Expression and Regulation of

Leucocyte Immunoglobulin-Like Receptors

in the Human Colonic Environment

A thesis submitted in fulfilment of the requirements for the degree of Doctor of Philosophy

by

Greta Shao Chu LEE

St George and Sutherland Clinical School

Faculty of Medicine

University of New South Wales

2015

PLEASE TYPE THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet

Surname or Family name: Lee

First name: Greta Other name/s: Shao Chu

Abbreviation for degree as given in the University calendar: PhD

School: St George and Sutherland Clinical School Faculty: Medicine

Title: Expression and Regulation of Leucocyte Immunoglobulin-Like Receptors in the Human Colonic Environment

Abstract 350 words maximum: (PLEASE TYPE)

Immune homeostasis in the healthy gastrointestinal tract is characterised by active suppression of inflammation and immune tolerance and, conversely, chronic inflammatory bowel disease [IBD] results from an imbalance of these tightly regulated signals. Macrophages are abundant immune cells in the colon that protect against foreign antigens and their down-regulated phenotype is influenced by the local stromal milieu. As leukocyte immunoglobulin-like receptors [LILRs] are primarily expressed on cells of the monocytic-macrophage lineage, they are likely important immunoregulatory candidates in the colon.

The experiments in this thesis aimed to describe the pattern of distribution of LILRs in the colon and the changes that occur in IBD, and to develop a cell culture model of colonic-like macrophages that enables further characterisation of these receptors.

To determine the presence of LILRs, tissue was obtained from subjects undergoing a colonic resection. Immunohistochemistry, immunofluorescence, flow cytometry, Western blotting and qRT-PCR experiments were performed on tissue sections or isolated colonic lamina propria mononuclear cells, protein or RNA. There was universal expression of LILRB1 and LILRA5, predominantly on macrophages, but also on T and B lymphocytes. The majority of subjects also expressed LILRA1 and LILRA3. Interestingly, LILRB1 and LILRB4 were increased in chronic IBD. The presence of both inhibitory and activating LILRs on colonic macrophages suggests that a balance between positive and negative signals is important in the colon.

Colonic stromal-derived conditioned media [SCM] generated from the lamina propria of 12 control subjects was used to differentiate peripheral blood monocytes in vitro. Cytokine production by the resulting colonic-like macrophages was suppressed; minimal amounts of TNF-α, IL-1, IL-2, IL-4, IFN-γ, TGF-β and IL-10 were produced, similar to colonic macrophages. Importantly, as assessed by ELISA, multiplex and qRT-PCR, although TNF-α was produced following LPS- stimulation, this was down-regulated by SCM in a dose- and time- dependent manner. SCM also inhibited their ability to produce TGF-β in response to stimulation. Compared with classically-differentiated macrophages, inhibitory LILRs were predominantly up-regulated and, in particular, elevated LILRB1 corresponded with the increase seen on colonic macrophages. This novel in vitro model of colonic-like macrophages allows further delineation of the role of LILRs in the colon.

Declaration relating to disposition of project thesis/dissertation

I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in 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 property 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 Abstracts International (this is applicable to doctoral theses only).

……………………………………………………... ……………………………………..………….. ……….……………………......

Signature Witness Date

The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

FOR OFFICE USE ONLY Date of completion of requirements for Award:

THIS SHEET IS TO BE GLUED TO THE INSIDE FRONT COVER OF THE THESIS

i

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 ......

ii

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 ......

Date ......

iii

Acknowledgements

I wish to express my gratitude to all who have made significant contributions to this project, without whom this PhD thesis would not have been written.

First and foremost to my supervisor, Professor Michael Grimm, who has guided me through the numerous components of this project and helped see this thesis through to completion. Thank you for your advice, patience, continued support and belief in my abilities; thank you also for the detailed reviewing of this thesis.

Thank you to my co-supervisor, Associate Professor Nicodemus Tedla, for providing access to the leucocyte immunoglobulin-like receptor antibodies, for advice in the early experimental stages of this study, and for performing the arduous task of enumerating positive LILR staining on multiple immunohistochemistry slides.

Without the collaboration of the colorectal surgeons this project would not have been possible. Thank you to the following surgeons for allowing access to your patients, assistance with recruitment of subjects and provision of suitable tissue specimens: the late Dr Philip Douglas, Dr Shing Wong, Dr Graham Newstead, Professor Denis King, Dr Shevy Perera, Dr Mark Muhlmann, Dr Sanjay Kariappa, Dr Steven Gan and Dr Carina Chow.

A special thank you to all the patients who willingly agreed to participate in this study.

Thank you to all the members of the Inflammation and Infection Research Group, UNSW, who have helped me, a novice in the lab, become experienced in a range of experimental techniques. There are too many of you to name individually; you know who you are. Thank you for the countless tips and suggestions, words of wisdom and practical help with the various lab experiments and procedures.

Thank you to the National Health and Medical Research Council (NHMRC) for providing a Medical Postgraduate Research Scholarship.

And to all the family, friends and colleagues near and far who have supported, encouraged, provided insights into research and tried to keep me sane throughout these years; without your support I would not be at the end of this chapter in my life. Thank you.

iv

Abstract

Immune homeostasis in the healthy gastrointestinal tract is characterised by active suppression of inflammation and immune tolerance and, conversely, chronic inflammatory bowel disease [IBD] results from an imbalance of these tightly regulated signals. Macrophages are abundant immune cells in the colon that protect against foreign antigens and their down-regulated phenotype is influenced by the local stromal milieu. As leukocyte immunoglobulin-like receptors [LILRs] are primarily expressed on cells of the monocytic-macrophage lineage, they are likely important immunoregulatory candidates in the colon.

The experiments in this thesis aimed to describe the pattern of distribution of LILRs in the colon and the changes that occur in IBD, and to develop a cell culture model of colonic-like macrophages that enables further characterisation of these receptors.

To determine the presence of LILRs, tissue was obtained from subjects undergoing a colonic resection. Immunohistochemistry, immunofluorescence, flow cytometry, Western blot and qRT-PCR experiments were performed on tissue sections or isolated colonic lamina propria mononuclear cells, protein or RNA. There was universal expression of LILRB1 and LILRA5, predominantly on macrophages, but also on T and B lymphocytes. The majority of subjects also expressed LILRA1 and LILRA3. Interestingly, LILRB1 and LILRB4 were increased in chronic IBD. The presence of both inhibitory and activating LILRs on colonic macrophages suggests that a balance between positive and negative signals is important in the colon.

Colonic stromal-derived conditioned media [SCM] generated from the lamina propria of 12 control subjects was used to differentiate peripheral blood monocytes in vitro. Cytokine production by the resulting colonic-like macrophages was suppressed; minimal amounts of TNF-α, IL-1, IL-2, IL-4, IFN-γ, TGF-β and IL-10 were produced, similar to colonic macrophages. Importantly, as assessed by ELISA, multiplex and qRT-PCR, although TNF-α was produced following LPS stimulation, this was down-regulated by SCM in a dose- and time- dependent manner. SCM also inhibited their ability to produce TGF-β in response to stimulation. Compared with classically-differentiated macrophages, inhibitory LILRs were predominantly up-regulated and, in particular, elevated LILRB1 corresponded with the increase seen on colonic macrophages. This novel in vitro model of colonic-like macrophages allows further delineation of the role of LILRs in the colon.

v

Publications arising from this research

1. Lee, GSC., Tedla, N., Borges, L., Grimm, MC. “Leucocyte Immunoglobulin-Like Receptors in the Human Colon” (manuscript in preparation)

2. Lee, GSC., Grimm, MC. “Recreating the Intestinal Macrophage in vitro: A Potential Role for Stromal Factors” (manuscript in preparation)

Abstract publications for conference presentations

1. Australian Gastroenterology Week – 2014 Lee, GSC., Grimm, MC. Recreating the Intestinal Macrophage in vitro: A Potential Role for Stromal Factors. Journal of Gastroenterology and Hepatology 2014; 29 (Suppl. 2): 20-21

2. European Crohn’s and Colitis Organisation IBD Congress – 2014 Lee, GSC., Grimm, MC. Recreating the Intestinal Macrophage in vitro: A Potential Role for Stromal Factors. Journal of Crohn's and Colitis 2014; 8 (Suppl. 1): S74-75

3. Digestive Diseases Week, USA – 2008 Lee, GSC., Tedla, N., Borges, L., Grimm, MC. Leukocyte Immunoglobulin-Like Receptors in the Human Gastrointestinal Tract. Gastroenterology 2008; 134(4): A511

vi Table of Contents

Thesis Dissertation Sheet ……………………………………………………………………………. i Copyright and Authenticity Statements ……………………………………………………… ii Originality Statement ………………………………………………………………………………... iii Acknowledgements …………………………………………………………………………………… iv Abstract ……………………………………………………………………………………………………… v Publications ………………………………………………………………………………………………. vi Table of Contents ……………………………………………………………………………………… vii List of Figures ………………………………………………………………………………………….. xiii List of Tables .……………………………………………………………………………………………. xv Abbreviations ………………………………………………………………………………………….. xvi

1 Leucocyte Immunoglobulin-Like Receptors and Their Potential Role in the Human Gastrointestinal Tract ...... 1 1.1 Introduction ...... 1 1.2 The Leucocyte Receptor Complex ...... 2 1.3 Nomenclature of the Leucocyte Immunoglobulin-Like Receptors ...... 4 1.4 Structure of the Leucocyte Immunoglobulin-Like Receptors ...... 5 1.4.1 Extracellular Domain ...... 6 1.4.2 Transmembrane and Intracellular Domains ...... 7 1.4.2.1 Activating Receptors ...... 7 1.4.2.2 Inhibitory Receptors ...... 8 1.4.3 Soluble LILRs ...... 10 1.5 Genetic Polymorphisms ...... 11 1.6 Expression ...... 12 1.7 Ligands ...... 13 1.7.1 MHC Class 1 Antigens ...... 13 1.7.2 Non-MHC Class 1 Self Antigens ...... 14 1.7.2.1 LILRB1 ...... 14 1.7.2.2 LILRB2 ...... 14 1.7.2.3 LILRA4 ...... 15 1.7.3 Microbial Antigens ...... 15

1.8 Function ...... 16 1.8.1 Inhibitory LILRs ...... 16 1.8.2 Activating LILRs ...... 17 1.8.3 Soluble LILRs ...... 18 1.9 Leucocyte Immunoglobulin-Like Receptors in Clinical Conditions ...... 19 1.9.1 LILRs in Chronic Inflammatory Bowel Disease ...... 19 1.9.2 LILRs in Colorectal Cancer ...... 19 1.9.3 LILRs in Other Malignancies ...... 19 1.9.4 LILRs in Pregnancy ...... 20 1.9.5 LILRs in the Post-Transplant Setting ...... 20 1.9.6 LILRs in Infection ...... 21 1.9.7 LILRs in Autoimmune Conditions ...... 21

2 The Structure and Immunology of the Human Gastrointestinal Tract and Changes that Occur in Chronic Inflammatory Bowel Disease ...... 23 2.1 Introduction ...... 23 2.2 Normal Immunological Tolerance in the Colon ...... 23 2.2.1 The Normal Gastrointestinal Tract ...... 23 2.2.2 Layers of the Colon ...... 24 2.2.3 Immunological Tolerance in the Normal Gastrointestinal Tract ...... 25 2.2.4 Innate Immune Cells of the Lamina Propria ...... 26 2.2.4.1 Macrophages ...... 26 2.2.4.2 Dendritic Cells ...... 26 2.2.4.3 Mast Cells ...... 27 2.2.4.4 Eosinophils ...... 27 2.2.4.5 Neutrophils ...... 27 2.2.4.6 Innate Lymphoid Cells ...... 28 2.2.5 Acquired Immune Cells of the Lamina Propria ...... 29 2.2.5.1 T Lymphocytes ...... 29 2.2.5.2 B Lymphocytes and Plasma Cells ...... 30 2.2.6 Non-Haematopoietic Cells and the Colonic Stroma ...... 31 2.3 Chronic Intestinal Inflammation: Crohn’s Disease and Ulcerative Colitis .... 33 2.3.1 Epidemiology ...... 33 2.3.2 Clinical Aspects ...... 34 2.3.3 Genetics ...... 34 2.3.4 Dysbiosis ...... 35 2.3.5 Increased Intestinal Permeability ...... 36

2.3.6 Immunological Changes ...... 37 2.3.6.1 Innate Immune Cell Changes ...... 40 2.3.6.2 Acquired Immune Cell Changes ...... 41 2.3.6.3 Stromal Cell Changes ...... 42 2.4 Hypotheses and Aims ...... 44 2.4.1 Hypotheses ...... 44 2.4.2 Aims ...... 44

3 Subject Recruitment and Clinical Details ...... 45 3.1 Introduction ...... 45 3.2 Recruitment of Subjects ...... 45 3.2.1 Subjects Undergoing a Colonic Resection ...... 45 3.2.2 Medical History ...... 46 3.2.3 Blood Samples ...... 47 3.2.4 Colonic and Ileal Tissue ...... 47 3.3 Ethics ...... 48 3.4 Subjects Undergoing a Colonic Resection ...... 49 3.5 Subjects with Chronic Inflammatory Bowel Disease ...... 50 3.6 Deviations from Initial Plans ...... 56 3.6.1 Recruitment of Subjects ...... 56 3.6.2 Obtaining the Structured Medical History ...... 56

4 Presence and Expression Patterns of Leucocyte Immunoglobulin-Like Receptors in the Human Colon ...... 57 4.1 Introduction ...... 57 4.2 Materials and Methods ...... 57 4.2.1 Specimen Collection ...... 57 4.2.2 Histopathology ...... 57 4.2.3 Immunohistochemistry ...... 58 4.2.3.1 Immunohistochemical Staining with Anti-LILR Antibodies ...... 58 4.2.3.2 Semi-Quantative Scoring of LILRB1 and LILRA5 ...... 59 4.2.3.3 Other Cell Markers ...... 59 4.2.4 Immunofluorescence ...... 60 4.2.5 Statistics ...... 60 4.3 Results ...... 61 4.3.1 Patient Demographics and Histological Grading ...... 61 4.3.2 Leucocyte Immunoglobulin-Like Receptor Expression ...... 61 4.3.3 LILRB1 and LILRA5 Expression ...... 62

4.3.4 Cellular Expression of Leucocyte Immunoglobulin-Like Receptors ...... 65 4.4 Discussion ...... 79 4.5 Summary ...... 82

5 LILRA5 and LILRB1 Protein Expression in the Colon ...... 83 5.1 Introduction ...... 83 5.2 Materials and Methods ...... 83 5.2.1 Flow Cytometry ...... 83 5.2.1.1 Lamina Propria Mononuclear Cells ...... 83 5.2.1.2 Comparison with Peripheral Blood Mononuclear Cells ...... 86 5.2.2 Western Blotting and Immunoprecipitation ...... 88 5.2.2.1 Protein Extraction ...... 88 5.2.2.2 Western Blotting ...... 89 5.2.2.3 Immunoprecipitation ...... 91 5.2.2.4 Further Immunoprecipitation Experiments ...... 92 5.2.2.5 Experiments to Test the Anti-LILRA5 Antibody ...... 93 5.3 Results ...... 94 5.3.1 Flow Cytometry ...... 94 5.3.1.1 Lamina Propria Mononuclear Cells ...... 94 5.3.1.2 Peripheral Blood Mononuclear Cells ...... 100 5.3.2 Western Blotting and Immunoprecipitation ...... 104 5.3.2.1 Subjects for Western Blotting or Immunoprecipitation ...... 104 5.3.2.2 Western Blotting on Colonic Lamina Propria ...... 104 5.3.2.3 Immunoprecipitation ...... 107 5.3.2.4 Further Immunoprecipitation Experiments ...... 109 5.3.2.5 Experiments to Test the Anti-LILRA5 Antibody ...... 110 5.4 Discussion ...... 111 5.5 Summary ...... 114

6 Leucocyte Immunoglobulin-Like Receptor mRNA Expression in the Human Colon ...... 115 6.1 Introduction ...... 115 6.2 Materials and Methods ...... 115 6.2.1 RNA Extraction from Colon or Terminal Ileum ...... 115 6.2.2 RNA Extraction from Peripheral Blood Mononuclear Cells ...... 116 6.2.2.1 Isolation of Peripheral Blood Mononuclear Cells ...... 116 6.2.2.2 Extraction of RNA from Peripheral Blood Mononuclear Cells ...... 116 6.2.3 RNA Extraction from Lamina Propria Mononuclear Cells ...... 117 6.2.3.1 Disaggregation of Colon ...... 117

6.2.3.2 Extraction of RNA from Lamina Propria Mononuclear Cells ...... 117 6.2.4 Conversion of RNA to cDNA ...... 117 6.2.5 Quantitative Real Time PCR ...... 118 6.3 Results ...... 120 6.3.1 Subject Demographics ...... 120 6.3.2 LILR RNA Expression in Inflammatory Bowel Disease ...... 120 6.3.2.1 Inflamed vs. Non-Inflamed Colon ...... 120 6.3.2.2 PBMCs in Inflammatory Bowel Disease vs. Controls ...... 121 6.3.2.3 Inflamed Ileum vs. Inflamed Colon ...... 121 6.3.2.4 Inflamed Colon and Ileum vs. PBMCs from Subjects with IBD ...... 121 6.3.3 LILR RNA Expression in Lamina Propria ...... 122 6.3.3.1 Lamina Propria Tissue ...... 122 6.3.3.2 Lamina Propria Mononuclear Cells ...... 123 6.3.4 TNF-α Expression ...... 124 6.3.4.1 In Inflammatory Bowel Disease Subjects ...... 124 6.3.4.2 In Control Subjects ...... 124 6.4 Discussion ...... 125 6.5 Summary ...... 126

7 Replicating Human Colonic-Like Macrophages in vitro ...... 127 7.1 Introduction ...... 127 7.2 Materials and Methods ...... 127 7.2.1 Cytokine Production by Lamina Propria Mononuclear Cells ...... 127 7.2.1.1 Cytokine Response of Normal LPMCs to Lipopolysaccharide Stimulation ...... 127 7.2.1.2 Cytokine Production by LPMCs in Response to Other Stimuli ...... 129 7.2.2 Stromal-Derived Conditioned Media ...... 129 7.2.2.1 Generating Stromal-Derived Conditioned Media ...... 129 7.2.2.2 Assessing Stromal-Derived Conditioned Media ...... 130 7.2.3 Cultured Peripheral Blood Mononuclear Cells ...... 131 7.2.3.1 Culturing Monocytes with Stromal-Derived Conditioned Media ...... 131 7.2.3.2 Analysis of Cytokine Production in Cell Culture Supernatants ...... 133 7.2.3.3 Flow Cytometry ...... 135 7.3 Results ...... 136 7.3.1 Cytokine Production by Isolated Lamina Propria Mononuclear Cells ...... 136 7.3.1.1 Response of Normal LPMCs to LPS Stimulation ...... 136 7.3.1.2 Cytokine Production in Response to a Range of Other Stimuli ...... 138 7.3.2 Stromal-Derived Conditioned Media ...... 141 7.3.2.1 Subjects and Specimens ...... 141

7.3.2.2 Endotoxin Level ...... 141 7.3.2.3 Protein Quantification ...... 142 7.3.2.4 Cytokine Analysis ...... 142 7.3.3 Effect of Stromal-Derived Conditioned Media on Cultured Peripheral Blood Mononuclear Cells ...... 146 7.3.3.1 Cell Morphology ...... 146 7.3.3.2 Phenotype Determined by Flow Cytometry ...... 149 7.3.3.3 Cytokine Production ...... 150 7.4 Discussion ...... 163 7.5 Summary ...... 170

8 Leucocyte Immunoglobulin-Like Receptors and TNF-α Expression of in vitro Human Colonic-Like Macrophages ...... 171 8.1 Introduction ...... 171 8.2 Materials and Methods ...... 171 8.2.1 Extraction of RNA from Monocytes and Cultured PBMCs ...... 171 8.2.1.1 Preliminary Experiments ...... 171 8.2.1.2 Culturing Monocytes and Extraction of mRNA for qRT-PCR ...... 173 8.2.2 Synthesis of cDNA ...... 174 8.2.3 qRT-PCR Array ...... 174 8.3 Results ...... 175 8.3.1 PBMCs ...... 175 8.3.2 LILR mRNA Expression of Culture Monocytes ...... 176 8.3.2.1 Effect of Stromal-Derived Conditioned Media on LILR Expression of Unstimulated Cultured Monocytes ...... 176 8.3.2.2 Effect of Cell Stimulation on LILR Expression of Cultured Monocytes ...... 180 8.3.3 TNF-α mRNA Expression of Cultured Monocytes ...... 185 8.3.3.1 Effect of Stromal-Derived Conditioned Media on TNF-α Expression of Unstimulated Cultured Monocytes ...... 185 8.3.3.2 Effect of Cell Stimulation on TNF-α Expression of Cultured Monocytes ...... 186 8.4 Discussion ...... 188 8.5 Summary ...... 191

9 Discussion and Future Directions ...... 192

10 Appendices ...... 198 10.1 Structured Medical Questionnaire ...... 198 10.2 Additional Cytokine Graphs from the Cell Culture Model ...... 206

11 References ...... 215

List of Figures

Figure 1.1: The human leucocyte receptor complex on chromosome 19q13.4 2 Figure 1.2: Activating, inhibitory and soluble leucocyte immunoglobulin-like receptors 5 Figure 1.3: Extracellular, transmembrane and intracellular structure of different LILRs 10 Figure 2.1: Layers of the colonic wall 24 Figure 2.2: Proposed paradigm of chronic inflammatory bowel diseases pathogenesis 33 Figure 4.1: Proportion of subjects positively expressing LILRs in the lamina propria 61 Figure 4.2: LILRB1 and LILRA5 cell counts in the lamina propria 62 Figure 4.3: Correlation between LILRB1 and LILRA5 density 63 Figure 4.4: Correlation between LILRB1 / LILRA5 and CD68 63 Figure 4.5: LILRB1 and LILRA5 cell counts in the submucosa 64 Figure 4.6: Immunohistochemistry – Colonic lamina propria 66 Figure 4.7: Immunohistochemistry – Lymphoid follicle in lamina propria 67 Figure 4.8: Immunohistochemistry – Endothelium 68 Figure 4.9: LILRA3 and tryptase expression in lamina propria 68 Figure 4.10: LILRB1, LILRB2 and LILRB3 expression in lamina propria 69 Figure 4.11: LILR expression in inflamed colon 70 Figure 4.12: Representative immunofluorescence images of lamina propria 72 Figure 5.1: LILRA5 expression vs. CD14 and LPMC viability 95 Figure 5.2: LPMC viability vs. CD14 95 Figure 5.3: Rested LPMCs – viability, CD14 and LILRA5 96 Figure 5.4: LILRB1 expression of LPMCs 96 Figure 5.5: Co-expression of LILRB1 and LILRA5 on LPMCs 97 Figure 5.6: Co-expression of LILRB1 and LILRA5 on PBMCs 97 Figure 5.7: Phenotype of isolated LPMCs as assessed by flow cytometry 98 Figure 5.8: CD68, CD14 and CD33 expression on isolated LPMCs 99 Figure 5.9: Phenotype of lamina propria macrophages before / after negative selection 100 Figure 5.10: LILRA5 expression on PBMCs 101 Figure 5.11: LILRB1 expression on PBMCs 101 Figure 5.12: Co-expression LILRA5 and CD68 on permeabilised PBMCs 102 Figure 5.13: Western blot – Lamina propria, probed with anti-LILRA5 antibody 105 Figure 5.14: Western blot – Lamina propria from IBD and control subjects, probed with 106 anti-LILRA5 antibody Figure 5.15: PBMCs and lamina propria immunoprecipitated with anti-LILRA5 or mouse 107 IgG1 antibodies, probed with polyclonal rabbit anti-LILR antibody Figure 5.16: PBMCs immunoprecipitated with anti-LILRA5 or mouse IgG1 antibodies, 108 probed with streptavidin HRP; and corresponding Western blot of PBMCs and lamina propria, probed with anti-LILRA5 antibody

xiii Figure 5.17: PBMCs immunoprecipitated with anti-LILRA5, anti-LILRB4 and mouse IgG1 109 antibodies, probed with polyclonal rabbit anti-LILR antibody Figure 5.18: Silver stain of rhLILRA5 and PBMCs 110 Figure 7.1: TNF-α levels in cultured LPMCs and PBMCs 136 Figure 7.2: IL-8, IL-1β and TGF-β levels in cultured LPMCs and PBMCs 137 Figure 7.3: IL-1β, IL-8, TGF-β and TNF-α production by cultured LPMCs, unstimulated 139 or stimulated for up to 36 hours Figure 7.4: Endotoxin levels in pooled and individual SCM 141 Figure 7.5: Total protein levels in individual and pooled SCM 142 Figure 7.6: IL-8, IL-1β, TGF-β and TNF-α levels in individual SCM 143 Figure 7.7: Cytokines, chemokines and growth factors in the SCM 144 Figure 7.8: Cell morphology after 5 days of culture 146 Figure 7.9: Cell morphology in the presence or absence of SCM, LPS and GM-CSF 147 Figure 7.10: Differences in cell morphology between adherent standard or ultra-low 148 attachment cell culture plates Figure 7.11: CD45/CD14 expression on MACS-isolated peripheral blood monocytes, 149 monocytes cultured with GM-CSF and monocytes cultured with SCM and GM-CSF Figure 7.12: TNF-α levels in cell culture supernatants comparing GM-CSF vs. M-CSF 150 differentiation of monocytes, varying amounts of SCM and 6 or 24 hours of LPS stimulation Figure 7.13: TNF-α levels in cell culture supernatants comparing monocyte selection via 151 adherence or MACS Figure 7.14: TNF-α levels in cell culture supernatants showing varying cytokine 152 response of monocytes to different amounts of SCM Figure 7.15: TNF-α levels in cell culture supernatants of monocytes cultured in the 154 presence or absence of SCM for 5 days or with SCM for 3 or 5 days in the presence or absence of GM-CSF, followed by stimulation with LPS for 6 or 24 hours Figure 7.16: TNF-α levels in response to IFN-γ and IL-10 stimulation 155 Figure 7.17: IL-1β levels in response to LPS, IFN-γ and IL-10 stimulation 156 Figure 7.18: TGF -β levels in response to LPS, IFN-γ and IL-10 stimulation 157 Figure 7.19: IL-6 levels in response to LPS, IFN-γ and IL-10 stimulation 158 Figure 7.20: Comparison of TNF-α levels following cell differentiation with either SCM or 161 LPS, in response to LPS stimulation Figure 7.21: Comparison of TNF-α levels following cell differentiation with either SCM or 162 LPS, in response to IFN-γ and IL-10 stimulation Figure 8.1: Cell differential of PBMCs prior to MACS negative selection 175 Figure 8.2: LILR mRNA expression in cell cultures 181 Figure 8.3: LILR mRNA expression following stimulation with LPS, IFN-γ or IL-10 183 Figure 8.4: TNF-α expression following stimulation with LPS 186

