells and IL-21 conversely effect PD-1 expression on antigen specific T cells CFSE+ OTII T cells were adoptively transferred at the time of immunisation with NP-OVA, and spleens examined at day 2.5. (A) Representative histograms showing PD-1 expression and (B) quantification of mean fluorescence values +/- SEM of PD-1 expression on gated OTII T cells from different genotypes. Data is representative of 2 experiments in which n=3.

We speculated that Il21r-/- antigen specific CD4+ T cells might be more susceptible to cell death when not bound to B cells displaying cognate antigen. This may explain why many more Il21r-/- OTII T cells could be detected in conjugates, as their unbound counterparts may be dying. Indeed, we found that at day 2 of the response to NP-OVA, OTII T cells that could not respond to IL-21 had slightly decreased levels of the pro-survival molecule Bcl-xL, while no difference in Bcl- 2 could be detected (Figure 4-34A and B). This defect in Bcl-xL expression could be rescued by treatment with an antagonistic anti-PD-1 antibody, which blocked ligation with B7 family

119 molecules (Figure 4-34C). Taken together, these data showed that proliferation and long-term survival of antigen specific CD4+ T cells involved in T dependent GC reactions could be directly influenced by IL-21. Surprisingly, Annexin VI levels were not different between these populations, perhaps because cell death during the cell isolation and flow cytometry masks small difference in cell viability generated in vivo (data not shown).

Figure 4-34 IL-21 maintains survival signals in antigen specific CD4+ T cells Representative histograms from flow cytometry of OTII T cells and host WT T cells showing levels of survival molecules (A) Bcl-2 and (B) Bcl-xL. (C) Quantification of the mean Bcl-xL expression of OTII T cells 2.5 days after immunisation with or without treatment with an anti-PD-1which prevents ligation with B7 family on APC. Data is representative of 2 experiments in which 3-4 mice were pooled.

120 In order to see if IL-21 was also capable of sustaining BCL-xL in a polyclonal CD4+ T cell response, we immunised WT and Il21r-/- mice with SRBC and looked at pro-survival molecule levels at day 2. Even at this early timepoint, we were able to see an IL-21 dependent increase in Bcl-xL expression in the CXCR5+ICOS+ Tfh subset, and to a less extent in other ICOS+ CD4+ T helper cells (Figure 4-35A). The MFI of Bcl-xL within the early ICOS+ population at day 2 after SRBC was significantly lower in Il21r-/- T cells (Figure 4-35

p=0.01

Figure 4-35 IL-21 sustains Bcl-xL in polyclonal Tfh cells

(A) Representative histograms showing Bcl-xL staining of ICOS+ or ICOS+CXCR5+ Tfh CD4+ T cells from WT or Il21r-/- at day 2 post immunisation with SRBC. (B) Representative MFI values from above at days 2 or 4 post SRBC immunisation. These data represent 2 experiments in which n=2-3.

121 4.3 Discussion

This study demonstrates that IL-21 is critically important for the generation of Tfh cells and that responsiveness of Tfh cells to IL-21 drives the formation of the GC reaction. IL-21 modulated the expression of chemokine receptors and costimulation molecules in vivo that direct Tfh cells, expressing the IL-21R, into the GC. Unexpectedly, IL-21 responsiveness was required on CD4+ T cells but not B cells for the generation of the GC and production of IgG1. Our data indicate that IL-21 exerts its effects by costimulation of the TCR. IL-21-driven costimulation was required for the high expression levels of CXCR5 that guides Tfh cells into the CXCL13-rich GC. It is likely that after the initial activation by antigen-loaded antigen-presenting cells in the T cell zone, T cells destined to become Tfh cells must receive additional signals provided by cells located in or close to the B cell follicles (Ebert, 2004). Il21r-/- OTII are currently being backcrossed onto a CXCR5 transgenic line to determine whether IL-21 induced retention of CXCR5 expression on Tfh cells is the dominant defect of Il21r-/- CD4+ T cells that prevents their entry into the follicle. Endogenous Il21r-/- CD4+ T cells, Il21r-/- and WT OTII T cells formed a continuum, which represented a gradient their ability to provide T cell help to B cells. Polyclonal Il21r-/- did not support the required threshold of T cell help needed to produce GC reactions in Il21r-/- mice. Both WT and Il21r-/- OTII T cells could restore this defect by histological measures in the primary immune response to NP-OVA, due to increased T cell frequency in the follicle. The quantitative difference in OTII T cell numbers in the follicle depended on sensitivity to IL-21, and the importance of this became clear during a secondary response where WT, but not Il21r-/- OTII T cells were able to restore an appropriate recall response to antigen. Therefore IL-21 dependent Tfh cells perform an essential role in generating memory B cells and long lived plasma cells from the GC in response to immunisation. Data in the previous chapter showed that B cells could costimulate CD4+ T cells for optimal production of IL-21 through ICOS:ICOSL interactions. The level of expression of ICOS has been previously suggested to define a qualitative aspect of T helper cell function based upon cytokine production (Lohning, 2003) and in this context, the highest production of IL-21 is also associated with the highest expression of ICOS in Tfh cells. However, while IL-21 was an absolute requirement for ICOS expression in vitro, it was not necessary for ICOS expression on CXCR5- CD4+ T helper cells in vivo. This discrepancy possibly reflects that our in vitro culture conditions incompletely mimic T cell activation events in vivo and indicates that factors in the lymphoid

122 microenvironment driving ICOS expression could compensate for the lack of IL-21. Thus, IL-21 acted as a soluble costimulator but Tfh cells were further dependent on IL-21 for their growth or survival since increased costimulation per se (via CD28) could not recover Tfh cells in vivo. BrdU incorporation studies on the increased numbers of Tfh and Treg cells following in vivo CD28 stimulation would be an interesting addition in order to determine whether this phenomenon reflects increased recruitment or cell division. Previous studies indicate that IL-21 is a switch factor for IgG1 (Pene, 2006). In contrast, our data indicate that IL-21 responsiveness by B cells was not necessary for the switch to IgG1, but it clearly impacted on the amount of antibody produced. Our findings support a quantitative effect of IL-21 on IgG1 production insofar as GC B cells required IL-21 for their expansion rather than for switching to the IgG1 isotype. These findings are consistent with the notion that the induction of isotype switch recombination is imprinted prior to the GC reaction (Chan, 2009; McHeyzer- Williams, 2005). This study was concerned with whether B cell responsiveness to IL-21 influences the affinity maturation of antibody responses. The data described above have shown that IL-21 is necessary for T cells to acquire a Tfh phenotype and enter the GC reaction, therefore removing what is known to be an important stimuli for proliferating B cells in the light zone (Allen, 2007a). As a result, restoring IL-21 sensitive antigen specific T cells in the form of OTII cells was sufficient to increase high affinity antibody to WT levels as assessed by ELISA. Yet interestingly, this was not due to an absence of SHM in the defective GC that remain in Il21r-/- without WT T cell help, suggesting that the defect in high affinity antibody levels in the serum in fact lies down stream, perhaps at the point of differentiation into antibody secreting cells after SHM has taken place. Thus, IL-21 expressed by Tfh cells and acting on B cells is not necessary for the process SHM per se, but is critical for the amplification of the resulting clones into antibody producing cells. Data shown here from BM chimera experiments confirmed a non-redundant role for IL-21 in plasma cell formation in response to NP-OVA. The high affinity antibody titres restored by WT OTII T cells in Il21r-/- hosts, was likely to have been secreted by the residual remaining plasma cells, or perhaps by antibody secreting cells derived from memory B cells, as it was improved after secondary immunisation. Further research is required to characterise the presence of BM emigrant plasma cells and their requirement for IL-21 for their generation and survival. Recent characterisation of Il21-/- and Il21r-/- CD8+ T cells during chronic viral infections showed IL-21 was critical in preventing a PD-1+ ‘exhausted’ phenotype arising (Elsaesser, 2009; Frohlich, 2007). A similar phenomenon may be seen here in the Tfh cell subset where IL-21 acting on T cells was shown to prevent PD-1 expression in response to intense TCR signals from B cells

123 during the provision of CD4+ T cell help. This resulted in increased antigen specific T cell numbers at all stages of the immune response to NP-OVA. Further, this study suggests a possible mechanism for this protective effect of IL-21 specifically on Tfh cells is maintenance of Bcl-xL upregulation. The use of FACS analysis of T-B conjugates appearing after immunisation is an interesting new technique for quantifying germinal center reactions, used here to highlight a trend toward decreased interactions with B cells and Il21r-/- OTII CD4+ T cells (Figure 4-32). A larger study with more mice was not possible due to limited breeding during my PhD, but would be an interesting addition to confirm whether this trend is truly significant, as it would reflect a more functional read out of T cell activity in the GC than donor cell numbers alone. We propose a model where autocrine IL-21 antagonises inhibitory signals delivered via PD-1 ligation leads to increased cell survival mediated by increased Bcl-xL. Further studies will be required to delineate the biochemical pathways involved this mechanism, but preliminary data presented here suggests that this is indeed the case. We saw no differences in the pro-survival molecule Bcl-2 between WT and Il21r-/- (data not shown), and it would be interesting to see whether a reciprocal increase in pro-apoptotic molecules such as Min and Puma also characterises the Tfh subset in Il21r-/- mice. Although we attempted to quantify the apoptosis of this subset immediately ex-vivo using flow cytometry analysis of Annexin-V, we were not able to see any differences between genotypes (data no shown). The extreme sensitivity of the Tfh population to apoptosis (high PD-1 and CTLA-4, fas in humans) make it difficult to detect differences in viability that may arise in vivo, due to the background levels of death caused by cell isolation. In the future, it would be interesting to study this phenomenon in situ using histological analysis of apoptosis of T cells in and surrounded the GC in histological sections.

One important issue pertaining to Tfh cells is their relationship to other T helper cell subsets, such as Th1, Th2, Th17, and T regulatory (Treg) cells. Th17 cells express IL-21 in response to IL-6 but are unlikely to be the predominant source of IL-21 since RoRt-/- mice express normal levels of IL-21 (Zhou, 2007). In this study, Tfh cells were shown express IL-21 and high levels of the IL-21R but do not produce IL-17. These data support the notion that Tfh and Th17 cells are indeed distinct subsets. The IL-21-driven autocrine loop established by Tfh cells determines their phenotype and fate and distinguishes them from other Th cells, including Th17 cells.

This study uncovers important information on the biology of IL-21 and IL-21R for Tfh cell generation. B cells have been thought to be the predominant recipient of IL-21’s effects, driving class switch and antibody production. This study challenges that notion by revealing an IL-21- driven autocrine loop in Tfh cells that controls isotype switching and GC formation. These findings

124 reveal a fundamental, but previously unknown, aspect of T helper cell differentiation, and the central role of IL-21 and IL-21R in Tfh cell biology.

125 5 IL-21 in autoimmunity

5.1 Introduction

The two c-chain family cytokines IL-2 and IL-21 lie adjacently in a region of linkage disequilibrium on chromosome 3 in mice and chromosome 4 in humans. Although the transcription of these two cytokine genes may be coordinately regulated (McGuire, 2009), IL-2 and IL-21 are crucial growth factors for distinct T helper subsets with regulatory and effector functions, respectively. Both cytokines are released following TCR ligation and yet seem to have distinct and sometimes opposing autocrine effects on T cells. IL-2 is a survival factor for peripheral forkhead family transcription factor Foxp3 expressing Treg cells which are vital for regulating immune responses and tolerance to self (Fontenot, 2005; Furtado, 2002; Malek, 2002). Conversely, IL-21 seems to play an important role in acquisition of effector phenotypes by T cells (Kim, 2005). As we have shown in previous chapters, IL-21 acts on T cells to promote a Tfh phenotype, which can in turn promote GC formation and high affinity antibody response (Nurieva, 2008; Vogelzang, 2008). IL-21 has also been reported to have important role in supporting long term CD8+ T cells activity and survival during chronic viral infections (Elsaesser, 2009; Frohlich, 2007; Yi, 2009). Mice made genetically deficient in IL-2 or its high affinity receptor chain (CD25) suffer from a fatal autoimmune disease characterized by ulcerative colitis and haemolytic anaemia (Sadlack, 1993; Willerford, 1995). The observed autoimmunity and associated lymphoid hyperplasia are thought to be caused, in part, by a loss of regulation of effector T cells due to a deficit of IL-2-dependent Foxp3 regulatory T cell survival or expansion in the periphery (Fontenot, 2005). We have recently observed that Il2-/- mice have large numbers of circulating IL-21- producing CD4+ T cells. IL-21 is thought to deliver a costimulatory signal to lymphocytes (Parrish- Novak, 2000) and molecular studies are beginning to reveal pathways downstream of the IL-21R that might account for its costimulatory function. For instance in T cells, JAK-STAT, MAPK and PI3K pathways are all involved in IL-21 signalling through its receptor (Zeng, 2007). Consistent with its actions on lymphocyte populations, IL-21 contributes to development of autoimmune diseases in a number of animal models such as Systemic lupus erythematosus (SLE or lupus), experimental autoimmune encephalomyelitis and rheumatoid arthritis (Andersson, 2008; Daha, 2009; Leonard and Spolski, 2005; Liu, 2008; Spolski and Leonard, 2008; Vollmer, 2005); moreover, studies describing the sanroque mouse strain that develops a Lupus-like disease reveal

126 excessive levels of IL-21 in the serum (Vinuesa, 2005b). However, the mechanisms explaining IL- 21’s function in autoimmune disease pathogenesis remain unknown. Since IL-21 is known to facilitate the development of a number of autoimmune diseases, the possible contribution of IL-21 to the autoimmune pathology observed in Il2-/- mice was investigated in this chapter by analysing double knock-out Il2-/- and Il21r-/- (Il2 Il21r-/-) mice. Il2 Il21r-/- mice exhibited reduced morbidity and mortality compared with Il2-/- mice, which was characterized by an altered CD4+ T cell effector cytokine phenotype and reduced serum antibody levels and, paradoxically, fewer Foxp3+ Treg cells. Splenomegaly and lymphadenopathy were unchanged in Il2-/- mice in the absence of IL-21:IL-21R signalling. However, CD8+ T cells numbers were reduced in Il2 Il21r-/- mice, although they retained a highly activated phenotype. Histological analysis revealed that pancreatitis might play a more important role than colitis in the fatal wasting disease observed in the absence of IL-2. These findings demonstrate that IL-21:IL- 21R signalling contributes to the destruction of tissues attributed to autoimmunity in Il2-/-mice and suggests that IL-21-producing T cells are significant targets of immune system regulation.

127 5.2 Results

5.2.1 Autoimmune Il2-/- mice produce high levels of IL-21

Il2-/- mice display a diffuse autoimmune T cell activation that leads to early mortality in this strain (Kramer, 1995; Sadlack, 1993). This disease, which is thought to result from the escape of the adaptive immune system from regulation (Fontenot, 2005; Sadlack, 1993). Deletion of the IL-2 gene removes a vital survival signal for peripheral Foxp3+ Treg cells leaving only a vestigial population that fails to regulate the size and function of the effector T cell pool (Fontenot, 2005). Mice made genetically deficient in both the high affinity IL-2Ra subunit (CD25) and CD4+ T cells were shown to lack the severe colitis present in CD25-/- mice, and thus CD4+ T cells are thought to be responsible for much of the immune infiltration in mucosal tissues in Il2-/- mice (Simpson, 1995). We found that Il2-/- CD4+ T cells produced higher levels of IL-21 mRNA than WT CD4+ T cells when lymph node cells were stimulated in vitro with anti-CD3 and CD28, and increased transcription was also induced more quickly (Figure 5-1A). This finding correlated with an increased number and intensity of IL-21 protein detected by flow cytometry in CD4+ T cells ex- vivo. Positive staining for IL-21 was concentrated in the CD44hi population of CD4+ T cells that was enlarged in Il2-/- mice, and was not seen in CD8+ T cells, which were used as a negative control (Figure 5-1B and data not shown). Unfortunately it was not possible to detect IL-21 in serum using the commercial reagents currently available.

Figure 5-1 IL-21 production is enhanced in autoimmune Il2-/- CD4+ T cells (a) IL-21 mRNA levels from WT and Il2-/- LN cells ex-vivo and after stimulation with soluble anti- CD3 and CD28 measured by Real Time PCR. IL-21 mRNA expression is presented as fold modulation compared to WT ex-vivo levels. b) Representative histograms showing intracellular IL- 21 staining of the CD44hi subset of CD4+ T cells compared to CD8+ T cells as a control. Data is representative of 2 experiments in which n=3.

128 5.2.2 IL-21 drives morbidity in autoimmune Il2-/- mice

To determine what role excessive IL-21 might play in driving autoimmune pathology in Il2- /- mice, we crossed them onto the Il21r-/- background to create Il2 Il21r-/- mice. It was immediately evident from the healthy appearance of Il2 Il21r-/- mice, compared to their Il2-/- littermates, that removing IL-21 had mitigated autoimmune disease in these mice. Indeed, morbidity and mortality was significantly delayed in this strain as only 2/7 aged Il2 Il21r-/- mice had succumbed to disease by 30 weeks of age, whereas the equivalent 50% of Il2-/- mice had been euthanised by 12 weeks of age (Figure 5-2).

Figure 5-2 Improved survival of Il2 Il21-/- mice compared to Il2-/- littermates Percent survival of Il2 Il21r-/- and Il2-/- mice was measured using euthanasia as an endpoint when mice lost 10% of their weight, or when severe morbidity was observed. The chi squared logrank test was used to compare survival curves.

129 The improved outlook for Il2 Il21-/- mice was also evident in the increased weight of both sexes compared to Il2-/- littermates (p<0.0001, by one way ANOVA), beginning at 6 weeks when we started our measurements and lasting throughout the period of study up to 40 weeks (Figure 5-3A). Surprisingly Il2 Il21r-/- had comparable weight to WT or Il21r- /- mice that have no autoimmune phenotype. When the weights of both sexes were compiled, Il2-/- trends reflected early wasting disease with an almost universal downwards trend of individual mice, which was only apparent is some Il2 Il21r-/-individuals much later at the 25 weeks timepoint (Figure 5-3B).

Figure 5-3 Reduced wasting disease in Il2 Il21r-/- mice

(a) Cumulative weights of WT, Il21r-/-, Il2-/- mice and Il2 Il21r-/- littermates, n values as indicated in the legends, data is the mean +/- SEM. (b) Weights of individual mice of both sexes from Il2-/- and Il2 Il21r-/- strains.

130 Despite an improvement in lifespan and weight loss, it was surprising to find that the secondary lymphoid organs in the Il2 Il21r-/- displayed similar splenomegaly and lymphadenopathy to Il2-/- littermates (Figure 5-4 and Figure 5-5).