xiv List of Tables

Table 1.1: Alternative nomenclature of the leucocyte immunoglobulin-like receptors 4 Table 1.2: Leucocyte immunoglobulin-like receptor proteins 6 Table 1.3: Expression of classical ITAM-associating receptors on immune cells 8 Table 1.4: Expression of ITIM-containing receptors on immune cells 9 Table 1.5: Cellular expression of LILRs 12 Table 4.1: Histological scoring system 58 Table 5.1: PBMCs in colonic disaggregation solutions 103 Table 5.2: Protein bands detected by anti-LILRA5 antibody 104 Table 5.3: PBMCs immunoprecipitated with anti-LILRA5 antibody 109 Table 6.1: Primers for qRT-PCR 119 Table 6.2: LILRs over-expressed in inflamed colon cf. non-inflamed colon 120 Table 6.3: LILR expression in inflamed colon and ileum cf. PBMCs in subjects with IBD 121 Table 6.4: LILR expression in lamina propria cf. full thickness colon 122 Table 6.5: LILR expression in colon and lamina propria cf. PBMCs in control subjects 122 Table 6.6: LILR expression in LPMCs cf. colon and lamina propria 123 Table 6.7: LILR expression in LPMCs cf. PBMCs in control subjects 123 Table 6.8: TNF-α in colon or lamina propria cf. PBMCs and LPMCs in control subjects 124 Table 7.1: Cytokine, chemokines and growth factors in the SCM 144 Table 7.2: Expression of CD45, CD14 and CD3 on cultured monocytes 149 Table 8.1: Genes over-expressed in cultured monocytes, in cells differentiated with 176 SCM with or without GM-CSF cf. cells differentiated with GM-CSF without SCM Table 8.2: Expression of LILRs in MACS-isolated monocytes cf. PBMCS in controls 177 Table 8.3: LILR mRNA expression in cultured monocytes cf. MACS-isolated monocytes 178 Table 8.4: LILR mRNA expression in cultured monocytes cf. colonic lamina propria 179 Table 8.5: Genes over-expressed in monocytes differentiated with GM-CSF after 24 180 hours of stimulation with LPS, IFN-γ or IL-10 cf. unstimulated cultured monocytes Table 8.6: Genes over-expressed in monocytes differentiated with GM-CSF and SCM or 182 SCM after 24 hours of stimulation with LPS, IFN-γ or IL-10 cf. unstimulated cultured monocytes Table 8.7: TNF-α expression in unstimulated cells 185 Table 8.8: TNF-α expression in unstimulated cells differentiated with GM-CSF, SCM or 185 both cf. colonic lamina propria or MACS-isolated monocytes Table 8.9: TNF-α expression following 6 and 24 hour stimulation 187

xv Abbreviations

p-ANCA Perinuclear anti-neutrophil cytoplasmic antibody

ASCA Anti-Saccharomyces cerevisiae antibody

BMI Body mass index

BSA Bovine serum albumin

ELISA Enzyme-linked immunosorbent assay

GAPDH Glyceraldehyde-3-phosphate dehydrogenase

HBSS Hank's buffered salt solution

HLA Human leucocyte antigen

IBD Inflammatory bowel disease

IFN γ Interferon-γ

IL-10 Interleukin-10

ILCs Innate lymphoid cells

ITAMs Immunoreceptor tyrosine-based activating motifs

ITIMs Immunoreceptor tyrosine-based inhibitory motifs

KIRs Killer cell immunoglobulin-like receptors

LAIRs Leucocyte-associated immunoglobulin-like receptors

LILRs Leucocyte immunoglobulin-like receptors

LPMCs Lamina propria mononuclear cells

LPS Lipopolysaccharide

MHC Major histocompatibility complex

NK Natural killer

PIRs Paired immunoglobulin-like receptors

PBMCs Peripheral blood mononuclear cells

D-PBS Dulbecco's phosphate-buffered saline

PMA Phorbol myrisate acetate

PVDF Polyvinylidene difluoride

ROR-γT Retinoic acid receptor-γ related orphan receptor

SCCAI Simple clinical colitis activity index

SCM Stromal-derived conditioned media

xvi TBS Tris-buffered saline

TNF-α Tumour necrosis factor-α

Th Helper T lymphocytes

Treg Regulatory T lymphocytes

xvii 1 Leucocyte Immunoglobulin-Like Receptors and Their Potential Role in the Human Gastrointestinal Tract

1.1 Introduction

The chronic inflammatory bowel diseases, Crohn’s disease and ulcerative colitis, have been recognised for more than a century1-3 and their incidence and prevalence have increased rapidly in the past few decades. The cause of these conditions is still unknown, however, various genetic, environmental and immunological factors are thought to play a role in their pathogenesis.

The immune homeostasis of the gastrointestinal tract is tightly regulated and, in recent times, the role of various factors that help to maintain a healthy tolerogenic state has been characterised. Besides cross-talk between the numerous cells of both the innate and adaptive immune system via their associated cytokines, these cells interact with the luminal commensal microbiota and other contents, epithelium and intestinal stroma, to maintain an active immune tolerant environment (as detailed in Chapter 2).

Leucocyte immunoglobulin-like receptors are a newly described family of cell surface proteins expressed on a range of haematopoietic cells that are thought to play a role in immune regulation. There are two broad types of receptors which show high homology: inhibitory receptors that contain immunoreceptor tyrosine-based inhibitory motifs [ITIMs] and activating receptors that contain a positively charged arginine residue that associates with adapter proteins containing immunoreceptor tyrosine-based activating motifs [ITAMs](see Figure 1.2).

Both inhibitory and activating leucocyte immunoglobulin-like receptors may recognise the same or similar ligand(s) and, similar to other paired immune receptors4, counter- balancing inhibitory and activating signals may modulate the amplitude and threshold of immune responses.

Leucocyte immunoglobulin-like receptors have been implicated in a variety of clinical conditions, including infections, autoimmune conditions, pregnancy, post-transplantation and malignancies (see section 1.9). They are also likely to be present in the human gastrointestinal tract, a major site of immune tolerance in the body, and an imbalance or

1 alteration of their density, type or function could play a role in chronic inflammatory bowel disease.

In order to understand how these receptors may play a role in the human colon, this chapter will first provide an overview of the family of leucocyte immunoglobulin-like receptor proteins, including their genomic organisation, extracellular and intracellular structure, expression, function and proposed ligands. The current knowledge about clinical conditions associated with the leucocyte immunoglobulin-like receptors will be outlined.

1.2 The Leucocyte Receptor Complex

Leucocyte immunoglobulin-like receptors [LILRs] are members of the immunoglobulin superfamily of proteins. They are encoded by 13 genes on 2 loci within the human leucocyte receptor complex on chromosome 19q13.45-7. The 2 loci are separated by a region of approximately 200 Kb that is flanked by 2 genes encoding the leucocyte- associated immunoglobulin-like receptors [LAIRs]. The centromeric cluster encodes LILRB3, LILRA6, LILRB5, LILRB2, LILRA3, LILRA5, LILRA4 and LAIR-1. The telomeric cluster encodes LAIR-2, LILRA2, LILRA1, LILRB1, LILRB4 and the pseudogenes LILRP1 and LILRP2. Within the leucocyte receptor complex, just telomeric to the LILR genes, are the genes encoding killer cell immunoglobulin-like receptors [KIRs]8.

Figure 1.1 The human leucocyte receptor complex on chromosome 19q13.4

2 There is homology between the LILRs and other members of the immunoglobulin superfamily. This functionally diverse group of proteins includes the KIRs, platelet glycoprotein VI, IgA FcαR, NKp46/Ly94 and the LAIRs. Further details about these are beyond the scope of this thesis, but they have been reviewed by Barrow and Trowsdale9.

Some of the LILR genes are conserved in animal species. The LILRA5, LILRA6, LILRB3, LILRB4 and LILRB5 genes are conserved in chimpanzees, the LILRA5, LILRA6, LILRB3 and LILRB4 genes in rats, the LILRA5 and LILRB3 genes in dogs, the LILRA5, LILRB3 and LILRB4 genes in mice, the LILRA6, LILRB4 and LILRB5 genes in rhesus monkeys and the LILRB5 gene in cows. Orthologs have been described in a number of species, mainly in primates such as chimpanzees, macaques and olive baboons, but also in other vertebrates such as cattle, bats, cats and chickens10-13.

The best characterised of the animal homologues are in mice. The paired immunoglobulin- like receptors [PIRs] share genetic sequence, expression profile, structure and function with the human LILRs. They map to the proximal end of mouse chromosome 7. At least 6 genes encode the activating receptors, PIR-A, whilst a single gene encodes the inhibitory receptor, PIR-B. PIR-B binds HLA-G and corresponds to human LILRB214. Another murine homologue is Gp49B, which corresponds to human LILRB415, 16.

3 1.3 Nomenclature of the Leucocyte Immunoglobulin-Like Receptors

The leucocyte immunoglobulin-like receptors were first identified in 1997 by a number of research groups17-23, and this has led them to be known by a variety of names. They were named immunoglobulin-like transcripts [ILTs] by the investigators at the Basel Institute for Immunology in Switzerland5-7, 17-19, 24-26, leucocyte immunoglobulin-like receptors [LIRs] by the investigators at the Immunex Coorporation in Seattle, Washington USA20, 21, human monocyte receptors [HM] by the investigators at the Harvard Medical School, Boston USA22 and monocyte/macrophage immunoglobulin-like receptors [MIRs] by the investigators at the National Institutes of Health in Maryland, USA23. The family of receptors were allocated CD85 numbers in 200027-29 and the current official Human Genome Organization Nomenclature classification is Leucocyte Immunoglobulin-Like Receptors [LILRs].

Table 1.1 Alternative nomenclature of the leucocyte immunoglobulin-like receptors30

LILRA1 LIR6, CD85i

LILRA2 ILT1, LIR7, CD85h

LILRA3 ILT6, LIR4, CD85e, HM31, HM43

LILRA4 ILT7, CD85g

LILRA5 ILT11, LIR9, CD85f, LILRB7

LILRA6 ILT8, CD85b, LILRB3, LILRB6

LILRB1 ILT2, LIR1, CD85j, MIR7, CD85

LILRB2 ILT4, LIR2, CD85d, MIR10

LILRB3 ILT5, LIR3, CD85a, HL9, LILRA6

LILRB4 ILT3, LIR5, CD85k, HM18

LILRB5 LIR8, CD85c

LILRP1 ILT9, CD85l, LILRA6P

LILRP2 ILT10, CD85m, LILRA5

4 1.4 Structure of the Leucocyte Immunoglobulin-Like Receptors

The LILRs are classified into membrane-bound (activating or inhibitory) and soluble receptors.

Figure 1.2 Activating, inhibitory and soluble leucocyte immunoglobulin-like receptors

Membrane-bound LILRs are characterised by the presence of 2 or 4 homologous extracellular domains, a single transmembrane domain and a variable intracellular region. Soluble LILRs have neither transmembrane nor cytoplasmic domains.

5 Table 1.2 Leucocyte immunoglobulin-like receptor proteins30

Protein Amino acids Isoforms Sub-cellular location Ig-like C2- type domains

LILRA1 Transcript 489 2 Membrane 4

LILRA2 Protein 483 2 Membrane 4

LILRA3 Protein 439 3 Secreted 4

LILRA4 Protein 499 2 Membrane 4

LILRA5 Protein 299 4 Membrane: 2 isoforms 2 Secreted: all isoforms

LILRA6 Transcript 481 2 Membrane 2

LILRB1 Protein 650 4 Membrane 4

LILRB2 Protein 598 2 Membrane 4

LILRB3 Transcript 631 3 Membrane 4

LILRB4 Protein 448 3 Membrane 2

LILRB5 Protein 590 3 Membrane 4

1.4.1 Extracellular Domain

The LILRs contain 2 or 4 highly homologous C2-type immunoglobulin-like extracellular domains19-21 and, although ligands for most of the LILRs remain to be identified, the striking similarities between these receptors suggest that they may recognise similar ligands. Indeed, LILRB1, LILRB2, LILRB4, LILRA1 and LILRA3 have all been shown to bind various MHC class 1 molecules31-34, albeit with varying affinity. Binding of these molecules to inhibitory LILRB1 and LILRB2 leads to down-regulation of the immune response19, 20, 35.

The extracellular domains of the LILRs may undergo post-translational modification, as all except LILRB4 have multiple N-linked glycosylation sites36-38. Binding of a complex sugar molecule to these sites may affect the receptor’s ligand-binding specificity and affinity, and prevent non-specific protein-protein interactions and proteolysis by proteases39, 40. N- glycosylation is essential for ligand binding to LILRA341.

6 1.4.2 Transmembrane and Intracellular Domains

1.4.2.1 Activating Receptors

Activating LILRs (LILRA1, LILRA2, LILRA4-A6) have a basic arginine residue within the hydrophobic transmembrane domain and a short cytoplasmic domain that lacks signalling motifs42, 43. This is similar to other activating receptors that have been characterised in recent decades, such as T and B cell receptors, several activating natural killer cell receptors, some Fc receptors, scavenging receptors and other activating receptors on haematopoietic cells44.

Upon activation, these receptors become non-covalently associated with a transmembrane adapter molecule that contains a negatively-charged aspartic acid residue in the transmembrane domain and one or more cytoplasmic ITAM units.45, 46. The positively charged arginine residues of the transmembrane domain of LILRA2, LILRA4 and LILRA5 have been shown to associate with the Fc receptor common γ chain, an adapter molecule that contains a negatively charged aspartic acid residue in the transmembrane domain and a single ITAM in the cytoplasmic domain38. The structural similarity between the activating LILRs suggests that LILRA1 and LILRA6 are likely to also associate with the Fc receptor common γ chain, although this is yet to be demonstrated.

This association between receptor and adapter molecule results in activation of the Src family kinases that phosphorylate tyrosine residues in the ITAM of the adapter molecule. This leads to recruitment of Zap-70 or Syk, triggering activation of a cascade of signalling pathways. These result in effector functions such as induction of cytokine production, calcium mobilisation, phagocytosis, cell migration, proliferation and survival47.

Of note, different immune activating receptors can share the same adapter molecule, highlighting the importance and versatility of ITAMs in immune receptor signalling.

In addition to cell activation, several ITAM-associating receptors have been shown to have the ability to exert inhibitory effects and, whilst the mechanisms have not been fully elucidated, this may be due to factors such as degree of phosphorylation48, 49.

7 Table 1.3 Expression of classical ITAM-associating receptors on immune cells50, 51

Surface receptor Adapter ITAM sequence Signalling Expression pattern protein molecule

NKG2D-S, KIR-2D, DAP12 ESPYQELQGQR Syk or Zap- Myeloid and natural Ly49, TREM1,2,3 SDVYSDL 70 killer cells

FcαR, FcγR, OSCAR, FcRγ DGVYTGLSTR Syk Broad haematopoietic PIR-A, Dectin-2, NQETYETL expression, myeloid TCR (integrins) cells

BCR, MHC-II Igα ENLYEGLNLD Syk B lymphocytes DCSMYEDI

BCR, MHC-II Igβ DHTYEGLDID Syk B lymphocytes QTATYEDI

TCR, CXCR4 CD3ζ DGLYQGLSTAT Zap-70 T lymphocytes KDTYDAL

1.4.2.2 Inhibitory Receptors

In distinction to the activating receptors, a number of inhibitory receptors have been shown to be critical in counteracting excessive cell activation by ITAM-associating receptors52. These receptors occur on both haematopoietic and non-haematopoietic cells and are characterised by the presence of several ITIMs in their intracellular domains47, 53.

These inhibitory receptors require co-engagement of ITAM-associating receptors for phosphorylation. This results in recruitment of either Src homology 2 domain-containing phosphatases, SHP-1 and SHP-2, or Src homology 2-containing inositol polyphosphate 5- phosphatase, SHIP, which inhibit kinases that trigger cellular activation54, 55.

Of note, many of these inhibitory receptors have putative activating counterparts within the same family of receptors4. Ligands have not been determined for most of these paired receptors, however those that have been identified are commonly widely expressed host moieties that bind more weakly to activating receptors than to inhibitory receptors56-59. These include HLA-A, B and C for KIR on chromosome 19 and HLA-E for NKG2 on chromosome 12, sialic acids for SIGLEC on chromosome 19, and CD47 for SIRP on chromosome 2060.

8 Table 1.4 Expression of ITIM-containing receptors on immune cells61

Surface Chromosomal Number Expression pattern Ligands Putative Receptor location of ITIMs activating receptors

FcγRIIB 1q23-24 1 Myeloid, mast, B IgG FcγRIIA, lymphocytes complexes FcγRIII Programmed 2q37.3 1 T, B, NK PD-1 Death lymphocytes ligands receptor-1

PILRα 7q22 2 Myeloid cells PILRβ NKG2A 12q13.1-13.2 2 NK, T lymphocytes HLA-E NKG2C, NKG2E CD31 17q23 1 Myeloid, platelet, CD31 endothelial cells, subset of T and NK lymphocytes CD22 19q13.1 4 B lymphocytes Sialic acid CD66a 19q13.2 2 Epithelial cells, CD66, granulocytes, T CD62E lymphocytes, subset of B and NK lymphocytes SIGLEC 5,6,7 19q13.3 1 Myeloid, B, NK Sialic acid lymphocytes, cytotrophoblast LILRB1-5 19q13.4 2-4 Myeloid, B MHC-1 for LILRA1-6 lymphocytes, subset B1 and B2 of T and NK lymphocytes LAIR-1 19q13.4 2 Leucocytes KIR2,3DL 19q13.4 2 NK, T lymphocytes HLA-A,B,C KIR2,3DS

SIRPα 20q13 2 Myeloid, non- CD47 SIRPβ haematopoietic cells

9 The cytoplasmic tail of the inhibitory LILRs (LILRB1-B5) has 2 to 4 immunoreceptor tyrosine-based inhibitory motifs or ITIMs which, when phosphorylated, lead to recruitment of src homology 2 domain–containing phosphatase, SHP-1, which terminates or inhibits signalling42, 43, 61.

Figure 1.3 Graphical representation of extracellular, transmembrane and intracellular structure of different LILRs. All LILRs have 2 or 4 highly homologous extracellular domains. Activating LILRs have a positively charged arginine residue in the transmembrane region and a short cytoplasmic tail. LILRA2, LILRA4 and LILRA5 have been shown to signal through association with the Fc receptor common γ chain, which contains a cytoplasmic ITAM. Alternative gene splicing results in LILRA1a and LILRA1b. Inhibitory LILRs have 2 to 4 ITIMs in the cytoplasm domain. Figure modified from43.

1.4.3 Soluble LILRs

LILRA3 is the only member of the LILR family that has a termination codon in the stalk, leading to a deletion of the transmembrane domain. It is, therefore, expressed only as a soluble protein with no transmembrane or intracellular domains19, 21, 22, 62. As it has four C2-type Ig-like domains similar to most of the other LILRs, it may act as a competitive agonist or antagonist of the other LILRs.

Alternative mRNA splicing of the LILR genes may generate soluble isoforms in the majority of LILRs. Transcripts for putative soluble proteins for LILRA1, LILRA2, LILRA5 and LILRB1-4 have been reported62-65. These have a cryptic stop codon before the transmembrane domain, but protein expression is yet to be demonstrated.

High levels of soluble LILRA4 protein have been documented in serum from patients with melanoma, colorectal or pancreatic cancer65. 10 1.5 Genetic Polymorphisms

Genetic polymorphisms have been detected for most of the LILRs and these may affect their expression and function. Of all the LILRs, however, only LILRA3 and LILRA6 exhibit copy number variation66, 67.

LILRA3 exhibits a gene presence or absence variation consisting of a 6.7-kb deletion that comprises the first 6 of the 7 exons. This removes all of the Ig-like domains of the gene, suggesting that the putative truncated protein would not be functional7, 68-70. The frequency of the LILRA3 deletion polymorphism varies widely between different populations, the non-functional LILRA3 variation is higher in Northeast Asians (Korean 84%, Japanese 71%) than in Thais (21%), Europeans (17%) or Africans (10%)70-72.

LILRA6 has either a deletion of one copy of the gene or a duplication variation resulting in three or four copies of the gene. In a study that detected a 40% LILRA3 gene deletion rate, there was a LILRA6 deletion rate of 8.3% and a duplication rate of 33.3%. The remaining 58.4% had the normal two copies of the gene66. LILRA6 also displays significant genetic diversity, with at least 29 polymorphic sites66.

LILRB3 is the most highly polymorphic of the LILRs, particularly when compared with LILRB1 or LILRB219, having at least 40 polymorphic sites66. Despite its high rate of allelic variation, LILRB3 has two constant copies per genotype66, 73.

11 1.6 Expression

The LILRs are expressed on a broad range of leucocytes, predominantly myeloid antigen presenting cells such as monocytes, macrophages and dendritic cells, as well as to a lesser degree on T lymphocytes, B lymphocytes, natural killer [NK] lymphocytes, basophils, eosinophils, neutrophils and mast cells17-21, 38, 43, 63, 73-88.

Table 1.5 Cellular expression of LILRs

High expression levels Lower expression levels

LILRA1 Monocytes B lymphocytes

LILRA2 Monocytes, Macrophages, NK lymphocytes, Eosinophils, Basophils, Dendritic cells Mast cells, T lymphocytes, B lymphocytes

LILRA3 Monocytes, B lymphocytes NK lymphocytes, T lymphocytes, B lymphocytes

LILRA4 Plasmacytoid dendritic cells Eosinophils, Neutrophils, Monocytes

LILRA5 Monocytes, Neutrophils

LILRA6 Monocytes

LILRB1 B lymphocytes, Monocytes, T lymphocytes, NK lymphocytes, Dendritic cells Neutrophils, Eosinophils

LILRB2 Monocytes, B lymphocytes Dendritic cells, NK lymphocytes, Neutrophils, Eosinophils, Basophils

LILRB3 Monocytes, Myeloid type 1 Neutrophils, Eosinophils, Basophils, B dendritic cells lymphocytes

LILRB4 Monocytes, Macrophages, NK lymphocytes, Neutrophils, B Plasmacytoid > Myeloid lymphocytes dendritic cells

LILRB5 Mast cells, NK lymphocytes Monocytes

12 1.7 Ligands

Knowledge about the natural ligands for the LILRs is limited, and this has hindered identification of the in vivo functions of these receptors. As LILRs are heavily glycosylated, production of these proteins in prokaryotic cells such as E. coli is challenging89, 90. In vitro studies have detected some potential ligands, although these overall bind with low affinity to the LILRs31.

1.7.1 MHC Class 1 Antigens

The inhibitory LILRs, LILRB1, LILRB2 and LILRB4, have been shown to bind to major histocompatibility complex [MHC] class 1 antigens. The MHC is a genomic region on the short arm of chromosome 6 that encodes a large number of cell surface and secreted proteins. These are broadly grouped into three classes of molecules: MHC classes I, II and III. MHC class I and II encode human leucocyte antigen [HLA], cell surface proteins mainly responsible for immune activation and autoimmunity. Class 1 antigens include the classic proteins, HLA-A, HLA-B and HLA-C, as well as the non-classical proteins, HLA-E, HLA-F and HLA-G. MHC class III encode secreted proteins with immune functions related to the HLA antigens, including components of the complement cascade, tumour necrosis factor-α [TNF-α] and heat shock proteins91.

LILRB1 was initially discovered by its binding to UL1820, 31, 92, a MHC class 1 homologue expressed by human cytomegalovirus. LILRB1 also recognises a broad spectrum of HLA-A, HLA-B, HLA-C and HLA-G alleles, although it binds to UL18 with a >1000-fold higher affinity than to these MHC class 1 proteins31.

LILRB2 and LILRB4 also recognise a broad spectrum of HLA-A, HLA-B, HLA-C and HLA-G alleles19-21, 34, 35, 93-98. In particular, LILRB2 binds with highest affinity to HLA-G99 and this leads to a down-regulation of the immune response, including inhibition of serotonin release, intracellular calcium mobilisation and phosphorylation of cellular proteins, and the development of tolerance.

HLA-G has 4 membrane-bound and 3 soluble isoforms100 and its expression is normally restricted to a few immune privileged tissues, in particular the placenta101, 102, thymus103, cornea104, endocrine pancreas105 and proximal nail apparatus106. Ectopic expression of

13 HLA-G has been observed in tumour escape mechanisms, allograft acceptance, viral infections, autoimmune diseases and inflammation100.

It is possible that, due to the similarity between extracellular domains, the other inhibitory LILRs, LILRB3 and LILRB5, also bind to MHC class 1 antigens.

Soluble LILRA3 binds both classical and non-classical HLA class I molecules but with reduced affinities compared to LILRB1 or LILRB2107, and LILRA1 binds HLA-B27 and the HLA-C free H chain32, 33. The significant structural differences in MHC class 1 binding sites between inhibitory and activating LILRs suggest that distinct ligand recognition may occur between the two groups and, indeed, LILRA2 and LLRA5 do not bind a range of MHC class 1 molecules37.

PIRs, the murine orthologues of the LILRs, bind mouse MHC class 1 molecules, H-2K, H-2D and H-2L108.

1.7.2 Non-MHC Class 1 Self Antigens

1.7.2.1 LILRB1

S100A8 and S100A9, calcium-binding proteins of the S100 family expressed on monocyte derived dendritic cells, have been shown to be ligands for LILRB1 and binding of S100A9 to peripheral NK lymphocytes can stimulate anti-HIV-1 activity through TNF-α production109.

1.7.2.2 LILRB2

LILRB2 may be involved in regulation beyond the innate immune system. LILRB2 and the murine ortholog, PIR-B, bind the myelin-derived inhibitory protein, Nogo66, that is involved in preventing axon regeneration110, 111. No benefit, however, was shown in axon or neuron regeneration in PIR-B knockout mice 112, 113.

LILRB2 and PIR-B were also demonstrated to bind to soluble beta-amyloid oligomers and enhance cofilin signalling in Alzheimer’s disease. In a transgenic mouse model of

14 Alzheimer’s disease, PIR-B mediated loss of synaptic plasticity in the juvenile visual cortex and contributed to memory deficits111.

LILRB2 and PIR-B can also bind angiopoietin-like proteins to support hematopoietic stem cell development and, in mice, differentiation of leukemic cells114.

1.7.2.3 LILRA4

LILRA4 binds bone marrow stromal cell antigen 2, which is expressed by bone marrow stromal cells, plasma cells and multiple types of cancer cells. This may regulate plasmacytoid dendritic cell function in response to viral infections115, 116 and in malignancy117, although its biological role has not been fully determined118.

1.7.3 Microbial Antigens

Inhibitory LILRs are up-regulated in a number of different infections. LILRB1 is increased on HIV-1, CMV, EBV -specific CD8+ T lymphocytes119-122, and the level of LILRB1 on CMV and EBV -specific memory CD8+ T lymphocytes increases with time123-125. NK lymphocytes expressing LILRB1 suppressed viral replication in HIV-1-infected monocyte derived dendritic cells in vitro suggesting LILRB1 plays a role in controlling early infection126. LILRB2 and LILRB4 are also up-regulated by interleukin-10 [IL-10] in HIV-1 infection127, 128. HIV-1-infected non-progressors who have undetectable levels of the virus in the absence of anti-retroviral therapy have a decreased ability to secrete pro-inflammatory cytokines, which is associated with up-regulation of LILRB1 and LILRB3129. Conversely, the HLA-B*3503 molecule associated with progressive HIV-1 disease binds significantly more strongly than the HLA-B*3501 molecule to LILRB2, and this B*3503-LILRB2 binding corresponds with higher degrees of LILRB2-mediated dendritic cell dysfunction130. Binding of soluble HLA-G to LILRB2, but not LILRB1, is also associated with rapidly progressive HIV-1 infection, inhibition of antigen-presentation and enhancement of pro- inflammatory cytokines by dendritic cells131.

LILRB1 also binds Staphylococcus aureus and Escherichia coli, but not Pseudomonas aeruginosa, Listeria monocytogenes, Helicobacter pylori or Bacillus subtilis –expressing NIH3T3 cells132. Additionally, dengue virus evades the immune system through co-ligating with LILRB1, resulting in inhibition of Fc receptor signalling and reduced interferon-γ

15 [IFN-γ] expression133. LILRB2 and LILRB4 are increased in Salmonella typhimurium infection134.

1.8 Function

The expression of LILRs on myeloid and lymphoid cells and their binding to endogenous MHC class-1 molecules suggests they have a role in immune regulation. Functional studies in vivo have been limited by the lack of knowledge about natural ligands.

1.8.1 Inhibitory LILRs

LILRB1 is the most extensively studied of the LILRs in vitro. It is expressed by a variety of haematopoietic cells and its activation by MHC class 1 molecules and homologs results in inhibitory signalling. On T lymphocytes, it suppresses signalling by inhibiting phosphorylation of the T cell receptor, inhibits rearrangement of the actin cytoskeleton, antigen-induced T cell proliferation and IL-2 and IFN-γ production and increases IL-10 and TGF-β production135-137. LILRB1 is also expressed on NK, B and myelomonocytic cells and, when activated, inhibits killing by NK and T lymphocytes, calcium mobilisation by B lymphocytes and monocytes, proliferation, differentiation and immunoglobulin secretion by B lymphocytes, and the ability to produce cytokines, up-regulate costimulatory molecules and stimulate T cell proliferation by dendritic cells19, 138-140.

LILRB2, also expressed on a broad range of haematopoietic cells, inhibits activation of T lymphocytes, B lymphocytes and monocytes19, 35 and importantly, along with LILRB4 on monocytes and dendritic cells, LILRB2 binds HLA-G resulting in tolerisation of these antigen presenting cells141, 142. LILRB2 and HLA-G expression on dendritic cells is induced by high doses of IFN-γ143. LILRB2 is implicated in maternal-foetal tolerance and preventing allogenic graft rejection as described below.

The function of LILRB3, expressed on the surface of innate immune cells, has been studied in basophils where coligation of LILRB3 resulted in inhibition of histamine, cysteinyl leukotriene and IL-4 release78.

Up-regulation of LILRB4 results in tolerisation of dendritic cells, which have reduced expression of the costimulatory molecules, CD80 and CD86 and promote regulatory and suppressor T lymphocytes, contributing to immune tolerance141, 144. A LILRB4 knockdown 16 model of dendritic cells showed increased TLR responsiveness, secretion of CXCL10 and CXCL11 chemokines, T cell migration and proliferation and synthesis of pro-inflammatory cytokines such as IL-1, IL-6, IL-17A and IFN-γ145.

LILRB5 mRNA was discovered in NK lymphocytes and LILRB5 protein was described in mast cell granules, released following activation of these cells by IgE82, but the functional significance is yet to be explored.

1.8.2 Activating LILRs

Less is known about the in vivo functions of activating LILRs. LILRA1-transfected HEK 293T lymphocytes were able to bind HLA-B2732 and the HLA-C free H chain33, although the functional significance of these interactions is unknown.

Cells of the monocyte lineage abundantly express LILRA2, but its natural ligand is unknown. Activation of monocytes by antibody crosslinking induced the secretion of pro- inflammatory cytokines, TNF-α, IL-6, IL-8 and IL-12, and impaired dendritic cell differentiation in vitro146. Antigen presentation and phagocytic ability were also reduced147. However the physiological conditions that trigger LILRA2 activation are yet to be determined.

Up-regulation of LILRA2 on CD14+CD68+ cells from lesions of lepromatous patients shifts the cytokine profile towards a Th2 profile, with an increase in IL-10 and a decrease in IL- 12146, 148.

Crosslinking LILRA4 experiments on human plasmacytoid dendritic cells can lead to intracellular calcium mobilisation and IFN-α production, conversely costimulation with TLR7 or TLR9 can significantly down-regulate IFN-α80. With these dual roles, LILRB4 may tightly regulate plasmacytoid dendritic function115.

LILRA5 exists as a membrane-bound and a soluble receptor63. Crosslinking of the membrane-bound isoform on monocytes induced the production of pro-inflammatory cytokines, TNF-α, IL-1β and IL-6, and the immunomodulatory cytokine, IL-10149. The function of the soluble isoform has not been determined, but it could function as a soluble antagonist to the membrane-bound receptor.

17 Little is known about the function of LILRA6, although it is closely related to LILRB3; they are encoded on neighbouring genes within the LILR complex and possess indistinguishable extracellular domains.

1.8.3 Soluble LILRs

As a soluble protein, LILRA3 may have stimulatory or inhibitory functions. It has high structural homology with LILRB1 and LILRA121. It has been shown to induce CD8+ T cell and NK lymphocyte proliferation in culture, stimulating secretion of IL-6, IL-8, IL-1β and IL-10150. Conversely, when bound to primary monocytes, LILRA3 significantly suppressed lipopolysaccharide mediated TNF-α production by these cells41. These opposing roles, whilst not well defined, support the possibility that LILRA3 may act as an agonist or antagonist to the other LILRs.

18 1.9 Leucocyte Immunoglobulin-Like Receptors in Clinical Conditions

LILRs are increasingly recognised in the pathogenesis of various diseases.

1.9.1 LILRs in Chronic Inflammatory Bowel Disease

LILRB1 and LILRB4 have been described in the colonic mucosa of both healthy children and paediatric patients with ulcerative colitis151. LILRB1 was expressed on CD68+ and CD68- mononuclear cells and LILRB4 was localised to the crypt epithelium. In a different study to determine predictor genes for inflammation using microarray gene expression profiling of colonic biopsies from Crohn’s disease and ulcerative colitis patients, LILRA2 was one of ten genes included in the gene set that predicted the need for surgery (sensitivity and specificity of 50% and 85%, respectively)152.

1.9.2 LILRs in Colorectal Cancer

LILRB4 was present in CD68-positive tumour-associated macrophages in colorectal cancer, and soluble LILRB4 was present in the serum of these patients65.

1.9.3 LILRs in Other Malignancies

LILRB4 was also detected in pancreatic cancer and melanoma65 and LILRB2 in lung cancer153.

HLA-G expressed by solid tumours is associated with inhibition of anti-tumour immune cells, higher histological grades and advanced clinical stages, a poorer prognosis and resistance to immunotherapy154. Elevated levels of soluble HLA-G have also been described in B-cell malignancies such as multiple myeloma, non-Hodgkin B cell lymphoma and B-cell chronic lymphocytic leukaemia155-158 however a correlation with unfavourable clinical outcomes is less clear and this may be due to the balance and function of LILRB1 or LILRB2 on these malignant immune cells154.

Besides LILRB1 up-regulation in B-cell chronic lymphocytic leukaemia158, LILRB2 and LILRA4 are also expressed159. LILRB1 is also increased in CD8+ cells from T-cell large

19 granular leukaemia160. HLA-G and LILRB1 are expressed in cutaneous T cell lymphoma cells161-163. The non-functional gene deletion of LILRA3 is also associated with B-cell non- Hodgkin’s lymphoma150.

1.9.4 LILRs in Pregnancy

Soluble HLA-G is secreted by trophoblasts164, 165 and binds to LILRB1 and LILRB2, modulating the immune tolerance of the mother to the growing foetus. This is impaired in pre-eclampsia166 and HLA-G dysregulation may also play a role in spontaneous abortion.

In cell culture, Toxoplasma gondii infection resulted in higher LILRB1 and KIR2DL4 expression on NK-92 cells and HLA-G on human placental BeWo cells within 48 hours167, and this may be related to the adverse pregnancy outcomes seen with T. gondii infection.

1.9.5 LILRs in the Post-Transplant Setting

Inhibitory LILRs play a role in preventing rejection following organ transplants. This has been studied most following heart, kidney or liver transplantation, and recipients who were rejection-free for at least 1 year developed allospecific CD8+ T lymphocytes that induced up-regulation of LILRB2 or LILRB4 on donor antigen presenting or endothelial cells141, 168. Young kidney donors (< 55 years old) were shown to induce a higher level of tolerogenic LILRB2+ dendritic cells in the graft recipients169. HLA-G isoforms bind these inhibitory LILRs, leading to induction of regulatory T lymphocytes, down-regulation of costimulatory molecules and graft tolerance170, 171. A synthetic soluble HLA-G-derived molecule that binds LILRB2 and induces tolerance has been produced with the aim of developing a therapeutic molecule useful in preventing allograft rejection172.

In haematopoietic stem cell transplants, the role of LILRB3 variants between donor and recipients has been studied, and these may induce an antibody response affecting graft tolerance173.

20 1.9.6 LILRs in Infection

Up-regulation of inhibitory LILRs, as described above, occurs in both chronic infections such as HIV-1, CMV, EBV and leprosy, as well as in acute infections such as dengue fever, Staphylococcus aureus, Escherichia coli and Salmonella typhimurium.

Additionally, during acute inflammation, in response to neutrophil degranulation rapid translocation of intracellular LILRB2 occurs and this up-regulation on neutrophils inhibits phagocytosis. In sepsis, however, this up-regulation is impaired86 and conversely, LILRB2 is up-regulated on monocytes, resulting in reduced IL-12 and increased IL-10 production, and reduced expression of the costimulatory molecule CD86174.

With diverse populations of microorganisms present in the human gastrointestinal tract, it is possible that other microbial components, in addition to those identified so far, may control the immune response through these paired receptors.

1.9.7 LILRs in Autoimmune Conditions

LILRs are also expressed in various autoimmune conditions. In active rheumatoid arthritis, increased expression of LILRA2, LILRA5, LILRB2 and LILRB3 has been shown in synovial tissue macrophages, fibroblast-like synoviocytes, and endothelial cells76, 149. Levels of these LILRs correlated with disease severity175 and were reduced in response to treatment176. Serum and synovial fluid LILRA3 protein levels also correlated with disease activity84. In a collagen-induced arthritis mouse model, dimers of HLA-G had a significant anti- inflammatory effect, due to their binding with the mouse homolog, PIR-B177.

In systemic lupus erythematosis, LILRB1 was decreased on circulating CD19+ B lymphocytes, and CD4+ and CD8+ T lymphocytes had diminished inhibitory activity178. Whilst there were increased levels of plasmacytoid dendritic cells, correlating with disease activity, these cells had low LILRB1 expression179.

Relapsing remitting multiple sclerosis was associated with a reduction in LILRB4+ monocytes. These increased following treatment with interferon β-1b180. LILRB1 was expressed predominantly on microglia and macrophages in chronic active plaques in the central nervous system, as well as in the lymphocytic infiltrates in acute multiple sclerosis181. 21 Psoriatic skin lesions showed HLA-G5 present in macrophages and the receptor, LILRB1, on CD4+ infiltrating T lymphocytes182. LILRB1 was also expressed in the inflammatory myopathies; in polymyositis and idiopathic inclusion body myositis, LILRB1 was detected on CD8+ T lymphocytes and infiltrating macrophages and in dermatomyositis LILRB1 was expressed in a proportion of macrophages, B lymphocytes and CD3+CD8- T lymphocytes183.

The LILRs show genetic diversity and links between polymorphisms and diseases have been studied. Functional LILRA3, but not the non-functional gene deletion, was strongly associated with susceptibility to rheumatoid arthritis and joint destruction, particularly in males184. Conversely, the non-functional LILRA3 gene deletion was associated with Sjogren’s syndrome185 but not psoriasis186. In Spanish and German but not Polish patients, the non-functional LILRA3 gene deletion was associated with multiple sclerosis susceptibility187-189. Polish patients who were negative for the deletion, however, had an earlier onset of disease187.

Loss of function LILRB4 polymorphisms have been demonstrated in systemic lupus erythematosis and these are associated with higher levels of type 1 interferon and TNF- α190, 191. Activating LILRA2 polymorphisms were associated with systemic lupus erythematosis, as well as with microscopic polyangiitis190. One of five polymorphisms of LILRB1 was associated with susceptibility to rheumatoid arthritis, but none of these correlated with systemic lupus erythematosis susceptibility192.

In summary, leucocyte immunoglobulin-like receptors have been implicated in a number of clinical conditions, particularly infections, autoimmune diseases and malignancies, as well as immune tolerant situations such as pregnancy and post-allogeneic transplantation. This family of inhibitory and activating receptors are expressed on a variety of cell types, especially cells of the innate immune system, and whilst functional studies are limited by the paucity of knowledge about their ligands, they appear to have heavily immune- modulatory roles. They are likely to be important in an immune tolerant environment such as the colon that is constantly exposed to a diverse range of commensal microorganisms.

22 2 The Structure and Immunology of the Human Gastrointestinal Tract and Changes that Occur in Chronic Inflammatory Bowel Disease

2.1 Introduction

In order to determine the role of leucocyte immunoglobulin-like receptors in the human colon, it is essential to understand the normal immunological tolerance in the gastrointestinal tract, and the immunological changes that occur in chronic inflammatory bowel disease [IBD]. In this chapter, the structure of the colon and the resident immune and stromal cells of the mucosa are described. Next, the immunological changes that occur in chronic inflammatory bowel disease are outlined. Finally, the hypotheses and aims of this thesis are presented.

2.2 Normal Immunological Tolerance in the Colon

2.2.1 The Normal Gastrointestinal Tract

The human gastrointestinal tract is one of the most crucial organs in the body. Its surface area is approximately 400 m2 and it is in constant contact with an enormous microbial load; it is estimated that there are 1012 microorganisms per gram of luminal content193. It contains 60-70% of the body’s immune cells and is therefore the largest lymphoid organ of the body. The gastrointestinal tract is the interface between the body and the outside world and has two essential functions; first, digestion of food, absorption of nutrients and water and excretion of waste and, secondly, immunological protection against microbial and other antigens.

23 2.2.2 Layers of the Colon

The colonic mucosa has three layers. There is a single epithelial cell layer, composed primarily of absorptive columnar enterocytes and mucus-producing goblet cells in a ratio of 4:1, which provides a physical barrier against the luminal contents. Located just below the epithelium is the lamina propria. This contains vast numbers of unique immune cells from both the innate and acquired immune systems, as well as various non- haematopoietic stromal cells. The third layer is the muscularis mucosae, a thin smooth muscle layer that assists with peristalsis.

Below the mucosa lies the submucosa, a layer of connective tissue that contains blood vessels, nerves and lymphatics. The outer layers of the colonic wall are the circular and longitudinal smooth muscle layers of the muscularis externa involved in peristalsis, and the subserosa and serosa, the final loose connective tissue layers.

Figure 2.1 Schematic representation of the layers of the colonic wall. Image sourced from AJCC: Cancer Staging Manual, 7th edition, October 2009. American Joint Committee on Cancer

24 2.2.3 Immunological Tolerance in the Normal Gastrointestinal Tract

Under normal circumstances the gastrointestinal tract is in a state of immune tolerance, consisting of active suppression and controlled inflammation, in order to sustain homeostasis.

This healthy state is maintained in a number of ways. Whilst the colonocytes facilitate absorption of water, the tight junctions between these cells prevent pathogens and foreign antigens from entering194-199. A mucus layer secreted by goblet cells protects the epithelial cell layer. It is mainly composed of complex mucin glycoproteins and it contains secreted IgA, peptides, enzymes and other immunologically active substances. In addition to this are the beneficial effects of peristalsis and the colonisation of the gastrointestinal tract by commensal bacteria.

Within the epithelial layer are intraepithelial lymphocytes, a distinct group of lymphocytes that have an active protective role in the intestine. These predominantly CD8 positive T lymphocytes are extremely heterogeneous, antigen-specific and phenotypically different from peripheral or lamina propria lymphocytes. They provide immediate immune protection from luminal pathogens, secrete modulatory cytokines such as TGF-β1, are cytotoxic and have been implicated in antibody class switching and IgA production. They maintain the epithelial cell barrier by clearing necrotic epithelium, secreting keratinocyte growth factor in response to damage and controlling cell growth and turnover200, 201.

The lamina propria contains a high density of phenotypically and functionally distinct immune cells that can maintain control when the above measures are breached. These include unique resident macrophages that reside particularly in the subepithelial area of the lamina propria, dendritic cells, small numbers of mast cells, eosinophils, neutrophils and innate lymphoid cells. These innate immune cells function in conjunction with a variety of adaptive immune cells, such as regulatory T lymphocytes that secrete large amounts of IL-10 and TGF-β, to actively maintain the tolerogenic state. IgA-secreting plasma cells are also prominent. Within the lamina propria are organised lymphoid follicles, which contain predominantly T and B lymphocytes and, to a lesser extent, macrophages.

25 2.2.4 Innate Immune Cells of the Lamina Propria

Innate immune cells are critical in providing rapid responses to mucosal pathogens.

2.2.4.1 Macrophages

Resident intestinal macrophages comprise 10-20% of the mononuclear cells of the lamina propria202. They are preferentially localised to the sites of antigen entry; the lamina propria and lymphoid follicles. They differ phenotypically and functionally from circulating monocytes and other tissue macrophages.

They are unique in that, whilst they are highly phagocytic, they are down-regulated cells that, unlike other macrophages, have constitutively high levels of the anti-inflammatory cytokine, IL-10203, and consequently have suppressed production of pro-inflammatory cytokines. This helps achieve a healthy state of immune tolerance in the pathogen-rich environment. Despite expressing high levels of CD33, CD68, CD13 and HLA-DR, the majority of colonic macrophages do not express the lipopolysaccharide receptor, CD14, and express only low levels of the Fcγ receptors, CD16, CD32 and CD64, the complement receptors, CD11b and CD11c, the integrin LFA-1 receptor, CD11a, and the receptors for the costimulatory B7 molecules that play an important role in T cell activation, CD80 and CD86204-209.

Similarly, jejunal macrophages have been shown to be down-regulated for both CD14 and the IgA receptor, CD89210, 211, as well as a range of other innate response receptors212.

2.2.4.2 Dendritic Cells

Dendritic cells are important in maintaining tolerance to food microbial pathogens. Those that reside in the lamina propria fall broadly into two subsets: CD11chighCD103+CD11b+ and CD11chighCX3CR1+ cells. These immature dendritic cells can take up antigens by a variety of pathways. CD11chighCD103+CD11b+ process antigens, migrate to draining lymphoid follicles and induce systemic tolerance by priming naïve T lymphocytes to become CD4 and CD8 regulatory T lymphocytes. CD11chighCX3CR1+ dendritic cells do not migrate, but can penetrate through the epithelial cell tight junctions to sample gut luminal contents213 and through the endothelium to survey circulatory antigens214. Dendritic cells

26 also activate B lymphocytes and promote their isotype class-switching and differentiation into IgA-producing plasma cells215.

2.2.4.3 Mast Cells

Mast cells comprise 2 to 5% of lamina propria mononuclear cells and are located either in the lamina propria or adjacent to nerve terminals. They maintain homeostasis by expressing proteases that protect against invasive pathogens and regulating claudin-3 expression that helps to control epithelial cell turnover. In response to acute infections, they rapidly release an array of pre-formed inflammatory mediators such as histamine, serotonin, cytokines and proteases. They also play a role in intestinal motility and nociception216-220.

2.2.4.4 Eosinophils

Eosinophils are present in normal colon as a defence against helminths. The number varies between individuals and geographical variation has also been reported221. In two paediatric studies, there were up to 50 eosinophils per high power field with a proximal to distal gradient; a mean of 20-35 eosinophils/HPF were counted in the ascending colon- caecum and 8-10/HPF in the rectum222, 223. As eosinophils are elevated in atopy, there may also be variations in eosinophil counts in response to seasonal antigenic changes or other allergic triggers.

2.2.4.5 Neutrophils

Neutrophils are rarely present in non-inflamed intestinal mucosa. During bacterial infection or acute inflammation, there is vasodilation and increased capillary permeability and, within minutes, circulating neutrophils are rapidly recruited to the intestine. Neutrophils respond to the pathogen by phagocytosis, release of granular antimicrobial peptides and proteolytic enzymes, and formation of neutrophil extracellular traps. They also release monocyte chemoattractants, including CAP18, cathepsin G and azurocidin, which recruit CD14+ macrophages from circulating monocytes to the site of inflammation, thus enabling a second-wave inflammatory response224-227.

27 2.2.4.6 Innate Lymphoid Cells

Innate lymphoid cells [ILCs] are a group of recently characterised primitive lymphocytes that are important in immune homeostasis. They are of a common lymphoid origin and require the γ-chain of the interleukin-2 receptor and the inhibitor of DNA binding 2 (ID2) transcription factor for their development. They lack myeloid and dendritic cell phenotypic markers, as well as recombination activating gene -dependent rearranged antigen receptors228.

ILCs can be categorised by the transcription factors that regulate their development and function and their cytokine profile. Group 1 ILCs include natural killer cells, they require the transcription factor T-bet and they produce IFN-γ in response to IL-12 and IL-18. Natural killer cells are predominantly CD3-CD56+CD16-, they do not express perforin and, unlike most natural killer cells, they have a predominant helper rather than a cytotoxic function229-231. Group 2 ILCs, also known as nuocytes232, require GATA binding protein 3 and are able to respond rapidly to helminth infections by producing IL-5 and IL-13 in response to IL-25 and IL-33. Group 3 ILCs include lymphoid tissue inducer cells233, 234, are dependent on retinoic acid receptor-related orphan receptor [RORγt] and produce IL-17 and/or IL-22 and GM-CSF in response to IL-23 or IL-1β 235, 236.

ILCs have functionally similar counterparts in T helper cells: ILC1s and Th1 cells, ILC2s and Th2 cells, and ILC3s and Th17 and Th22 cells, respectively237, and therefore ILCs are likely to produce the initial rapid cytokine response before the acquired immune system is activated. As knowledge about ILCs expands, this classification system is becoming further refined238, 239.

NKp44+RORγt+ILC3 (NK22) cells, in particular, have been implicated in regulation of the gut microenvironment. They are the main source of intestinal IL-22 which stimulates epithelial cells to proliferate, secrete IL-10, generate antimicrobial products and suppress the reactivity of commensal bacteria-specific T lymphocytes, thus helping to maintain the epithelial cell barrier in the healthy state240, 241.

28 2.2.5 Acquired Immune Cells of the Lamina Propria

2.2.5.1 T Lymphocytes i. Regulatory T lymphocytes

T lymphocytes with regulatory functions are abundant in the intestinal lamina propria. These cells are important in maintaining tolerance in the gastrointestinal tract and include CD4+CD25+FOXp3+ regulatory T lymphocytes [Treg]242, 243 and FOXp3 negative CD4+ TGF- β-secreting Th3 cells244, CD4+ IL-10-secreting Tr1 cells245 and CD8+ suppressor T lymphocytes246. CD4+CD25+FOXp3+ Tregs account for the main regulatory activity in the colon247-249. Approximately half are natural thymic-derived Tregs that express neuropilin- 1 and the rest are peripherally-derived antigen-induced Tregs250-253. Tregs express the IL- 2 receptor, CD25, and the transcription factor, forkhead box protein 3 [FOXp3], both of which play crucial roles in Treg differentiation, regulation, metabolism and suppressive functions243, 249, 254. Ongoing FOXp3 has been shown to be critical in maintaining Tregs in a dose- and time- dependent manner255, 256. Tregs suppress effector T lymphocyte functions257, 258 though the production of large amounts of immunosuppressive cytokines such as IL-10, TGF-β259 and in mice, IL-35256, 260, 261. They also have directly cytolytic capacity, can disrupt IL-2 or cAMP pathways necessary for effector cell metabolism and suppress dendritic cell maturation and function through CTLA-4 and LAG3262, 263. Colonic bacterial antigens have been shown to induce generation of Tregs, resulting in oral tolerance253. In particular, Clostridium-derived butyrate has recently been demonstrated to induce peripherally derived Tregs 264, 265. ii. Helper T lymphocytes

CD4+ helper T lymphocytes [Th] are critical in mediating adaptive immunity. They are neither cytotoxic nor phagocytic; instead they recognise antigens associated with MHC class II molecules on antigen-presenting cells and, in response, secrete cytokines that have a wide range of actions. Similar to innate lymphoid cells, there are 3 types of Th cells distinguishable by the transcription factors controlling their differentiation from naïve CD4+ T lymphocytes and by their subsequent cytokine profile266. Th1 cells require the transcription factor T-bet, secrete IFN-γ, lymphotoxin-α and IL-2 and activate macrophages to defend against intracellular pathogens. Th2 cells are regulated by GATA-3

29 and mediate host defence against parasites by inducing mast cell, basophil and eosinophil activation through production of IL-4, IL-5 and IL-13. In addition, they activate IgE production, thus playing a role in allergic conditions. Th17 cells differentiate under the influence of ROR-γt, produce IL-17 and IL-22, activate mononuclear phagocytes, recruit neutrophils and induce epithelial antimicrobial responses, thus protecting against extracellular pathogens267, 268. Recently, functional plasticity of the CD4+ T cell subsets has been described, particularly between the Th1 and Th17 lineages and this may play a key role in different autoimmune and inflammatory conditions269-273. iii. Other T lymphocytes

Cytotoxic CD8+ T lymphocytes are important in direct pathogen killing. They recognise antigens associated with MHC class 1 molecules and when activated, express and release perforin and granzymes that mediate direct cytolysis of target cells274-276.

NKT lymphocytes are present in small numbers in the gastrointestinal tract. Whilst they have features of NK lymphocytes, they also recognise self and microbial glycolipid antigens presented by the atypical MHC class 1 molecule, CD1d. Whilst their role in the gastrointestinal tract is not fully elucidated, they interact bi-directionally with microorganisms in the lumen and thus are important regulators of intestinal inflammation and homeostasis277-280.

2.2.5.2 B Lymphocytes and Plasma Cells

Plasma cells comprise approximately 10 to 40% of the cells in the lamina propria, depending on the method of analysis202, 281, 282. They terminally differentiate from mucosal B lymphocytes under the influence of TGF-β, IL-10, IL-4, IL-5 and IL-6. They produce approximately 40 mg IgA dimers and polymers per kg body weight per day that are selectively transported across the epithelial cell layer by Fc receptor-mediated transcytosis to provide humoral mucosal immunity. In the lumen IgA binds to microorganisms, reducing their motility and preventing adhesion and penetration through the epithelial cell layer. IgA in the lamina propria can bind antigen that has breached the epithelial barrier, leading to their excretion into the lumen, and can also inhibit virus production283-286.