Figure 5-4 Enlarged secondary lymphoid organs in both Il2-/- strains Representative photographs of spleens, MLN and inguinal LN of indicated mouse lines at 10 weeks of age (n=<10).

When cell numbers in the spleen, lymphoid organs and tissues were quantified, there was a slight increase in the spleen sizes of Il2 Il21r-/- mice compared to the Il2-/- mice (Figure 5-5). This trend was not evident in peripheral immune sites, including the MLN, or in the epithelium and lamina propria of the colon, which is the site of cellular infiltrate in Il2-/- mice, where cell numbers were similar (Figure 5-5).

*** *** * ***

Figure 5-5 Increased cell numbers is spleen but not mucosal sites in Il2 Il21r-/-mice Cell numbers from the spleen, MLN, and lymphocyte isolations from intra epithelial sites (IEL) and the lamina propria (LPL). Data represent values from individual mice between 8-14 weeks, and the mean +/- SEM from 5 pooled experiments. Genotypes were compared using one-way ANOVA and Bonferroni’s post-test to compare all Il2-/- and Il2 Il21r-/- values to WT, *p<0.05, ***p<0.001.

131 5.2.3 Histological analyses of colitis and pancreatitis in Il2-/- mice

To discover whether the observed changes in lymphocyte distribution and lifespan in the absence of IL-21 corresponded with reduced tissue damage, histological analyses were carried out on several organs. We began by examining colon sections for signs of immune destruction, as the colon has been previously described as the target of the most severe cellular infiltrate in Il2-/- mice (Sadlack, 1993). There were some examples of crypt branching where cell junctions had been compromised (Figure 5-6C and D), and increased mononuclear lymphocytes found throughout both the mucosa and lamina propria (Figure 5-6E and F). The presence of some lymphocytes within the mucosa is normal, but tends to be found towards the surface exposed to the contents of the gut (Figure 5-6A and B) and not throughout the mucosa (Figure 5-6E and F). We also found that the frequent gut-associated lymphoid structures were enlarged in both Il2-/- and Il2 Il21r-/- strains (Figure 5-7).

132 Figure 5-6 Crypt branching and mononuclear cell infiltrate in Il2-/- strains. Representative H&E stain histological sections of the distal colon from indicated genotypes showing (a, b) Healthy WT and Il21r-/- mucosa, (c, d) crypt branching indicated by arrows and (e, f) increased mucosal mononuclear infiltrate in Il2-/- and Il2 Il21r-/- mice. All samples were obtained from 9-12 week old mice, n=3 for WT and Il21r-/- samples and >10 for both Il2-/- strains.

133 Despite registering these frequent signs of inflammation, we could see no ulceration or erosive destruction of the mucosa in any of the samples studied and large areas retained structural, apparently functional integrity (Figure 5-7). Indeed,clinical signs of colitis was rarely observed in live mice, and were restricted to mice that survived beyond 18 weeks, which was very rare in the Il2-/- strain because only 20% of mice had survived to this timepoint (Figure 5-2).

Figure 5-7 Mucosal barrier integrity despite mononuclear infiltration of Il2-/- strains Representative H&E stain histological sections of the distal colon from indicated genotypes. All samples were obtained from 9-12 week old mice, n=3 for WT and Il21r-/- samples and >10 for Il2- /- strains.

This assessment led us to examine other tissues for damage that may have been able to explain the early morbidity in Il2-/- mice but not Il2 Il21r-/- mice. Pancreatitis can lead to an inability to digest dietary fats and in light of the severe wasting in Il2-/- mice, the pancreas was a candidate organ for pathology in these mice. Although there is no published observation of pancreatitis in this strain in the literature, we found that in our colony and SPF housing conditions, 90% of samples from both strains at the age of 9-12 weeks contained diffuse lymphocytic aggregates that were both parenchymal and perivascular. However, it was clear that in the absence of IL-21 there was an improvement in the level of parenchymal damage inflicted with minimal to 134 sporadic damage of exocrine tissue, whereas atrophied serous acini and loss of the lobular architecture was widespread in all Il2-/- pancreata (Figure 5-8). Pancreatic islets remained undamaged by infiltrate in both strains (Figure 5-8). Several other organs such as the kidney, liver, and lung were also examined and were found to contain sporadic infiltrating lymphocytes but not the large aggregates and tissue damage found in the colon and pancreas respectively.

Figure 5-8 Severe pancreatitis in Il2-/- mice is improved in the absence of IL-21 Representative H&E stained histological sections of pancreata from indicated genotypes. All samples were obtained from 9-12 week old mice, n=3 for WT and Il21r-/- samples and >10 for Il2- /- strains.

In order to determine whether pathology was delayed rather than absent when IL-21 was removed from the autoimmune process, we examined histological sections from both the colon and pancreas of aged Il2 Il21r-/- mice. We found that while aged Il2 Il21r-/- colons were characterised by a enlarged mucosa and frequent lymphoid aggregates, there were still very few crypt abscesses or eroded mucosal surfaces and much uninterrupted, apparently functional surface could be observed. It therefore appeared that colitis was not a major factor in the observed morbidity in these strains within the timeframe studied within our colony (Figure 5-9).

Figure 5-9 Cellular infiltrates at mucosal sites of aged Il2 ll21r-/- mice Representative H&E stained histological sections of the distal colon and pancreas from Il2 Il21r-/- mice over the age of 30 weeks (n=5).

135 The pancreata of older mice contained an increase in parenchymal damage with some but not all lobes displaying atrophy of exocrine tissues. Even many weeks after the death of Il2-/- littermates, the pancreas of Il2 Il21r-/- mice retained lobes with only sporadic infiltrating mononuclear cells and healthy parenchymal acini, though regular deposits of fibrosis in these samples were evidence that inflammation had reached an endpoint in some areas (Figure 5-9).

5.2.4 Exacerbation of Treg defect in the absence of IL-21

As the autoimmune phenotype of Il2-/- mice is driven by a large reduction in the proportion of the suppressive Treg CD4+ T cell subset relative to effector T cells (Fontenot, 2005), we examined our double knockout strain for Foxp3 expression to see IL-21 played any role in sustaining the remaining Foxp3+ population. We found that in contrast to our expectation that the decreased morbidity of Il2 Il21r-/- mice may be explained by an improved Treg phenotype, this strain was characterised by a further decrease in the proportion of Foxp3+ cells among the CD4+ T cell population (Figure 5-10A and B). This was most marked in the spleen, which may reflect the increased cell number in this organ in particular (Figure 5-5), but was also evident at mucosal sites.

Figure 5-10 Further decrease in T reg cells in Il2 Il21r-/- mice (a) Quantification of Foxp3+ cells as a proportion of CD4+CD3+ T cells shown as individual mice and mean +/- SEM. (b) Representative dot plots from MLN showing Foxp3+CD25+ T reg cells gated on CD4+CD3+ T cells. Data from 5 pooled experiments in which n=6-10. P values less then 0.05 calculated with Student’s t test are indicated in the figure.

136 It is unlikely that the observed reduction in the ratio of Treg to T effector cells was due to direct agonistic signalling through IL21R on Tregs, but rather due to an increase of CD4+ effectors as absolute numbers of Foxp3+ cells were similar in WT, Il2-/- and Il2 Il21r-/- tissues due to the large expansion of the CD4+ T cell subset in the latter 2 groups (data not shown). Also, Il21r-/- Tregs were found to be capable of suppressing the proliferation of both WT and Il21r-/- CD25- CD4+ effector cells in an in vitro proliferation assay where suppression of CSFE- dilution by Il21r- /- Tregs was equivalent to WT Tregs (Figure 5-11). These assays indicated that much like IL-2 (Fontenot, 2005), IL-21 was not directly driving Foxp3+ Treg development or function, at least in an in vitro setting, but was rather modulating homeostatic Treg to T effector ratios in Il2 Il21r-/- mice.

Figure 5-11 Il-21 is not necessary for in vitro Treg suppression (a) Suppressive activity of WT and Il21r-/- CD25+CD4+ T regs (Treg) was measured by culture in reducing ratios with 2x105 CFSE labelled CD4+ CD25- T effectors (Teff), anti-CD3 at 2μg/ml and irradiated APCs for 72 hours. (b) Representative histograms showing WT or Il21r-/- T reg suppression of WT effector population at various ratios. Representative of 2 experiments in which n=3

5.2.5 Altered CD4+ T cell effector phenotype in Il2-/- mice in the absence of IL-21

The importance of thymocytes in the Il2-/- strains’ lymphoproliferative disorder was previously established in experiments that utilized athymic Il2-/- mice to demonstrate that disease progression in this model is dependent on T cells (Kramer, 1995). CD4+ T cells, in particular, were shown to be central to the disease progress when Il2-/- mice made genetically deficient in MHC 137 class I exhibited enhanced colitis. Infiltrating Il2-/- TCR CD4+ T cells were also shown to possess cytotoxic activity in vitro (Simpson, 1995). By introducing IL-21R deficiency to the Il2-/- strain, we had hoped to pinpoint roles for IL-21 in developing CD4+ T cell effector phenotypes that mediate autoimmune tissue damage in this model. However, aside from the altered ratio of Treg cells to effector CD4+ T cells, we could identify few differences between the numbers of T cells with an activated or memory surface phenotype in Il2-/- autoimmune mice in the presence or absence of IL-21 (Figure 5-12A and B).

Figure 5-12 Expanded CD4+ T cells in Il2-/- strains with a mucosal homing phenotype both in the presence and absence of IL-21. (a) Quantification of CD4+ CD3+ T cells in Spleen, MLN and lamina propria of genotypes indicated between 8-12 weeks of age. Data are values from individual mice from 5 experiments and the mean -/+ SEM. (b) Representative images of CD3+CD4+ T cells showing activated mucosal homing CD44hi 47+ subset. Non parametric ANOVA was used to compensate for different variances, using Dunn’s post test to compare all groups. N.d indicates no significant difference, ** indicates a p value less than 0.01, *** less than 0.001.

The CD4+ T cell compartment was dominated by CD44hi cells and approximately 20% of these cells also expressed the mucosal homing marker 47, which drives homing of these activated cells to the mucosal surfaces where much of the damage in Il2-/- mice reportedly occurs (Simpson, 1995 #458; Sadlack 1993; (Hsu, 2009)(Figure 5-12B). No difference in cytotoxic Granzyme B production could be detected in the spleen or gut by flow cytometry (data not shown).

138 Although priming, cytotoxicity and homing seemed unaffected, IL-21 did seem to play a role in shaping the T helper profile of the resulting effector T cells, both in lymphoid organs and sites of inflammation. Il2 Il21r-/- CD4+ T cells showed a reduced output of the pro-inflammatory cytokine IL-17A, particularly at inflamed mucosal sites such as the lamina propria (Figure 5-13A and B). IL- 17A was initially identified as a driving force in autoimmune inflammation due to its ability to recruit neutrophils and lymphocytes to tissues, and has since been described in protection against mucosal Gram-negative bacterial infections such S. pneumoniae, fungi and protozoa (Ivanov, 2009; Ma, 2009b; Pepper, 2009; van de Veerdonk, 2009). IL-21 is produced by the Th17 CD4+ T subset, and has beenshown to create a positive feedback loop that enlarges this population (Korn, 2007a; Wei, 2007), and this finding is supported by reduced IL-17A production in our model (Figure 5-13).

LPL *

Figure 5-13 Ameliorated Th17 T cell phenotype in the absence of IL-21 (a) Quantification of intracellular IL-17A+ staining in CD4+ CD3+ T cells in Spleen, MLN and among intra –epithelial (IEL) and the lamina propria (LPL) lymphocyte populations between 8-12 weeks of age. Data are values from individual mice from 3 experiments with the mean -/+ SEM. (b) Representative flow cytometry images of staining of cells isolated from the lamina propria and (c) pooled quantification of CD3+CD4+ T cells positive for intracellular IL-17A and stimulated with PMA and ionomycin for 4 hours. P values were calculated using one-way ANOVA and Bonferroni’s post-test compared to WT values (*p<0.05).

Another difference between Il2-/- and Il2 Il21r-/- T cells was observed when we performed intracellular flow cytometry analysis of IFN and TNF, two typical Th1 inflammatory cytokines commonly associated with autoimmune disease (Liblau, 1995). Whereas IL-21 seemed to drive increased IL-17A in Il2-/-, this correlated with a trend towards increased production of IFN and 139 TNF within the CD4+ T cell compartment (Figure 5-14A and B). These characteristic cytokine profiles accompany a reduction of the Th17 response towards a more classical Th1 cytokine repertoire, which may be less harmful to self tissues (Irmler, 2007; Rosloniec, 2002).

Figure 5-14 Limited influence of IL-21 on Th1 cytokines at mucosal sites Quantification (left) and representative dot plots (right) of intracellular (a) IFN and (b) Intracellular TNF flow cytometry staining in CD4+ CD3+ T cells in from the LPL populations between 8-12 weeks of age. Lymphocytes were isolated from the lamina propria and stimulated with PMA and ionomycin for 4 hours. Data are pooled values from individual mice from 5 experiments and the mean -/+ SEM. Groups were analysed by one way non-parametric ANOVA for unequal variance and no significant difference between groups was found.

5.2.6 IL-21 shapes the serum cytokine profile during autoimmunity

CD4+ T cells can shape an immune response to a particular pathogen by differentiation into various helper populations in response to antigen. Gene-targeted “knock-out” mouse models of cytokines such as IL-4, which helps drive a Th2 response, or IFN, which is important for Th1 responses, demonstrate that an altered immune phenotype arises at many levels of the innate and adaptive immune systems to influence CD4+ T cell differentiation (Mosmann, 1991). To dissect how IL-21 shapes the global serum cytokine profile during colitis we measured levels of a variety of inflammatory cytokines using a flow cytometry cytokine bead array (CBA) method that compared relative mean fluorescence intensity at discrete detection intervals in serum samples. Reduced CD4+ T cell derived IL-17A had previously been detected in the absence of IL-2 and IL-21R in our model by flow cytometry (Figure 5-13), and this was confirmed by a reduction in serum levels detected by CBA. Surprisingly, this was not the case with the Th17 associated

140 cytokine IL-22, which was significantly higher in Il2 Il21r-/-, and to a lesser extent in Il21r-/-, implying that IL-21 may limit IL-22 production (Figure 5-15). IL-22 is a newly described cytokine produced by CD4+ T cells that can act on both immune and epithelial cells to promote angiogenesis and fibrosis. IL-22 production without IL-17 has also recently been described in human lymphocytes in the skin, activated by Langerhans cells (Fujita, 2009; Trifari, 2009), and human IL- 22 and TNF producing Th22 cells have also been recently described to induce genes that modulate innate immune responses, as well as enhancing wound healing and tissue remodelling in the skin (Eyerich, 2009; Pickert, 2009). This altered IL-17A:IL-22 balance in our Il2-/- strains suggests that IL-21 drives IL-17A production, perhaps at the cost of IL-22, which can help regenerate tissues damaged by infiltrating immune cells in the mucosa. In support of this notion, we observed enhanced fibrosis in Il2 Il21r-/- in our histological analyses of the pancreas though it is unclear whether this fibrosis was evidence of tissue healing or rather scarring as a result of ongoing inflammation (Figure 5-9). We attempted to perform intracellular cytokine staining for IL-22 but were not able to detect staining in any genotype above background levels due to a paucity of specific antibody binding.

Figure 5-15 Altered serum cytokine profile in Il2-/- mice Cytokine bead array using a flow cytometry assay for serum cytokine levels. Data represents mean fluorescence intensity (MFI) from serum samples from individual mice for each cytokine with mean +/- SEM mice vary in age between 6-28 weeks, but only 6-18 weeks in Il2-/- due to increased mortality. P values were calculated using one-way ANOVA and Bonferroni’s post-test to compare all groups. (*p<0.05, **p<0.01***p<0.001).

141 IL-21 also seems to drive the immune response away from classic Th1 cytokines such as TNF and IFN, that are associated with inflammation and sometimes tissue pathology (Liblau, 1995), but seem to correlate with improved outlook in the absence of IL-21 (Figure 5-15). These cytokines may well be derived primarily from CD4+ T cells as the serum levels reflect earlier intracellular staining of this population ex vivo (Figure 5-14). These two cytokines were significantly higher in Il2 Il21r-/- serum, and may be the lesser of two evils in our model (Figure 5-15). IL-10 is often characterised as a regulatory cytokine, due to its ability to downregulate antigen presenting cell activation and inflammatory cytokine secretion (Wilson, 2005, Hunter, 1997). IL-21 and other STAT3 signalling cytokines have been seen to drive an IL-10 production in vitro, which can ameliorate disease due to its immunosuppressive properties (Hunter, 1997; Pot, 2009; Spolski, 2009; Stumhofer, 2007). Several regulatory CD4+ T cell subtypes such as Tr1 cells and Tregs utilise IL-10 as a mechanism to limit inflammation (Groux, 1997), so we measured relative serum IL-10 to see whether the IL-21 was influencing this cytokine in our model of colitis driven by an absence of peripheral regulation in Il2-/- mice.

Figure 5-16 IL-21 does not effect IL-10 production by CD4+ T cells Representative flow cytometry images of CD3+CD4+ T cells showing intracellular IL-10 staining of cells isolated from the MLN of 9 week old mice and stimulated with PMA and ionomycin for 4 hours. Representative of 3 experiments in which n=3.

IL-10 mfi were not significantly different between healthy and diseased genotypes in our model, although we saw a slight increase in IL-10 in the healthier Il2 Il21r-/- strain compared to Il2- /- sera (Figure 5-15). We were unable to detect any differences in IL-10 secretion ex-vivo in the CD4+ T cell population, so the observed differences must be attributed to alternative T cell or myeloid sources of IL-10 (Figure 5-16). The pro-inflammatory cytokines GMCSF and IL-1 were also increased in the absence of IL-21, despite the improved lifespan in these mice (Figure 5-15).

142 5.2.7IL-21 drives expansion of the autoimmune CD8+ T cell compartment

Double CD25-/- and MCHI-/- or cd8-/- knockout mice incapable of priming CD8 T cells to antigen still suffered from early mortality and colitis in previous studies (Hsu, 2009) (Simpson, 1995). However, although they do not cause colitis in this model, CD8+ T cells were shown to drive cytotoxic tissue destruction at other sites such as the bile ducts of the liver, so we decided to examine how the CD8 cells phenotype in Il2 Il21r-/- had been affected by the absence of IL- 21(Hsu, 2009). Perhaps driven by the changes in CD4+ T cell activation and altered serum cytokine levels, activated CD44hi CD8+ T cells were reduced in some mucosal sites, such as the MLN and the epithelium in the absence of IL-21 (Figure 5-17A). This was accompanied by a slight reduction in the proportion of CD8+ cells that displayed CD122+ CD44hi central memory phenotype in the absence of IL-21 in both IL-2-/- strains (Figure 5-17B). We found this particularly interesting, as recent reports have identified IL-21 as a survival factor for CD8+ T cells in chronic as opposed to acute viral infections, which this model resembled in terms of the longevity of the immune activation against microflora in the mucosa and the surrounding tissues (Elsaesser, 2009; Yi, 2009).