30 Besides producing antibodies, B lymphocytes have other immune modulatory functions, including antigen presentation and cytokine production287. Recently IL-10-producing regulatory B lymphocytes have been shown in mice to be up-regulated by CD1d in chronic intestinal inflammation, where they limit excessive immune responses288-291.

2.2.6 Non-Haematopoietic Cells and the Colonic Stroma

The haematopoietic cells of the lamina propria lie within an extracellular matrix of proteoglycans, collagens, elastins, fibronectins glycoproteins, glycosaminoglycans and tenascins292. Multiple populations of non-haematopoietic stromal cells produce the structural framework of the intestine, synthesizing and secreting these connective tissue elements. They comprise at least 20% of the cells of the lamina propria202 and include fibroblasts, myofibroblasts, pericytes, vascular and lymphatic endothelial cells and smooth muscle cells. They are negative for lymphocyte common antigen, CD45; instead they express the intermediate filaments, vimentin and α-smooth muscle actin, and the thymocyte differentiation antigen 1 glycoprotein, CD90. 293, 294. Interstitial cells, comprising the interstitial cells of Cajal, telocytes and fibroblast-like cells, are also present in the subepithelial compartment, and these cells assist with normal gastrointestinal motor function295.

The role of the intestinal stromal cells, however, is more than just provision of a physical scaffold; this dense network of cells also has important immune functions interacting with both epithelial and haematopoietic immune cells. They respond to a variety of cytokines such as TNF-α, IL-1, IL17A, IL-4, IL-31, IFN-γ, TGF-β296 and produce a diverse range of cytokines, chemokines, growth factors and other mediators including IL-6, IL-8, CCL2, IL- 11, IL-33, GM-CSF, M-CSF and MCP-1297-300.

Stromal cells maintain gut homeostasis through bidirectional cross-talk with epithelial cells. They promote proliferation of the adjacent epithelial cells and repair and regeneration of the epithelial barrier. Stromal cell-derived IL-24 activates mucin production in vitro301. Stromal cells, in turn, are tightly regulated through feedback from Indian hedgehog homolog produced by the epithelial cells. This controls their development and orderly arrangement within the mucosa302-306.

They have an important homeostatic role in limiting antigen-specific immune responses through strong expression of B7 negative costimulators PD-L1 and PD-L2, regulation of

31 activated CD4+ T lymphocytes via direct cell contact and soluble mediators and induction of regulatory T lymphocytes, thus controlling the extent of the inflammatory response307- 310. They also produce IL-10 and prostaglandin E2, factors that can enhance generation of regulatory dendritic cell populations311.

Additionally, they are able to function as innate immune cells, directly responding to pathogens that breach the epithelial barrier. Fibroblasts and myofibroblasts express mRNA to the nucleotide-binding oligomerisation domain-containing proteins, NOD1 and NOD2, and the Toll-like receptors, TLR2 and TLR4, that bind bacterial components such as lipopolysaccharide, leading to expression of pro-inflammatory mediators and antigen presenting cells and T lymphocytes312-314.

32 2.3 Chronic Intestinal Inflammation: Crohn’s Disease and Ulcerative Colitis

Dysregulation of these tightly controlled homeostatic mechanisms may lead to development of the chronic inflammatory bowel diseases, Crohn’s disease and ulcerative colitis. When this occurs, it is believed that genetically predisposed individuals mount an excessive immunological response to an altered intestinal environment, resulting in chronic inflammation and tissue damage. These conditions are influenced by a variety of environmental factors and are characterised by dysbiosis and increased mucosal permeability.

Figure 2.2 Proposed paradigm of chronic inflammatory bowel diseases pathogenesis

2.3.1 Epidemiology

Crohn's disease and ulcerative colitis are estimated to affect approximately 1 in 300, or >75 000 Australians315. There are ethnic and regional differences in incidence and prevalence, highlighting the importance of genetics, environmental factors and gut microbiota. Recent studies estimate the age-standardised incidence of inflammatory bowel disease in Australia to be between 24.5 and 29.6 per 100 000316, 317. This rate is similar to other high-incidence areas, including Canada (29.2 per 100 000)318, 319, New

33 Zealand (25.2 per 100 000)320, Denmark (23.1 per 100 000)321, the United Kingdom (22.2 per 100 000)322 and North America (17.7 to 27.9 per 100 000)323, 324. It is, however, rare in indigenous populations in Australia325, 326, New Zealand320 and Canada319 and in ethnic minority and Hispanic Americans327. There are regional variations within countries and continents, with broadly a north-south328, 329 and west-east330 gradient in Europe, and a north-south gradient in North America331. The incidence of inflammatory bowel disease is increasing in some countries, particularly in Asia, North Africa and Eastern Europe317, 332- 334 attributed to urbanisation and improved hygiene and socioeconomic status. The ratio of Crohn’s disease to ulcerative colitis cases varies in different regions worldwide and, in the Australia, the incidence of Crohn's disease is 1.6 to 2.0 times that of ulcerative colitis316, 317.

Of the proposed environmental risk factors, the strongest link is with current cigarette smoking and Crohn’s disease335; this is associated with a more aggressive disease course, frequent relapses, earlier post-operative recurrence and higher frequency of extraintestinal manifestations336-339, particularly in females340. The detrimental effects of smoking may be mediated by nicotine, carbon monoxide or hypoxia341, 342, changes in intestinal microbiota343 or possibly through impairment of autophagy in macrophages as has been demonstrated in alveolar macrophages344. Smoking cessation fortunately improves the clinical outcomes in these patients345.

2.3.2 Clinical Aspects

Whilst Crohn’s disease and ulcerative colitis are both incurable chronic inflammatory diseases of the intestine, there are significant clinical differences between the two conditions. In ulcerative colitis, there are superficial inflammatory changes of the mucosa and submucosa of the colon, involving the rectum and extending proximally in a continuous fashion, although periappendiceal inflammation may also be present. The inflammation is deeper in Crohn’s disease and is associated with fistulae, strictures and abscesses. Although it most commonly involves the terminal ileum, any part of the gastrointestinal tract can be affected in a discontinuous distribution pattern.

2.3.3 Genetics

Both Crohn’s disease and ulcerative colitis are complex polygenic disorders. Genome wide association studies have identified over 160 non-overlapping risk loci associated with the 34 inflammatory bowel diseases, with around 30 shared between Crohn's disease and ulcerative colitis346-349. Despite this, these susceptibility genes explain less than 15% of total disease variance348. The concordance rate in monozygotic twins in European twin cohort studies is 30-70% in Crohn's disease, but only 10-20% in ulcerative colitis350, 351, suggesting that other genetic and environmental factors may be important co-factors influencing disease penetrance.

The strongest associations with specific single gene polymorphisms are linked to the innate immune system347. The best characterised of these are the caspase recruitment domain-containing protein 15 [CARD15] gene mutations on chromosome 16 that are associated with Crohn's disease352-354. CARD15 encodes NOD2, a cytosolic receptor on intestinal epithelial cells, monocytes, macrophages, T and B lymphocytes, Paneth cells and dendritic cells that is involved in bacterial recognition355-359. Although more than 25 mutations of the CARD15 gene have been described, there are three common functional mutations accounting for the majority of susceptibility, the missense mutations, R702W and G908R, and a frameshift mutation, 1007fs353, 354. Certain CARD15 mutations are associated with a younger age at onset and a more aggressive and complicated disease course with a stricturing phenotype and increased need for surgery360-363.

Other susceptibility genes that code for components of the innate immune system, include the autophagy-related 16-like 1 (ATG16L1)364, 365 and autophagy366-368 genes. The adaptive immune response is also important in inflammatory bowel disease, as highlighted by IL-23 receptor polymorphisms369. The R381Q variant which impairs IL-23-induced Th17 effector responses370 is protective against Crohn’s disease in children371 whereas the rs10889677 variant results in dysregulated IL-23R signalling372.

2.3.4 Dysbiosis

The luminal microbiota play an important role in inflammatory bowel disease. Experimental mouse models do not develop colitis in a germ-free environment373-375, patients with inflammatory bowel disease respond to antibiotic treatment and, in patients with Crohn’s disease, diversion of the faecal stream induces remission and mucosal healing376, 377 and postoperative recurrence is triggered by luminal contents378, 379.

The microflora of the intestine has been studied using stool culture techniques, ribosomal RNA gene sequencing380 and, more recently, metagenomic sequencing381, 382. It is well-

35 established that in inflammatory bowel disease there is an imbalance in the normal microbial environment of the colon383, 384, with some differences between Crohn’s disease and ulcerative colitis385, 386. There is reduced biodiversity387, and a decrease in phyla Firmicutes, particularly Faecalibacterium prausnitzii388-391, which may be predictive of relapse390, 392. In vitro, F. prausnitzii induced the production of anti-inflammatory cytokines, IL-10, IL-12 and TGF-β1, inhibited IL-17, and up-regulated CD25+Foxp3+ regulatory T lymphocytes393, 394. In addition, an increase in subdominant and potentially pathogenic bacteria is often observed, such as phyla Proteobacteria391, particularly adherent and invasive Escherichia coli395-399 which has been linked with dysregulated autophagy400, Bacteroidetes401-403, Lactobacillus and Bifidobacterium404. The role of Campylobacter concisus and its potential for increasing intestinal permeability in inflammatory bowel disease405 is also being explored. An increase in diversity of fungi has also been demonstrated406, correlating with serum C-reactive protein and Crohn’s disease activity index and associated with increased expression of TNF-α, IFN-γ and IL-10407. Recently, alterations in the virome have been shown; not only are there significant increases in Caudovirales bacteriophages, there are also interesting differences in bacteriophage richness between Crohn’s disease and ulcerative colitis408.

Variations in intestinal microbial composition and diversity between ethnicities and regions within the same country support the premise that genetic and environmental factors shape the intestinal microbiome in both health and inflammatory bowel disease409, 410. Other ecological variables are being explored, such as the influence of cigarette smoking 343, 411 and diet and intake412, as well as potential management strategies that alter the gut microflora413 including probiotics414 and, possibly, faecal transplantation415- 417.

2.3.5 Increased Intestinal Permeability

Increased epithelial permeability has been described in patients with Crohn’s disease418, 419, as well as in healthy first-degree relatives420-422 and some spouses of patients with Crohn’s disease423-425. There is an associated frameshift 3020insC mutation in the NOD2 gene426, 427, which may lead to decreased expression of α–defensins by Paneth cells428, 429, allowing dysbiosis to occur. Conversely, dysregulation of the intestinal barrier may be induced or perpetuated by luminal contents, resulting in changes in tight junction claudins and other barrier proteins430, 431 and subsequent reductions in α–defensins.

36 Increased permeability is linked to a higher risk of relapse432-434, particularly in small intestinal Crohn’s disease435. It resolves following successful anti-tumour necrosis factor treatment436, 437 suggesting that TNF-α may possibly trigger alterations in the para- and transepithelial pathways that lead to loss of normal barrier function, dysregulation of gut microbiota and subsequent intestinal inflammation. This has been demonstrated in SCID mice, where CD4+ T lymphocyte transfer stimulated production of TNF-α, increased para- cellular permeability and decreased epithelial resistance and net ion transport, even before overt microscopic inflammation occurred438.

Increased permeability also occurs in ulcerative colitis439, 440 but is less well- characterised. Lamina propria mononuclear cell-derived IL-13 up-regulated epithelial cell apoptosis and expression of the pore-forming tight junction protein, claudin-2, and decreased epithelial restitution velocity441, 442. This cytokine is increased in ulcerative colitis443 and is likely to be a key effector molecule in barrier dysfunction.

2.3.6 Immunological Changes

There are three reasons to believe that Crohn’s disease and ulcerative colitis have a strong immunological basis.

1. The role of the appendix a) Epidemiology

Evidence for the protective role of early appendicitis and appendicectomy on subsequent ulcerative colitis risk initially came from a multicenter paediatric study in 1987444 and, whilst confirmed in further case-control studies445, 446, subsequent large cohort studies provided divergent results447, 448. A prior appendicectomy in patients with ulcerative colitis may have a protective effect resulting in a more benign disease course449. Appendicectomy following diagnosis of ulcerative colitis is controversial and has been noted to be of benefit in small case reports and case series only450-452. Overall, these studies suggest a protective effect of appendicitis and appendicectomy on ulcerative colitis.

In variance to ulcerative colitis, there is possibly an increased risk of developing Crohn’s disease in the first few years after an appendicectomy. This was most pronounced in the first 6 to 12 months, with a systematic review of observational studies estimating the

37 relative risk to be 6.69 (95% CI = 5.42-8.25) within the first year 453 and a large cohort study estimating the standardised incidence ratio to be 8.69 (95% CI = 7.68–9.84) in the first 6 months454 following surgery. The risk diminished rapidly to reach background levels after 5 years and this initial peak may be partly due to a diagnostic confounding bias. The presence or absence of intra-abdominal inflammation may affect the risk of later Crohn’s disease, however, only some of these studies confirmed the histopathological diagnosis of appendicitis. Prior appendicectomy, however, has been linked to increased ileal disease, it negatively correlates with anal fistulae, whilst having no association with Crohn’s disease severity455. b) Mouse models of colitis and appendicitis

To support the epidemiological findings, the role of the appendix in various mouse models of colitis has been studied. The appendix is embryologically derived from the caecum and is characterized by a large number of lymphoid follicles in the lamina propria that contain germinal centres456. IgG-secreting B lymphocytes are abundant in the appendix457, which is also a source of extrathymically-derived T lymphocytes457, 458. The protective effect of appendicectomy on subsequent development of colitis was demonstrated in T-cell receptor-α(-/-) mutant mice459, a dextran sulphate sodium-induced mouse colitis model460 and a CD62L+CD4+ T lymphocyte adoptive transfer model of colitis in severe combined immunodeficient mice461.

A recent mouse model of appendicitis462 has demonstrated significant increases in CD8+Foxp3+ regulatory T lymphocytes in younger mice462, and protection against TNBS- induced colitis partly through suppression of T helper-17 pathway and autophagy gene expression in the distal colon463, 464. c) Peri-appendiceal inflammation

Peri-appendiceal inflammation may occur as a “skip lesion” in patients with ulcerative colitis465, 466, particularly with distal colonic involvement467, 468, histologically resembling typical changes seen in the actively inflamed colon469 although the significance of this is yet to be determined470.

38 2. The presence of antibodies

Serological autoantibodies associated with the chronic inflammatory bowel diseases highlight the role of the immune system in these conditions. The two in current clinical use are perinuclear anti-neutrophil cytoplasmic antibody [p-ANCA]471, 472, possibly directed against a neutrophilic nuclear histone473, and anti-Saccharomyces cerevisiae antibody [ASCA]474, 475, directed against the cell wall mannan of the yeast. These are present in ulcerative colitis and Crohn's disease, respectively, and are most useful in distinguishing between the two conditions476. A number of other antibodies against components of bacterial and fungal cell walls have been identified477, including anti- Escherichia coli outer membrane porin C (Omp-C)478, anti-flagellin (cBir1)479 and anti- Pseudomonas aeroginosa (I2)480 with varying sensitivities and specificities for these conditions481. Levels of these antibodies are stable over time regardless of disease acitivity481, 482 and their presence in Crohn’s disease is associated with a more severe disease course that includes stricturing or penetrating disease and need for small bowel or early surgery482-485, hence a panel of antibodies may be useful for predicting prognosis.

3. The immune basis for medical treatments for Crohn’s disease and ulcerative colitis

The chronic inflammatory bowel diseases are incurable. The mainstay of treatment targets the immune system and current management aims to modulate the clinical disease course486 with treatment endpoints that include endoscopic and mucosal healing487 and the decreased need for hospitalization and surgery488. For mild disease489, management is based upon the 5-aminosalicylic acid group of compounds, for which a number of anti- inflammatory mechanisms have been proposed, including inhibition of antibody secretion490, NF-κB in macrophages491, prostaglandin E2 production492, synthesis of chemoattractants by neutrophils493, neutrophil degranulation494, reactive oxygen formation495 and IL-2 production by activated T lymphocytes 496. Broad suppression of cell-mediated immunity with corticosteroids497, 498 is useful for control of active inflammation. Immunomodulatory medications such as the antimetabolites, azathioprine, 6-mercaptopurine and methotrexate499, and the calcineurin inhibitors, cyclosporine and tacrolimus, inhibit synthesis of nucleic acids and inhibit T lymphocyte activation, respectively. These, too, result in general suppression of the immune system.

39 Newer medications focussed on specific immune targets, the prototype of these being the anti-TNF antibodies, infliximab and adalimumab. Anti-integrin antibodies against α4 (natalizumab)500, 501 or α4β7 (vedolizumab)502, 503 are useful for Crohn’s disease and ulcerative colitis, although natalizumab has been withdrawn from clinical use due to its lack of gastrointestinal specificity and resultant cases of progressive multifocal leukoencephalopathy504. A number of promising therapies are in the pipeline505, some currently in phase 2 or 3 clinical trials, which target smad-7506-508, the IL-12/IL-23 cytokine component p40509, IL-23 p19510 and Janus kinase511, 512.

2.3.6.1 Innate Immune Cell Changes

Epithelial cells, innate immune cells and stromal cells sense invading microbial pathogens through Toll-like and NOD-like pattern recognition receptors513 and initiate rapid inflammatory responses. In inflammatory bowel disease, this may be triggered by pathogens that readily breach the more permeable epithelial barrier. In addition to this, abnormal sensing of and response to commensal organisms has been demonstrated in vitro and in animal models, particularly by dendritic cells514. As increasing understanding of dysbiosis occurs, it is likely that these abnormal pro-inflammatory responses to the altered intestinal microbial composition, partly through enhancement of selected commensal organisms, will be better characterised and the implications for therapies will be explored.

Neutrophils are rapidly recruited from the blood stream226 and, in addition, there are increased numbers of activated mast cells and, consequently, higher levels of tryptase in intestinal tissue and peripheral circulation. This induces release of inflammatory mediators such as IL-1β, IL-6, CXCL1 and the matrix metalloproteinases, MMP3 and MMP13515, 516.

Intestinal macrophages are naturally down-regulated203 and this enables them to maintain homeostasis in the healthy quiescent state. In Crohn’s disease, whilst they have normal mRNA levels of TNF-α, intracellular levels are low, suggesting intracellular breakdown517. These cells are unable to mount an acute inflammatory response, leading to impaired antigen clearance and likely formation of granulomas518, one of the pathognomonic features of Crohn’s disease519, 520.

40 In inflammatory bowel disease, an effective antimicrobial macrophage response ensues from an increase in newly recruited blood monocytes that express CD14 and calprotectin521, 522, generate reactive oxygen metabolites523, 524, show enhanced antigen presentation525 and secretion of cytokines526.

Cytolytic IFN-γ producing NK lymphocytes have been shown to be increased in the colon in Crohn’s disease and ulcerative colitis, and are suppressed by azathioprine527. Recently, type 3 innate lymphoid cells producing IL-17 and IFN-γ in response to IL-23528 and type 1 innate lymphoid cells producing IFN-γ in response to IL-12 and IL-15 have been implicated in inflammatory bowel disease529.

2.3.6.2 Acquired Immune Cell Changes

Multiple alterations in T lymphocyte expression and function are seen in inflammatory bowel disease. In Crohn’s disease, there is an imbalance of T regulatory lymphocytes and recently this has, in mice, been shown to be due to IL-23 inhibition of IL-33, leading to suppression of T regulatory lymphocyte differentiation and accumulation in inflamed tissues530. This altered T regulatory lymphocyte function leads to excessive pro- inflammatory Th1 and Th17 -type responses by CD4+ T lymphocytes531.

Naïve T lymphocytes expressing the transcription factor, T-bet, in the presence of IL-12, preferentially differentiate into Th1-type CD4+ T lymphocytes532. The Th1 immune response has been well-characterised in Crohn’s disease, the main effector cytokines include IL-1, IL-6, TNF-α and IFN-γ533-535.

In the presence of TGF-β and IL-6, however, naïve T lymphocytes differentiate into Th17 lymphocytes. IL-1 and TNF-α enhance this cell differentiation, whilst IL-23 is important in maintaining Th-17 survival and expansion. Th17 lymphocytes produce IL-17A, IL-17F, IL- 22 and IL-26 in inflammatory bowel disease and in Crohn’s disease, a subset also secrete IFN-γ515, 532, 536-541.

There is also persistence of activated T lymphocytes that fail to undergo apoptosis542, and this disturbed clearance of over-reactive or auto-reactive T lymphocyte populations in Crohn’s disease is one of the mechanisms targeted by therapeutic anti-tumour necrosis factor antibodies543, 544.

41 There is great overlap and redundancy in the production and actions of the multiple cytokines, chemokines, growth factors and other immune mediators, that enable interactions between the epithelial, innate and acquired immune and stromal cells. Balance of these signals occurs in health whilst dysregulation is a crucial component in the pathogenesis of Crohn’s disease and ulcerative colitis.

Whilst Crohn’s disease is predominantly Th1/Th17-mediated, and ulcerative colitis is broadly Th2/Th17-mediated, less is known about the immunopathology of ulcerative colitis. The Th2 cytokine profile in ulcerative colitis is atypical, and increased levels of IL-5, IL-6 and IL-10 have been demonstrated. In addition, increased levels of IL-13 from activated NKT lymphocytes may contribute to colonic inflammation through changes in epithelial permeability443, 545, 546.

Besides changes in levels of pro-inflammatory cytokines mentioned above, important immune-regulatory cytokines in the intestine, particularly IL-10, IL-4, IL-13 and TGF-β, are suppressed in inflammatory bowel disease547.

As discussed in the preceding section 2.3.6, B lymphocytes produce antibodies to both bacterial and non-bacterial antigens. They are non-universally expressed and their functions are unclear, although in Crohn’s disease, the presence of multiple antibodies signifies a poorer prognosis possibly signifying broader dysregulation and loss of tolerance in these patients.

2.3.6.3 Stromal Cell Changes

IL-33 is a cytokine of the IL-1 family that is expressed on a wide range of lymphocytes, has a number of mainly Th2-associated functions, including promoting regulatory T lymphocyte function in the intestine. IL-33 levels are increased in ulcerative colitis and correlate with disease activity and, whilst it is preferentially expressed by colonic stromal myofibroblasts, its role whether protective or pathological is not yet elucidated530, 548, 549.

In a mouse model of Crohn’s disease, intestinal stromal cells were activated by TNF-α, leading to up-regulation of intercellular adhesion molecule 1 (ICAM-1) and matrix metalloproteinase550. Colonic myofibroblasts from patients with active Crohn’s disease expressed higher levels of membrane-bound TNF-α and in vitro increased tissue inhibitor of metalloproteinase 1 (TIMP-1) production in response to anti-TNF agents551.

42 As intestinal stromal cells are also able to respond to IL-1 and IL-17A299, 552, 553, these pro- inflammatory cytokines are likely to trigger immune responses in these cells in inflammatory bowel disease.

Therefore, as the body’s largest lymphoid organ, multiple complex interactions occur between the numerous cells in the gastrointestinal tract, both in health and in inflammation. In an environment juxtaposed with commensal microorganisms and foreign luminal antigens, immune tolerance is crucial. The diverse epithelial, stromal and innate and acquired mucosal immune cells signal in a concerted manner via various cytokines, chemokines, growth factors and other cell mediators to maintain homeostasis. Whilst the pathogenic mechanisms of Crohn’s disease and ulcerative colitis are not fully understood, dysregulation and dysfunction of these tightly controlled mechanisms result in abnormal immune responses that drive the chronic inflammation and consequent tissue damage.

43 2.4 Hypotheses and Aims

2.4.1 Hypotheses

Mechanisms by which the gastrointestinal tract down-regulates its immune responsiveness have not been fully elucidated. The presence and function of leucocyte immunoglobulin-like receptors in the colon are not well-characterised, but there is a strong likelihood that these receptors are involved in constitutive gastrointestinal immune tolerance and an imbalance in their number or function contributes to the pathogenesis of chronic inflammatory bowel disease.

Leucocyte immunoglobulin-like receptors are predominantly expressed on cells of the monocytic-macrophage lineage and, in the intestine, resident tissue macrophages are down-regulated cells that have an important role in maintaining immune homeostasis. Local factors in the colonic milieu are likely to influence the differentiation of newly recruited blood monocytes into immune-tolerant intestinal-specific macrophages, and this may be able to be replicated in vitro using colonic stromal-derived conditioned media.

2.4.2 Aims

The aims of this thesis were 1. to describe the pattern of leucocyte immunoglobulin-like receptor expression in human colonic tissue in “normal” controls and patients with inflammatory bowel disease. In particular, to describe which leucocyte immunoglobulin-like receptors are present, which cell types express these leucocyte immunoglobulin-like receptors and the differences that occur in the chronic inflammatory bowel diseases; and 2. to develop a cell culture model of colonic-like macrophages to enable further in vitro characterisation of leucocyte immunoglobulin-like receptors in these cells.

44 3 Subject Recruitment and Clinical Details

3.1 Introduction

This research focussed on the presence and role of leucocyte immunoglobulin-like receptors in the human colon. Subjects with chronic inflammatory bowel disease, as well as control subjects undergoing a colonic resection for other causes, were recruited to provide intestinal and blood samples. Buffy coats from healthy Australian Red Cross Blood Service donors were obtained for cell culture experiments. Some of the colonic tissue used in the immunohistochemistry experiments was archival.

3.2 Recruitment of Subjects

Patients having a colonic resection were invited to participate ↓ Written consent from participating subjects ↓ Structured medical history ↓ Blood sample prior to the operation ↓ Colon +/- ileum collected at the time of surgery ↓ Follow-up questionnaire for subjects with inflammatory bowel disease

3.2.1 Subjects Undergoing a Colonic Resection

Patients undergoing a colonic resection were recruited through collaborating colorectal surgeons at two hospitals, Prince of Wales Hospital, Randwick and St George Hospital, Kogarah. Subjects were required to be at least 18 years of age, able to give informed consent and speak English adequately to provide a medical history.

A total of 43 subjects were recruited for the various experimental studies. Six had chronic inflammatory bowel disease (4 Crohn's disease and 2 ulcerative colitis). The other subjects underwent a colonic resection for colon cancer, non-malignant colonic polyps, diverticular disease or angiodysplasia.

45 Participating subjects gave informed written consent before undergoing a structured medical interview. A blood sample and a portion of the resected colon were collected at the time of operation. Subjects with chronic inflammatory bowel disease were contacted 2 months after the operation for a follow-up questionnaire.

3.2.2 Medical History

A medical history was obtained from subjects prior to their operation using a structured template (see Appendix 1). Clinical details were confirmed using the patient’s medical record. Demographic data, previous operations, medications, smoking history and family history of medical conditions were obtained (see section 3.4). Just over half the subjects were Australian-born; their parents were also born in Australia. One subject born in Germany had Australian-born parents. All other subjects born overseas had parents born in those countries. There were no subjects of Aboriginal or Torres Strait Islander descent. As expected, subjects with inflammatory bowel disease were significantly younger than control subjects, p = 0.0003, and subjects undergoing colonic resection for colonic polyps or cancer were predominantly male.

Details of other medical conditions were collected; besides common medical conditions such as cardiovascular disease, hypertension and diabetes mellitus, the questionnaire included a variety of conditions that are associated with the chronic inflammatory bowel diseases. These include the extraintestinal manifestations (primary sclerosing cholangitis, ankylosing spondylitis, iritis/uveitis, pyoderma gangrenosum, erythema nodosum), autoimmune conditions (e.g. thyroid disease, psoriasis) and other noted associations (e.g. arthritis, demyelinating disorders)554-561.

Current clinical status using the Simple Clinical Colitis Activity Index [SCCAI] and body mass index [BMI] data were collected. The SCCAI is a validated scoring system of disease activity in colitis562 encompassing bowel frequency, urgency of defaecation, blood in the stool, general well-being and 4 extra-colonic features (arthritis, pyoderma gangrenosum, erythema nodosum and uveitis). Of the established colitis activity indices, the SCCAI does not involve laboratory or endoscopic parameters. It was shown to correlate strongly with the Powell-Tuck Index563, the Seo complex integrated disease activity index564 and laboratory parameters (albumin, haemoglobin, platelet count, erythrocyte sedimentation rate). It was designed for use in exacerbations of colitis and, whilst the scores range from

46 0 (lowest activity) to 19 (highest activity), remission has been defined as ≤2565, relapse as ≥5566 and a significant improvement as a decrease of ≥2565.

Subjects with chronic inflammatory bowel disease completed a post-operative health questionnaire that included current medications and clinical status. Five of the 6 subjects returned their questionnaire.

3.2.3 Blood Samples

A blood sample for baseline blood tests including a full blood count, electrolytes, renal function and liver function tests, and for isolation of PBMCs for laboratory studies was obtained pre-operatively.

3.2.4 Colonic and Ileal Tissue

At the time of colonic resection, a portion of the resected colon was collected. In subjects in whom terminal ileum was also available, such as those undergoing a right hemicolectomy, a portion of the resected ileum was also collected. The inflamed intestine was collected from subjects with chronic inflammatory bowel disease, and macroscopically normal intestine at least 2 cm away from the pathological lesion was collected from control subjects undergoing a colonic resection for other reasons.

For immunohistochemical and immunofluorescence studies, a small portion of full- thickness colon was washed twice in Dulbecco’s phosphate-buffered saline [D-PBS, Gibco, New York, USA], dried and immersed in Tissue-Tek OCT compound [Sakura Finetek, Zoeterwoude, The Netherlands] and then snap frozen on dry ice. For isolation of lamina propria mononuclear cells and generation of stromal-derived conditioned media, lamina propria was dissected off the underlying muscularis mucosae, washed twice in phosphate- buffered saline, and then transported back to the laboratory in cold Hank’s buffered salt solution [HBSS, Sigma-Aldrich, Missouri, USA]. Full thickness tissue and lamina propria was also collected in RNAlater solution [Ambion, Life Technologies, California, USA] at room temperature for RNA isolation, or snap frozen on dry ice for protein extraction.

47 3.3 Ethics

Ethics approval to perform this study was granted from the Human Research Ethics Committees at the University of NSW, and the South Eastern Sydney and Illawarra Area Health Services, covering the St George Hospital, Kogarah and Prince of Wales Hospital, Randwick. All subjects gave informed written consent to participate.

48 3.4 Subjects Undergoing a Colonic Resection

Inflammatory Bowel Disease Control Number of subjects 6 37 Gender 2 M, 4 F 23 M, 14 F Age 43 ± 7 years 66 ± 2 years Indication for Crohn’s disease: 4 Colorectal cancer: 22 surgery Ulcerative colitis: 2 Colonic polyp: 12 Angiodysplasia: 1 Diverticular disease: 2 Born in Australia Yes: 4 Yes: 18 No: United Kingdom = 2 No: New Zealand = 1 United Kingdom = 1 Ireland = 1 Western Europe = 3 Southern-Eastern Europe = 5 North Africa = 1 South East Asia = 1 South America = 1 N/A: 5 Smoking Never: 2 Ulcerative colitis Never: 7 Ex-smoker: 3 Crohn’s disease Ex-smoker: 17 Current: 1 Crohn’s disease Current: 7 N/A: 6 SCCAI 5.3 ± 1.3 1.8 ± 0.4 BMI 22.7 ±2.7 25.6 ± 0.9 Appendicectomy Yes, ≤ 18 years: 2 Yes, ≤ 18 years: 2 Yes, > 18 years: 2 Yes, > 18 years: 6 No: 2 No: 22 N/A: 7 Results expressed as counts, percentages or mean ± SEM; N/A = data not available SCCAI = Simple clinical colitis activity Index; BMI = Body mass index

49 3.6 Deviations from Initial Plans

3.6.1 Recruitment of Subjects

A sample size of 25 subjects with chronic inflammatory bowel disease and 50 controls was calculated a priori for the various components of this study, based on initial immunohistochemistry experiments. The number of subjects with IBD recruited overall was lower than expected. Of the first 21 subjects recruited in the initial 3 months, 6 had IBD. However, as there were difficulties with the leucocyte immunoglobulin-like receptor antibodies during the Western blotting and flow cytometry experiments, the latter experiments focussed on developing and studying an in vitro model of colonic-like macrophages using stromal-derived conditioned media from control subjects, and further subjects with IBD were not recruited.

All subjects with IBD had already been on long-term medical treatment; no subject presented with acute severe ulcerative colitis. The colectomies were performed for refractory disease and prior medical treatment is likely to have influenced leucocyte immunoglobulin-like receptor expression and/or function. The rate of colonic resection for IBD is lower than historical data suggested486, 567, likely due to advances in medical therapeutic options for these conditions. To overcome this, obtaining colonic mucosal biopsy specimens during endoscopic procedures would provide tissue samples from subjects with a broader spectrum of activity, stages of disease and medical therapies.

3.6.2 Obtaining the Structured Medical History

Obtaining a medical history from subjects, particularly control subjects, was more difficult than anticipated in the preoperative setting. Subjects with inflammatory bowel disease were more willing in general to talk about their illness and their medical history, as would be expected from subjects who have learnt to live with a chronic illness. Control subjects, however, often had a short lead time between the diagnosis of colon cancer or a significant colonic polyp requiring surgery and the operation, mostly within 2 weeks. This is likely to have influenced their responses to these questions, particularly their estimation of their general well-being. However, due to the change in direction of the research caused by technical limitations of the planned experiments, significant correlations between the clinical phenotypic data and experimental findings could not be made.

56 4 Presence and Expression Patterns of Leucocyte Immunoglobulin-Like Receptors in the Human Colon

4.1 Introduction

Vast numbers of phagocytic and antigen presenting cells are located in the human colon at the interface with the foreign food and microbial antigens, and occasionally pathogens, of the gastrointestinal lumen. These cells are crucial in maintaining homeostasis and responding rapidly to external insults. Leucocyte immunoglobulin-like receptors [LILRs] are expressed predominantly on these cells and, whilst their expression has been characterised in other tissues, little is known about their presence in the human colon.

In the following three chapters, the various experimental techniques used to determine the expression of leucocyte immunoglobulin-like receptors in the human colon are described. In this initial chapter, immunohistochemistry and immunofluorescence were performed on frozen tissue sections to demonstrate the histological pattern of leucocyte immunoglobulin-like receptors in the human colon and to make preliminary quantitative assessments.

4.2 Materials and Methods

4.2.1 Specimen Collection

Patients having a bowel resection were invited to participate in the study. At the time of colonic resection, a portion of the resected colon was snap frozen in Tissue-Tek OCT compound [Sakura Finetek, Zoeterwoude, The Netherlands]. Inflamed colon from 8 subjects with chronic IBD (Crohn’s disease or ulcerative colitis) was compared with macroscopically normal colon from 16 control subjects undergoing surgery for colonic polyps or cancer, diverticular disease or angiodysplasia.

4.2.2 Histopathology

Histological grading on routine hematoxylin and eosin stained sections was performed using a scoring system adapted from D’Haens et al379 and Riley et al568 to ensure the

57 absence of colonic inflammation in the control subjects. The items scored included the continuity of the epithelium and infiltration of polymorphonuclear and mononuclear leucocytes in the epithelium, lamina propria and submucosa, with a maximum possible score of 8.

Table 4.1 Histological scoring system

Category Score

A. Continuity of surface epithelium 0 = Normal 1 = Discontinuous and/or mucin depletion 2. Ulcers

B. Infiltration of polymorphonuclear 0 = Normal leucocytes in the lamina propria and 1 = Moderate increase epithelium 2 = Crypt abscess

C. Infiltration of mononuclear leucocytes in 0 = Normal the lamina propria 1 = Moderate increase 2 = Severe increase

D. Inflammatory infiltrate in the 0 = Normal submucosa 1 = Moderate increase 2 = Severe increase

Moderate increase = up to twice the normal number of cells; Severe increase = more than twice the normal number of cells.

4.2.3 Immunohistochemistry

4.2.3.1 Immunohistochemical Staining with Anti-LILR Antibodies

Immunohistochemistry was performed on sections of these frozen colons using monoclonal mouse anti-human antibodies against 8 of the LILRs (LILRB1, LILRB2, LILRB3, LILRB4, LILRA1, LILRA2 and LILRA5 [Amgen, Washington, USA] and LILRA3 [Novus Biologicals, Colorado, USA]) along with irrelevant mouse IgG1 controls [Dako, Glostrup, Denmark]. Five-micron thick sections of colon were cut onto silane-covered slides using a cryomicrotome. These were air-dried for 30 minutes and then acetone-fixed for 10 minutes. The slides were stored at -80°C until ready to use. The slides were equilibrated to room temperature with tris-buffered saline [TBS], pH 7.6, and then blocked with 20% goat

58 serum at room temperature for 30 minutes. Excess serum was removed and the tissue sections were incubated with primary antibody or irrelevant IgG1 controls at 4°C overnight. The following morning, the slides were washed 4 times each for 5 minutes with TBS prior to incubation with biotinylated goat anti-mouse antibody at room temperature for 60 minutes. Binding of the LILR antibodies was detected using an ABC-alkaline phosphatase enzyme-substrate reaction [Vector Laboratories, California, USA] and then the sections were counterstained with hematoxylin [Dako, Glostrup, Denmark]. The stained sections were dehydrated with 100% ethanol twice followed by xylene twice and then coverslips were applied. Subjects expressed a LILR if there was definite cell- associated staining in the lamina propria of >10 cells per high power field (x400 magnification).

4.2.3.2 Semi-Quantative Scoring of LILRB1 and LILRA5

Semi-quantitative scoring of LILRB1 and LILRA5 expression was performed by a separate investigator masked to the clinical diagnosis and antibody used. The numbers of LILR- positive cells in tissue sections were evaluated by counting immunoreactive cells in 3 randomly selected fields across the whole section as described elsewhere149, 175, 176, 569. In brief, after ensuring that sections stained with isotype control exhibited no significant immunoreactivity, the number of positive cells per field (×250 magnification) was enumerated.

4.2.3.3 Other Cell Markers

Serial sections were stained with antibodies to T and B lymphocyte, neutrophil, macrophage and mast cell markers along with irrelevant isotype controls (mouse monoclonal anti-human CD4, CD20, CD68, neutrophil elastase, mast cell tryptase, mouse IgG1, mouse IgG2a, and rabbit polyclonal anti-human CD3 and rabbit Ig [Dako, Glostrup, Denmark]). Semi-quantitative scoring of CD68 expression was performed in 4 subjects with IBD and 10 control subjects and these scores were correlated with LILRB1 and LILRA5 expression.

Background staining was an issue in the immunohistochemistry experiments and various experiments to determine the cause were performed. Different washing steps, blockers, substrate times and antibody concentrations were tested.

59 4.2.4 Immunofluorescence

Immunofluorescence using fluorochrome labelling of the primary antibodies was performed on 4 control subjects to confirm expression of LILRA5 on lamina propria macrophages. Five-micron thick tissue sections were cut onto silane-covered slides, air- dried and then acetone-fixed. The slides were equilibrated to room temperature with TBS and then blocked with 20% goat serum at room temperature for 30 minutes. Primary antibodies, LILRB1, LILRA1, LILRA5 [Amgen, Washington, USA], CD3 and CD68 [Dako, Glostrup, Denmark] or their irrelevant IgG controls, mouse IgG1 and rabbit Ig [Dako, Glostrup, Denmark] were labelled with Zenon Alexa Fluor 488 nm, 555 nm, 647 nm or 700 nm [Invitrogen, California, USA]. The slides were incubated with these fluorochrome- labelled antibodies at room temperature for 2 hours in the dark. The slides were washed 4 times each for 5 minutes with TBS, and then mounted in ProLong Gold Antifade with DAPI (4’,6’-diamidino-2-phenylindole) mounting media [Invitrogen, California, USA]. Glass coverslips were applied and the slides were dried at room temperature for 24 hours in the dark. The sections were imaged with an Olympus FV1000 inverted confocal laser microscope [Olympus, Tokyo, Japan].

4.2.5 Statistics

Two-sided t tests or Mann Whitney U tests, where appropriate, were calculated to compare frequency of LILR-positive cells between IBD and controls, and Spearman’s correlation coefficient, rs, to correlate LILR density and histological scores; statistical analyses were performed using GraphPad Prism 5.0b [GraphPad Software, California, USA].

60 4.3 Results

4.3.1 Patient Demographics and Histological Grading

Eight subjects with chronic inflammatory bowel disease (ulcerative colitis or Crohn’s disease) and 16 control subjects undergoing surgery for colonic polyps or cancer, diverticular disease or angiodysplasia were included. Subjects undergoing surgery for IBD were younger than control subjects (mean ± SEM age: IBD 34.5 ± 4.1 vs. controls 68.3 ± 3.0 years, p = 0.002).

All subjects in the control group scored zero on the histological grading score, consistent with the absence of inflammation in these samples. All subjects with IBD scored between 4 and 7 points, which corresponded with moderately severe inflammation.

4.3.2 Leucocyte Immunoglobulin-Like Receptor Expression

All subjects expressed LILRB1 and LILRA5 in the lamina propria of the colon and the majority of subjects also expressed LILRA1 and LILRA3. Although absent in control subjects, 25% of subjects with IBD expressed LILRB4 (p-value = 0.04). There were no significant differences in the expression of any of the other individual LILRs between IBD and control groups.

Figure 4.1 Proportion of subjects positively expressing LILRs in the lamina propria 61 4.3.3 LILRB1 and LILRA5 Expression

The number of positively stained lamina propria cells per high power field was significantly higher for LILRB1 in subjects with IBD, compared with the control subjects (mean ± SEM, 276.4 ± 69.9 vs. 104.5 ± 15.8, p = 0.004). There was no difference in LILRA5 density between groups (IBD 201.3 ± 48.5 vs. control 132.6 ± 18.4, p = 0.12).

Figure 4.2 LILRB1 and LILRA5 cell counts in the lamina propria

62 There was a strong positive correlation between LILRB1 and LILRA5 density, particularly in those with IBD (whole cohort R2 = 0.65, p = 0.0005; IBD R2 = 0.90, p = 0.005; control R2 = 0.53, p = 0.03).

Figure 4.3 Correlation between LILRB1 and LILRA5 density

There was also a strong positive correlation between CD68 positive cells and LILRB1 (R2 = 0.39, p = 0.02) and LILRA5 (R2 = 0.66, p = .0004) expression.

Figure 4.4 Correlation between LILRB1 / LILRA5 and CD68

There was, however, no linear correlation between LILR density and histological score (data not shown). 63 In subjects where the submucosa was also evaluable, a smaller number of cells per high power field was positive for LILRB1 (mean ± SEM, 12.0 ± 3.7) or LILRA5 (20.5 ± 4.0). There was, however, no significant difference in LILRB1 or LILRA5 receptor expression between IBD and control subjects (LILRB1: IBD 9.8 ± 5.3 vs. control 12.9 ± 4.9, p = 0.83; LILRA5: IBD 23.6 ± 9.0 vs. control 18.6 ± 3.9, p = 0.95).

Figure 4.5 LILRB1 and LILRA5 cell counts in the submucosa

The majority of subjects strongly expressed LILRA5 in the endothelium (81.2%); both LILRB1 and LILRA5 were highly expressed in lymphoid aggregates (83.3% and 92.9%, respectively).

64 4.3.4 Cellular Expression of Leucocyte Immunoglobulin-Like Receptors

Based on immunohistochemical staining of serial sections, LILRB1 and LILRA5 were predominantly expressed on CD68+ macrophages and CD3+ T lymphocytes of the lamina propria (Figure 4.6), as well as in CD20+ B lymphocytes, CD3+ T lymphocytes and CD68+ macrophages within the lymphoid follicles (Figure 4.7). LILRA5 was also expressed on endothelial cells (Figure 4.8). LILRA1 was predominantly expressed on lamina propria macrophages (Figure 4.6). LILRA3 showed a different staining pattern, being expressed mainly on tryptase+ mast cells (Figure 4.9). In some subjects, there was expression of LILRB2 and LILRB3 in the lamina propria (Figure 4.10).

The following 6 figures are photos of immunohistochemical sections from representative subjects. The antibody of interest is detected with alkaline phosphatase (pink). Cell nuclei are counterstained with haematoxylin (blue).

65

LILRB1 LILRA1

LILRA5 LILRA3

CD20 CD3

CD68 Mouse IgG1 Isotype Control Figure 4.6 Immunohistochemistry – Colonic lamina propria

66

LILRB1 LILRA1

LILRA5 LILRA3

CD20 CD3

CD68 Mouse IgG1 Isotype Control Figure 4.7 Immunohistochemistry – Lymphoid follicle in lamina propria

67

LILRA5 LILRB1

Mouse IgG1 Isotype Control Figure 4.8 Immunohistochemistry – Endothelium

LILRA3 Tryptase

Mouse IgG1 Isotype Control Figure 4.9 LILRA3 and tryptase expression in lamina propria 68

LILRB1 LILRB2

LILRB3 Mouse IgG1 Isotype Control Figure 4.10 LILRB1, LILRB2 and LILRB3 expression in lamina propria

69 In active inflammatory bowel disease, there was gross distortion of the normal colonic architecture with loss of crypts, ulceration of the mucosa and a heavy superficial lymphoid infiltrate. In this representative subject, there was abundant LILRB1, LILRB4 and LILRA1 expression.

LILRB1 LILRB4

LILLRA1 Mouse IgG1 Isotype Control Figure 4.11 LILR expression in inflamed colon

70 Immunofluorescence confirmed expression of LILRA5 on CD68 positive macrophages in the lamina propria (Figure 4.12a) and in the lymphoid follicles (Figure 4.12b). LILRA5 and LILRB1 were co-expressed in the lamina propria (Figure 4.12c) as were LILRB1 and LILRA1 (Figure 4.12d). CD3 positive T lymphocytes expressed LILRA5 in the lamina propria (Figure 4.12e) and lymphoid follicle (Figure 4.12f). T lymphocytes also expressed LILRB1 (Figure 4.12g).

Figure 4.12 Representative immunofluorescence images of lamina propria (following 7 pages). Cells are stained with fluorochrome-conjugated LILRA5, LILRB1, LILRA1, CD68, and CD3. Cell nuclei are counterstained with DAPI (blue). Top images show fluorochrome detected at 488 nm (green), middle images show fluorochrome detected at 555 nm, 647 nm or 700 nm (red) and bottom images are merged images showing co-expression (orange)

71 Figure 4.12a Lamina propria: LILRA5 is expressed on CD68+ macrophages

LILRA5 488

CD68 647

Merge

72 Figure 4.12b Lymphoid follicle: LILRA5 is expressed on CD68+ macrophages

LILRA5 488

CD68 647

Merge

73 Figure 4.12c LILRA5 is co-expressed on LILRB1+ cells in the lamina propria. Not all LILRA5 cells are LILRB1+

LILRA5 488

LILRB1 555

Merge

74 Figure 4.12d High power image of lamina propria showing co- expression of LILRB1 and LILRA1

LILRB1 488

LILRA1 555

Merge

75 Figure 4.12e LILRA5 is expressed on CD3+ T lymphocytes in the lamina propria

LILRA5 488

CD3 700

Merge

76 Figure 4.12f LILRA5 is expressed on CD3+ T lymphocytes in the lymphoid follicle

LILRA5 488

CD3 700

Merge

77 Figure 4.12g LILRB1 is expressed on CD3+ T lymphocytes in the lamina propria

LILRB1 488

CD3 700

Merge

78 4.4 Discussion

Expression of multiple leucocyte immunoglobulin-like receptors in the human colon has been demonstrated in these experiments. There was universal expression of LILRB1 and LILRA5 and, in the majority of subjects, expression of LILRA1 and LILRA3. Notably, LILRB4 was absent in control subjects. There was heterogeneity of expression of the other LILRs, in both subjects with inflammatory bowel disease and in controls.

Ethnic differences may account for some of the differences in LILR expression. Varying frequency of the LILRA3 deletion polymorphism depending upon ethnicity has been shown7, 68-70. Many of the other LILRs also show genetic polymorphisms and mutation sites5, 190, 192, 570-572 and these, too, may show ethnic differences. This study population reflected multicultural Sydney, with just under half of the subjects coming from diverse countries world-wide. Of note, there were no first generation Australians. A larger study population, however, would be required to determine the true effect of ethnic differences on LILR expression in the colon.

Regarding differences of LILR expression with age, LILRB1 expression in T lymphocytes has specifically been studied in subjects aged between 13 and 94 years124. With increasing age there were higher expression levels of both LILRB1 and CD57, a marker of antigen- experienced effector memory T lymphocytes, and a reciprocal decrease in levels of the central memory T lymphocyte marker, CCR7. A 1 year increase in age was associated with a 0.53% increase in LILRB1 expression on CD8+ T lymphocytes.

In this study, subjects with inflammatory bowel disease were younger than subjects with colon polyps or cancer. This age discrepancy was not unexpected as inflammatory bowel disease is usually diagnosed in young adults; the peak age for the occurrence of Crohn’s disease is 20-30 years and of ulcerative colitis is 30-40 years332, whilst the conditions leading to surgery in the control subjects occur predominantly in adults over the age of 50 years573, 574. Interestingly, LILRB1 was higher in inflammatory bowel disease in our cohort, demonstrating that other factors besides increasing age are important in determining LILRB1 expression.

LILRA1 was expressed by lamina propria macrophages; previous studies have shown its mRNA to be expressed on peripheral blood monocytes21, 575 and B lymphocytes21, but not

79 on T or NK lymphocytes. Our group has demonstrated expression of LILRA1 in synovial macrophages in a small proportion of patients with rheumatoid arthritis175.

There was more extensive expression of LILRB1 in the colon, in keeping with the wider cellular expression pattern of LILRB1 documented by other groups17, 20. The expression of LILRB1 on colonic lamina propria macrophages is consistent with the findings of Munitz et al in their paediatric cohort151. In healthy pregnancy, LILRB1 is predominantly expressed on decidual macrophages where it promotes an immune tolerant state through binding to its ligand HLA-G, thus preventing foetal rejection576-578. Similarly, LILRB1 binding to yet- unidentified ligands in the healthy colon is likely to help promote immune tolerance to luminal pathogens.

Extensive expression of LILRA5 by macrophages and lymphocytes, as well as neutrophils and endothelial cells was seen in the colon. LILRA5 has been previously demonstrated on CD14 positive monocytes, but not in T lymphocytes, B lymphocytes, or NK lymphocytes in peripheral blood62, 63. In human tissue, LILRA5 is expressed in synovial macrophages and endothelial cells in patients with rheumatoid arthritis149, 175. In this study, we demonstrate more extensive cellular LILRA5 expression in the human colon than was seen in the synovium.

In contrast to the predominant expression of the other LILRs on intestinal macrophages, LILRA3 was expressed predominantly on colonic mast cells. This is the only LILR that does not have a transmembrane or cytoplasmic domain, and therefore is predicted to be a secreted protein19, 21, 22. The function of this soluble LILR is unknown, but an increased serum protein level was present in patients with rheumatoid arthritis and correlated with disease activity84, and the non-functioning gene deletion polymorphism is associated with multiple sclerosis187-189 and Sjogren’s syndrome185.

Interestingly, other inhibitory LILRs that have key functions in other immune tolerant conditions were not highly expressed in the normal colon. In particular, inhibitory LILRB2 and LILRB4 play critical tolerogenic roles in preventing graft rejection in the transplant setting141, 168-170 and LILRB3 along with LILRB1 are important in controlling HIV infection 126, 129.

In the healthy non-inflamed state, co-expression of activating and inhibitory LILRs may serve to maintain homeostatic balance in the colon, in a mechanism similar to other paired

80 receptors. These paired receptor families have both activating and inhibitory receptors encoded by different genes within the particular gene cluster. These receptors are co- expressed on immune cells and have highly homologous extracellular domains that are thought to regulate the immune response by recognising the same ligand, albeit with differing affinities4.

In this study, we demonstrate expression of inhibitory LILRB1 and activating LILRA1 and LILRA5 on lamina propria macrophages. Interestingly, more activating, rather than inhibitory LILRs, were highly expressed in the normal colon. It has been suggested that, as a response to the down-regulation by pathogens through inhibitory receptors, activating receptors have evolved to counterbalance this effect579 and binding of host antigens to these paired receptors is likely to regulate the host immune response in the intestine. More importantly, in the pathogen-rich environment of the colon, microbial components may potentially hijack these paired receptors in order to manipulate the host response, particularly with the diverse populations of microorganisms present in the colon.

In chronic inflammatory bowel disease, there was a significantly higher mean density of LILRB1+ but not LILRA5+ cells than in controls. In addition, there was a positive correlation between LILRB1 and LILRA5 density, suggesting that there is a relative over- expression of LILRB1 compared with LILRA5 in chronic inflammatory bowel disease.

Interestingly, other inflammatory conditions have been also associated with increased expression of the inhibitory LILRB1 receptor. These include T and B lymphocytes and macrophages infiltrating the muscle lesions in the idiopathic inflammatory myopathies183, infiltrating CD4+ T lymphocytes in psoriatic skin lesions182 and macrophages and microglia of chronic active plaques as well as acute lymphocytic infiltrates in active multiple sclerosis181. The over-expression of LILRB1 may be a compensatory mechanism in order to maintain regulatory immunological balance in the setting of colonic inflammation. Indeed, a study of oxidative stress demonstrated that nitrated LILRB1 bound more HLA-G than untreated LILRB1 and this helped to maintain its inhibitory function580.

LILRB4 was only expressed in the lamina propria of 25% of the subjects with inflammatory bowel disease, whilst it was absent in all control subjects. Although this finding was no longer statistically significant after correcting for multiple comparisons, this needs to be pursued in a larger cohort. A previous study had reported LILRB4 in colonic epithelium but not in lamina propria mononuclear cells in both inflamed and

81 normal colon151, although the positive staining in the immunofluorescence image presented is not convincingly epithelial, whilst in this study LILRB4 was expressed in cells of the inflammatory infiltrate.