Figure 5-17 IL-21 drives CD8+ T cell expansion in Il2-/- mice (a) Quantification of CD8+ CD3+ T cells analysed by flow cytometry in various genotypes between 8-12 weeks of age. Data are pooled from 5 separate experiments and are represented as values from individual mice -/+ SEM. (b) Representative dot plots from experiments described above showing expansion of the CD122+CD44hi population gated on CD3+ CD8+ T cells from the spleen. P values were calculated using one-way ANOVA and Bonferroni’s post-test to compare to WT values (***=p<0.001).

143 IL-21 has also been shown to be critical for the homeostatic proliferation of CD8+ T cells observed previously in diabetic NOD mice (King, 2004). To explore this phenomenon further, we performed adoptive transfer experiments of 106 CD122+ CD44hi CD8+ T cells into WT, Il21r-/-, IL-2-/- or IL2 Il21r-/- hosts to address whether IL-21 could drive CD8+ T cell proliferation in vivo. Donor cells were isolated from IL-7tg donors as their T cell compartment is rich in CD122+ CD44hi CD8+ T cells due to the heightened levels of this homeostatic cytokine (Mertsching, 1996; Schluns, 2000). In this strain, IL-7 is expressed under the mouse MHC II promoter, which drives constitutive expression, mainly by thymic stromal cells. Interestingly, even in healthy syngeneic Il21r-/- hosts, some transferred CD8+ T cells reproducibly underwent 3-4 rounds of division within 7 days, possibly driven by high circulating IL-21 levels in the absence of the host absorbance via IL21R (Figure 5-18), although we were not able to confirm this due to the previously described difficulty in measuring IL-21 in serum by ELISA. This response was amplified in autoimmune Il2 Il21r-/- hosts, where barely any undivided donor cells could be detected at the end of the 7 day study, presumably due to even higher levels of IL-21 in the serum that may be expected to arise from the increased Il21 mRNA transcript found in lymphoid tissues of Il2-/- mice compared to WT (Figure 5-1). We speculate that this may be produced by the constitutively activated and expanded CD4+ T cell effector population in these mice combined with an inability to absorb ligand in the absence of the receptor (Figure 5-18 and Figure 5-12). This phenomenon was also seen when normal C57BL/6 CD8+ T cells were transferred, and the CD44hi population analysed for CFSE dilution within the same timeframe, ruling out the need for prior conditioning of memory cells within the artificial IL7tg donor.

Figure 5-18 Il21r-/- hosts drive proliferation of adoptively transferred memory CD8+ T cells 1x106 CFSE labelled Thy1.1+CD122+CD44hi congenic CD8+ T cells were purified from IL-7tg hosts and adoptively transferred into various 8 week old hosts as labelled above and analysed after 7 days for proliferation by flow cytometry. (a) Quantification of the number of donor CD8+ T cells present in the MLN after 7 days and (b) representative histograms gated on the CD44hi cells of donor CD8+ T cells showing the percent of dividing cells. Data is representative of 2 experiments in which n=2-3. P values were calculated using one-way ANOVA and Bonferroni’s post-test compared to WT values (***=p<0.001).

144 Il21r-/- CD8+ T cells were then adoptively transferred into WT, Il2-/- and Il21r-/- hosts to determine whether IL-21 was responsible for homeostatic expansion in resting and autoimmune mice. A mixed CD8+ T cell donor cohort was used due to limited numbers of memory T cells in this mice, CFSE dilution was analysed on the gated CD44hi population to simulate the memory cells were able to purify from IL-7tg mice in previous experiment. Removing the ability of donor cells to respond to IL-21 completely abrogated the proliferation previously observed in Il21r-/- hosts, suggesting IL-21 alone drives this expansion (Figure 5-19). However, Il2 Il21r-/- hosts still drove CFSE dilution, albeit to a far lesser extent, implying that both IL-21 and other environmental factors present during inflammation, but not in the resting mouse, were promoting expansion of donor CD8+ T cells in these mice (Figure 5-19).

Figure 5-19 High circulating IL-21 drives homeostatic proliferation of CD8+ T cells 1x106 purified and CFSE labelled Il21r-/- CD8+ T cells were adoptively transferred into various 8 week old hosts as labelled above and analysed after 7 days for proliferation by flow cytometry. Representative histograms gated on donor CD8+ T cells in the MLN showing the percent of dividing cells. Data is representative of 2 experiments in which n=2-3.

It was interesting to note that single Il2-/- hosts did not support donor CD8+ T cell proliferation (Figure 5-18 and Figure 5-19) even though transcript IL-21 is highly expressed in these mice (Figure 5-1). This could be explained by two different hypotheses; Firstly, IL-21 may be quickly utilised in IL21R competent mice so that even ongoing immune responses do not lead to high circulating IL-21 levels that could drive global CD8+ T cell expansion. Alternatively, IL-21 may drive expansion of CD8+ T cells in an IL-2 dependent manner, by increasing IL-2 production or sensitivity rather than directly action through the IL21R. To this end, we adoptively transferred Il2-/- CD8+ T cells to see whether cell autocrine IL-2 signals were necessary to respond to high IL- 21 levels. Il2-/- donor CD8+ T cells underwent several rounds of division in Il21r-/- hosts, to around 80% of WT CD8+ T cell levels implying that host IL-2 was sufficient to support expansion of these cells (Figure 5-20). However, the Il2 Il21r-/- hosts which provide copious IL-21 but no IL- 2 had half the amount of divided cells after 7 days compared to WT donors (Figure 5-20),

145 suggesting that IL-2 from either host or donor is required for CD8+ T cells to undergo the large expansion observed in this host with WT donors (Figure 5-18). In terms of cytotoxic activity of CD8+ T cells, IL-21 did not seem to play a role in Il2-/- pathology as both Il2-/- and Il2 Il21r-/- cells produced equivalent amounts of TNF, IFN and granzyme B ex vivo in response to stimulation with PMA and ionomycin (data not shown).

Figure 5-20 Both IL-2 and IL-21 are necessary for CD8+ T cell proliferation 1x106 purified and CFSE labelled Il2-/- CD8+ T cells were adoptively transferred into various 8 week old hosts as labelled above and the MLN analysed after 7 days for proliferation by flow cytometry. Representative histograms gated on donor CD8+ T cells showing the percent of dividing cells. Data is representative of 2 experiments in which n=2-3.

5.2.8 IL-21 promotes T dependent antibody production in Il2-/- mice

Some B cells express the high affinity IL-2 receptor CD25, but the progressive loss of B cells in Il2-/- mice is independent from its expression, and therefore direct action of IL-2 on B cells (Schultz, 2001). In the Il2-/- model, B cells are initially stimulated to make large amounts of class switched antibody, particularly the IgG1, IgG2a and IgE isotypes that are characteristic of a classic Th2 response. This antibody seems to target self-antigens, indeed 80% of mice that died between 6- 9 weeks were diagnosed with haemolytic anaemia on autopsy (Sadlack, 1993). This harmful B cell activation is ameliorated in surviving older mice as the B cell population undergoes a rapid dissolution, perhaps due to overcrowding of the bone marrow with activated thymocytes, inhibiting survival or development of IgM-B220+ B cell precursors (Schultz, 2001). We looked at B cell numbers and activity in Il2 Il21r-/- mice to determine how IL-21 may affect B cells and antibody production in this model of autoimmunity.

146 Figure 5-21 B cell numbers in Il2-/- mice increase in the absence of IL-21 Quantification of CD19+ B cell numbers in spleen, MLN, intra-epithelial lymphocytes (IEL) and the colon lamina propria analysed by flow cytometry in various genotypes between 8-12 weeks of age. Data are pooled from 6 separate experiments and are represented as values from individual mice -/+ SEM. P values of <0.05 (*) or <0.001(***) were calculated by one-way ANOVA and Bonferroni’s post test compared to WT values.

B cell numbers remained at slightly higher levels in Il2 Il21r-/- mice than Il2-/- between 9 and 12 weeks of age, both in the spleen, MLN and also at mucosal sites (Figure 5-21). The phenotype of the B cells at this timepoint was also quite different, as both strains of Il2-/- mice were rich in GC and plasma phenotype cells, despite the depleted total B cell population in these mice. However, GC B cells and plasma cells were reduced in the absence of IL-21 in both number (Figure 5-22) and as a proportion of B cells (Figure 5-23).

147 Figure 5-22 IL-21 drives GC B cell and plasma cell generation or survival in Il2-/- mice

Quantification of GL7+B220+ GC (left) and B220lo Syndecam-1+ plasma cell (right) numbers in MLN analysed by flow cytometry in various genotypes between 8-12 weeks of age. Data are pooled from 2 separate experiments in which n =2-3 and are represented as values from individual mice -/+ SEM. * = p value below 0.05 calculated by non parametric ANOVA and Dunn’s post test to compare groups.

Figure 5-23 IL-21 enhances GC and plasma B cell proportions in Il2-/- mice (A) Representative flow cytometry showing (A) GL7+B220+ GC and (B) B220lo Syndecan1+ plasma cell populations gated on total B220+ and lymphocytes respectively from the MLN at 11 weeks of age. Representative of 2 separate experiments (n=4-5).

148 Our previous work demonstrated that IL-21 was an important survival factor for Tfh cells during responses to various antigens, which led us to investigate Tfh cell numbers in the Il2-/- model where they could drive antibody production against self antigen or harmless commensal micro-organisms. Several key distinguishing Tfh markers including CXCR5, ICOS (Figure 5-24) and PD-1 were highly expressed on Il2-/- CD4+ T cells in an IL-21 dependent manner as these markers were reduced in Il2 Il21r-/- littermates. This unusual upregulation of Tfh markers in the absence of immunisation may suggest these T cells are selecting B cells in GC inappropriately in response to self or commensal organisms in the gastro-intestinal tract. It should be noted that these markers are all also designated as early response markers and may alternatively label CD4+ T cells that have recently been exposed to antigen.

Figure 5-24 IL-21 supports Tfh like cells in Il2-/- mice Representative flow cytometry showing ICOS+CXCR5 Tfh populations gated on total CD3+CD4+ T cells from the MLN in various genotypes at 11 weeks of age. Representative of 2 separate experiments (n=4-5).

However, the measurement of serum antibody levels using ELISA suggested a functional IL-21 dependent, autoimmune Tfh population in Il2-/- mice. These findings showed that despite large ongoing immune responses in Il2 Il21r-/- mice, the T dependent immunoglobulin isotypes IgG1 and IgG2c were comparable to resting WT levels rather than the high circulating levels observed in the Il2-/- strain (Figure 5-25). This was true both at the peak of the B cell immune response in Il2-/- mice under 11 weeks of age when significant numbers of B cells remained, and also for the few long lived antibody secreting cells that had survived at the later timepoints studied. IgA is typically produced at mucosal sites due to its stable dimer formation that allows transport into the GI tract and resistance to high pH, and can be produced in response to T dependent and independent antigens (Suzuki, 2009). IL-21 also seemed to play a role in IgA production as levels were significantly reduced in Il2 Il21r-/- serum (Figure 5-25). IgM was particularly affected by the reduced B cell number in Il2-/- strains with typically low levels observed both in the presence and absence of IL-21 (Figure 5-25). IgE was similarly unaffected by

149 IL-21 and remained high in both Il2-/- strains, with no evidence for a further increase in IgE production that may have been expected considering the previously described increase seen in single Il21r-/- knockouts. This isotype does not seem affected by the increase in Tfh like cells, which agrees with reports that IgE class switch and production can result from both T dependent and independent pathways (Geha, 2003).

Figure 5-25 IL-21 promotes T dependent antibody isotype production in Il2-/- mice

Quantification of the log10 of serum antibody isotype concentrations or titre measured by ELISA in young (<11 weeks) and mature (11-30 weeks) mice. Data indicate the mean and values from individual mice pooled from 2 separate experiments. P values were calculated comparing similar age groups using Student’s t test.

150 5.3 Discussion

Our current understanding of human inflammatory bowel disease proposes a combination of genetic predisposition, and environmental interactions with the mucosa and immune system. The observation that germ free Il2-/- mice have delayed and milder intestinal infiltration indicates that this model of colitis is not autoimmune insofar as self-reactive T cells are still deleted in the thymus (Kramer, 1995), but rather a defect in peripheral tolerance mechanisms towards mucosal commensal antigens (Contractor, 1998; Schultz, 1999). The caveat inherent in the study of gastrointestinal disorders in germ free mice is that commensal micro-organisms are required for the normal development of the GI mucosal immune system, including gut associated lymphoid tissue (GALT) and peyers’ patches (Rhee, 2004). Subsequent analyses of germ free Il2-/- mice with mono-cultures of specific commensals supported the notion that gut microbes modulate the progression of colitis and T cell activation in this strain (Hans, 2000b) (Muller, 2008; Waidmann, 2003). Mild histological evidence of colitis and severe pancreatitis observed in our Il2-/- colony may reflect this important role for commensal organisms, which differ between research institutes despite identical genetic B6 background. In our colony the mortality rate only reached 50% at 12 weeks rather than the 9 weeks described by Sadlack et al when this strain was first developed (Figure 5-2). Differences between Il2-/- and Il2 Il21r-/- strains however, can be compared within the same SPF conditions to dissect the specific role of IL-21 within the unique conditions of our colonies. Certainly both strains showed that increased cellularity in lymphoid organs consisting largely of activated IL-21 producing CD44hiCD4+ T cells that co-expressed the mucosal homing integrin 47. This finding was in accordance with mucosal priming driving the enlarged activated CD4+ T cell population in both strains (Figure 5-12). The extensive mononuclear cell infiltrate, atrophy of parenchymal cells and fibrosis we observed in the pancreas of Il2-/- were a novel finding, and it would be particularly interesting to see if it is this is present in other colonies with different SPF conditions (Figure 5-8 and Figure 5-9). Both Il2 and Il2 Il21r-/- strains showed increased circulating inflammatory cytokine profiles, which was modulated by the presence of IL-21 in the system. The endocrine and exocrine tissue of the pancreas are known to be sensitive to chronic inflammatory antigen exposure, for example, IFN, TNF and IL-1 can induce apoptosis of pancreatic beta cells in vitro (Chong, 2002; Sadlack, 1993; Thomas, 2002). Introduction of constitutive IFN production in the pancreas via an islet specific transgene also promoted development of a severe pancreatitis (Sarvetnick, 1988). Interestingly, our Il2 Il21r-/- model showed decreased IL-17A in serum compared to Il2-/- littermates, and a despite increasing of systemic levels of IFN, IL-1 and TNF (cytokines with 151 proven harmful effects on the pancreas), the presence of IL-21 seemed to exacerbate pancreatic pathology. Further studies are required to characterise the infiltrate, for example histology and flow cytometry could be used to assess the numbers and location of component CD4+ and CD8+ T cells, and also B cells which could produce antibody complexes which have also been attributed to some models of pancreatic pathology, particularly in humans (Endo, 2009; Takahashi, 2009). Our assay of suppressive function of Il21r-/- Tregs suggested that IL-21 was not necessary for in vitro function, so it was surprising to discover that in spite of increased longevity and weight gain in Il2 Il21r-/- mice, there was a further decrease in the remaining CD25-foxp3+ cells as a proportion of total CD4+ T cells compared to Il2-/- (Figure 5-10). CD25-foxp3+ cells from Il2-/- mice have previously been shown to retain suppressive function in in vitro assays, but fail to thrive and prevent the lymphoproliferation that characterises Il2-/- mice in vivo (Fontenot, 2005). Foxp3 dependent genes such as TGF, granzyme A and perforin were reduced in Il2-/- T regs (Fontenot, 2005) and may explain their deficient expansion or fitness in the periphery in vivo, and further study is required to see if these factors might be further affected by the lack of IL-21 in our model. Clearly other circulating c-chain cytokines, especially in this autoimmune setting could somehow promote Foxp3 expression, as Il2-/- and Il2 Il21r-/- mice retained similar Foxp3+ Treg numbers whereas they are absent in c-chain KO mice (Fontenot, 2005). These results challenge the relevance of previous reports that IL-21 plays a role in Foxp3 transcript upregulation (Bucher, 2009; Kim-Schulze, 2009; Li and Yee, 2008; Peluso, 2007; Piao, 2008) In accordance with previous reports that IL-21 and other STAT3 signalling cytokines can enhance IL-10 production, resting serum IL-10 levels tended to be slightly decreased in Il21r-/- compared to WT (Spolski 2009, Pot C 2009, Stumhofer JS 2007). Unexpectedly IL-21 also seemed to negatively regulate IL-10 production in certain conditions, as serum IL-10 was similar in Il2 Il21r-/- mice despite having the lowest proportion of Foxp3+ cells. IL-10 deficiency bears some striking resemblances to the Il2-/- model of colitis, as this strain responds to mucosal T. gondii infection by over-production of inflammatory cytokines resulting in necrosis derived from T cell infiltrates (Gazzinelli, 1996). IL-10 producing Tr1 cells have also been shown to ameliorate colitis (Groux, 1997). Therefore it could be significant that removing IL-21 led to slightly increased circulating IL-10 levels in Il2 Il21r-/- mice compared to Il2-/- littermates as this correlates with increased survival and weight gain. As this increase could not be ascribed to CD4+ T cells, further study is required to identify which other cell populations may be differentially producing IL-10 in Il2 Il21r-/- mice. IL-10 from a variety of sources, even derived from genetically modified bacteria can ameliorate colitis symptoms (Steidler, 2000). The observation that CD4+ T cells from Il2-/- and

152 Il2 Il21r-/- mice showed no significant differences in IL-10 levels, agrees with previous reports that Il10 mRNA in Tregs was not dependent on IL2:IL2R interactions (Fontenot, 2005). IL-21 is largely produced by CD4+ T cells, which can also express IL-21R. It is thus perhaps unsurprising that many of the changes observed when IL-21 deficiency was introduced into Il2-/- seemed to reflect autocrine effects on CD4+ T cells and the flow on effects of these may drive other changes in downstream effector cell subtypes such as CD8+ T cells and T dependent antibody responses (Leonard and Spolski, 2005; Parrish-Novak, 2000). Initial T cell priming, proliferation and homing seemed unaffected by IL-21’s presence or absence. By removing Il-21 from this system we could infer that IL-21 drives IL-17A production, particularly at mucosal sites of inflammation, and that correlates with a slight drop in classic Th1 type cytokines such as IFN and TNF (Figure 5-13 and Figure 5-14). IL-17A producing Th17 cells produce IL-21 which contributes to autocrine expansion of this cell subset, explaining why IL- 17A may be reduced in our Il2 Il21r-/- model (Korn, 2007a; Wei, 2007). It would be interesting to analyse mucosal CD4+ T cells in this strain for the ROR transcription factor expression that characterises the Th17 to see if it is also decreased. A reduction in Th17 cells could contribute to decreased morbidity, as IL-17A or IL-17F producing cells can induce chronic intestinal inflammation when adoptively transferred into RAG-/- mice (Leppkes, 2009), and IL-17 transcript was increased in the mucosa of Crohns’ disease patients (Veny, 2009). IL-22 is expressed by Th17 cells in concert with IL-17A, and was shown to be important for ameliorating tissue damage in models of liver toxin exposure and colitis (Marks BR 2009). IL- 22 has also recently been identified in a new subset of human skin-derived CD4+ T cells that do not co-express IL-17 (Eyerich, 2009; Fujita, 2009). We are particularly interested in the latter phenomenon as systemic cytokine analysis showed that IL-22 and IFN were strongly induced in Il2 Il21r-/- without IL-17A induction. IFN has complex influences in the immune system and can both enhance or prevent autoimmune diseases depending on the model (Irmler, 2007; Rosloniec, 2002). Further study is required to see if blocking IL-21 in this autoimmune setting may promote tissue healing via IL-22 while reducing the cytotoxic effects of IL-17. Studies in previous chapters strongly suggest that IL-21 has the ability to sustain a CXCR5+ICOS+ Tfh population in response to immunisation (Nurieva, 2007; Vogelzang, 2008), and this may also seems to be the case in Il2-/- mice where T dependent class switched antibodies contribute to haemolytic anaemia. Certainly Tfh markers were increased in Il2-/- lymph nodes and spleen in an IL-21 dependent manner (Figure 5-24). Balb/c Il2-/- mice die from a more severe and earlier onset haemolytic anaemia which could be prevented by anti-CD40L treatment indicating a strong causative link with B-T cell interactions and disease progression (Sadlack, 1995). In this

153 paper, the T cells were described as CD69+ CD44hi, which may correlate with our expanded CD44hi CXCR5+ ICOS+ population, and suggests these cells must be a target of immune suppression in healthy mice. These studies will be followed by analysis of the Tfh transcription factor Bcl-6 to confirm that these cells represent bona fide Tfh rather than merely recently antigen exposure by CD4+ T cells (Johnston, 2009; Nurieva, 2009; Yu, 2009). We expect to find these are Tfh, as T dependent isotype antibody responses in the MLN draining the inflamed region in Il2-/- mice were strongly affected by IL-21, which also supported expanded GC B cells and plasma cells (Figure 5-22) though it remains unclear whether this is through signalling via T or B cells. Further bone marrow chimera studies are underway to assess the contribution of IL-21 signalling through IL-21R on either B or T cells drives antibody production in this model. Despite lower antibody concentrations in serum, Il2 Il21r-/- B cells survived longer than their Il2-/- counterparts (Figure 5-21). It is not known why B cell disappear in Il2-/- but it may be due to overcrowding of B cell progenitors in the bone marrow which becomes infiltrated with mature T cells in this model (Schultz, 2001). Future experiments planned in our lab include phenotyping the bone marrow infiltrate in Il2 Il21r-/- to discover whether IL-21 may affect T cell homing to the bone marrow and the resulting B cell attrition. High circulating IFN, IL-4 and IL-10 may drive the antibody response in Il2 Il21r- /-, as they have all been implicated in class switching and plasma cell formation and may explain how IgE remains elevated in the absence of IL-21 (Caven, 2005; Caven, 2007; Geha, 2003). However, IL-21 is non redundant for IgG1 and IgG2c production, which may reflect its important role in sustaining T cell help for these T dependent isotypes, but also expansion and survival of plasma cells. Work in previous chapters has outlined an important role for IL-21 in the generation of GC, memory and plasma B cells via its actions supporting the Tfh phenotype in CD4+ T cells. The Il2-/- anaemia is independent of microbiota, as it is present in germ-free mice and thus constitutes true autoimmune activation in absence of regulation (Contractor, 1998). Further studies are required to ascertain whether GC structures in peripheral lymph nodes differ in Il2-/- and Il2 Il21r-/- mice, and whether autoantibody production is reduced. In future we plan to measure hematocrit levels in Il2-/- and Il2 Il21r-/- to ascertain whether the IL-21 dependent IgG1 and IgG2c causes the haemolytic anaemia in Il2-/- mice. It is probable that expanded Il2-/- memory phenotype CD8+ T cells are also a target of immune regulation by Treg cells in the periphery in B6 mice. Memory CD8+ T cells are CD44hiCD122+ and require IL-7 and IL-15 for survival and homeostasic proliferation in resting hosts (Surh and Sprent, 2008). Yet much like transgenic IL-21 over-expressing mice, IL-21 seems to support an unusual accumulation of memory phenotype CD8+ T cells that are CD44hiCD122+ in

154 autoimmune Il2-/- mice (Allard, 2007). BrdU incorporation studies would be useful to see if there is constitutive IL-21 dependent turnover of the Il2-/- memory CD8+ T cells. CD8+ T cells do not express the high affinity IL-2 receptor CD25, but they do express CD122, the low affinity IL-2 receptor which signals in concert with the c-chain or the IL-15R. IL-2 can utilise this receptor on CD8+ T cells, as injection of an agonistic IL-2:monoclonal antibody complex can cause proliferation of WT memory phenotype CD8+ T cells (Boyman, 2006). By utilising adoptive transfer into our panel of Il2-/- and Il21r-/- mice, we were able to identify a role for IL-21 in driving appreciable CD8+ T cell proliferation in vivo in Il21r-/- hosts, in a manner that also strongly relied on the presence of IL-2 but also other unidentified stimuli (Figure 5-18). The transferred cells were derived from a non-autoimmune background, so presumably had been previously tolerised to B6 self antigens. We plan to address whether interactions with self-MHC molecules are necessary by crossing Il21r-/- mice onto a MHC II deficient background. It is possible to measure IL-21 levels in vitro by ELISA, so it would be interesting to see if activated Il21r-/- T cell cultures accumulate IL-21, as we suspect may be occurring in vivo leading to IL-2 dependent turnover of adoptively transferred WT CD8+ T cells.

155 6 General Discussion

6.1 Research outcomes

The adaptive immune response is generated by T and B lymphocytes, which can provide specific protection against particular pathogens as well as protective memory against reinfection. The cytokine network coordinates this response by controlling the differentiation of lymphocytes through their acquisition of specialised effector functions to establish immunity. The c-chain cytokine IL-21 is produced almost exclusively by CD4+ T cells but can act on a broad range of immune cells due to widespread expression of its unique receptor chain, namely IL-21R alpha (Leonard and Spolski, 2005). The broad aims of this study were to define how IL-21 acts during protective immune responses and also to examine its role in autoimmunity. We examined the ability of CD4+ T cell derived IL-21 to act in an autocrine manner to enhance TCR signals, and in a paracrine fashion to support B lymphocyte driven humoral as well as CD8+ T cell responses. These studies began with experiments in Il21-/- and Il21r-/- mice that were designed to confirm previous incomplete reports that IL-21 does not play a role in development of the immune system (Kasaian, 2002; Ozaki, 2002). Indeed, normal development and homeostasis of T and B cell mature compartments was observed in the lymphoid and myeloid compartments of these strains. IL- 21 producing CD4+ T cells were characterised in normal healthy B6 mice, and were most commonly found in the lymphoid structures draining the gut mucosa, the MLN and especially the peyers’ patches rather than in the spleen. IL-21 is produced by Th17 cells generated in the mucosa in response to TGF-, IL-6 secreting APC (Monteleone, 2009), and potentially, CD4+ T helper cells that stimulate IgA production. The importance of these two subsets in humoral and cellular protection against microbes crossing epithelial barriers in the gut may drive this increased IL-21 expression in the peyers’ patches. IL-21 producing cells differed in their surface marker phenotype depending upon their organ of origin; those found in the spleen and peripheral lymph nodes displayed a CD44+ memory phenotype combined with the expression of both ICOS and CD62L, which suggested they were sessile central memory B cell helpers, whereas those found in the peyers’ patches were a mixture of recently activated migratory, effector memory CD4+ T cells, expressing CD44 and CD69 but lacking CD62L. However, the vast majority of IL-21+ cells in healthy mice seemed to be primed in non-mucosal secondary lymphoid organs or destined for non-mucosal tissue sites as only 10-20% of IL-21-producing cells expressed the mucosal homing markers CCR9 and 47 (Campbell and Butcher, 2002; Papadakis, 2000). Hence, IL-21 expression by CD44+ memory phenotype cells in 156 the normal resting immune system in SPF conditions was acquired by T cells primed in organised lymphoid tissues and was also associated with a helper T cell phenotype in peripheral lymphoid organs. This study of IL-21 protein expression directly ex vivo, reflects previous studies showing Il21 transcript produced at very high levels by Tfh cells (Chtanova, 2004; Korn, 2007a), and also recent reports that these cells can form a long lived memory population that remains in the lymph node near deposits of antigen (Fazilleau, 2007a). Data from human in vitro assays suggest early IL- 12 signalling can upregulate IL-21 production and other Tfh markers by naive CD4 T cells (Ma, 2009a; Schmitt, 2009), but further study is required to determine exactly what signals derived from various subsets of DC are required to induce sustained IL-21 production upon priming in vivo. Findings presented in chapter 3 and 4 suggested that IL-21 acted in an autocrine fashion to drive the differentiation of CD4+ T cells, supporting the phenotypic changes in the expression of molecules that define helper T cells such as ICOS, CTLA-4 and CCR7. We also saw a marked requirement for IL-21 for the retention of CXCR5 expression, with consequent downstream effects detected on CD4+ T cell positioning in the GC (Vogelzang, 2008). Further studies are planned in our laboratory to dissect whether IL-21 dependent CXCR5 expression is the dominant role of IL- 21’s actions on Tfh cells. We plan to study the migration of Il21r-/- OTII cells that have been manipulated to transgenically induce constitutive CXCR5 expression (Haynes, 2007), to see whether this rescues defects in Il-21r-/- Tfh cell migration. This study confirms previous reports that IL-21 costimulates the TCR-induced proliferation of CD4+ T cells (Parrish-Novak, 2000), indicating that IL-21 may mediate its effects on T helper generation by potentiating signals through the TCR. IL-21 can modulate TCR signal strength in mature peripheral T cells through activation of Vav1, which we predict might intersect to increase P13K signalling and cement effector phenotype differentiation (Figure 6-1). Similar to studies with Vav1 specific T cells (Gulbranson-Judge, 1999; Villalba, 2001), downstream effects on humoral responses were shown to have a significant CD4+ T cell intrinsic component. However this modulation of TCR signal did not appear to play a role in thymic selection, as CD5 levels, which reflect TCR signalling strength in the thymus, were normal in IL-21 deficient mice. Redundancies between the Vav1 family proteins may explain why thymocytes were unaffected by this IL-21 deficiency, indeed in Vav1 deficient mice there is only a partial block in thymic development at the DN stage, and reduced thymic output only occurs in complete Vav1/2/3 deficiency (Fischer, 1995; Kong, 1998; Tarakhovsky, 1995b; Turner, 1997; Zhang, 2003). The effects of IL-21 on T cells began to be apparent at the point of acquisition of effector function by CD4+ T cells.

157 Figure 6-1 Schematic diagram showing that IL-21:IL21R interactions activate Vav1 and downstream chemokine receptors and co-stimulatory molecules, which define the Tfh cell subset.

The GC requirement for IL-21 was found to reflect a CD4+ T cell-intrinsic requirement for

IL-21-driven Tfh cell generation. The data presented in chapter 3 and 4 definitively demonstrated a need for IL-21 in the GC reaction, specifically signalling through the IL-21R on T cells. This thesis outlines a role for IL-21 to optimise TCR signals in mature CD4+ T cells, driving sustained expression of Tfh cell markers (Figure 6-1). Vav1 is vital for the activation, differentiation and cytokine production by mature T cells, and hence T dependent antibody responses and cytotoxic protection from viruses (Gulbranson-Judge, 1999; Penninger, 1999; Tybulewicz, 2003). By increasing Vav1 phosphorylation alongside TCR signals, IL-21 could enhance Ca2+ flux and ERK MAPK pathway activation, which have been shown to be downstream of Vav1 phosphorylation (Charvet, 2002; Tybulewicz, 2003). The high expression of T cell costimulation factors such as ICOS and IL-21 by Tfh cells supports research that demonstrated the selective generation of Tfh cells by high affinity or sustained TCR interactions with antigen (Fazilleau, 2009). This may also be important later in the maintenance of the Tfh cell phenotype later in the GC reaction where these same strong co-stimulatory signals may deflect PD-1 induced ‘exhaustion’ and death of this fragile T cell subset due to constant BCR:TCR engagement (Marinova, 2006; Okazaki and Honjo, 2006). There are 2 pathways through which Vav1 signalling is known to control ERK; Reduced phospholipase C1 (PLC1) activation and diacylglycerol second messenger leads to reduced activation of Ras, MEK and finally, ERK (Reynolds, 2004; Reynolds, 2002). Vav1 is also necessary for optimal recruitment of guanosine exchange factors to the linker for activation of T cells (LAT),

158 perhaps by increasing phosphorylation of LAT (Reynolds, 2004). Downstream of ERK activation are a number of transcription factors and pro-survival molecules, including Bcl-xL, that are important for the survival of antigen experienced CD4+ T cells. It would be interesting to see in future studies whether IL-21 sustains elements of the ERK phosphorylation pathway, antagonising PD-1 activity by maintaining Bcl-xL upregulation by antigen specific T cells. ICOSL was found to be necessary for maximum induction of IL-21, and interestingly, given the importance of these two molecules for B cell help, Tfh cells exhibited very high expression of both IL-21 and ICOS. This may reflect the ability of ICOS and IL-21 to form a positive feedback loop enhancing their reciprocal expression. Creation of the autoimmune sanroque strain demonstrated that strong ICOS signalling makes IL-21 redundant for T helper cell activation as Il21-/- knockout mice that also carried the mutant sanroque allele showed no alteration in increased Tfh numbers (Linterman, 2009). These studies show that while IL-21 is needed for upregulation of ICOS on T cells activated by TCR signals in vitro, other factors could compensate for this defect in vivo. The combination of these two studies implied that IL-21 may act before ICOS in Tfh cell differentiation, or that unnaturally strong ICOS costimulation present in the sanroque strain could compensate for any need for costimulation through the IL-21R. The sanroque model and those presented in this thesis coalesce by demonstrating that IL-21 was not necessary for B cells to participate in pathogenic GC formation (Linterman, 2009) or GC reactions following T dependent immunisations. We observed that adoptively transferred Il21r-/- OTII cells could restore primary but not secondary IgG1+ GC formation to Il21r-/- B cells in response to re-immunisation, suggesting that CD4+ T cell responsiveness to IL-21 governed the ability of B cells to differentiate into memory cells. Data from the individual analysis of B cell clones sequences generated with IL-21 responsive or deficient T cells suggested that the role of CD4+ T cells was the delivery of differentiation signals to B cells rather than competition for the purpose of SHM, which was unaffected by IL-21 deficiency on B cells in the primary immune response. Therefore, the strong TCR:BCR interactions in the light zone may promote survival and differentiation signals, whereas SHM may be based on competition for the collection of antigen from FDC and other B cells bearing weaker BCR, which is unaffected by IL-21 deficiency and is likely to be under the control of other factors including IL-4 (Reinhardt, 2009). These data suggest that Il21r-/- B cells exhibit an inability to survive in the GC and differentiate into long-lived antibody forming cells without strong T cell help, rather than a defect in SHM per se. During the primary response, we were also interested to note the apparent paradox of a trend towards greater selection of the high affinity L>W mutation at position 33 in CDR1 of the

159 cannonical VH186.2 sequence (Bothwell, 1981; Cumano and Rajewsky, 1985; Kelsoe, 1996; Weiss, 1992) in the presence of Il21r-/- T cell help, despite these groups having little high affinity NP-binding antibody present in serum. It is possible that the absence of IL-21 survival signals for Il21r-/- B cells, compounded by poor provision of T cell help in some groups, led to only very few cells surviving in the GC that had undergone strong selection pressure for the L>W mutation. High affinity clones are generally selected to enter the memory B cell pool (Weiss, 1990) during the response to NP, therefore this trend may represent an arrest in the development of GC Il21r-/- B cells without appropriate T cell help before memory differentiation. This could explain the exacerbated defects in memory recall responses that were observed when measuring high affinity antibody after secondary immunisation, especially considering memory B cells have been shown to differentiate into IgG1+ plasma cells upon re-immunisation (Tao, 1993). Our experiments measuring antibody affinity maturation by ELISA after secondary immunisation showed a gross defect in serum IgG1 that could bind NP at high affinity in the absence of IL-21 acting on CD4+ T cells, especially after a boost of antigen. The inability to produce plasma cells without IL-21 may account for this and, indeed, our data and previous studies indicate that plasma cells are critically dependent on IL-21 (Ettinger, 2005; Kuchen, 2007; Ozaki, 2004; Tao, 1993). However, the total absence of IgG1+ staining in follicles after secondary immunisation when Il21r-/- but not WT OTII T cells were adoptively transferred, suggested that a combination of both defective memory and plasma cell differentiation results from a paucity of survival signals provided by Il21r-/- OTII T cells. The data generated using protein immunisations in chapter 4 was confirmed by our findings in autoimmune disease, where IL-21 was required to generate the large population of autoreactive GC and plasma cells in Il2-/- mice. Further study will be required to confirm the T cell intrinsic nature of this autoimmune GC and antibody production in Il2-/- mice, but novel data presented here delineate the increased Tfh subset in this strain and, as discussed later in this discussion, dysregulated generation of Tfh cells have been shown to underpin the development of antibody mediated autoimmunity (Bubier, 2009; Linterman, 2009; Vinuesa, 2009). Research outlined in chapter 5 shows that IL-21 derived from CD4 T cells was necessary for provision of help to effector cells during autoimmunity. IL-21 was found to be important for generating a cohort of IL-17A producing cells at mucosal sites and draining lymph nodes, confirming earlier reports of IL-21 driving this pathogenic effector phenotype in autoimmunity (Deenick, 2007; Nalbandian, 2009; Sutton, 2009). Removing IL-21 drove a marked change of serum cytokines towards a Th1 profile, with increased IFN and TNF, which correlated with increased health and lifespan. IFN has complex effects on multiple arms of the immune system,