The antibodies used to detect the LILRs in our study did not distinguish between membrane-bound and soluble forms and there may be a difference in the ratio of membrane-bound to soluble receptors in inflammatory bowel disease, particularly of LILRA5. There may also be qualitative differences in function of these molecules between health and inflammation. The higher density of LILRB1 correlated with an increase in CD68 positive macrophages, but not with overall inflammatory score. In chronic inflammatory bowel disease, there is an increase in tissue macrophages that express CD14 and calprotectin (S100A8/A9) due to the recruitment of blood monocytes521, 522. These newly enlisted macrophages display antimicrobial activity through generation of reactive oxygen metabolites523, 524, secretion of cytokines526 and enhanced antigen presentation525. In addition, there is dysbiosis in chronic inflammatory bowel disease581-583 and, although a single causative organism has not been identified, pathogens can bind to LILRs with high affinity and hence may affect their function in Crohn’s disease and ulcerative colitis.

4.5 Summary

In summary, by using an extensive panel of anti-leucocyte immunoglobulin-like receptor antibodies LILRB1 expression in the human colon was confirmed. LILRA1, LILRA3 and LILRA5 are highly expressed in both normal and inflamed human colon. Leucocyte immunoglobulin-like receptor expression occurred predominantly in the macrophages and lymphocytes of the lamina propria, and there was a strong positive correlation between CD68 and LILRB1 and LILRA5 expression. These data suggest that, in the healthy non-inflamed state, co-expression of both inhibitory and activating receptors may help maintain normal immune tolerance, particularly in this pathogen-rich environment.

82 5 LILRA5 and LILRB1 Protein Expression in the Colon

5.1 Introduction

In the previous chapter, expression patterns in full thickness colon of eight of the leucocyte immunoglobulin-like receptors were determined. This chapter focusses on protein expression of the two main leucocyte immunoglobulin-like receptors in the human colon, LILRA5 and LILRB1, in an attempt to confirm their presence and to study their expression levels further. Lamina propria mononuclear cells [LPMCs] were isolated and analysed via flow cytometry and total protein was extracted and studied via Western blotting techniques from colonic tissue. Peripheral blood mononuclear cells [PBMCs] were isolated and analysed for comparison.

5.2 Materials and Methods

5.2.1 Flow Cytometry

5.2.1.1 Lamina Propria Mononuclear Cells i. Disaggregation of colon

Lamina propria mononuclear cells were isolated from the colonic lamina propria using a modified mechanical and enzymatic disaggregation technique as previously described584. At the time of colonic resection, the lamina propria was dissected off the muscularis mucosae, cut into small strips and washed with Dulbecco’s Phosphate Buffered Saline [D- PBS, Gibco, New York, USA] before being transported back to the laboratory in cold Hank’s buffered salt solution [HBSS, Sigma-Aldrich, Missouri, USA]. The colon was washed several times in HBSS with 0.75 mM EDTA, 100 U/mL penicillin-streptomycin and 0.05 mg/mL gentamicin at 160 rpm, 37°C, for 20 minutes per cycle on a Bioline orbital shaker incubator [Edwards Instrument Company, NSW, Australia] to remove the epithelium. A final wash in HBSS was performed to remove residual EDTA. The colonic tissue was finely cut using a 2-scalpel cross-blades technique, and the resultant slurry transferred to the enzymatic disaggregation solution (20 mL per gram tissue) for overnight incubation at 160 rpm, 37°C. This solution contained RPMI-1640, 10% heat inactivated foetal bovine

83 serum, 8 U/mL collagenase type 1, 10 U/mL DNAse II, 4 mM L-glutamine, 100 U/mL penicillin-streptomycin, 50 µg/mL gentamicin and 20 mM HEPES. Twelve to 16 hours later, the digest was filtered through layers of sterile gauze. The colonic stroma was reserved to make stromal-derived conditioned media (see section 7.2.2), and the liquid containing the LPMCs was centrifuged to pellet the cells. The cells were washed twice in D- PBS prior to a Lymphoprep 1.077 g/mL density gradient separation [Axis-Shield, Oslo, Norway]. The lamina propria mononuclear cell layer was harvested to a separate tube, washed twice with D-PBS and then counted and assessed for viability using 0.1% Trypan blue exclusion. ii. LILRA5 and LILRB1 expression

LPMCs were stained for flow cytometry to determine LILRA5 and LILRB1 expression. Isolated LPMCs were resuspended in D-PBS and blocked with 10% human AB serum [Sigma-Aldrich, Missouri, USA] for 15 minutes at room temperature. Three to 5 x 105 cells per tube were incubated with 0.5 µg mouse monoclonal anti-human IgG1 LILRA5 (n=10 control subjects) or LILRB1 (n=4 control subjects) antibodies [both Amgen, Washington, USA], or irrelevant mouse IgG1 [Dako, Glostrup, Denmark]. Either the primary antibodies were conjugated with Zenon AlexaFluor 488 or 647 [Invitrogen, California, USA] prior to incubation with the LPMCs, or the LPMCs were sequentially incubated with the unconjugated primary antibodies then labelled with FITC-conjugated F(ab’)2-specific goat anti-mouse IgG [Jackson Immunoresearch, Pennsylvania, USA]. LPMCs were incubated with the antibodies for 30 minutes on ice in the dark and then washed three times with cold PAB buffer (D-PBS/0.01% Sodium Azide/2% BSA). The cells were fixed in 500 µL 1% paraformaldehyde overnight and analysed the following day on a FACSCalibur flow cytometer [Becton Dickinson, New Jersey, USA]. The percentages of gated cells that were positive for LILRA5 or LILRB1 are expressed as mean ± standard error of the mean percentages.

As LILRA5 receptors may have been internalised by the LPMCs during the overnight colonic tissue disaggregation, 12 x 106 cells were rested in 20 mL RPMI with 10% autologous serum at 37° C for 2 or 4 hours prior to antibody labelling to determine if subsequent re-expression of these receptors occurred.

In addition, to determine if there was intracellular localisation of LILRA5, LPMCs from 2 control subjects were permeabilised with 0.1% saponin and 1% bovine serum albumin

84 [BSA] in D-PBS, 50 µL per tube, prior to incubation with LILRA5 or IgG1 antibodies conjugated with AlexaFluor 488. Subsequent washes were performed using the saponin- based permeabilisation solution. iii. Expression of other cell markers

LPMCs isolated from 9 control subjects were assessed for expression of a range of phenotypic markers: CD45, CD13, HLA-DR, CD14, CD33, CD3, CD4, CD16, CD56, CD19. LPMCs were washed, blocked with 10% human AB serum, incubated with fluorochrome- conjugated primary antibodies or their irrelevant mouse Ig controls, washed with PAB buffer and fixed with 1% paraformaldehyde as described above. The primary antibodies were commercially conjugated to FITC, PE, PerCP, PE-Cy5 or APC [BD Biosciences, New Jersey, USA]. Up to 4 fluorochromes were assessed at a time and, in the various experiments, antibodies conjugated to different fluorochromes were used to achieve the required combinations.

To determine the presence of intracellular CD68 expression, LPMCs from 5 control subjects were permeabilised and incubated with PE-conjugated CD68 or IgG2b antibodies [BD Biosciences, New Jersey, USA]. iv. Macrophage-enriched fraction of the isolated lamina propria mononuclear cells

Colonic macrophages were isolated from LPMCs of 3 control subjects by magnetic negative selection using a MACS human monocyte isolation kit [Miltenyi Biotec, Cologne, Germany]. The LPMCs were resuspended in cold degassed autoMACS running buffer [Miltenyi Biotec, Cologne, Germany] and blocked with Fc blocking reagent. The cells were indirectly labelled using a cocktail of biotinylated antibodies against CD3, CD7, CD16, CD19, CD56, CD123 and Glycophorin A, followed by the addition of magnetic anti-biotin microbeads. The cells were washed with cold degassed autoMACS running buffer, filtered (30 µm) and then separated using a MACS LS column. Labelled cells were retained in the column placed in the magnetic field of a MACS separator, and the unlabelled macrophage- enriched cell fraction was collected as it passed through the column.

To determine differences in expression levels and degree of macrophage purification between lamina propria mononuclear cells and the subsequent macrophage-enriched cell

85 fraction, cells were assessed by flow cytometry for expression of CD45, HLA-DR, CD14, CD33 and CD3 before and after MACS negative separation.

5.2.1.2 Comparison with Peripheral Blood Mononuclear Cells i. Freshly isolated peripheral blood mononuclear cells

Freshly isolated peripheral blood monocytes were also assessed for LILRA5 expression via flow cytometry. Whole blood from 3 healthy subjects was collected in acid-citrate- dextrose anticoagulant and centrifuged to separate the plasma. This was removed and the remainder was diluted with D-PBS. Lymphoprep 1.077 g/mL was underlayed to generate a density gradient following a further centrifugation. The mononuclear cell layer was harvested to a separate tube, washed with twice with D-PBS and then counted and assessed for viability using 0.1% Trypan blue exclusion.

Three to 5 x 105 PBMCs per tube were blocked with 10% human AB serum for 15 minutes at room temperature. Cells were incubated with 0.5 μg LILRA5 or irrelevant mouse IgG1 control antibodies for 30 minutes on ice in the dark, washed with cold PAB buffer and then incubated with FITC-conjugated F(ab’)2-specific goat anti-mouse IgG for a further 30 minutes on ice in the dark. After a further three washes with cold PAB buffer, peripheral blood mononuclear cells were fixed in 500 μL 1% paraformaldehyde overnight and analysed the following day on a FACSCalibur flow cytometer. LILRA5 and IgG1 antibodies directly conjugated with Zenon AlexaFluor 488 and 647 were also used to stain PBMCs for comparison.

To test for co-expression, PBMCs were co-stained with LILRA5 conjugated with Zenon AlexaFluor 647 and LILRB1 conjugated with Zenon AlexaFluor 488. PBMCs were also assessed for CD45, CD14, CD13 and CD33 expression using primary fluorochrome- conjugated (FITC, PE, PE-Cy5 and APC) antibodies [BD Biosciences, New Jersey, USA].

The effects of permeabilisation on LILRA5 expression of PBMCs were also assessed. PBMCs were incubated with LILRA5 or mouse IgG1 followed by FITC-conjugated IgG antibodies, before and after permeabilisation.

To assess expression of monocyte-macrophage markers, PBMCs were co-stained with CD14-FITC and CD33-PE-Cy5 or the corresponding irrelevant mouse IgG antibodies [BD 86 Biosciences, New Jersey, USA], permeabilised with fixation-permeabilisation buffers [eBioscience, California USA] and then incubated with PE-conjugated CD68 or mouse IgG2b antibodies [BD Biosciences, New Jersey, USA]. ii. Peripheral blood mononuclear cells in colonic disaggregation solution

As LILR expression was unexpectedly low in isolated lamina propria mononuclear cells, the next step was to determine if the disaggregation process was the cause of this reduction. Peripheral blood mononuclear cells were tested after incubation in the two solutions used to isolate the lamina propria mononuclear cells from colonic tissue: HBSS with EDTA or the collagenase-DNase mixture.

Peripheral blood mononuclear cells from 3 healthy subjects were isolated as described above. 30 x 106 PBMCs from the first subject were incubated in 200 mL HBSS with EDTA at 37°C on a Bioline orbital shaker incubator. After 120 minutes, only 0.3% of the starting number of peripheral blood mononuclear cells remained, although these were 90% viable by 0.1% trypan blue exclusion. There were not enough cells available to assess LILRA5 expression by flow cytometry. The experiment was repeated with PBMCs from a second subject with an incubation time of 60 minutes. Again the yield was only 0.3% of the starting cell number.

The experiment was repeated with PBMCs from a third subject. PBMCs were incubated in HBSS with EDTA for 30 or 60 minutes or in the collagenase-DNase mixture for up to 16 hours. Following this, the PBMCs were washed, counted and stained for flow cytometry with LILRA5 or mouse IgG1 antibodies then FITC-conjugated goat anti-mouse IgG as described above. Differential cell counts were performed using a Coulter Counter [Beckman Coulter, California, USA] and manual cell counts and viability with 0.1% trypan blue exclusion on a haemocytometer. The PBMCs were analysed on a FACSCalibur flow cytometer and LILRA5 expression was determined on a monocyte gate where 1% of the corresponding control isotype was negative.

87 5.2.2 Western Blotting and Immunoprecipitation

5.2.2.1 Protein Extraction i. Colonic lamina propria

At the time of colonic resection, lamina propria from 3 subjects with inflammatory bowel disease (2 ulcerative colitis and 1 Crohn’s disease) and 8 control subjects were dissected off the muscularis mucosae and snap frozen on dry ice. Approximately 100 mg of tissue was lysed in 2 mL cold lysis buffer and homogenised using a Pro200 homogeniser with a 10 mm generator [ProScientific, Connecticut, USA]. As a comparison, two different lysis buffers were used: RIPA buffer [Cell Signalling, Massachusetts, USA] and an in-house cell lysis buffer (10 mM Tris HCl, 150 mM NaCl, 2 mM EDTA, 0.5% NP40). The following protease and phosphatase inhibitors were added to the lysis buffers: Complete Protease Inhibitor Cocktail (2 mg/mL), PhosStop Phosphatase Inhibitor Cocktail (1 tablet per 10 mL) [both Roche, Basel, Switzerland] and phenylmethylsulfonyl fluoride (1 mM) [PMSF, Sigma- Aldrich, Missouri, USA]. The homogenates were incubated on ice at 4°C for 30 minutes and then centrifuged at 10 000 g for 20 minutes at 4°C to pellet insoluble material. The supernatant was removed and stored at -80°C. ii. Peripheral blood mononuclear cells

PBMCs from healthy control subjects were isolated from whole blood using a Lymphoprep 1.077 g/mL density gradient as described above. The cells were washed twice with D-PBS and then either lysed in cold RIPA buffer with PMSF (2 x 107 PBMCs per mL lysis buffer to yield approximately 1 μg/µL protein) as above for Western blotting or biotinylated prior to lysis for immunoprecipitation as described below. iii. Protein quantification

Protein quantification was performed on lysed samples using the Bicinchoninic Acid Protein Assay Kit as per manufacturer’s instructions [Thermo-Scientific, Illinois, USA] and read at 562 nm on a SpectraMax M2 microplate reader [Molecular Devices, California, USA].

88 5.2.2.2 Western Blotting i. SDS-page gel and Western transfer

An initial Western blot was performed using protein extracted from the lamina propria of 4 control subjects. Both of the lysis buffers described above (RIPA and NP40 in-house buffer) were used on each sample and 20 µg of total protein was loaded per lane. The protein samples were reduced by the addition of 10 mM Dithiothreitol (DTT) and Tricine Sample Buffer 2x [Bio-Rad, California, USA] and heating to 95°C for 5 minutes. These reduced samples were run on a 10% SDS-page gel for 30 minutes at 40 volts, and then for 80 minutes at 100 volts using a MiniProtean Electrophoresis System [Bio-Rad, California, USA]. The gel was transferred to a 0.2 µm polyvinylidene difluoride [PVDF] membrane [Millipore, Massachusetts, USA] for 45 minutes at 75 volts for subsequent Western blotting. A standard ladder was also loaded on each gel to determine molecular weight of the proteins (Precision plus protein standards dual color [Bio-Rad, California, USA]). ii. Probing of the Western blot

After blocking with 5% skim milk powder in TBS, the membrane was incubated with 1 µg/mL mouse monoclonal anti-human IgG1 LILRA5 antibody [Amgen, Washington, USA], followed by HRP-conjugated goat anti-mouse IgG [Bio-Rad, California, USA] using a 1:4000 dilution with TBS. In-between, the membrane was rinsed with TBS-Tween20. The Western blot was developed using the Western Lightning Chemiluminescence Reagent Plus Kit [Perkin-Elmer, Massachusetts, USA] and analysed on a Luminescent Image Analyser LAS 3000 [Fuji Photo Film, Tokyo, Japan]. iii. Reprobing of the membrane

The membrane was stripped using Restore Western Blot Stripping Buffer [Thermo- Scientific, Illinois, USA] as per the manufacturer’s instructions and reprobed with mouse monoclonal ß-actin antibody [Sigma-Aldrich, Missouri, USA] to assess for protein loading.

89 iv. Further Western blot

As a result of the Western blot being indeterminate, blotting was repeated using 40 µg total protein per lane from lamina propria of the same 4 control subjects. A 0.1% Ponceau S in 5% acetic acid stain [Sigma-Aldrich, Missouri, USA] was performed on the membrane prior to blotting to confirm effective transfer of the proteins. v. Confirmatory Western blot

To verify these findings, a third Western blot was performed with colonic lamina propria protein lysed in RIPA buffer from these same 4 subjects using 10 µg per lane, along with freshly isolated PBMCs lysed in RIPA buffer as a positive control. The membrane was probed with mouse monoclonal anti-human IgG1 LILRA5 antibody [Amgen, Washington, USA], and then reprobed using the in-house polyclonal rabbit anti-LILR antibody. Prior to reprobing with the second antibody, the membrane was stripped using Restore Western Blot Stripping Buffer [Thermo-Scientific, Illinois, USA] as per the manufacturer’s instructions. vi. Coomassie blue staining

Coomassie Blue (2.5% Brilliant Blue R in 10% acetic acid and 40% methanol) staining of a corresponding 10% SDS page gel using 5 µg of protein per lane was performed to confirm the presence of protein. The gel was stained with Coomassie Blue solution for 30 minutes on a Bioline platform rocker [Edwards Instrument Company, NSW, Australia], destained overnight with 10% acetic acid and 40% methanol, and then washed with destain solution until protein bands were visualised. vii. Subjects with chronic inflammatory bowel disease

Western blotting was performed with lysed colonic lamina propria from 3 subjects with IBD and 5 control subjects, 15 µg of protein per lane, with PBMCs as a positive control. The first membrane was probed with anti-LILRA5 antibody and the second with the in-house polyclonal rabbit anti-LILR antibody.

90 vii. Steps to assess high background

There was high background on the membranes and various antibody concentrations, number and types of washing steps and chemiluminescence exposure times were assessed. Different blockers were also compared: 5% skim milk powder, 3% BSA, 1% ovalbumin or 3% ovalbumin in TBS, TBS/0.05% Tween20, D-PBS or D-PBS/0.05% Tween 20.

5.2.2.3 Immunoprecipitation i. Biotinylation of peripheral blood mononuclear cells

Freshly isolated PBMCs were biotinylated prior to immunoprecipitation. The cells (1 x 107

PBMCs/mL) were washed twice with cold D-PBS containing 1mM MgCl2 and 0.1 mM CaCl2, and then incubated with 0.5 mg Sulfo-NHS-LC-Biotin [Thermo-Scientific, Illinois, USA] for 30 mins at 4°C. Unbound biotin was quenched with serum-free RPMI-1640 and the cells were washed twice with D-PBS/MgCl2/CaCl2 prior to lysis. ii. Immunoprecipitation of protein

Lamina propria from one subject with IBD, one control subject with colon cancer, and biotinylated PBMCs from a healthy donor, were immunoprecipitated with anti-LILRA5 or mouse IgG1 antibody. 1.5 mg protein of each specimen was lysed in cold RIPA buffer containing protease and phosphatase inhibitors. This was precleared with protein G sepharose 4B-coupled goat anti-mouse IgG beads [Zymed, California, USA]. The lysates were incubated overnight with mouse monoclonal anti-human LILRA5 antibody [Amgen, Washington, USA] or irrelevant control mouse IgG1 antibody [Sigma-Aldrich, Missouri, USA]. Protein G sepharose 4B-coupled beads were used to immunoprecipitate the antibody-protein complexes. After incubation with the beads the mixture was centrifuged and then heated at 100°C for 5 minutes to dissociate the immunocomplexes from the beads. The supernatants were run on a 10% SDS-page gel as described above.

91 iii. Transfer of proteins and Western blotting of the PVDF membrane

The proteins were transferred from the 10% SDS page gel to a 0.2 µm PVDF membrane. The SDS page gel was stained with Coomassie Blue to confirm complete transfer of the proteins from the gel. The PVDF membrane was blocked with 5% skim milk powder and then incubated with the in-house polyclonal rabbit anti-LILR antibody. After washing with TBS-Tween20, the membrane was incubated with HRP-conjugated goat anti-rabbit IgG [Bio-Rad, California, USA] using a 1:4000 dilution with TBS/2% skim milk and then developed and analysed via chemiluminescence as described above.

To check for biotinylation, the lanes with biotinylated peripheral blood mononuclear cells were washed, blocked with TBS/2% BSA and then reprobed with Streptavidin HRP [R&D Systems, Minnesota, USA] using a 1:3000 dilution with TBS/1% BSA. iv. Corresponding Western blot

A Western blot with protein from the same 3 subjects was performed as described above. The membrane was probed with the mouse monoclonal anti-LILRA5 antibody.

5.2.2.4 Further Immunoprecipitation Experiments

The findings of a large immunoprecipitated protein between ~75 kDa and ~110 kDa using the anti-LILRA5 antibody was unexpected, and further experiments were performed in order to explain these results.

The immunoprecipitation procedure was repeated using freshly isolated biotinylated PBMCs from a second healthy donor. The cells were lysed with cold RIPA buffer, immunoprecipitated with either anti-LILRA5 or mouse IgG1 antibodies and run on a 10% SDS page gel. The proteins were transferred to a PVDF membrane, which was blotted with the in-house polyclonal rabbit anti-LILR antibody.

A further immunoprecipitation was performed using freshly isolated biotinylated PBMCs from a third healthy donor. The biotinylated PBMCs were immunoprecipitated with anti- LILRA5, mouse monoclonal anti-human IgG1 LILRB4 [Amgen, Washington, USA] or irrelevant control mouse IgG1 antibodies. The mouse monoclonal anti-human IgG1 LILRB4 antibody was used as a positive control to check that the immunoprecipitation 92 protocol was working, as this antibody has been shown in our laboratory to reliably immunoprecipitate a protein of ~70 kDa size from PBMCs585. The membrane was first probed with Streptavidin HRP and then reprobed with the in-house polyclonal rabbit anti- LILR antibody.

The remaining lysis buffers post-immunoprecipitation were also run on a gel, transferred to a PVDF membrane, and probed with Streptavidin HRP, to ensure biotinylated proteins were present in the lysate.

5.2.2.5 Experiments to Test the Anti-LILRA5 Antibody

As the immunoprecipitation experiments on biotinylated PBMCs failed to demonstrate an appropriate-sized protein with the anti-LILRA5 antibody, the integrity of this antibody was questioned.

A further Western blot using PBMCs from 4 different healthy donors was performed as described above to test the anti-LILRA5 antibody. A corresponding SDS-page gel run at the same time was stained with Coomassie Blue to confirm that intact proteins were present in these specimens.

Recombinant human LILRA5 protein with a theoretical molecular weight of 59.2 kDa [Abnova, Taipei, Taiwan] was purchased to test the anti-LILRA5 antibody. A 10% SDS- page gel was run with 100 ng rhLILRA5 and PBMCs and a silver stain was performed to ensure protein was present. The gel was fixed with 45% acetic acid/50% methanol followed by 50% methanol, washed with water, and then sensitised with 0.05% sodium thiosulfate solution. After washing with water, 0.1% silver nitrate was added for 45 minutes. The gel was again washed with water, and then developed with 2.5% sodium carbonate/0.025% formaldehyde solution before fixation with 5% acetic acid solution. A second gel was made, the proteins transferred to a membrane and incubated with mouse monoclonal anti-human LILRA5 antibody as per Western blotting protocol above.

93 5.3 Results

5.3.1 Flow Cytometry

5.3.1.1 Lamina Propria Mononuclear Cells i. Lamina propria mononuclear cells and LILRA5 expression

Six male and 4 female subjects, average age 72 ± 4 years, undergoing a colonic resection for colonic polyps (n=6), colon cancer (n=3) or diverticular disease (n=1) were included. Specimens were from the left colon in 4 subjects and the right colon in 6 subjects. An average of 15.3 ± 1.4 grams of lamina propria per subject was obtained and 2.2 ± 0.3 x 106 LPMCs per gram of lamina propria were isolated, with a cell viability of 79.1 ± 4.0%.

Overall, isolated lamina propria mononuclear cells expressed minimal levels of LILRA5. One subject had a particularly high LILRA5 expression level of 54.7%. The colonic resection was performed for right-sided colonic cancer, and there were no important clinical differences between this subject and the rest of the cohort to account for the high LILRA5 level. After excluding this subject, the average LILRA5 level was 5.4 ± 0.8%. There were no significant differences in LILRA5 levels between the different fluorochromes.

There were strong positive correlations between LILRA5 expression and CD14 expression (r = 0.81, p = 0.002) and LPMC viability (r = 0.62, p = 0.03), see Figure 5.1. There was no correlation between LILRA5 expression and age of the subjects (r = 0.14, p = 0.70, data not shown).

There was a strong non-linear positive correlation between CD14 and viability. There was an exponential rise in CD14 expression with increasing isolated LPMC viability, see Figure 5.2. LPMCs with > 90% viable cells had CD14 levels > 20%, and LPMCs with > 95% viable cells had CD14 levels > 40%.

94

Figure 5.1 LILRA5 expression vs. CD14 and LPMC viability

Figure 5.2 LPMC viability vs. CD14

LPMCs that were rested in the incubator for 2 or 4 hours after isolation did not show a significant change in cell viability, LILRA5 or CD14 expression. LILRA5 levels were 3.6 ± 0.7%.

95

Figure 5.3 Rested LPMCs – viability, CD14 and LILRA5

Following permeabilisation, there appeared to be dissociation of the AlexaFluor 488 from the primary antibodies. Only 2% of permeabilised LPMCs expressed LILRA5. ii. Lamina propria mononuclear cells and LILRB1 expression

LILRB1 expression levels were similarly low at 2.9 ± 1.1%.

Figure 5.4 LILRB1 expression of LPMCs (representative data showing a small positive population)

96 When co-expression of LILRB1 and LILRA5 was assessed, all LILRB1+ LPMCs were LILRA5+, and a further 3% of cells were LILRA5+LILRB1–. This was in distinct contrast to the high co-expression levels of LILRA5 and LILRB1 on peripheral blood mononuclear cells.

Figure 5.5 Co-expression of LILRB1 (LIR1) and LILRA5 (LIR9) on LPMCs

Figure 5.6 Co-expression of LILRB1 (LIR1) and LILRA5 (LIR9) on PBMCs

97 iii. Lamina propria mononuclear cells and other cell markers

Lamina propria mononuclear cell viability was 86.3 ± 3.2% (mean ± SEM). CD45 was expressed on the majority of LPMCs (92.5 ± 2.9%). These cells were mainly CD4+ T lymphocytes (CD3 = 80.0 ± 2.9%, CD4 = 60.0 ± 10.2%). CD19 was expressed on 6.5 ± 1.6% and CD56 on 4.8 ± 0.5%. Approximately 10-15% of LPMCs expressed monocyte- macrophage markers; in the macrophage gate CD14 = 14.8 ± 7.1%, CD33 = 11.2 ± 5.8% and CD68 = 9.9 ±2.0%. HLA-DR (23.1 ± 10.3%) and CD13 (41.5 ± 17.3%) were expressed in a higher proportion of macrophages, and CD16 expression was low (1.9 ± 0.9%).

Figure 5.7 Phenotype of isolated LPMCs as assessed by flow cytometry

98 After permeabilisation, all CD14 positive LPMCs expressed CD33 and CD68. There was, however, expression of CD33 on other cells, in addition to those that were CD14 or CD68 positive.