160 and deficiencies in either ligand or receptor lead to conflicting outcomes in both cellular and humoral autoimmune diseases. IFN deficient mice have increased susceptivity to EAE (Pal, 2002), and IFN and its receptors has previously been associated with protection from autoimmune arthritis in experiments using knockout mice (Manoury-Schwartz, 1997; Vermeire, 1997), perhaps due to antagonism of pathological IL-17 mediated neutrophil recruitment (Irmler, 2007) or by reducing autoantibody production (Guedez, 2001). Certainly, it will be interesting to determine whether increased IL-17A is directly linked with disease pathology in this colitis model as has been shown in RAG-/- colitis (Leppkes, 2009), especially since the progressive neutropenia in Il2-/- mice indicates other pathways of inflammation mediated by Th17 may be responsible. In addition to Tfh cells, it is also likely that Th17 cells could contribute to both autoantibody and IL-21 production, as IL-17 has been decisively linked with antibody production in both a murine model of Lyme disease (Blaho, 2009) and autoimmune BXD2 mice (Hsu, 2008). Tfh cells and Th17 cells could be distinguished following NP-OVA and SRBC immunisation as we were not able to detect IL-17 expression by CXCR5+ Tfh cells. However, we have yet to determine whether such a distinction exists in the setting of chronic self-tissue destructive inflammation. Further study is required to dissect whether the improved morbidity and mortality in Il2 Il21r-/- mice was due to the absence of the pathogenic Th17 population driving antibody production or neutrophil recruitment, or the ability of Th1 cells to maintain and contain breaches of epithelial barriers (Blaho, 2009) (Hsu, 2008) (Vermeire 1997, Manoury-Schwatz 1997) (Imler 2007). The integrity of the mucosal epithelium may also be influenced by IL-21 in our model by the increased IL-22 found in the circulation. This was a particularly interesting finding as IL-22 is thought to aid tissue remodelling and healing (Aujla and Kolls, 2009; Eyerich, 2009; Pickert, 2009) and little is known about its induction in CD4+ T cells. The ability of IL-22 derived from CD4+ T cells to act on epithelial cells forms an interesting link between the immune system and mucosal tissues. The increase of IL-22 at the cost of IL-17 observed in our Il2 Il21r-/- strain suggests that IL-21 drives IL-17A production, while inhibiting IL-22. This could create a more pathogenic CD4+ T cell phenotype instead of IL-22 production helping regenerate tissues damaged by infiltrating immune cells in the mucosa as it has in other models of colitis (Marks, 2009). However, since IL-22 has also been associated with promoting pro-inflammatory mediators produced by keratinocytes (Boniface, 2005), further study is required to determine whether neutralising IL-21 in this autoimmune setting may promote tissue healing via IL-22. One interesting future experiment that could detect differences in barrier function in the mucosa is the adoptive transfer of CFSE labelled T cells bearing transgenic TCRs into Il2-/- and Il2 Il21r-/- mice, followed by oral administration of their cognate antigen. Differences in the resulting proliferation would potentially reflect the

161 integrity of the mucosal epithelium and its ability to separate the immune system from commensal antigens. The studies described in this thesis have uncovered a possible role for high IL-21 production in promoting the maintenance of polyclonal CD8+ T cell numbers, but not effector phenotype in a model of autoimmune disease. Previous studies using transgenic CD8+ T cells and in vitro assays of CD8+ T function have hinted at a requirement for IL-21 in the acquisition of effector or memory function (Allard, 2007; Li, 2005), but it is novel in this autoimmune setting. This study also outlined the novel finding that high circulating IL-21 levels could induce turnover of memory phenotype CD8+ T cells in the absence of other inflammatory signals in a resting Il21r- /- mouse on the B6 background. Further research is required to identify other signals apart from IL- 2, such as MHC-1 or antigen that lead to this cell division. This is of particular interest considering the correlation of inflated circulating IL-21 levels reported in human autoimmune diseases (Wang, 2007; Yamamoto-Furusho, 2009).

6.2 Clinical Relevance

Immunological tolerance in T cells is of overriding importance due to its ability to control downstream cytotoxic and humoral effector subsets (Holst, 2008). Understanding the ability of Tfh cells to drive high affinity antibody production in particular presents a unique challenge for manipulation in the design of protein vaccinations, and also in the treatment of antibody mediated immune diseases. Autoantibodies can interfere with normal tissue functions, and deposits of immune complexes can trigger inflammation through Fc receptors (Martin, 2006). It is possible that Tfh cells can deliver costimulation to B cells that have acquired autoimmune antibody specificities while undergoing BCR rearrangements in the GC (King, 2008). Indeed, several murine models of autoimmunity can be ameliorated by inhibition of the function of Tfh cell associated molecules such as CD40L, ICOS, SAP, and IL-21, resulting in reduced autoantibody production (Bubier, 2009; Jang, 2009; Linterman, 2009; Vinuesa, 2005a). This has even been observed in human disease, where the blocking CD40:CD40L interactions reduced autoantibody levels and decreased the numbers of GC and plasma cells that were present in the circulation of SLE patients (Hu, 2009). The data presented in this thesis showed that IL-21 delivers a potent costimulatory signal, which is vital for optimal Tfh cell function and survival, reinforcing the current interest in modulating IL-21 actions in autoimmune clinical settings. The speculated redundancy of IL-21 with other co-stimulation factors makes this cytokine an attractive therapeutic target for modulating chronic B cell activation in autoimmune disease, perhaps without compromising the adaptive immune system and antibody production as much as some other therapies such as B cell depletion.

162 In fact, exogenous IL-21 has already been used in clinical trials to restore IgG and IgA production in patients with common variable immunodeficiency and selective IgA deficiency (Borte, 2009), but it remains to be seen if antagonising IL-21 in humans will have similarly effective inverse effects in autoimmunity. Another important clinical setting where interrupting IL-21:IL21R signals could produce beneficial results is certain types of T cell lymphomas. Angioimmunoblastic T cell lymphomas (AITL) express many Tfh cell associated cell surface markers such as CXCR5, CD40L and PD-1 (Rodriguez Pinilla 2006, Rodriguez-Justo 2009) and they are also reported to express the transcription factor Bcl-6 (Ree 1999, de Leval 2001) and to secrete CXCL13 (Grogg 2006, Yu 2009). Several studies have proposed that these malignant Tfh cells may be able to move into the follicles and cause widespread B cell activation leading to the harmful hypergammaglobulinemia observed in patients (Krenacs 2006, Ferenczi 2009). It would be interesting to see if IL-21 levels are also elevated in these patients, and whether interrupting signalling could abrogate disease. The role of IL-21 in Th17 differentiation is controversial and further analyses presented here shows that IL-21 contributes to Th17 cell differentiation, skewing away from both the high IL- 22 and IFN levels that were associated with improved morbidity and mortality in Il2 Il21r-/- mice. However, a number of questions remain that are highly relevant to the pathogenesis of certain autoimmune diseases, including the manner in which IL-21 is induced in T cells during immunization or infection outside of ICOS ligation, and how IL-21 affects the migration of T cell subsets to lymphoid or tissue sites. Recent studies utilising cytokine and transcription factor reporter mice have proved to be effective tools with which to follow T helper cells from activation in the T cell zone to differentiation and migration to follicular sites and the periphery (Fontenot, 2005; Im, 2005; Reinhardt, 2006; Reinhardt, 2009). The development of IL-21 reporter mice could potentially answer many questions about the timing and migration of IL-21 producing cells. The early role for IL-21 modulating TCR signals suggests that Il-21 may play a role for different T helper subsets that arise downstream of antigen priming. An interesting new study that analysed the chromatin state of resting and effector T cells including Th1, Th2, Th17 and Tregs revealed bivalent permissive and repressive transcription factor binding associated with the signature genes of helper subtypes, which suggests that the potential exists for revision of T helper differentiation decisions (Wilson, 2009). However, the Il21 loci in Th1, Th2 and Tregs was strongly suppressed through epigenetic modifications, indicating heritable IL-21 production by Th17 and Tfh may diverge at early timepoints from other helper T cells (Wilson, 2009). Future studies will

163 need to determine whether CD4+ T cells differentiated in vivo, including Tfh cells, have similar potential for plasticity at the level of epigenetic modifications. The effects of IL-21 on T and B cell mediated autoimmune disease described herein reflect its ability to co-stimulate the activation of antigen-driven responses. Specific mechanisms explaining a qualitative or quantitative effect of IL-21 on immune responses had been unclear, and the several attempts to define a role for IL-21 in the immune system have been confounded by a considerable overlap of the actions of the c family of cytokines. This thesis demonstrated the important role of IL-21 in CD4+ T cell function and differentiation and showed that by acting as a soluble helper molecule, IL-21 supported both B cell as well as CD8+ T cell responses during chronic self-tissue destructive inflammation. In addition, IL-21 supported the survival of antigen specific precursors and their subsequent differentiation into Tfh cells following T dependent immunization as well as IL-17A production and Th17 cell generation in the inflamed intestinal mucosa during autoimmune disease. CD4+ T cells that had access to IL-21 were shown to exhibit improved fitness during T dependent responses and the resulting generation of Tfh cells was shown to be critical for optimal antibody production during both immunity and disease. Through the analyses of the genesis of Tfh cells and how they can be manipulated, the findings presented in this thesis provide important information that can be exploited for the design of better vaccines, and for the development of new therapies for autoimmune diseases.

164 7 References

Abbas, AK, et al. (1990). Heterogeneity of helper/inducer T lymphocytes. IV. Stimulation of resting and activated B cells by Th1 and Th2 clones. J Immunol 144, 2031-2037.

Acosta-Rodriguez, EV, et al. (2007). Surface phenotype and antigenic specificity of human interleukin 17-producing T helper memory cells. Nat Immunol 8, 639-646.

Adams, EJ, et al. (2005). Structure of a gammadelta T cell receptor in complex with the nonclassical MHC T22. Science 308, 227-231.

Akiba, H, et al. (2005). The role of ICOS in the CXCR5+ follicular B helper T cell maintenance in vivo. J Immunol 175, 2340-2348.

Alam, R, et al. (2003). 3. Lymphocytes. J Allergy Clin Immunol 111, S476-485.

Allard, EL, et al. (2007). Overexpression of IL-21 promotes massive CD8+ memory T cell accumulation. Eur J Immunol 37, 3069-3077.

Allen, CD, et al. (2007a). Germinal-center organization and cellular dynamics. Immunity 27, 190- 202.

Allen, CD, et al. (2007b). Imaging of germinal center selection events during affinity maturation. Science 315, 528-531.

Allman, D, et al. (2001). Resolution of three nonproliferative immature splenic B cell subsets reveals multiple selection points during peripheral B cell maturation. J Immunol 167, 6834-6840.

Alugupalli, KR (2008). A distinct role for B1b lymphocytes in T cell-independent immunity. Curr Top Microbiol Immunol 319, 105-130.

Alugupalli, KR, et al. (2004). B1b lymphocytes confer T cell-independent long-lasting immunity. Immunity 21, 379-390.

Alves, NL, et al. (2005). IL-21 sustains CD28 expression on IL-15-activated human naive CD8+ T cells. J Immunol 175, 755-762.

Amsen, D, et al. (2007). Direct regulation of Gata3 expression determines the T helper differentiation potential of Notch. Immunity 27, 89-99.

Anderson, MK, et al. (2002). Constitutive expression of PU.1 in fetal hematopoietic progenitors blocks T cell development at the pro-T cell stage. Immunity 16, 285-296.

Andersson, AK, et al. (2008). Neutralizing IL-21 and IL-15 inhibits pro-inflammatory cytokine production in rheumatoid arthritis. Scand J Immunol 68, 103-111.

Annacker, O, et al. (2005). Essential role for CD103 in the T cell-mediated regulation of experimental colitis. J Exp Med 202, 1051-1061.

165 Annunziato, F, et al. (2007). Phenotypic and functional features of human Th17 cells. J Exp Med 204, 1849-1861.

Ansel, KM, et al. (2006). Regulation of Th2 differentiation and Il4 locus accessibility. Annu Rev Immunol 24, 607-656.

Ansel, KM, et al. (1999). In vivo-activated CD4 T cells upregulate CXC chemokine receptor 5 and reprogram their response to lymphoid chemokines. J Exp Med 190, 1123-1134.

Asano, K, et al. (2007). Molecular scanning of interleukin-21 gene and genetic susceptibility to type 1 diabetes. Hum Immunol 68, 384-391.

Asao, H, et al. (2001). Cutting edge: the common gamma-chain is an indispensable subunit of the IL-21 receptor complex. J Immunol 167, 1-5.

Aujla, SJ, et al. (2009). IL-22: a critical mediator in mucosal host defense. J Mol Med 87, 451-454.

Avery, DT, et al. (2008). IL-21-induced isotype switching to IgG and IgA by human naive B cells is differentially regulated by IL-4. J Immunol 181, 1767-1779.

Azzam, HS, et al. (2001). Fine tuning of TCR signaling by CD5. J Immunol 166, 5464-5472.

Baan, CC, et al. (2007). Interleukin-21: an interleukin-2 dependent player in rejection processes. Transplantation 83, 1485-1492.

Bach, JF, et al. (1997). The NOD mouse. Res Immunol 148, 285-286.

Banchereau, J, et al. (1994). Molecular control of B lymphocyte growth and differentiation. Stem Cells 12, 278-288.

Banchereau, J, et al. (1998). Dendritic cells and the control of immunity. Nature 392, 245-252.

Barber, DL, et al. (2006). Restoring function in exhausted CD8 T cells during chronic viral infection. Nature 439, 682-687.

Barker, BR, et al. (2007). IL-21 induces apoptosis of antigen-specific CD8+ T lymphocytes. J Immunol 179, 3596-3603.

Bendelac, A, et al. (2001). Autoreactivity by design: innate B and T lymphocytes. Nat Rev Immunol 1, 177-186.

Bennett, CL, et al. (2001). IPEX is a unique X-linked syndrome characterized by immune dysfunction, polyendocrinopathy, enteropathy, and a variety of autoimmune phenomena. Curr Opin Pediatr 13, 533-538.

Benson, MJ, et al. (2007). All-trans retinoic acid mediates enhanced T reg cell growth, differentiation, and gut homing in the face of high levels of co-stimulation. J Exp Med 204, 1765- 1774.

Berek, C, et al. (1991). Maturation of the immune response in germinal centers. Cell 67, 1121-1129.

166 Berg, LJ, et al. (1989). Antigen/MHC-specific T cells are preferentially exported from the thymus in the presence of their MHC ligand. Cell 58, 1035-1046.

Bevilacqua, M, et al. (1991). Selectins: a family of adhesion receptors. Cell 67, 233.

Billingham, RE, et al. (1953). Desensitization to skin homografts by injections of donor skin extracts. Ann Surg 137, 444-449.

Biron, CA, et al. (1999). Natural killer cells in antiviral defense: function and regulation by innate cytokines. Annu Rev Immunol 17, 189-220.

Blaho, VA, et al. (2009). Cyclooxygenase-1 orchestrates germinal center formation and antibody class-switch via regulation of IL-17. J Immunol 183, 5644-5653.

Boise, LH, et al. (1995). Growth factors can enhance lymphocyte survival without committing the cell to undergo cell division. Proc Natl Acad Sci U S A 92, 5491-5495.

Bonasio, R, et al. (2006). Clonal deletion of thymocytes by circulating dendritic cells homing to the thymus. Nat Immunol 7, 1092-1100.

Boniface, K, et al. (2005). IL-22 inhibits epidermal differentiation and induces proinflammatory gene expression and migration of human keratinocytes. J Immunol 174, 3695-3702.

Borgulya, P, et al. (1992). Exclusion and inclusion of alpha and beta T cell receptor alleles. Cell 69, 529-537.

Borte, S, et al. (2009). Interleukin-21 restores immunoglobulin production ex vivo in patients with common variable immunodeficiency and selective IgA deficiency. Blood 114, 4089-4098.

Bossaller, L, et al. (2006). ICOS deficiency is associated with a severe reduction of CXCR5+CD4 germinal center Th cells. J Immunol 177, 4927-4932.

Bothwell, AL, et al. (1981). Heavy chain variable region contribution to the NPb family of antibodies: somatic mutation evident in a gamma 2a variable region. Cell 24, 625-637.

Boyman, O, et al. (2006). Selective stimulation of T cell subsets with antibody-cytokine immune complexes. Science 311, 1924-1927.

Brandt, K, et al. (2003). Interleukin-21 inhibits dendritic cell-mediated T cell activation and induction of contact hypersensitivity in vivo. J Invest Dermatol 121, 1379-1382.

Breitfeld, D, et al. (2000). Follicular B helper T cells express CXC chemokine receptor 5, localize to B cell follicles, and support immunoglobulin production. J Exp Med 192, 1545-1552.

Bretscher, PA (1999). A two-step, two-signal model for the primary activation of precursor helper T cells. Proc Natl Acad Sci U S A 96, 185-190.

Briere, F, et al. (1994). Human interleukin 10 induces naive surface immunoglobulin D+ (sIgD+) B cells to secrete IgG1 and IgG3. J Exp Med 179, 757-762.

Brugnera, E, et al. (2000). Coreceptor reversal in the thymus: signaled CD4+8+ thymocytes initially terminate CD8 transcription even when differentiating into CD8+ T cells. Immunity 13, 59-71. 167 Brunkow, ME, et al. (2001). Disruption of a new forkhead/winged-helix protein, scurfin, results in the fatal lymphoproliferative disorder of the scurfy mouse. Nat Genet 27, 68-73.

Brunner, T, et al. (1995). Cell-autonomous Fas (CD95)/Fas-ligand interaction mediates activation- induced apoptosis in T-cell hybridomas. Nature 373, 441-444.

Bryant, VL, et al. (2007). Cytokine-Mediated Regulation of Human B Cell Differentiation into Ig- Secreting Cells: Predominant Role of IL-21 Produced by CXCR5+ T Follicular Helper Cells. J Immunol 179, 8180-8190.

Bubier, JA, et al. (2007). Treatment of BXSB-Yaa mice with IL-21R-Fc fusion protein minimally attenuates systemic lupus erythematosus. Ann N Y Acad Sci 1110, 590-601.

Bubier, JA, et al. (2009). A critical role for IL-21 receptor signaling in the pathogenesis of systemic lupus erythematosus in BXSB-Yaa mice. Proc Natl Acad Sci U S A 106, 1518-1523.

Bucher, C, et al. (2009). IL-21 blockade reduces graft-versus-host disease mortality by supporting inducible T regulatory cell generation. Blood 114, 5375-5384.

Burnet, M, et al. (1965). Genetic investigations of autoimmune disease in mice. Nature 207, 368- 371.

Calabrese, GC, et al. (2009). Importance of CCL25 in the attraction of T cells and the role of IL-7 on the signaling pathways in intestinal epithelial cells. Immunobiol.

Campbell, DJ, et al. (2002). Rapid acquisition of tissue-specific homing phenotypes by CD4(+) T cells activated in cutaneous or mucosal lymphoid tissues. J Exp Med 195, 135-141.

Cannons, JL, et al. (2004). SAP regulates T(H)2 differentiation and PKC-theta-mediated activation of NF-kappaB1. Immunity 21, 693-706.

Cantrell, DA, et al. (1983). Transient expression of interleukin 2 receptors. Consequences for T cell growth. J Exp Med 158, 1895-1911.

Caruso, R, et al. (2009). Involvement of interleukin-21 in the epidermal hyperplasia of psoriasis. Nat Med 15, 1013-1015.

Caven, TH, et al. (2005). IL-21 dependent IgE production in human and mouse in vitro culture systems is cell density and cell division dependent and is augmented by IL-10. Cell Immunol 238, 123-134.

Caven, TH, et al. (2007). BCR ligation antagonizes the IL-21 enhancement of anti-CD40/IL-4 plasma cell differentiation and IgE production found in low density human B cell cultures. Cell Immunol 247, 49-58.

Chan, TD, et al. (2009). Antigen affinity controls rapid T-dependent antibody production by driving the expansion rather than the differentiation or extrafollicular migration of early plasmablasts. J Immunol 183, 3139-3149.

Charvet, C, et al. (2002). Vav1 couples T cell receptor to serum response factor-dependent transcription via a MEK-dependent pathway. J Biol Chem 277, 15376-15384.

168 Chen, W, et al. (2003). Conversion of peripheral CD4+CD25- naive T cells to CD4+CD25+ regulatory T cells by TGF-beta induction of transcription factor Foxp3. J Exp Med 198, 1875-1886.

Chen, Z, et al. (2008). Th17 cells: a new fate for differentiating helper T cells. Immunol Res 41, 87- 102.

Cho, BK, et al. (1999). Functional differences between memory and naive CD8 T cells. Proc Natl Acad Sci U S A 96, 2976-2981.

Cho, JH, et al. (2007). An intense form of homeostatic proliferation of naive CD8+ cells driven by IL-2. J Exp Med 204, 1787-1801.

Chong, MM, et al. (2002). Suppressor of cytokine signaling-1 regulates the sensitivity of pancreatic beta cells to tumor necrosis factor. J Biol Chem 277, 27945-27952.

Chtanova, T, et al. (2004). T follicular helper cells express a distinctive transcriptional profile, reflecting their role as non-Th1/Th2 effector cells that provide help for B cells. J Immunol 173, 68- 78.

Cohn, L, et al. (2004). Asthma: mechanisms of disease persistence and progression. Annu Rev Immunol 22, 789-815.

Constant, SL, et al. (1997). Induction of Th1 and Th2 CD4+ T cell responses: the alternative approaches. Annu Rev Immunol 15, 297-322.

Contractor, NV, et al. (1998). Lymphoid hyperplasia, autoimmunity, and compromised intestinal intraepithelial lymphocyte development in colitis-free gnotobiotic IL-2-deficient mice. J Immunol 160, 385-394.

Coombes, JL, et al. (2007). A functionally specialized population of mucosal CD103+ DCs induces Foxp3+ regulatory T cells via a TGF-beta and retinoic acid-dependent mechanism. J Exp Med 204, 1757-1764.

Coquet, JM, et al. (2007). IL-21 is produced by NKT cells and modulates NKT cell activation and cytokine production. J Immunol 178, 2827-2834.

Cottrez, F, et al. (2004). Specialization in tolerance: innate CD(4+)CD(25+) versus acquired TR1 and TH3 regulatory T cells. Transplantation 77, S12-15.

Coyle, AJ, et al. (2001). The expanding B7 superfamily: increasing complexity in costimulatory signals regulating T cell function. Nat Immunol 2, 203-209.

Coyle, AJ, et al. (2000). The CD28-related molecule ICOS is required for effective T cell- dependent immune responses. Immunity 13, 95-105.

Cretney, E, et al. (2007). Cancer: novel therapeutic strategies that exploit the TNF-related apoptosis-inducing ligand (TRAIL)/TRAIL receptor pathway. Int J Biochem Cell Biol 39, 280-286.

Crowson, AN, et al. (2001). The cutaneous pathology of lupus erythematosus: a review. J Cutan Pathol 28, 1-23.

169 Cumano, A, et al. (1985). Structure of primary anti-(4-hydroxy-3-nitrophenyl)acetyl (NP) antibodies in normal and idiotypically suppressed C57BL/6 mice. Eur J Immunol 15, 512-520.

Cunningham, AF, et al. (2002). Th2 activities induced during virgin T cell priming in the absence of IL-4, IL-13, and B cells. J Immunol 169, 2900-2906.

Daha, NA, et al. (2009). Confirmation of STAT4, IL2/IL21, and CTLA4 polymorphisms in rheumatoid arthritis. Arthritis Rheum 60, 1255-1260.

Dal Porto, JM, et al. (1998). Antigen drives very low affinity B cells to become plasmacytes and enter germinal centers. J Immunol 161, 5373-5381. de Vinuesa, CG, et al. (2000). Germinal centers without T cells. J Exp Med 191, 485-494.

Deenick, EK, et al. (2005). Decision criteria for resolving isotype switching conflicts by B cells. Eur J Immunol 35, 2949-2955.

Deenick, EK, et al. (2007). Autoimmunity: IL-21: a new player in Th17-cell differentiation. Immunol Cell Biol 85, 503-505.

Denny, P, et al. (1997). Mapping of the IDDM locus Idd3 to a 0.35-cM interval containing the interleukin-2 gene. Diabetes 46, 695-700. di Carlo, E, et al. (2007). Role of IL-21 in immune-regulation and tumor immunotherapy. Cancer Immunol Immunother 56, 1323-1334.

Djuretic, IM, et al. (2007). Transcription factors T-bet and Runx3 cooperate to activate Ifng and silence Il4 in T helper type 1 cells. Nat Immunol 8, 145-153.

Dolfi, DV, et al. (2008). Late signals from CD27 prevent Fas-dependent apoptosis of primary CD8+ T cells. J Immunol 180, 2912-2921.

Dong, C, et al. (2001). Cutting edge: critical role of inducible costimulator in germinal center reactions. J Immunol 166, 3659-3662.

Durand, DB, et al. (1988). Characterization of antigen receptor response elements within the interleukin-2 enhancer. Mol Cell Biol 8, 1715-1724.

Ebert, LM, et al. (2004). B cells alter the phenotype and function of follicular-homing CXCR5+ T cells. Eur J Immunol 34, 3562-3571.

Eddahri, F, et al. (2009). Interleukin-6/STAT3 signaling regulates the ability of naive T cells to acquire B-cell help capacities. Blood 113, 2426-2433.

Eggena, MP, et al. (2004). Cooperative roles of CTLA-4 and regulatory T cells in tolerance to an islet cell antigen. J Exp Med 199, 1725-1730.

Egle, A, et al. (2004). VavP-Bcl2 transgenic mice develop follicular lymphoma preceded by germinal center hyperplasia. Blood 103, 2276-2283.

Elias, KM, et al. (2008). Retinoic acid inhibits Th17 polarization and enhances FoxP3 expression through a Stat-3/Stat-5 independent signaling pathway. Blood 111, 1013-1020. 170 Elsaesser, H, et al. (2009). IL-21 is required to control chronic viral infection. Science 324, 1569- 1572.

Endo, T, et al. (2009). Amylase alpha-2A autoantibodies: novel marker of autoimmune pancreatitis and fulminant type 1 diabetes. Diabetes 58, 732-737.

Ettinger, R, et al. (2005). IL-21 induces differentiation of human naive and memory B cells into antibody-secreting plasma cells. J Immunol 175, 7867-7879.

Eyerich, S, et al. (2009). Th22 cells represent a distinct human T cell subset involved in epidermal immunity and remodeling. J Clin Invest 119, 3573-3585.

Falcone, M, et al. (1999). A defect in interleukin 12-induced activation and interferon gamma secretion of peripheral natural killer T cells in nonobese diabetic mice suggests new pathogenic mechanisms for insulin-dependent diabetes mellitus. J Exp Med 190, 963-972.

Fantini, MC, et al. (2007). IL-21 regulates experimental colitis by modulating the balance between Treg and Th17 cells. Eur J Immunol 37, 3155-3163.

Fazilleau, N, et al. (2007a). Lymphoid reservoirs of antigen-specific memory T helper cells. Nat Immunol 8, 753-761.

Fazilleau, N, et al. (2007b). Local development of effector and memory T helper cells. Curr Opin Immunol 19, 259-267.

Fazilleau, N, et al. (2009). The function of follicular helper T cells is regulated by the strength of T cell antigen receptor binding. Nat Immunol 10, 375-384.

Festen, EA, et al. (2009). Genetic variants in the region harbouring IL2/IL21 associated with ulcerative colitis. Gut 58, 799-804.

Fina, D, et al. (2007). Interleukin-21 Contributes To The Mucosal T Helper Cell Type 1 Response In Celiac Disease. Gut.

Fina, D, et al. (2008). Regulation of gut inflammation and th17 cell response by interleukin-21. Gastroenterology 134, 1038-1048.

Finkelman, FD, et al. (2004). Interleukin-4- and interleukin-13-mediated host protection against intestinal nematode parasites. Immunol Rev 201, 139-155.

Fischer, KD, et al. (1995). Defective T-cell receptor signalling and positive selection of Vav- deficient CD4+ CD8+ thymocytes. Nature 374, 474-477.

Fisher, GH, et al. (1995). Dominant interfering Fas gene mutations impair apoptosis in a human autoimmune lymphoproliferative syndrome. Cell 81, 935-946.

Fontenot, JD, et al. (2003). Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat Immunol 4, 330-336.

Fontenot, JD, et al. (2005). A function for interleukin 2 in Foxp3-expressing regulatory T cells. Nat Immunol 6, 1142-1151.

171 Forster, R, et al. (1994). Expression of the G-protein--coupled receptor BLR1 defines mature, recirculating B cells and a subset of T-helper memory cells. Blood 84, 830-840.

Forte, GI, et al. (2006). Search for genetic factors associated with susceptibility to multiple sclerosis. Ann N Y Acad Sci 1067, 264-269.

Frohlich, A, et al. (2007). IL-21 receptor signaling is integral to the development of Th2 effector responses in vivo. Blood 109, 2023-2031.

Fujieda, S, et al. (1996). Direct evidence that gamma 1 and gamma 3 switching in human B cells is interleukin-10 dependent. Mol Immunol 33, 1335-1343.

Fujita, H, et al. (2009). Human Langerhans cells induce distinct IL-22-producing CD4+ T cells lacking IL-17 production. Proc Natl Acad Sci U S A.

Fukuda, T, et al. (1995). The murine BCL6 gene is induced in activated lymphocytes as an immediate early gene. Oncogene 11, 1657-1663.

Fuqua, CF, et al. (2008). Involvement of ERK-1/2 in IL-21-induced cytokine production in leukemia cells and human monocytes. Cytokine 44, 101-107.

Furtado, GC, et al. (2002). Interleukin 2 signaling is required for CD4(+) regulatory T cell function. J Exp Med 196, 851-857.

Garside, P, et al. (1998). Visualization of specific B and T lymphocyte interactions in the lymph node. Science 281, 96-99.

Gazzinelli, RT, et al. (1996). In the absence of endogenous IL-10, mice acutely infected with Toxoplasma gondii succumb to a lethal immune response dependent on CD4+ T cells and accompanied by overproduction of IL-12, IFN-gamma and TNF-alpha. J Immunol 157, 798-805.

Geha, RS, et al. (2003). The regulation of immunoglobulin E class-switch recombination. Nat Rev Immunol 3, 721-732.

Glas, J, et al. (2009). Novel genetic risk markers for ulcerative colitis in the IL2/IL21 region are in epistasis with IL23R and suggest a common genetic background for ulcerative colitis and celiac disease. Am J Gastroenterol 104, 1737-1744.

Good, KL, et al. (2006). Kinetics of human B cell behavior and amplification of proliferative responses following stimulation with IL-21. J Immunol 177, 5236-5247.

Green, EA, et al. (1999). The initiation of autoimmune diabetes. Curr Opin Immunol 11, 663-669.

Greenwald, RJ, et al. (2005). The B7 family revisited. Annu Rev Immunol 23, 515-548.

Griffiths, GM, et al. (1984). Somatic mutation and the maturation of immune response to 2-phenyl oxazolone. Nature 312, 271-275.

Groux, H, et al. (1997). A CD4+ T-cell subset inhibits antigen-specific T-cell responses and prevents colitis. Nature 389, 737-742.

172 Gu, Y, et al. (2006). Oncogenic Vav1 induces Rac-dependent apoptosis via inhibition of Bcl-2 family proteins and collaborates with p53 deficiency to promote hematopoietic progenitor cell proliferation. Oncogene 25, 3963-3972.

Guedez, YB, et al. (2001). Genetic ablation of interferon-gamma up-regulates interleukin-1beta expression and enables the elicitation of collagen-induced arthritis in a nonsusceptible mouse strain. Arthritis Rheum 44, 2413-2424.

Gulbranson-Judge, A, et al. (1999). Defective immunoglobulin class switching in Vav-deficient mice is attributable to compromised T cell help. Eur J Immunol 29, 477-487.

Habib, T, et al. (2002). The common gamma chain (gamma c) is a required signaling component of the IL-21 receptor and supports IL-21-induced cell proliferation via JAK3. Biochemistry 41, 8725- 8731.

Hans, W, et al. (2000a). Interleukin-12 induced interferon-gamma increases inflammation in acute dextran sulfate sodium induced colitis in mice. Eur Cytokine Netw 11, 67-74.

Hans, W, et al. (2000b). The role of the resident intestinal flora in acute and chronic dextran sulfate sodium-induced colitis in mice. Eur J Gastroenterol Hepatol 12, 267-273.

Hattori, N, et al. (2005). Cytotoxic T-lymphocyte antigen 4 gene polymorphisms in sarcoidosis patients. Sarcoidosis Vasc Diffuse Lung Dis 22, 27-32.

Hauser, AE, et al. (2007). In vivo imaging studies shed light on germinal-centre development. Nat Rev Immunol 7, 499-504.

Haynes, NM, et al. (2007). Role of CXCR5 and CCR7 in follicular Th cell positioning and appearance of a programmed cell death gene-1high germinal center-associated subpopulation. J Immunol 179, 5099-5108.

Heinzel, FP, et al. (1989). Reciprocal expression of interferon gamma or interleukin 4 during the resolution or progression of murine leishmaniasis. Evidence for expansion of distinct helper T cell subsets. J Exp Med 169, 59-72.

Hendriks, J, et al. (2000). CD27 is required for generation and long-term maintenance of T cell immunity. Nat Immunol 1, 433-440.

Herber, D, et al. (2007). IL-21 has a pathogenic role in a lupus-prone mouse model and its blockade with IL-21R.Fc reduces disease progression. J Immunol 178, 3822-3830.

Holm, C, et al. (2006). Interleukin-21 mRNA expression during virus infections. Cytokine 33, 41- 45.

Holst, J, et al. (2008). Scalable signaling mediated by T cell antigen receptor-CD3 ITAMs ensures effective negative selection and prevents autoimmunity. Nat Immunol 9, 658-666.

Hori, S, et al. (2003). Control of regulatory T cell development by the transcription factor Foxp3. Science 299, 1057-1061.

Hsieh, CS, et al. (2006). An intersection between the self-reactive regulatory and nonregulatory T cell receptor repertoires. Nat Immunol 7, 401-410. 173 Hsu, HC, et al. (2008). Interleukin 17-producing T helper cells and interleukin 17 orchestrate autoreactive germinal center development in autoimmune BXD2 mice. Nat Immunol 9, 166-175.

Hsu, W, et al. (2009). Differential mechanisms in the pathogenesis of autoimmune cholangitis versus inflammatory bowel disease in interleukin-2Ralpha(-/-) mice. Hepatology 49, 133-140.

Hu, YL, et al. (2009). B7RP-1 blockade ameliorates autoimmunity through regulation of follicular helper T cells. J Immunol 182, 1421-1428.

Hunter, CA, et al. (1997). IL-10 is required to prevent immune hyperactivity during infection with Trypanosoma cruzi. J Immunol 158, 3311-3316.

Hutloff, A, et al. (1999). ICOS is an inducible T-cell co-stimulator structurally and functionally related to CD28. Nature 397, 263-266.

Ikegami, H, et al. (2003). Congenic mapping and candidate sequencing of susceptibility genes for Type 1 diabetes in the NOD mouse. Ann N Y Acad Sci 1005, 196-204.

Im, W, et al. (2005). Cytokine reporter mouse system for screening novel IL12/23 p40-inducing compounds. Mol Cells 20, 288-296.

Irmler, IM, et al. (2007). Exacerbation of antigen-induced arthritis in IFN-gamma-deficient mice as a result of unrestricted IL-17 response. J Immunol 179, 6228-6236.

Itano, AA, et al. (2003). Antigen presentation to naive CD4 T cells in the lymph node. Nat Immunol 4, 733-739.

Ivanov, II, et al. (2009). Induction of intestinal Th17 cells by segmented filamentous bacteria. Cell 139, 485-498.

Ivanov, II, et al. (2007). Transcriptional regulation of Th17 cell differentiation. Semin Immunol 19, 409-417.

Jacob, J, et al. (1993). In situ studies of the primary immune response to (4-hydroxy-3- nitrophenyl)acetyl. III. The kinetics of V region mutation and selection in germinal center B cells. J Exp Med 178, 1293-1307.

Janeway, CA, Jr., et al. (2002). Innate immune recognition. Annu Rev Immunol 20, 197-216.

Jang, E, et al. (2009). A positive feedback loop of IL-21 signaling provoked by homeostatic CD4+CD25- T cell expansion is essential for the development of arthritis in autoimmune K/BxN mice. J Immunol 182, 4649-4656.

Jin, H, et al. (2004). Distinct activation signals determine whether IL-21 induces B cell costimulation, growth arrest, or Bim-dependent apoptosis. J Immunol 173, 657-665.

Jin, H, et al. (2006). Redundant and unique regulation of activated mouse B lymphocytes by IL-4 and IL-21. J Leukoc Biol 80, 1416-1423.

Johnston, RJ, et al. (2009). Bcl6 and Blimp-1 are reciprocal and antagonistic regulators of T follicular helper cell differentiation. Science 325, 1006-1010.