Figure 5.8 CD68, CD14 and CD33 expression on isolated LPMCs

99 iv. Lamina propria macrophages

Following purification of lamina propria macrophages via negative magnetic separation, CD45 and CD33 expression were slightly higher and CD3 was no longer expressed. HLA- DR, a marker of B lymphocytes and activated T lymphocytes as well as monocytes, macrophages and dendritic cells586, was reduced.

Figure 5.9 Phenotype of lamina propria macrophages before and after magnetic negative selection

5.3.1.2 Peripheral Blood Mononuclear Cells i. Peripheral blood mononuclear cells and LILR expression

In the monocyte gate, 96.4 ± 1.6 % of PBMCs were positive for LILRA5. There were no differences in expression levels between the 3 fluorochromes: AlexaFluor 488, AlexaFluor 647 or FITC. Monocytes were 98% CD45, CD14, CD13, CD33 positive and 88.2% co- expressed LILRA5 and LILRB1. Peripheral blood lymphocytes expressed negligible LILRA5, but 25.8% of these cells expressed LILRB1.

100

Figure 5.10 LILRA5 expression on PBMCs. Representative flow cytometry histogram showing LILRA5 (and mouse IgG1 isotype control) with secondary goat anti-mouse FITC antibody on PBMCs

Figure 5.11 LILRB1 expression on PBMCs

101 ii. Permeabilised peripheral blood mononuclear cells

LILRA5 expression was slightly reduced when cells were permeabilised following antibody staining (from 97% to 84%). When PBMCs were permeabilised first, and then incubated with LILRA5, expression was markedly reduced to 27%, with a 4-fold reduction in fluorescence intensity.

Of the monocyte-macrophage markers, there was a wide variation in CD68 expression between both the 4 subjects and with different experimental runs, ranging from 1.3 to 94.5%; the average CD68 expression level was 48.6 ± 14.7%. CD33 was a more consistent marker of these cells, with high expression levels of 93.4 ± 4.7%. CD14 expression was variable between subjects, with an average expression level of 69.9 ± 16.1%. There was co-expression of CD68 and LILRA5.

Figure 5.12 Co-expression LILRA5 and CD68 on permeabilised PBMCs

102 iii. Peripheral blood mononuclear cells in colonic disaggregation solutions

There was marked cell loss when PBMCs were in the HBSS with EDTA. This step is used in the colonic disaggregation process to remove epithelial cells and mucus from intestinal tissue and the process takes between 2 and 3 hours with changes of media every 20 minutes. Only 5–10% of initial PBMCs remained at 1 hour, although these remained highly viable. LILRA5 expression fell to 78.6% at 30 mins and 50.3% at 1 hour.

Similarly, there was loss of PBMCs in the collagenase-DNase solution, but to a lesser degree than PBMCs in the HBSS with EDTA. This step was performed to mimic the overnight step where minced lamina propria is placed in this solution and LPMCs are gradually released from the stroma. Cell loss occurred particularly in the first hour then plateaued; approximately 55% of PBMCs were lost after the overnight incubation in this media. Again, the remaining cells were highly viable. There was, however, only a 10% reduction in LILRA5 expression.

Table 5.1 PBMCs in colonic disaggregation solutions

White cells Monocytes % of initial Viability LILRA5 per tube per tube PBMCs

Initial PBMCs 22.4 x 106 1.8 x 106 90% 96.4%

HBSS-EDTA 2.0 x 106 0.2 x 106 8.9% 90% 78.6% 30 mins

HBSS-EDTA 1.1 x 106 N/A 4.9% 91% 50.3% 60 mins

Collagenase-DNase 15.5 x 106 1.9 x 106 69.2% 97% 86.1% 30 mins

Collagenase-DNase 10.5 x 106 1.2 x 106 46.9% 93% 90.3% 60 mins

Collagenase-DNase 7.7 x 106 1.3 x 106 34.4% 88% 90.6% 3 hours

Collagenase-DNase 9.6 x 106 1.1 x 106 42.3% 92% 85.8% 16 hours

103 5.3.2 Western Blotting and Immunoprecipitation

5.3.2.1 Subjects for Western Blotting or Immunoprecipitation

All 3 subjects with IBD were female. There were 4 males and 4 females in the control group. Again, the subjects with IBD were significantly younger than the control subjects (mean ± SEM: IBD 43 ± 12 years, control 69 ± 4 years, p-value = 0.03).

5.3.2.2 Western Blotting on Colonic Lamina Propria i. Control Subjects

Three Western blots with protein extracted from colonic lamina propria of 4 control subjects were performed to assess for the presence of LILRA5. For comparison, protein extracted from PBMCs was included in the third experiment (see Table 5.2). Reprobing of the first membrane with β-actin confirmed that equal amounts of protein were present in each lane.

Both lysis buffers were used in the first 2 blots; more prominent bands were seen with lamina propria lysed in RIPA buffer compared with the NP40 in-house cell lysis buffer; hence the RIPA buffer was used for future experiments.

Table 5.2 Protein bands detected by anti-LILRA5 antibody

~15 kDa ~55 kDa ~70 kDa ~100 kDa

WB 1: lamina propria, RIPA buffer Faint

WB 1: lamina propria, NP40 buffer Faint

WB 2: lamina propria, RIPA buffer 2 of 4 Yes Yes

WB 2: lamina propria, NP40 buffer 2 of 4 Yes

WB 3: lamina propria, RIPA buffer Yes

WB 3: PBMCs, RIPA buffer Yes Yes

104

Figure 5.13 Lamina propria lysed in RIPA or NP40 buffer, probed with anti-LILRA5 antibody

When the third membrane was reprobed with the polyclonal rabbit anti-LILR antibody, multiple distinct protein bands at ~110 kDa, ~70 kDa and ~35 kDa in lamina propria were detected. These presumably correlate with some of the other LILRs, possibly LILRB1, LILRA1/LILRA3/LILRB5 and LILRA5, respectively. In PBMCs, there were additional protein bands detected at ~100 kDa and weakly at ~60 kDa, possibly correlating with LILRB2/LILRB3 and LILRB4, respectively. Interestingly, the strongest band was at ~110 kDa in both lamina propria and PBMCs. Coomassie Blue staining showed multiple distinct protein bands, particularly prominent at ~70 kDa in the lamina propria.

105 ii. Subjects with Inflammatory Bowel Disease

When probed with anti-LILRA5 antibody, lamina propria from all subjects showed protein bands at ~70 kDa and ~55 kDa, and 2 of the 3 subjects with IBD had an addition protein band at ~25 kDa. On the PBMC positive control lane, there was a single protein band at ~55 kDa.

Figure 5.14 Lamina propria from IBD (n=3) and control (n=5) subjects, probed with anti-LILRA5 antibody

On the membrane probed with polyclonal rabbit anti-LILR antibody, there were distinct protein bands at ~110 kDa, ~100 kDa, ~65 kDa, ~40 kDa and ~25 kDa in lamina propria from subjects with IBD, and protein bands at ~110 kDa, ~100 kDa, ~35 kDa and ~25 kDa in lamina propria from control subjects. One of the control subjects had an additional protein band at ~40 kDa.

106 5.3.2.3 Immunoprecipitation

In the lamina propria and PBMC samples immunoprecipitated with the anti-LILRA5 antibody, but not the mouse IgG1 antibody, protein bands between ~75 kDa and ~110 kDa were detected by the polyclonal rabbit anti-LILR antibody.

Figure 5.15 PBMCs and lamina propria immunoprecipitated with anti-LILRA5 or mouse IgG1 antibodies, probed with polyclonal rabbit anti-LILR antibody

Streptavidin HRP detected protein bands at ~55 kDa in PBMCs immunoprecipitated with anti-LILRA5 antibody, in addition to bands at ~75 kDa and ~25 kDa in PBMCs immunoprecipitated with both the anti-LILRA5 and mouse IgG1 antibodies.

The corresponding Western blot probed with anti-LILRA5 antibody showed protein bands at ~110 kDa, ~55 kDa and ~25 kDa from colonic lamina propria from the subject with IBD. Similar bands at ~55 kDa and ~25 kDa, as well as a larger protein at ~130 kDa were present in colonic lamina propria from the control subject. PBMCs showed weak protein bands at ~100 kDa and ~55 kDa.

107

Figure 5.16 PBMCs immunoprecipitated with anti-LILRA5 or mouse IgG1 antibodies, probed with streptavidin HRP (left); Corresponding Western blot of PBMCs and lamina propria, probed with anti-LILRA5 antibody (right)

108 5.3.2.4 Further Immunoprecipitation Experiments

As the findings of a wide protein band in proteins immunoprecipitated with anti-LILRA5 antibody in the initial immunoprecipitation were unexpected, this was repeated on biotinylated PBMCs from 2 further subjects.

Table 5.3 PBMCs immunoprecipitated with anti-LILRA5 antibody, probed with polyclonal anti-LILR antibody or streptavidin HRP (3 healthy donors)

Polyclonal anti-LILR antibody Streptavidin HRP

IP 1 75-110 kDa ~55 kDa

IP 2 75-110 kDa ~55 kDa

IP 2 nil ~55 kDa

Findings in the second experiment were similar. In the third immunoprecipitation experiment, immunoprecipitation with monoclonal mouse anti-human LILRB4 produced a protein band of ~70 kDa as expected, confirming success of the immunoprecipitation procedure. In PBMCs immunoprecipitated with anti-LILRA5 antibody, streptavidin HRP again detected 55 kDa protein. The polyclonal anti-LILR antibody, however, did not detect any protein and the integrity of the anti-LILRA5 antibody was questioned.

Figure 5.17 PBMCs immunoprecipitated with anti-LILRA5, anti-LILRB4 and mouse IgG1 antibodies, probed with polyclonal rabbit anti-LILR antibody

109 5.3.2.5 Experiments to Test the Anti-LILRA5 Antibody i. Western blot and Coomassie Blue stain

The subsequent Western blot of protein from PBMCs from 4 different healthy donors probed with anti-LILRA5 antibody failed to show any protein bands, unlike previously, where a protein band of ~55 kDa was consistently detected, in addition to smaller and larger protein bands. A Coomassie Blue stain of the corresponding SDS-page gel showed multiple protein bands in all 4 subjects. ii Recombinant human LILRA5 protein

The silver stain gel showed a single band at ~55 kDa on the lane with recombinant human LILRA5 and multiple bands from PBMCs. However, the membrane blotted with LILRA5 antibody did not detect any bands. This led to the conclusion that the anti-LILRA5 antibody was no longer working.

L 1 2 3 4

~55 kDa

Figure 5.18 Silver stain showing rhLILRA5 present at ~55 kDa and multiple proteins present in PBMCs. L = Ladder; Lane 1 = recombinant human LILRA5; Lanes 2, 3 and 4 = PBMCs

110 5.4 Discussion

LILRA5 and LILRB1 expression on colonic lamina propria mononuclear cells was expected to be demonstrable by flow cytometry, as these receptors were abundantly expressed on lamina propria macrophages in colonic tissue sections using immunohistochemistry and immunofluorescence techniques. There was, however, low LILRA5 and LILRB1 expression in these isolated lamina propria mononuclear cells.

LPMCs were more heterogeneous than PBMCs and determination of the macrophage gate based on the broader forward-side scatter patterns was not always straightforward. In conjunction with the size and granularity of the cells, the various cell markers, CD45, CD14, CD13, CD33, HLA-DR, CD3, were used to capture the majority of macrophages. The majority of isolated LPMCs were T lymphocytes, consistent with previous studies587-589. There is no single phenotypic marker that reliably identifies colonic macrophages. CD68, the macrophage marker used in previous immunohistochemistry experiments, showed highly variable expression on both the isolated LPMCs and PBMCs following permeabilisation; this variability may be an artefact of the permeabilisation process. Of the other monocyte-macrophage markers used, previous research has shown that CD14 expression, whilst high on PBMCs, is down-regulated on normal immune tolerant colonic macrophages206. CD33 may prove to be a better surface marker of colonic macrophages than CD14 or intracellular CD68. It is, however, present on a broader range of cells, including activated T lymphocytes, myeloid progenitors and mast cells590, 591, and there were distinct CD33 positive populations of LPMCs that were CD68 and CD14 negative. Regardless of the population of cells chosen, however, LILRA5 expression was low.

LILRA5-positivity on lamina propria mononclear cells correlated with cell viability and CD14 expression, suggesting that the cells most sensitive to LILRA5 loss were those that were CD14 positive. This assumes that all subjects had similar CD14 starting levels. This is interesting as CD14 might be a marker of relatively recent recruitment from the circulation522, where circulating monocytes express high levels of LILRA5. CD14 expression was not tested in PBMCs placed in the overnight disaggregation solution.

Similar to LILRA5, there was low expression of LILRB1 on isolated lamina propria mononuclear cells, despite universal expression on colonic macrophages demonstrated by immunohistochemistry. All LILRB1 positive macrophages were LILRA5 positive, but not vice versa.

111 In one subject, there was also a discrete small population of LILRA1 positive cells detected in the isolated LPMCs (data not shown), which corresponds to the earlier immunohistochemistry findings where LILRA1 was non-universally expressed in lamina propria macrophages. A larger sample size would be necessary to confirm LILRA1 expression in these cells and to help determine what factors lead to this differential expression in the colon.

To determine why these receptors were poorly expressed in isolated LPMCs, a number of different factors were assessed. Peripheral blood mononuclear cells showed high LILRA5 and LILRB1 expression levels when stained in this way, confirming that these antibodies are useful flow cytometry markers.

Interestingly, LILRB1 and LILRA5 were co-expressed on peripheral blood monocytes, a characteristic that has not previously been demonstrated.

To determine if LILRA5 was expressed intracellularly, LPMCs were permeabilised and stained with fluorochrome-conjugated LILRA5. There was, however, dissociation of the conjugated antibody-fluorochrome complex making this difficult to interpret. When the effect of permeabilisation was tested on PBMCs using unconjugated LILRA5 followed by FITC-conjugated IgG antibodies, there was a deleterious effect of prior permeabilisation on LILRA5 expression. This may be due to the loss of membrane surface area or cleavage of the receptors by the permeabilisation process. Despite this, there was a low level of LILRA5 expression on these cells. When PBMCs were stained with unconjugated anti- LILRA5 and then FITC-conjugated IgG antibodies prior to permeabilisation, LILRA5 and CD68 co-expression was demonstrated. This method was not retried on isolated LPMCs, as their starting LILRA5 expression level was low and initial permeabilisation attempts showed that fewer than 2% of cells expressed LILRA5.

LPMCs were rested in the incubator for 2 or 4 hours following disaggregation to test if internalisation of receptors and subsequent re-expression occurred. LILRA5 expression, however, was still low. A longer resting time was not evaluated due to the loss of viability of the isolated LPMCs. This, however, is likely to be helpful in determining if the LPMCs were able to produce and express new LILRA5 receptors.

In an attempt to determine the stage during the disaggregation process at which expression of LILRA5 was lost, peripheral blood mononuclear cells were incubated in the

112 different colonic disaggregation media, HBSS with EDTA or collagenase-DNase. The initial plan was to sequentially incubate cells in HBSS and EDTA then collagenase-DNase, but there was significant cell loss with HBSS and EDTA. Cells that were retained were highly viable, but there was a differential cell loss seen, with a reduction in LILRA5 expression of approximately 50% resulting in a skewed population that was predominantly LILRA5 negative. One possible explanation is that the lamina propria mononuclear cells are not usually directly exposed to this solution during the disaggregation process; instead they are embedded within the stroma during this step and this supporting structure helps to maintain cell integrity.

After overnight incubation in collagenase-DNAse, there was a lesser degree of cell loss. In the remaining cells, which were again highly viable, LILRA5 expression was reduced by only 10-15%. During this step of the intestinal disaggregation, the LPMCs initially lie embedded within the supporting colonic stroma and are also likely to remain there for a large part of this process. This experiment represents a worst-case scenario where isolated cells are continuously bathed in media for 16 hours. From these experiments, although LILRA5 was present on PBMCs, its expression was reduced after exposure to both types of the media used in the disaggregation process.

As LILRA5 and LILRB1 were not demonstrable in significant levels by flow cytometry in these isolated LPMCs, another method to detect protein expression of these LILRs was pursued. Western blotting of synovial fluid using the same LILRA5 antibody had previously detected a ~25 kDa protein, consistent with soluble LILRA5, in subjects with rheumatoid arthritis but not those with osteoarthritis. In addition, biotin-labelled PBMCs that were immunoprecipitated with anti-LILRA5 antibody produced a ~40 kDa protein band in both subjects with rheumatoid arthritis and healthy controls149.

In all colonic lamina propria samples, there was a ~55 kDa protein band detected with the anti-LILRA5 antibody. There was some variability of LILRA5 expression; there were additional protein bands of ~15 kDa and ~70 kDa in control colon and ~25 kDa and ~70 kDa in inflamed colon. Control PBMCs showed protein bands at ~55 kDa, as well as a larger ~100 kDa protein in one subject.

Following immunoprecipitation of protein from colonic lamina propria and PBMCs with anti-LILRA5 antibody, unexpectedly large protein bands between ~75 kDa and ~110 kDa

113 were detected using the in-house polyclonal rabbit anti-LILR antibody. This finding was confirmed in PBMCs from a second healthy subject.

Immunoprecipitation of biotinylated PBMCs with anti-LILRA5 antibody resulted in detection of proteins of ~55 kDa with streptavidin HRP in all three experiments, larger in size than previously demonstrated149, but consistent with the above Western blot results. Colonic lamina propria was unable to be biotinylated. Successful immunoprecipitation technique was verified with immunoprecipitation of PBMCs with monoclonal anti-LILRB4 antibody, resulting in an appropriate single protein band of ~70 kDa.

Subsequent Western blots and immunprecipitation attempts with anti-LILRA5 antibody, however, failed to detect protein in colonic lamina propria, PBMCs and commercial recombinant human LILRA5 protein. Therefore, further experiments to detect LILRA5 protein in the colonic lamina propria and to determine differences in expression in inflammatory bowel disease were aborted, as further anti-LILRA5 antibody was not commercially available.

5.5 Summary

Universal LILRB1 and LILRA5 protein expression was unable to be confirmed by either flow cytometry on isolated lamina propria mononuclear cells and Western blot on extracted lamina propria protein. The overnight extraction of LPMCs resulted in survival of a non-representative LILRA5low population of cells, and the isolation process is likely to have altered expression levels of the LILRs. Western blot and immunoprecipitation showed larger than expected proteins most consistently at ~55 kDa, although the integrity of the available antibody was questionable and proved subsequently to be no longer working. This resulted in an inability to perform functional studies requiring antibody for cross-linking and a consequent shift in the direction taken to establish the presence of leucocyte immunoglobulin-like receptors in the human colon.

114 6 Leucocyte Immunoglobulin-Like Receptor mRNA Expression in the Human Colon

6.1 Introduction

In this chapter, quantitative real time PCR on extracted RNA was performed to verify the presence of leucocyte immunoglobulin-like receptors in the colon, particularly as flow cytometry or Western blot techniques were unable to confirm LILR expression. Expression of the eight LILRs assessed in Chapter 4 and of TNF-α was determined using quantitative real time PCR. Comparisons were made between inflamed and control colon and ileum, peripheral blood mononuclear cells and isolated lamina propria mononuclear cells.

6.2 Materials and Methods

6.2.1 RNA Extraction from Colon or Terminal Ileum

Intestinal tissue was collected from the colon from 4 subjects with IBD and 9 control subjects, from the terminal ileum from 3 subjects with IBD and from the colonic lamina propria from 11 control subjects. At the time of colonic resection, a small portion of colonic lamina propria, full-thickness colon or terminal ileum was cut into fragments (0.5 cm maximum thickness), immersed in 5 to 10 volumes of RNAlater solution [Ambion, Life Technologies, California, USA], incubated at 4°C overnight and then stored at -80°C as per protocol.

Prior to use, samples were placed at -20°C overnight. The tissue was retrieved from the RNAlater solution and blotted to remove excess RNAlater solution. Twenty to 30 mg of tissue was placed in 900 μL of the tissue lysis solution, Buffer RLT with 10 μL/mL β- mercaptoethanol [RNeasy Mini Kit, Qiagen, Hilden, Germany], in a 2 mL tube containing 2.8 mm ceramic zirconium oxide beads [Precellys, Bertin Technologies, Montigny-le- Bretonneux, France]. The colon was homogenised in a Precellys 24 Dual Tissue Homogenizer at 6000 rpm for 15 seconds x2 with 10 seconds break in-between cycles. The tubes were placed on ice to cool and then were centrifuged. The supernatant was transferred to a fresh tube and re-centrifuged. Eight hundred μL of supernatant was

115 transferred to a fresh tube and mixed with an equal volume of 75% ethanol. This was added to a RNEasy spin column and the RNA extracted as per RNeasy Mini Kit protocol [Qiagen, Hilden, Germany]. DNA digestion was performed with DNase I on-column, and the RNA was eluted in 30 μL DNase-free RNase-free water [Gibco, New York, USA]. The quality and quantity of the extracted RNA was analysed using a NanoDrop ND-1000 [Thermo Fischer Scientific, Massachusetts, USA] and an Agilent 2100 electrophoresis bioanalyser [Agilent Technologies, California, USA].

6.2.2 RNA Extraction from Peripheral Blood Mononuclear Cells

6.2.2.1 Isolation of Peripheral Blood Mononuclear Cells

RNA was extracted from peripheral blood mononuclear cells [PBMCs] from 6 subjects with IBD and 7 control subjects for comparison. Twenty mL blood was collected in acid-citrate- dextrose anticoagulant prior to the colonic resection. After the blood was centrifuged, the plasma was removed and the remainder diluted with D-PBS. PBMCs were separated by a Lymphoprep 1.077 gm/L density gradient as described in the previous chapter. The cells were washed twice with D-PBS and assessed for viability using 0.1% Trypan blue exclusion before storage in liquid nitrogen at a concentration of 5-10 x 106 PBMCs in 1 mL 10% DMSO, 40% RPMI-1640 and 50% autologous plasma per cryovial until use.

6.2.2.2 Extraction of RNA from Peripheral Blood Mononuclear Cells

PBMCs were rapidly thawed in a 37°C water bath and then resuspended in warm RPMI- 1640 [Gibco, New York, USA]. The cells were washed, pelleted and then lysed in 500 μL Buffer RLT with 10 μL/mL β-mercaptoethanol [RNeasy Mini Kit, Qiagen, Hilden, Germany]. The cells were homogenised via a combination of pipetting and a short vortex. 500 μL 75% ethanol was added and RNA was extracted using an RNeasy spin column as per RNeasy Mini Kit protocol [Qiagen, Hilden, Germany].

116 6.2.3 RNA Extraction from Lamina Propria Mononuclear Cells

6.2.3.1 Disaggregation of Colon

Lamina propria mononuclear cells [LPMCs] were isolated from colon of 7 control subjects using a mechanical and enzymatic disaggregation technique as described in section 5.2.2.1. Isolated LPMCs were lysed in TRIzol reagent [Ambion, Life Technologies, California, USA] at a concentration of 10 x 106 cells per mL and then stored at -80°C until use.

6.2.3.2 Extraction of RNA from Lamina Propria Mononuclear Cells

After the lysed LPMCs were thawed on ice, 250 μL cold chloroform per 1 mL TRIzol reagent was added and mixed, and then incubated at room temperature for 10 minutes. Following this, the cells were centrifuged for 15 minutes, the upper layer transferred to a separate tube and an equal amount of 75% ethanol added. This was applied to an RNeasy spin column and RNA was extracted as per RNeasy Mini Kit protocol [Qiagen, Hilden, Germany].

6.2.4 Conversion of RNA to cDNA

The extracted RNA was converted to cDNA using the RT2 HT First Strand Kit [Qiagen, Hilden, Germany]. In 0.2 mL PCR tubes, using barrier pipette tips [Zap Labcon, California, USA], 6 µL of the Genomic DNA Elimination Buffer was mixed with 2000 ng of RNA in 8 µL DNase-free RNase-free water [Gibco, New York, USA]. The samples were centrifuged for 1 minute in an E-centrifuge [Wealtec, Nevada, USA] and then incubated for 5 minutes at 37°C. For the reverse transcription reaction, 6 µL of the RT Master Mix was added to make a total of 20 µL and, after centrifugation, the samples were heated in a thermocycler [GeneAmp PCR System 9700, Applied Biosystems, New York, USA] at 42°C for 15 minutes then 95°C for 5 minutes before being cooled to 4°C. The samples were frozen at -80°C until ready for use in the qRT-PCR.

117 6.2.5 Quantitative Real Time PCR

384-well custom qRT-PCR plates [Qiagen, Hilden, Germany] were used to study RNA expression in colon, ileum, PBMCs and LPMCs. Primers to the eight LILRs used in earlier immunohistochemistry experiments and TNF-α were preloaded onto the plates (see Table 6.1).

Two reference housekeeping genes, glyceraldehyde-3-phosphate dehydrogenase [GAPDH] and β-actin, were included for normalisation of the data. The plate also contained three qRT-PCR controls: a human genomic DNA contamination control, a reverse transcription control in duplicate and a positive qRT-PCR control in duplicate.

For the qRT-PCR array, using barrier pipette tips, the cDNA samples were diluted by a factor of 3 in DNase-free RNase-free water and mixed with 2x RT2 SYBR Green 1 qRT-PCR Mastermix [Qiagen, Hilden, Germany]. Ten µL of each sample was loaded into the 16 wells containing the 16 individual primers. The qRT-PCR plates were sealed with Optical Adhesive Film and centrifuged for 2 minutes at 1000 g. The plates were heated for 10 minutes at 95°C to activate the HotStart DNA Taq Polymerase, and then the qRT-PCR reaction was performed on a Roche LightCycler 480 [Roche, Basel, Switzerland] under the following cycling conditions: 15 seconds at 95°C then 1 minute at 60°C for 45 cycles. After the fluorescence data were acquired, dissociation-melting curves were generated to assess qRT-PCR specificity.

Analysis was performed using the RT2 Profiler PCR Array Data Analysis programme version 3.5 [Qiagen, Hilden, Germany]. Calculated Ct values (absolute quantification, second derivative max, high confidence) were imported and the quality of the qRT-PCR was checked for PCR reproducibility, reverse transcription efficiency and the presence of genomic DNA contamination. Ct values were normalised to the housekeeping gene, GAPDH, and changes in gene expression were calculated. A fold regulation of ≥ 4 and a p-value of < 0.005, to account for the comparison of multiple genes, were deemed to be significant differences.