174 Josefowicz, SZ, et al. (2009). Control of regulatory T cell lineage commitment and maintenance. Immunity 30, 616-625.

Jungel, A, et al. (2004). Expression of interleukin-21 receptor, but not interleukin-21, in synovial fibroblasts and synovial macrophages of patients with rheumatoid arthritis. Arthritis Rheum 50, 1468-1476.

Kaisho, T, et al. (2006). Toll-like receptor function and signaling. J Allergy Clin Immunol 117, 979-987; quiz 988.

Kappler, JW, et al. (1989). Consequences of self and foreign superantigen interaction with specific V beta elements of the murine TCR alpha beta. Cold Spring Harb Symp Quant Biol 54 Pt 1, 401- 407.

Kappler, JW, et al. (1987). T cell tolerance by clonal elimination in the thymus. Cell 49, 273-280.

Kasaian, MT, et al. (2002). IL-21 limits NK cell responses and promotes antigen-specific T cell activation: a mediator of the transition from innate to adaptive immunity. Immunity 16, 559-569.

Kelsoe, G (1995). The germinal center reaction. Immunol Today 16, 324-326.

Kelsoe, G (1996). The germinal center: a crucible for lymphocyte selection. Semin Immunol 8, 179- 184.

Khader, SA, et al. (2009). Th17 cells at the crossroads of innate and adaptive immunity against infectious diseases at the mucosa. Mucosal Immunol 2, 403-411.

Kim, BO, et al. (2004). Induction of C chemokine XCL1 (lymphotactin/single C motif-1 alpha/activation-induced, T cell-derived and chemokine-related cytokine) expression by HIV-1 Tat protein. J Immunol 172, 1888-1895.

Kim, HP, et al. (2005). Calcium-dependent activation of interleukin-21 gene expression in T cells. J Biol Chem 280, 25291-25297.

Kim-Schulze, S, et al. (2009). Local IL-21 promotes the therapeutic activity of effector T cells by decreasing regulatory T cells within the tumor microenvironment. Mol Ther 17, 380-388.

King, C, et al. (2004). Homeostatic expansion of T cells during immune insufficiency generates autoimmunity. Cell 117, 265-277.

King, C, et al. (2008). T follicular helper (TFH) cells in normal and dysregulated immune responses. Annu Rev Immunol 26, 741-766.

Kishimoto, H, et al. (1999). Strong TCR ligation without costimulation causes rapid onset of Fas- dependent apoptosis of naive murine CD4+ T cells. J Immunol 163, 1817-1826.

Kisielow, P, et al. (1988). Tolerance in T-cell-receptor transgenic mice involves deletion of nonmature CD4+8+ thymocytes. Nature 333, 742-746.

Kong, YY, et al. (1998). Vav regulates peptide-specific apoptosis in thymocytes. J Exp Med 188, 2099-2111.

175 Korn, T, et al. (2007a). IL-21 initiates an alternative pathway to induce proinflammatory T(H)17 cells. Nature 448, 484-487.

Korn, T, et al. (2007b). Th17 cells: effector T cells with inflammatory properties. Semin Immunol 19, 362-371.

Kouskoff, V, et al. (2001). Role of receptor editing and revision in shaping the B and T lymphocyte repertoire. Life Sci 69, 1105-1113.

Kramer, S, et al. (1995). Immunopathology of interleukin (IL) 2-deficient mice: thymus dependence and suppression by thymus-dependent cells with an intact IL-2 gene. J Exp Med 182, 1769-1776.

Kuchen, S, et al. (2007). Essential role of IL-21 in B cell activation, expansion, and plasma cell generation during CD4+ T cell-B cell collaboration. J Immunol 179, 5886-5896.

Kunkel, EJ, et al. (2003). Chemokines in lymphocyte trafficking and intestinal immunity. Microcirculation 10, 313-323.

Laloux, V, et al. (2001). NK T cell-induced protection against diabetes in V alpha 14-J alpha 281 transgenic nonobese diabetic mice is associated with a Th2 shift circumscribed regionally to the islets and functionally to islet autoantigen. J Immunol 166, 3749-3756.

Langrish, CL, et al. (2005). IL-23 drives a pathogenic T cell population that induces autoimmune inflammation. J Exp Med 201, 233-240.

Lasky, LA (1992). Selectins: interpreters of cell-specific carbohydrate information during inflammation. Science 258, 964-969.

Leonard, WJ (2000). X-linked severe combined immunodeficiency: from molecular cause to gene therapy within seven years. Mol Med Today 6, 403-407.

Leonard, WJ, et al. (2005). Interleukin-21: a modulator of lymphoid proliferation, apoptosis and differentiation. Nat Rev Immunol 5, 688-698.

Leppkes, M, et al. (2009). RORgamma-expressing Th17 cells induce murine chronic intestinal inflammation via redundant effects of IL-17A and IL-17F. Gastroenterol 136, 257-267.

Li, J, et al. (2006). Interleukin-21 induces T-cell activation and proinflammatory cytokine secretion in rheumatoid arthritis. Scand J Immunol 64, 515-522.

Li, Y, et al. (2005). IL-21 influences the frequency, phenotype, and affinity of the antigen-specific CD8 T cell response. J Immunol 175, 2261-2269.

Li, Y, et al. (2008). IL-21 mediated Foxp3 suppression leads to enhanced generation of antigen- specific CD8+ cytotoxic T lymphocytes. Blood 111, 229-235.

Liblau, RS, et al. (1995). Th1 and Th2 CD4+ T cells in the pathogenesis of organ-specific autoimmune diseases. Immunol Today 16, 34-38.

Linterman, MA, et al. (2009). Follicular helper T cells are required for systemic autoimmunity. J Exp Med 206, 561-576.

176 Lipsky, PE, et al. (1997). Analysis of CD40-CD40 ligand interactions in the regulation of human B cell function. Ann N Y Acad Sci 815, 372-383.

Liu, R, et al. (2008). IL-21 receptor expression determines the temporal phases of experimental autoimmune encephalomyelitis. Exp Neurol 211, 14-24.

Liu, S, et al. (2007). IL-21 synergizes with IL-7 to augment expansion and anti-tumor function of cytotoxic T cells. Int Immunol 19, 1213-1221.

Liu, YJ, et al. (1996). Normal human IgD+IgM- germinal center B cells can express up to 80 mutations in the variable region of their IgD transcripts. Immunity 4, 603-613.

Liu, ZJ, et al. (2009). Potential role of Th17 cells in the pathogenesis of inflammatory bowel disease. World J Gastroenterol 15, 5784-5788.

Lohning, M, et al. (2003). Expression of ICOS in vivo defines CD4+ effector T cells with high inflammatory potential and a strong bias for secretion of interleukin 10. J Exp Med 197, 181-193.

Ludewig, B, et al. (1998). Dendritic cells induce autoimmune diabetes and maintain disease via de novo formation of local lymphoid tissue. J Exp Med 188, 1493-1501.

Ma, CS, et al. (2009a). Early commitment of naive human CD4(+) T cells to the T follicular helper (T(FH)) cell lineage is induced by IL-12. Immunol Cell Biol 87, 590-600.

Ma, HL, et al. (2003). IL-21 activates both innate and adaptive immunity to generate potent antitumor responses that require perforin but are independent of IFN-gamma. J Immunol 171, 608- 615.

Ma, J, et al. (2009b). Morphine Disrupts the IL-23/IL-17 Mediated Pulmonary Mucosal Host Defense Against S. pneumoniae Infection. Infect Immun.

MacLennan, IC (1994). Germinal centers. Annu Rev Immunol 12, 117-139.

Malek, TR, et al. (2002). CD4 regulatory T cells prevent lethal autoimmunity in IL-2Rbeta- deficient mice. Implications for the nonredundant function of IL-2. Immunity 17, 167-178.

Malisan, F, et al. (1996). Interleukin-10 induces immunoglobulin G isotype switch recombination in human CD40-activated naive B lymphocytes. J Exp Med 183, 937-947.

Manel, N, et al. (2008). The differentiation of human T(H)-17 cells requires transforming growth factor-beta and induction of the nuclear receptor RORgammat. Nat Immunol 9, 641-649.

Manoury-Schwartz, B, et al. (1997). High susceptibility to collagen-induced arthritis in mice lacking IFN-gamma receptors. J Immunol 158, 5501-5506.

Marinova, E, et al. (2006). Human germinal center T cells are unique Th cells with high propensity for apoptosis induction. Int Immunol 18, 1337-1345.

Marks, BR, et al. (2009). Thymic self-reactivity selects natural interleukin 17-producing T cells that can regulate peripheral inflammation. Nat Immunol 10, 1125-1132.

177 Marquez, A, et al. (2009). Novel association of the interleukin 2-interleukin 21 region with inflammatory bowel disease. Am J Gastroenterol 104, 1968-1975.

Martin, F, et al. (2006). B cell immunobiology in disease: evolving concepts from the clinic. Annu Rev Immunol 24, 467-496.

Martin, F, et al. (2000). B-cell subsets and the mature preimmune repertoire. Marginal zone and B1 B cells as part of a "natural immune memory". Immunol Rev 175, 70-79.

Martin, F, et al. (2001). B1 cells: similarities and differences with other B cell subsets. Curr Opin Immunol 13, 195-201.

McGuire, HM, et al. (2009). Loss of parity between IL-2 and IL-21 in the NOD Idd3 locus. Proc Natl Acad Sci U S A 106, 19438-19443.

McHeyzer-Williams, LJ, et al. (2005). Antigen-specific memory B cell development. Annu Rev Immunol 23, 487-513.

McHeyzer-Williams, MG, et al. (1993). Antigen-driven B cell differentiation in vivo. J Exp Med 178, 295-307.

Mebius, RE, et al. (2005). Structure and function of the spleen. Nat Rev Immunol 5, 606-616.

Medzhitov, R, et al. (2002). Decoding the patterns of self and nonself by the innate immune system. Science 296, 298-300.

Mehta, DS, et al. (2005). NFATc2 and T-bet contribute to T-helper-cell-subset-specific regulation of IL-21 expression. Proc Natl Acad Sci U S A 102, 2016-2021.

Mehta, DS, et al. (2003). IL-21 induces the apoptosis of resting and activated primary B cells. J Immunol 170, 4111-4118.

Mertsching, E, et al. (1996). Phenotypic and functional analysis of B lymphopoiesis in interleukin- 7-transgenic mice: expansion of pro/pre-B cell number and persistence of B lymphocyte development in lymph nodes and spleen. Eur J Immunol 26, 28-33.

Miller, JF (1961). Immunological function of the thymus. Lancet 2, 748-749.

Miller, JF (1965). Effect of thymectomy in adult mice on immunological responsiveness. Nature 208, 1337-1338.

Mitchison, NA (2004). T-cell-B-cell cooperation. Nat Rev Immunol 4, 308-312.

Monson, NL, et al. (2001). The role of CD40-CD40 ligand (CD154) interactions in immunoglobulin light chain repertoire generation and somatic mutation. Clin Immunol 100, 71-81.

Monteleone, G, et al. (2006). Control of matrix metalloproteinase production in human intestinal fibroblasts by interleukin 21. Gut 55, 1774-1780.

Monteleone, I, et al. (2009). Interleukin-23 and Th17 cells in the control of gut inflammation. Mediators Inflamm 2009, 297645.

178 Moon, JJ, et al. (2007). Naive CD4(+) T cell frequency varies for different epitopes and predicts repertoire diversity and response magnitude. Immunity 27, 203-213.

Morelli, AE, et al. (2001). Cytokine production by mouse myeloid dendritic cells in relation to differentiation and terminal maturation induced by lipopolysaccharide or CD40 ligation. Blood 98, 1512-1523.

Moroz, A, et al. (2004). IL-21 enhances and sustains CD8+ T cell responses to achieve durable tumor immunity: comparative evaluation of IL-2, IL-15, and IL-21. J Immunol 173, 900-909.

Mosmann, TR, et al. (1986). Two types of murine helper T cell clone. I. Definition according to profiles of lymphokine activities and secreted proteins. J Immunol 136, 2348-2357.

Mosmann, TR, et al. (1991). Diversity of cytokine synthesis and function of mouse CD4+ T cells. Immunol Rev 123, 209-229.

Muller, M, et al. (2008). Intestinal colonization of IL-2 deficient mice with non-colitogenic B. vulgatus prevents DC maturation and T-cell polarization. PLoS One 3, e2376.

Nagata, S, et al. (1995). Fas and Fas ligand: lpr and gld mutations. Immunol Today 16, 39-43.

Nalbandian, A, et al. (2009). Interleukin-17 and systemic lupus erythematosus: current concepts. Clin Exp Immunol 157, 209-215.

Nie, X, et al. (1997). Immunization with immune complex alters the repertoire of antigen-reactive B cells in the germinal centers. Eur J Immunol 27, 3517-3525.

Nurieva, R, et al. (2007). Essential autocrine regulation by IL-21 in the generation of inflammatory T cells. Nature 448, 480-483.

Nurieva, RI, et al. (2008). Generation of T follicular helper cells is mediated by interleukin-21 but independent of T helper 1, 2, or 17 cell lineages. Immunity 29, 138-149.

Nurieva, RI, et al. (2009). Bcl6 mediates the development of T follicular helper cells. Science 325, 1001-1005.

Obermeier, F, et al. (2003). OX40/OX40L interaction induces the expression of CXCR5 and contributes to chronic colitis induced by dextran sulfate sodium in mice. Eur J Immunol 33, 3265- 3274.

Odegard, JM, et al. (2008). ICOS-dependent extrafollicular helper T cells elicit IgG production via IL-21 in systemic autoimmunity. J Exp Med 205, 2873-2886.

Odendahl, M, et al. (2005). Generation of migratory antigen-specific plasma blasts and mobilization of resident plasma cells in a secondary immune response. Blood 105, 1614-1621.

Okazaki, T, et al. (2006). The PD-1-PD-L pathway in immunological tolerance. Trends Immunol 27, 195-201.

Onoda, T, et al. (2007). Human CD4+ central and effector memory T cells produce IL-21: effect on cytokine-driven proliferation of CD4+ T cell subsets. Int Immunol 19, 1191-1199.

179 Ostiguy, V, et al. (2007). IL-21 promotes T lymphocyte survival by activating the phosphatidylinositol-3 kinase signaling cascade. J Leukoc Biol 82, 645-656.

Oxenius, A, et al. (1996). CD40-CD40 ligand interactions are critical in T-B cooperation but not for other anti-viral CD4+ T cell functions. J Exp Med 183, 2209-2218.

Ozaki, K, et al. (2000). Cloning of a type I cytokine receptor most related to the IL-2 receptor beta chain. Proc Natl Acad Sci U S A 97, 11439-11444.

Ozaki, K, et al. (2004). Regulation of B cell differentiation and plasma cell generation by IL-21, a novel inducer of Blimp-1 and Bcl-6. J Immunol 173, 5361-5371.

Ozaki, K, et al. (2002). A critical role for IL-21 in regulating immunoglobulin production. Science 298, 1630-1634.

Pal, E, et al. (2002). Autonomic regulation of experimental autoimmune encephalomyelitis: the role of interferon-gamma. Neuroimmunomodulation 10, 80-84.

Pankewycz, O, et al. (1991). Islet-infiltrating T cell clones from non-obese diabetic mice that promote or prevent accelerated onset diabetes. Eur J Immunol 21, 873-879.

Papadakis, KA, et al. (2000). The role of thymus-expressed chemokine and its receptor CCR9 on lymphocytes in the regional specialization of the mucosal immune system. J Immunol 165, 5069- 5076.

Parrish-Novak, J, et al. (2000). Interleukin 21 and its receptor are involved in NK cell expansion and regulation of lymphocyte function. Nature 408, 57-63.

Parrish-Novak, J, et al. (2002). Interleukin-21 and the IL-21 receptor: novel effectors of NK and T cell responses. J Leukoc Biol 72, 856-863.

Paus, D, et al. (2006). Antigen recognition strength regulates the choice between extrafollicular plasma cell and germinal center B cell differentiation. J Exp Med 203, 1081-1091.

Peluso, I, et al. (2007). IL-21 counteracts the regulatory T cell-mediated suppression of human CD4+ T lymphocytes. J Immunol 178, 732-739.

Pene, J, et al. (2004). Cutting edge: IL-21 is a switch factor for the production of IgG1 and IgG3 by human B cells. J Immunol 172, 5154-5157.

Pene, J, et al. (2006). IFN-gamma-mediated inhibition of human IgE synthesis by IL-21 is associated with a polymorphism in the IL-21R gene. J Immunol 177, 5006-5013.

Penninger, JM, et al. (1999). The oncogene product Vav is a crucial regulator of primary cytotoxic T cell responses but has no apparent role in CD28-mediated co-stimulation. Eur J Immunol 29, 1709-1718.

Pepper, M, et al. (2009). Different routes of bacterial infection induce long-lived TH1 memory cells and short-lived TH17 cells. Nat Immunol 11, 83-89.

Pesce, J, et al. (2006). The IL-21 receptor augments Th2 effector function and alternative macrophage activation. J Clin Invest 116, 2044-2055. 180 Piao, WH, et al. (2008). IL-21 modulates CD4+ CD25+ regulatory T-cell homeostasis in experimental autoimmune encephalomyelitis. Scand J Immunol 67, 37-46.

Pickert, G, et al. (2009). STAT3 links IL-22 signaling in intestinal epithelial cells to mucosal wound healing. J Exp Med 206, 1465-1472.

Pipkin, ME, et al. Interleukin-2 and inflammation induce distinct transcriptional programs that promote the differentiation of effector cytolytic T cells. Immunity 32, 79-90.

Pistoia, V, et al. (2009). IL-21: a new player in the control of isotype switch in Peyer's patches. J Leukoc Biol 85, 739-743.

Pot, C, et al. (2009). Cutting edge: IL-27 induces the transcription factor c-Maf, cytokine IL-21, and the costimulatory receptor ICOS that coordinately act together to promote differentiation of IL-10- producing Tr1 cells. J Immunol 183, 797-801.

Qi, H, et al. (2008). SAP-controlled T-B cell interactions underlie germinal centre formation. Nature 455, 764-769.

Radaev, S, et al. (2002). Recognition of immunoglobulins by Fcgamma receptors. Mol Immunol 38, 1073-1083.

Rasheed, AU, et al. (2006). Follicular B helper T cell activity is confined to CXCR5(hi)ICOS(hi) CD4 T cells and is independent of CD57 expression. Eur J Immunol 36, 1892-1903.

Rees, W, et al. (1999). An inverse relationship between T cell receptor affinity and antigen dose during CD4(+) T cell responses in vivo and in vitro. Proc Natl Acad Sci U S A 96, 9781-9786.

Refaeli, Y, et al. (1998). Biochemical mechanisms of IL-2-regulated Fas-mediated T cell apoptosis. Immunity 8, 615-623.

Reinhardt, RL, et al. (2006). T helper cell effector fates--who, how and where? Curr Opin Immunol 18, 271-277.

Reinhardt, RL, et al. (2009). Cytokine-secreting follicular T cells shape the antibody repertoire. Nat Immunol 10, 385-393.

Reynolds, LF, et al. (2004). Vav1 transduces T cell receptor signals to the activation of the Ras/ERK pathway via LAT, Sos, and RasGRP1. J Biol Chem 279, 18239-18246.

Reynolds, LF, et al. (2002). Vav1 transduces T cell receptor signals to the activation of phospholipase C-gamma1 via phosphoinositide 3-kinase-dependent and -independent pathways. J Exp Med 195, 1103-1114.

Rhee, KJ, et al. (2004). Role of commensal bacteria in development of gut-associated lymphoid tissues and preimmune antibody repertoire. J Immunol 172, 1118-1124.

Ribatti, D (2009). Sir Frank Macfarlane Burnet and the clonal selection theory of antibody formation. Clin Exp Med 9, 253-258.

Ron, Y, et al. (1987). T cell priming in vivo: a major role for B cells in presenting antigen to T cells in lymph nodes. J Immunol 138, 2848-2856. 181 Rosloniec, EF, et al. (2002). Paradoxical roles of IFN-gamma in models of Th1-mediated autoimmunity. Arthritis Res 4, 333-336.

Sadlack, B, et al. (1993). Ulcerative colitis-like disease in mice with a disrupted interleukin-2 gene. Cell 75, 253-261.

Sakaguchi, S (2000). Regulatory T cells: key controllers of immunologic self-tolerance. Cell 101, 455-458.

Sakaguchi, S, et al. (1995). Immunologic self-tolerance maintained by activated T cells expressing IL-2 receptor alpha-chains (CD25). Breakdown of a single mechanism of self-tolerance causes various autoimmune diseases. J Immunol 155, 1151-1164.

Salazar-Fontana, LI, et al. (2003). CD28 engagement promotes actin polymerization through the activation of the small Rho GTPase Cdc42 in human T cells. J Immunol 171, 2225-2232.

Sallusto, F, et al. (1999). Switch in chemokine receptor expression upon TCR stimulation reveals novel homing potential for recently activated T cells. Eur J Immunol 29, 2037-2045.

Sallusto, F, et al. (2009). Heterogeneity of CD4+ memory T cells: functional modules for tailored immunity. Eur J Immunol 39, 2076-2082.

Sansom, DM, et al. (2006). The role of CD28 and cytotoxic T-lymphocyte antigen-4 (CTLA-4) in regulatory T-cell biology. Immunol Rev 212, 131-148.

Sarvetnick, N, et al. (1988). Insulin-dependent diabetes mellitus induced in transgenic mice by ectopic expression of class II MHC and interferon-gamma. Cell 52, 773-782.

Sawalha, AH, et al. (2008). Genetic association of interleukin-21 polymorphisms with systemic lupus erythematosus. Ann Rheum Dis 67, 458-461.

Schaerli, P, et al. (2000). CXC chemokine receptor 5 expression defines follicular homing T cells with B cell helper function. J Exp Med 192, 1553-1562.

Schluns, KS, et al. (2000). Interleukin-7 mediates the homeostasis of naive and memory CD8 T cells in vivo. Nat Immunol 1, 426-432.

Schmitt, N, et al. (2009). Human dendritic cells induce the differentiation of interleukin-21- producing T follicular helper-like cells through interleukin-12. Immunity 31, 158-169.

Schultz, M, et al. (2001). Disrupted B-lymphocyte development and survival in interleukin-2- deficient mice. Immunology 104, 127-134.

Schultz, M, et al. (1999). IL-2-deficient mice raised under germfree conditions develop delayed mild focal intestinal inflammation. Am J Physiol 276, G1461-1472.

Schwickert, TA, et al. (2007). In vivo imaging of germinal centres reveals a dynamic open structure. Nature 446, 83-87.

Serfling, E, et al. (1989). Ubiquitous and lymphocyte-specific factors are involved in the induction of the mouse interleukin 2 gene in T lymphocytes. Embo J 8, 465-473.

182 Shiow, LR, et al. (2006). CD69 acts downstream of interferon-alpha/beta to inhibit S1P1 and lymphocyte egress from lymphoid organs. Nature 440, 540-544.

Silveira, PA, et al. (2004). B cell selection defects underlie the development of diabetogenic APCs in nonobese diabetic mice. J Immunol 172, 5086-5094.

Simpson, SJ, et al. (1995). Evidence that CD4+, but not CD8+ T cells are responsible for murine interleukin-2-deficient colitis. Eur J Immunol 25, 2618-2625.

Sonderegger, I, et al. (2008). IL-21 and IL-21R are not required for development of Th17 cells and autoimmunity in vivo. Eur J Immunol 38, 1833-1838.

Spolski, R, et al. (2008). IL-21 signaling is critical for the development of type I diabetes in the NOD mouse. Proc Natl Acad Sci U S A 105, 14028-14033.

Spolski, R, et al. (2009). IL-21 mediates suppressive effects via its induction of IL-10. J Immunol 182, 2859-2867.

Spolski, R, et al. (2008). Interleukin-21: Basic Biology and Implications for Cancer and Autoimmunity. Annual Review of Immunology 26.

Spolski, R, et al. (2008a). Interleukin-21: basic biology and implications for cancer and autoimmunity. Annu Rev Immunol 26, 57-79.

Starr, TK, et al. (2003). Positive and negative selection of T cells. Annu Rev Immunol 21, 139-176.

Stavnezer, J (1996). Antibody class switching. Adv Immunol 61, 79-146.

Steidler, L, et al. (2000). Treatment of murine colitis by Lactococcus lactis secreting interleukin-10. Science 289, 1352-1355.

Strengell, M, et al. (2004). IFN-alpha regulates IL-21 and IL-21R expression in human NK and T cells. J Leukoc Biol 76, 416-422.

Strengell, M, et al. (2006). IL-21 enhances SOCS gene expression and inhibits LPS-induced cytokine production in human monocyte-derived dendritic cells. J Leukoc Biol 79, 1279-1285.

Strengell, M, et al. (2002). IL-21 up-regulates the expression of genes associated with innate immunity and Th1 response. J Immunol 169, 3600-3605.

Stumhofer, JS, et al. (2007). Interleukins 27 and 6 induce STAT3-mediated T cell production of interleukin 10. Nat Immunol 8, 1363-1371.

Surh, CD, et al. (2002). Regulation of naive and memory T-cell homeostasis. Microbes Infect 4, 51- 56.

Surh, CD, et al. (2005). Regulation of mature T cell homeostasis. Semin Immunol 17, 183-191.

Surh, CD, et al. (2008). Homeostasis of naive and memory T cells. Immunity 29, 848-862.

Suto, A, et al. (2002). Interleukin 21 prevents antigen-induced IgE production by inhibiting germ line C(epsilon) transcription of IL-4-stimulated B cells. Blood 100, 4565-4573. 183 Sutton, CE, et al. (2009). Interleukin-1 and IL-23 induce innate IL-17 production from gammadelta T cells, amplifying Th17 responses and autoimmunity. Immunity 31, 331-341.

Suzuki, K, et al. (2009). Diverse regulatory pathways for IgA synthesis in the gut. Mucosal Immunol 2, 468-471.

Swat, W, et al. (1991). Clonal deletion of immature CD4+8+ thymocytes in suspension culture by extrathymic antigen-presenting cells. Nature 351, 150-153.

Tafuri, A, et al. (2001). ICOS is essential for effective T-helper-cell responses. Nature 409, 105- 109.

Takahashi, N, et al. (2009). Possible association between IgG4-associated systemic disease with or without autoimmune pancreatitis and non-Hodgkin lymphoma. Pancreas 38, 523-526.

Takeda, K, et al. Direct effects of IL-4/IL-13 and the nematode Nippostrongylus brasiliensis on intestinal epithelial cells in vitro. Parasite Immunol 32, 420-429.

Tan, JT, et al. (2002). Interleukin (IL)-15 and IL-7 jointly regulate homeostatic proliferation of memory phenotype CD8+ cells but are not required for memory phenotype CD4+ cells. J Exp Med 195, 1523-1532.

Tanaka, Y, et al. (2005). Impaired IL-4 and c-Maf expression and enhanced Th1-cell development in Vav1-deficient mice. Blood 106, 1286-1295.

Tao, W, et al. (1993). Extensive somatic mutation in the Ig heavy chain V genes in a late primary anti-hapten immune response. Mol Immunol 30, 593-602.

Tao, X, et al. (1997). Strength of TCR signal determines the costimulatory requirements for Th1 and Th2 CD4+ T cell differentiation. J Immunol 159, 5956-5963.

Tarakhovsky, A, et al. (1995a). A role for CD5 in TCR-mediated signal transduction and thymocyte selection. Science 269, 535-537.

Tarakhovsky, A, et al. (1995b). Defective antigen receptor-mediated proliferation of B and T cells in the absence of Vav. Nature 374, 467-470.

Tarlinton, DM, et al. (2000). Dissecting affinity maturation: a model explaining selection of antibody-forming cells and memory B cells in the germinal centre. Immunol Today 21, 436-441.

Teague, TK, et al. (1997). IL-6 rescues resting mouse T cells from apoptosis. J Immunol 158, 5791- 5796.

Tedesco, F (2008). Inherited complement deficiencies and bacterial infections. Vaccine 26 Suppl 8, I3-8.

Thomas, HE, et al. (2002). Interleukin-1 plus gamma-interferon-induced pancreatic beta-cell dysfunction is mediated by beta-cell nitric oxide production. Diabetes 51, 311-316.

Tivol, EA, et al. (1995). Loss of CTLA-4 leads to massive lymphoproliferation and fatal multiorgan tissue destruction, revealing a critical negative regulatory role of CTLA-4. Immunity 3, 541-547.

184 Todd, JA (1999). From genome to aetiology in a multifactorial disease, type 1 diabetes. Bioessays 21, 164-174.

Todd, JA, et al. (2007). Robust associations of four new chromosome regions from genome-wide analyses of type 1 diabetes. Nat Genet 39, 857-864.

Toellner, KM, et al. (1998). T helper 1 (Th1) and Th2 characteristics start to develop during T cell priming and are associated with an immediate ability to induce immunoglobulin class switching. J Exp Med 187, 1193-1204.

Trifari, S, et al. (2009). Identification of a human helper T cell population that has abundant production of interleukin 22 and is distinct from T(H)-17, T(H)1 and T(H)2 cells. Nat Immunol 10, 864-871.

Turner, M, et al. (1997). A requirement for the Rho-family GTP exchange factor Vav in positive and negative selection of thymocytes. Immunity 7, 451-460.

Tybulewicz, VL, et al. (2003). Vav1: a key signal transducer downstream of the TCR. Immunol Rev 192, 42-52.

Ueda, M, et al. (2005). Expression of functional interleukin-21 receptor on adult T-cell leukaemia cells. Br J Haematol 128, 169-176.

Ullman, KS, et al. (1991). Activation of early gene expression in T lymphocytes by Oct-1 and an inducible protein, OAP40. Science 254, 558-562. van de Veerdonk, FL, et al. (2009). Th17 responses and host defense against microorganisms: an overview. BMB Rep 42, 776-787.

Vella, A, et al. (1997). Interleukin 4 (IL-4) or IL-7 prevents the death of resting T cells: stat6 is probably not required for the effect of IL-4. J Exp Med 186, 325-330.

Veny, M, et al. (2009). Late Crohn's disease patients present an increase in peripheral Th17 cells and cytokine production compared with early patients. Aliment Pharmacol Ther.

Vermeire, K, et al. (1997). Accelerated collagen-induced arthritis in IFN-gamma receptor-deficient mice. J Immunol 158, 5507-5513.

Vieira, P, et al. (1990). Persistence of memory B cells in mice deprived of T cell help. Int Immunol 2, 487-494.

Villalba, M, et al. (2001). Vav1/Rac-dependent actin cytoskeleton reorganization is required for lipid raft clustering in T cells. J Cell Biol 155, 331-338.

Vinuesa, CG, et al. (2005a). A RING-type ubiquitin ligase family member required to repress follicular helper T cells and autoimmunity. Nature 435, 452-458.

Vinuesa, CG, et al. (2009). Dysregulation of germinal centres in autoimmune disease. Nat Rev Immunol 9, 845-857.

Vinuesa, CG, et al. (2005b). Follicular B helper T cells in antibody responses and autoimmunity. Nat Rev Immunol 5, 853-865. 185 Vogelzang, A, et al. (2008). A fundamental role for interleukin-21 in the generation of T follicular helper cells. Immunity 29, 127-137.

Vollmer, TL, et al. (2005). Differential effects of IL-21 during initiation and progression of autoimmunity against neuroantigen. J Immunol 174, 2696-2701.

Waidmann, M, et al. (2003). Bacteroides vulgatus protects against Escherichia coli-induced colitis in gnotobiotic interleukin-2-deficient mice. Gastroenterology 125, 162-177.

Walker, LS, et al. (1999). Compromised OX40 function in CD28-deficient mice is linked with failure to develop CXC chemokine receptor 5-positive CD4 cells and germinal centers. J Exp Med 190, 1115-1122.

Wang, B, et al. (2001). CD1-restricted NK T cells protect nonobese diabetic mice from developing diabetes. J Exp Med 194, 313-320.

Wang, XF, et al. (2007). [Changes of serum BAFF and IL-21 levels in patients with systemic lupus erythematosus and their clinical significance.]. Xi Bao Yu Fen Zi Mian Yi Xue Za Zhi 23, 1041- 1042.

Wei, L, et al. (2007). IL-21 is produced by Th17 cells and drives IL-17 production in a STAT3- dependent manner. J Biol Chem 282, 34605-34610.

Weiss, U, et al. (1990). The repertoire of somatic antibody mutants accumulating in the memory compartment after primary immunization is restricted through affinity maturation and mirrors that expressed in the secondary response. J Exp Med 172, 1681-1689.

Weiss, U, et al. (1992). Accumulation of somatic mutants in the B cell compartment after primary immunization with a T cell-dependent antigen. Eur J Immunol 22, 511-517.

Werlen, G, et al. (2002). The T-cell receptor signalosome: a dynamic structure with expanding complexity. Curr Opin Immunol 14, 299-305.

White, L, et al. (2007). Differential effects of IL-21 and IL-15 on perforin expression, lysosomal degranulation, and proliferation in CD8 T cells of patients with human immunodeficiency virus-1 (HIV). Blood 109, 3873-3880.

Wiest, DL, et al. (1999). Control of early thymocyte development by the pre-T cell receptor complex: A receptor without a ligand? Semin Immunol 11, 251-262.

Willerford, DM, et al. (1995). Interleukin-2 receptor alpha chain regulates the size and content of the peripheral lymphoid compartment. Immunity 3, 521-530.

Williams, LM, et al. (2007). Maintenance of the Foxp3-dependent developmental program in mature regulatory T cells requires continued expression of Foxp3. Nat Immunol 8, 277-284.

Wilson, CB, et al. (2009). Epigenetic control of T-helper-cell differentiation. Nat Rev Immunol 9, 91-105.

Wong, P, et al. (2001). Dynamic tuning of T cell reactivity by self-peptide-major histocompatibility complex ligands. J Exp Med 193, 1179-1187.

186 Wong, P, et al. (2003). CD8 T cell responses to infectious pathogens. Annu Rev Immunol 21, 29- 70.

Wu, C, et al. (2001). SAP controls T cell responses to virus and terminal differentiation of TH2 cells. Nat Immunol 2, 410-414.

Wu, Z, et al. (2005). Interleukin-21 receptor gene induction in human T cells is mediated by T-cell receptor-induced Sp1 activity. Mol Cell Biol 25, 9741-9752.

Wurster, AL, et al. (2002). Interleukin 21 is a T helper (Th) cell 2 cytokine that specifically inhibits the differentiation of naive Th cells into interferon gamma-producing Th1 cells. J Exp Med 196, 969-977.

Yamamoto-Furusho, JK, et al. (2009). Interleukin 21 expression is increased in rectal biopsies from patients with ulcerative colitis. Inflamm Bowel Dis.

Yamanouchi, J, et al. (2007). Interleukin-2 gene variation impairs regulatory T cell function and causes autoimmunity. Nat Genet 39, 329-337.

Yao, Z, et al. (2007). Nonredundant roles for Stat5a/b in directly regulating Foxp3. Blood 109, 4368-4375.

Yasutomo, K, et al. (2000). The duration of antigen receptor signalling determines CD4+ versus CD8+ T-cell lineage fate. Nature 404, 506-510.

Yi, JS, et al. (2009). A vital role for interleukin-21 in the control of a chronic viral infection. Science 324, 1572-1576.

Yokosuka, T, et al. (2009). Dynamic regulation of T-cell costimulation through TCR-CD28 microclusters. Immunol Rev 229, 27-40.

Yoon, JS, et al. (2008). IL-21 enhances antitumor responses without stimulating proliferation of malignant T cells of patients with Sezary syndrome. J Invest Dermatol 128, 473-480.

Young, DA, et al. (2007). Blockade of the interleukin-21/interleukin-21 receptor pathway ameliorates disease in animal models of rheumatoid arthritis. Arthritis Rheum 56, 1152-1163.

Yu, D, et al. (2009). The transcriptional repressor Bcl-6 directs T follicular helper cell lineage commitment. Immunity 31, 457-468.

Zeng, R, et al. (2007). The molecular basis of IL-21-mediated proliferation. Blood 109, 4135-4142.

Zeng, R, et al. (2005). Synergy of IL-21 and IL-15 in regulating CD8+ T cell expansion and function. J Exp Med 201, 139-148.

Zhang, JL, et al. (2003). Human IL-21 and IL-4 bind to partially overlapping epitopes of common gamma-chain. Biochem Biophys Res Commun 300, 291-296.

Zhernakova, A, et al. (2007). Novel association in chromosome 4q27 region with rheumatoid arthritis and confirmation of type 1 diabetes point to a general risk locus for autoimmune diseases. Am J Hum Genet 81, 1284-1288.

187 Zhou, L, et al. (2007). IL-6 programs T(H)-17 cell differentiation by promoting sequential engagement of the IL-21 and IL-23 pathways. Nat Immunol 8, 967-974.

Ziegler, SF, et al. (1993). Molecular characterization of the early activation antigen CD69: a type II membrane glycoprotein related to a family of natural killer cell activation antigens. Eur J Immunol 23, 1643-1648.

188