118 Table 6.1 Primers for qRT-PCR

Gene Alias GenBank Official Full Name Symbol Reference Sequence

LILRB1 CD85 / CD85J / FLJ37515 NM_006669 Leucocyte immunoglobulin-like / ILT2 / LIR-1 / LIR1 / receptor, subfamily B (with TM MIR-7 / MIR7 and ITIM domains), member 1

LILRB2 CD85D / ILT4 / LILRA6 / NM_005874 Leucocyte immunoglobulin-like LIR-2 / LIR2 / MIR-10 / receptor, subfamily B (with TM MIR10 and ITIM domains), member 2

LILRB3 CD85A / HL9 / ILT5 / NM_006864 Leucocyte immunoglobulin-like LIR-3 / LIR3 / receptor, subfamily B (with TM MGC138403 / PIRB and ITIM domains), member 3

LILRB4 CD85K / HM18/ ILT3 / NM_006847 Leucocyte immunoglobulin-like LILRB5 / LIR-5 / LIR5 receptor, subfamily B (with TM and ITIM domains), member 4

LILRA1 CD85I / LIR-6 / LIR6 / NM_006863 Leucocyte immunoglobulin-like MGC126563 receptor, subfamily A (with TM domain), member 1

LILRA2 CD85H / ILT1 / LIR-7 / NM_006866 Leucocyte immunoglobulin-like LIR7 receptor, subfamily A (with TM domain), member 2

LILRA3 CD85E / HM31 / HM43 / NM_006865 Leucocyte immunoglobulin-like ILT6 / LIR-4 / LIR4 / e3 receptor, subfamily A (without TM domain), member 3

LILRA5 CD85 / CD85F / ILT11 / NM_021250 Leucocyte immunoglobulin-like LILRB7 / LIR9 receptor, subfamily A (with TM domain), member 5

TNF-α DIF / TNF-α / TNFA / NM_000594 Tumour necrosis factor TNFSF2

119 6.3 Results

6.3.1 Subject Demographics

RNA was obtained from 29 subjects undergoing colonic resection. Four subjects had Crohn’s disease (2 females, 2 males) and 2 subjects had ulcerative colitis (2 females); 4 of these subjects had inflamed colon and 3 had inflamed terminal ileum. In addition to intestinal tissue, all subjects with IBD provided peripheral blood samples. The average age of the subjects with IBD was 43.7 ± 6.8 years.

Uninflamed full thickness colon was obtained from 9 control subjects (all males), colonic lamina propria from a further 11 subjects (5 females and 6 males) and lamina propria mononuclear cells from 7 subjects (3 females and 4 males). Peripheral blood mononuclear cells were isolated from 7 control subjects (all male). The average age of the control subjects was 63.7 ± 2.6 years, which was significantly higher than that of the subjects with IBD (p = 0.003). There were no significant differences in age between the various groups of control subjects (p = 0.21).

6.3.2 LILR RNA Expression in Inflammatory Bowel Disease

6.3.2.1 Inflamed vs. Non-Inflamed Colon

When compared with uninflamed control colon (n=9), inflamed colon from patients with IBD (n=4) had 3.2 times higher expression of LILRB1. The increase in LILRB4 and LILRA3 in inflamed colon did not reach statistical significance after adjusting for multiple comparisons. LILRA5 was 5-fold higher but this, too, was not a statistically significant increase.

Table 6.2 LILRs over-expressed in inflamed colon cf. non-inflamed colon

Gene Symbol Fold Regulation p-value

LILRB1 3.25 0.0038 ***

LILRB4 2.69 0.0053 **

LILRA3 4.63 0.0129 *

LILRA5 5.01 0.0634 NS

120 6.3.2.2 PBMCs in Inflammatory Bowel Disease vs. Controls

There was no difference in expression of LILRs in peripheral blood mononuclear cells in patients with (n=6) or without (n=7) IBD.

6.3.2.3 Inflamed Ileum vs. Inflamed Colon

There were no significant differences in LILR expression between inflamed colon (n=4) and inflamed terminal ileum (n=3). LILRB1 was 2.3 times more highly expressed in inflamed colon than ileum, but this difference did not reach statistical significance, p=0.28.

6.3.2.4 Inflamed Colon and Ileum vs. PBMCs from Subjects with IBD

In subjects with active colitis (n=4), all 8 LILRs were more highly expressed in their peripheral blood mononuclear cells than in colon. In particular, LILRB1 was 7.5 times higher, LILRB4 was 3.9 times higher, LILRA1 was 79.5 times higher, LILRA2 was 20.7 times higher and LILRA3 was 34.0 times higher.

Similarly, LILRA1 was 37.2 times higher and LILRA2 was 36.5 times higher in peripheral blood mononuclear cells than in inflamed ileum in subjects with active ileitis (n=3).

Table 6.3 LILR expression in inflamed colon and inflamed ileum cf. PBMCs in subjects with IBD

Colon Ileum

Gene Symbol Fold Regulation p-value Fold Regulation p-value

LILRB1 -7.48 0.0002 *** -10.36 0.0469 *

LILRB2 -21.41 0.0095 ** -12.82 0.0284 *

LILRB3 -6.08 0.0088 ** -5.82 0.0215 *

LILRB4 -3.85 0.0002 *** -2.53 0.4543 NS

LILRA1 -79.48 0.0004 *** -37.19 0.00003 ***

LILRA2 -20.71 0.00004 *** -36.50 0.0038 ***

LILRA3 -34.00 0.0003 *** -16.41 0.1211 NS

LILRA5 -41.43 0.0073 ** -62.54 0.0225 *

121 6.3.3 LILR RNA Expression in Lamina Propria

6.3.3.1 Lamina Propria Tissue

There were no significant differences in expression of LILRs between full thickness colon (n=9) and lamina propria (n=11) in control subjects.

Table 6.4 LILR expression in lamina propria cf. full thickness colon

Gene Symbol Fold Regulation p-value

LILRB1 3.86 0.0230 *

LILRA2 -2.13 0.0231 NS

LILRA3 2.33 0.1097 NS

All 8 LILRs were much more highly expressed in peripheral blood mononuclear cells (n=7) than in uninflamed colon (n=7) or uninflamed lamina propria (n=11) from control subjects.

Table 6.5 LILR expression in colon (matched) and lamina propria cf. PBMCs in control subjects

Colon Lamina propria

Gene Symbol Fold Regulation p-value Fold Regulation p-value

LILRB1 -33.16 0.00005 *** -6.68 0.00002 ***

LILRB2 -50.51 0.0001 *** -30.76 0.000002 ***

LILRB3 -12.05 0.0003 *** -11.47 0.00002 ***

LILRB4 -11.68 0.0005 *** -6.91 0.00005 ***

LILRA1 -103.45 0.000001 *** -114.43 0.000000 ***

LILRA2 -31.00 0.000008 *** -63.25 0.000000 ***

LILRA3 -158.84 0.0013 *** -53.58 0.00008 ***

LILRA5 -210.42 0.000000 *** -135.53 0.000000 ***

122 6.3.3.2 Lamina Propria Mononuclear Cells

Isolated LPMCs (n=7) had higher levels of LILRB4 and LILRA5 than control colon (n=9) or lamina propria (n=11). These cells also had higher levels of LILRB1 than control colon, and higher levels of LILRB2 and LILRA3 than lamina propria.

Table 6.6 LILR expression in LPMCs cf. colon and lamina propria

Colon Lamina propria

Gene Symbol Fold Regulation p-value Fold Regulation p-value

LILRB1 4.24 0.0016 *** 1.10 0.95 NS

LILRB2 6.03 0.0052 ** 4.20 0.0036 ***

LILRB3 1.68 0.031 * 1.83 0.051 NS

LILRB4 9.88 0.0023 *** 7.29 0.0010 ***

LILRA1 -1.08 0.40 NS 1.14 0.28 NS

LILRA2 1.19 0.31 NS 2.53 0.046 *

LILRA3 62.91 0.0070 ** 26.95 0.0033 ***

LILRA5 10.19 0.0030 *** 6.19 0.0023 ***

LPMCs (n=7) expressed lower levels of LILRB1, LILRB2, LILRB3, LILRA1, LILRA2 and LILRA5 than PBMCs (n=7).

Table 6.7 LILR expression in LPMCs cf. PBMCs in control subjects

Gene Symbol Fold Regulation p-value

LILRB1 -6.08 0.0002 ***

LILRB2 -7.32 0.0006 ***

LILRB3 -6.27 0.0011 ***

LILRB4 1.05 0.47 NS

LILRA1 -100.53 0.000001 ***

LILRA2 -25.01 0.00001 ***

LILRA3 -1.99 0.27 NS

LILRA5 -21.90 0.000000 ***

123 6.3.4 TNF-α Expression

6.3.4.1 In Inflammatory Bowel Disease Subjects

TNF-α levels in IBD were similar in colon (fold regulation 1.24, p 0.25) and PBMCs (fold regulation 1.19, p 0.35) when compared to control subjects. There was also no difference in TNF-α level between inflamed colon and inflamed ileum (fold regulation -1.86, p 0.41).

There were lower levels of TNF-α in inflamed colon (fold regulation -4.90, p 0.047) and ileum (fold regulation -3.19, p 0.032), when compared with PBMCs from the same subjects with IBD.

6.3.4.2 In Control Subjects

In control subjects, there was no difference in TNF-α levels between lamina propria tissue and full thickness colon. TNF-α was higher in both PBMCs and LPMCs compared with uninflamed colon or uninflamed lamina propria, and LPMCs had 6.6-fold higher expression levels of TNF-α than PBMCs, p=0.46.

Table 6.8 TNF-α in colon or lamina propria cf. PBMCs and LPMCs in control subjects

Colon Lamina propria

Comparison Fold Regulation p-value Fold Regulation p-value

PBMCs -5.29 0.00007 *** -6.57 0.000003 ***

LPMCs -34.82 0.013 * -43.18 0.0061 **

124 6.4 Discussion

LILRB1 mRNA was more highly expressed in inflamed than non-inflamed colon, confirming the differential protein expression levels of LILRB1 in inflammatory bowel disease demonstrated in the earlier immunohistochemical experiments (see section 4.3.3). Interestingly, the increase in LILRB1 mRNA in inflammatory bowel disease was not paralleled in peripheral blood mononuclear cells.

Although there was a trend towards higher LILRB4, LILRA3 and LILRA5 expression in inflamed colon, these differences did not reach statistical significance. However, as a similar increase in LILRB4 expression in colitis was also shown in immunohistochemical experiments (see section 4.3.2), there is a possibility that this is a type II error and a larger study is needed to confirm or refute this.

No differences were seen in LILR mRNA expression between inflamed ileum and inflamed colon, although this may be due to the small number of samples available.

LILR expression was significantly higher in peripheral blood mononuclear cells than in intestinal tissue or in isolated colonic lamina propria mononuclear cells. PBMCs express high levels of LILRs, as seen in earlier flow cytometric analyses and as reviewed in Chapter 1. However, direct comparisons between LILR expression levels in tissue and peripheral blood have not previously been made.

LPMCs expressed higher levels of LILRs than colonic tissue, and this may partly reflect the higher concentration of macrophages present in the cell suspension. There may also be a relative enrichment of LILR-expressing cells during the disaggregation process. LILR mRNA expression was similar in full thickness colon and colonic lamina propria.

TNF-α is a pro-inflammatory cytokine that is important in chronic inflammatory bowel disease. Surprisingly, there was no difference in TNF-α mRNA levels between subjects with inflammatory bowel disease and control subjects either in colon or peripheral blood. TNF-α mRNA tissue levels are elevated in ulcerative colitis and Crohn’s disease; in ulcerative colitis, but not in Crohn’s disease, these correlate with disease activity and other inflammatory cytokines592-598.

125 Similar to the pattern of LILR expression, TNF-α mRNA levels were 3.2- to 6.6- fold higher in peripheral blood mononuclear cells than in intestinal tissue. The highest levels of TNF-α, however, were in isolated lamina propria mononuclear cells, 6.6-fold higher than in peripheral blood mononuclear cells. Although the peripheral blood and lamina propria mononuclear cells were not specifically stimulated before mRNA was extracted, the overnight disaggregation process and the subsequent density gradient separation to isolate lamina propria cells from colonic tissue may have stimulated TNF-α mRNA expression in these cells.

6.5 Summary

These three chapters demonstrate extensive leucocyte immunoglobulin-like receptor expression in the human colon, particularly on haematopoietic cells of the lamina propria. LILRB1 and LILRA5 were universally expressed predominantly on intestinal macrophages. There was differential expression of all the other 6 LILRs studied, but factors influencing this were not identified within the scope of this study. There was over-expression of LILRB1 in chronic inflammatory bowel disease seen by immunohistochemistry and confirmed with qRT-PCR.

Leucocyte immunoglobulin-like receptor protein expression was expected to be demonstrable by flow cytometry and Western blotting. However, leucocyte immunoglobulin-like receptor expression on isolated lamina propria mononuclear cells could not be verified with flow cytometry, possibly due to differential cell loss or receptor cleavage by the isolation process. The latter possibility, however, was not confirmed by analysing circulating monocytes incubated in disaggregation solutions. Whilst a protein band corresponding to LILRA5 was identified on early Western blot experiments, the lack of availability of further reliable antibody restricted further protein studies.

The next two chapters focus on development of a robust and reliable in vitro model of colonic macrophages in which leucocyte immunoglobulin-like receptor expression is determined.

126 7 Replicating Human Colonic-Like Macrophages in vitro

7.1 Introduction

In this chapter, lamina propria mononuclear cells were isolated and cultured, and their cytokine production in response to a range of stimuli was assessed. However, as colonic macrophages are difficult to isolate and culture in large numbers, there is wide inter- subject variation in cell yields and the long enzymatic and mechanical colonic digestion may negatively influence cell antigen expression and viability, an in vitro cell culture model of colonic-like macrophages was developed that could assist in studying the expression and function of leucocyte immunoglobulin-like receptors. As stromal factors are assumed to play a role in the differentiation of recruited blood monocytes into colonic macrophages, conditioned media derived from colonic stroma were generated. Peripheral blood monocytes were differentiated in the presence of conditioned media to simulate colonic macrophages, and their morphological features, cell surface markers and production of cytokines were assessed.

7.2 Materials and Methods

7.2.1 Cytokine Production by Lamina Propria Mononuclear Cells

The cytokine profiles of cultured lamina propria mononuclear cells in response to various pro– and anti– inflammatory stimuli were analysed.

7.2.1.1 Cytokine Response of Normal LPMCs to Lipopolysaccharide Stimulation

Lamina propria mononuclear cells were isolated by a process of enzymatic and mechanical disaggregation from 17 g of macroscopically normal lamina propria from a 71-year-old gentleman undergoing surgery for sigmoid colon cancer. LPMCs were separated using a Lymphoprep 1.077 g/mL density gradient [Axis-Shield, Oslo, Norway] and macrophage- enriched using the MACS monocyte isolation negative selection kit [Miltenyi Biotec, Cologne, Germany], as described in section 5.2.1.1. From an initial 55 x 106 LPMCs with 94% viability, 13.5 x 106 macrophage-enriched cells with 83% viability were obtained.

127 These cells were cultured in a 96 well flat bottom cell culture plates [Corning Costar, Sigma-Aldrich, Missouri, USA]. 5 x 104 cells per well in 200 µL RPMI-1640 [Invitrogen, California, USA] with 1% foetal calf serum [Sigma-Aldrich, Missouri, USA], with varying concentrations of LPS (Lipopolysaccharide from Salmonella minnesota) [Sigma-Aldrich,

Missouri, USA], were incubated at 37°C with 5% CO2. Lipopolysaccharide [LPS] concentrations ranged from 10 to 300 µg/mL. Supernatants were collected at 12, 24 and 36 hours and stored at -80°C for later cytokine analysis.

Peripheral blood mononuclear cells extracted from a Buffy Coat from a healthy Australian Red Cross Blood Service donor were cultured in the same way for comparison.

An enzyme-linked immunosorbent assay [ELISA] measuring TNF-α [DuoSet, R&D Systems, Minnesota, USA] was performed as per the manufacturer’s instructions. Briefly, 96 well flat bottom Nunc Maxisorp plates [Thermo-Scientific, Illinois, USA] were coated with mouse anti-human TNF-α (4 µg/mL) overnight at room temperature. The plates were washed and blocked with D-PBS/1% BSA and then 100 µL of supernatant or standard (recombinant human TNF-α) diluted in D-PBS/1% BSA, were added to each well. Samples were run in triplicates, and LPMC and PBMC supernatants were either used neat or diluted 1 in 2 or 1 in 5 in D-PBS/1% BSA. After 2 hours, the plate was washed with D-PBS/1% BSA, then incubated for a further 2 hours with 100 µL of biotinylated goat anti-human TNF-α (75 ng/mL) per well. A further wash step was performed prior to adding Streptavidin HRP for 20 minutes. This was followed by adding the chromagen 3-3’,5-5’- tetramethylbenzidine (TMB) [PanBio, Queensland, Australia]. The substrate reaction was stopped after 20 minutes by addition of 1 M sulphuric acid. The optical density was measured at 450 nm on a SpectraMax M2 microplate reader [Molecular Devices, California, USA].

ELISAs were also performed to measure IL-1β, IL-8 and TGF-β levels in cell culture supernatants from LPMCs that were either unstimulated or stimulated with 10 or 20 µg/mL LPS. Supernatants were tested in duplicate using a 1 in 5 dilution with D-PBS/1% BSA for IL-1β or TBS/0.1% BSA/0.05% Tween20 for IL-8, and a 1 in 3 dilution with D- PBS/0.05% Tween20/1.4% delipidized bovine serum for TGF-β, as per the manufacturer’s instructions [DuoSet, R&D Systems, Minnesota, USA].

128 7.2.1.2 Cytokine Production by LPMCs in Response to Other Stimuli

LPMCs were isolated from 30 g lamina propria of the ascending colon from a 76-year-old gentleman undergoing an extended right hemi-colectomy for transverse colon cancer. 197 x 106 LPMCs with a cell viability of 96% were obtained by overnight enzymatic and mechanical disaggregation, as above. To select macrophages by adherence, after 3 washes with D-PBS, LPMCs were incubated in 96 well flat bottom cell culture plates at 5 x 105 cells in RPMI-1640 with 10% heat inactivated human AB serum [Sigma-Aldrich, Missouri, USA] and 4 mmol L-glutamine [Invitrogen, California, USA] for 24 hours at 37°C with 5% CO2. The plates were washed twice with D-PBS and then the LPMCs were cultured, with or without stimulation, in RPMI with 5% human AB serum and 4 mmol L-glutamine (200 µL media per well) at 37°C with 5% CO2. Three different concentrations (10 ng/mL, 25 ng/mL and 100 ng/mL) of each of the following stimuli were used in triplicate wells: Phorbol Myrisate Acetate [PMA], IL-1β, TNF-α, IFN-γ, human GM-CSF, TGF-β, IL-10 and LPS. Supernatants were collected at 6, 12, 24 and 36 hours and stored at -80°C for later cytokine analysis.

IL-1β, IL-8, TGF-β and TNF-α. ELISAs were performed on pooled supernatants in triplicate using a 1 in 5 dilution with D-PBS/1% BSA for IL-1β or TBS/0.1% BSA/0.05% Tween20 for IL-8, a 1 in 3 dilution with D-PBS/0.05% Tween20/1.4% delipidized bovine serum for TGF-β, and a 1 in 2 dilution with D-PBS/1% BSA for TNF-α as per the manufacturer’s instructions [DuoSet, R&D Systems, Minnesota, USA].

7.2.2 Stromal-Derived Conditioned Media

To develop a cell culture model of colonic-like macrophages, stromal-derived conditioned media [SCM] were generated from the colonic stroma of 12 control subjects.

7.2.2.1 Generating Stromal-Derived Conditioned Media

After an overnight extraction of lamina propria mononuclear cells from the colonic lamina propria, the remaining stroma was incubated in a 75 cm2 tissue culture flask [Cellstar, Greiner Bio-One, Frickenhausen, Germany] with RPMI, 10% human AB serum, 100 U/mL penicillin-streptomycin, 50 µg/mL gentamicin, 3 mM L-glutamine and 15 mM HEPES at

37°C with 5% CO2. 1 mL of media was added for each gram of lamina propria based on the

129 pre-disaggregation starting weight. After 24 hours, the SCM from the 12 control subjects was sterile filtered (22 µm) and frozen at -80°C until ready for pooling. All specimens used for the generation of SCM had an extracted lamina propria mononuclear cell viability of > 75%.

7.2.2.2 Assessing Stromal-Derived Conditioned Media i. Endotoxin level

Given its derivation from colonic tissues, endotoxin was felt likely to be a significant component of the SCM. The endotoxin level was quantified using a Limulus Amebocyte Lysate assay, chromogenic endpoint method with diazo- coupling [Associates of Cape Cod, Massachusetts, USA] as per the manufacturer’s instructions. Individual and pooled SCM and endotoxin standards were diluted in sterile water [Baxter, Sydney, Australia] and 50 µL of each was added in duplicate to the inner 60 wells of a 96-well flat bottom microplate [Costar, Sigma-Aldrich, Missouri, USA]. 50 µL of LAL pyrochrome was added to each well, the plate was mixed and then incubated for 37 minutes at 37°C. The proteolytic cleavage reaction was stopped with HCl-reconstituted sodium nitrite solution. This was followed by the addition of ammonium sulfamate and N-(1-Naphthyl) ethylenediamine (NEDA) to form a diazotized magenta derivative of p-nitroaniline that absorbs at a range between 540–550 nm. The optical density was measured at 545 nm on a SpectraMax M2 microplate reader [Molecular Devices, California, USA]. ii. Protein quantification

Protein quantification was performed using the Bicinchoninic Acid Protein Assay Kit [Thermo-Scientific, Illinois, USA], as previously described, and the plate was read at 562 nm on a SpectraMax M2 microplate reader [Molecular Devices, California, USA]. iii. Cytokine analysis a. ELISA IL-8, IL-1β, TGF-β and TNF-α levels were measured in the individual SCM using ELISA DuoSet kits, as per the manufacturer’s instructions as previously described [R&D Systems, Minnesota, USA]. TGF-β and TNF-α levels were also measured in pooled SCM in the same

130 way. TGF-β results were adjusted for the baseline TGF-β level in 10% human AB serum/RPMI incubated at 37°C overnight with 5% CO2. b. Multiplex Pooled SCM was also analysed using a 27-analyte Bio-Plex magnetic bead-based human multiplex assay [Bio-Rad, California, USA] as per the manufacturer’s instructions. In brief, magnetic-coupled beads were added to a pre-wetted 96-well filter plate. Four different dilutions of SCM in RPMI were tested. Fifty µL of each was added to four individual wells and incubated at room temperature on a microplate shaker. After washing with a vacuum manifold, the samples were incubated with the detection antibodies. The plate was washed again before streptavidin-PE was added. After a further incubation, the plate was washed and read on a Bio-Plex 200 machine with Bio-Plex Manager software version 5.0 [Bio-Rad, California, USA].

7.2.3 Cultured Peripheral Blood Mononuclear Cells

For cell cultures, peripheral blood mononuclear cells were isolated from buffy coats from Australian Red Cross Blood Service donors using a Lymphoprep 1.077 g/mL density gradient. The cells were washed twice with D-PBS, and then stored in 10% DMSO in nitrogen vapour phase tanks until use.

7.2.3.1 Culturing Monocytes with Stromal-Derived Conditioned Media i. Monocytes isolated via adherence, differentiated with either GM-CSF or M-CSF and SCM, and stimulated with LPS

PBMCs were thawed and washed twice in RPMI, and then plated in 96-well flat-bottom cell culture plates [Corning Costar, Massachusetts, USA] at a concentration of 3.5–5 x 105 monocytes per mL of RPMI with L-glutamine and penicillin-streptomycin. 0.7–1.0 x 105 monocytes in 200 µL were added to each well. After 2 hours, the plates were washed three times with D-PBS to remove non-adherent cells.

The remaining cells were cultured for 3 days in RPMI and 10% human AB serum, with varying concentrations of SCM with or without GM-CSF (25 ng/mL) or M-CSF (50 ng/mL) at 37°C with 5% CO2. The cells were then washed twice with D-PBS and cultured for a

131 further 2 days in RPMI with 10% human serum with SCM, but without GM-CSF or M-CSF. Cell culture supernatants were collected for cytokine analysis either after 120 hours, or following stimulation for a further 6 or 24 hours with 100 ng/mL LPS (lipopolysaccharide from Salmonella minnesota) [Sigma-Aldrich, Missouri, USA]. ii. Monocytes isolated via magnetic negative selection, differentiated with GM-CSF and SCM, and stimulated with LPS

In a second experiment to compare monocytes selected either via adherence to cell culture plates or by magnetic negative selection using the MACS monocyte isolation method [Miltenyi Biotec, Cologne, Germany], the cell cultures were repeated using PBMCs from the same 2 donors.

After washing in RPMI, the PBMCs were resuspended in cold degassed autoMACS running buffer [Miltenyi Biotec, Cologne, Germany] and blocked with Fc blocking reagent. The cells were indirectly labelled using a cocktail of biotinylated antibodies against CD3, CD7, CD16, CD19, CD56, CD123 and Glycophorin A, followed by the addition of magnetic anti-biotin microbeads. The cells were washed with cold degassed autoMACS running buffer, filtered (30 µm) then separated using a MACS LS column. Labelled cells were retained in the column placed in the magnetic field of a MACS separator, and the unlabelled monocyte- enriched cell fraction was collected as it passed through the column. The monocyte viability after negative selection was > 98% as assessed by trypan blue exclusion, and the cells were > 95% CD45+CD14+CD3- as assessed by flow cytometry.

The isolated monocyte fraction was washed with autoMACS running buffer and then resuspended in RPMI with 10% human AB serum at a concentration of 9.0–10.5 x 105 monocytes per mL (1.8–2.1 x 105 monocytes per well). The monocytes were cultured in 96-well flat-bottomed cell culture plates [Corning Costar, Massachusetts, USA], with or without GM-CSF (25 ng/mL) and/or varying concentrations of SCM, for 3 days at 37°C with 5% CO2. The cells were then washed with D-PBS, and then cultured for a further 2 days in RPMI with 10% human serum without GM-CSF, with or without SCM. Cell culture supernatants were again collected for cytokine analysis after 5 days, or following a further 6 or 24 hours of 100 ng/mL LPS stimulation.

132 iii. Monocytes isolated via magnetic negative selection, differentiated with GM-CSF and SCM, and stimulated with either LPS, IFN-γ or IL-10

The cell cultures were repeated with PBMCs from another 3 healthy Australian Red Cross Blood Service donors, with monocytes isolated using the MACS negative selection method. The isolated cells were > 97% viable as assessed by trypan blue exclusion, and > 94% CD45+CD14+ as determined by flow cytometric analyses. The monocytes were cultured at a concentration of 1.6–1.8 x 105 monocytes per well as described above, and differentiated with GM-CSF (25 ng/mL) [R&D Systems, Minnesota, USA] and/or varying concentrations of SCM. After 5 days, cell culture supernatants were collected for cytokine analysis from unstimulated cells, or after stimulation for a further 6 or 24 hours with LPS (100 ng/mL), IFN-γ (25 ng/mL) [R&D Systems, Minnesota, USA] or IL-10 (25 ng/mL) [R&D Systems, Minnesota, USA]. iv. Monocytes isolated via magnetic negative selection, differentiated with GM-CSF and either SCM or LPS, and stimulated with LPS

To assess whether the effect of the SCM on cultured cells was due to LPS, peripheral blood monocytes isolated by MACS negative selection from four of the above five donors were re-cultured as above with either GM-CSF alone, GM-CSF and SCM (1:100) or GM-CSF and 0.0018 ng/mL LPS, the equivalent amount of LPS to that measured in the SCM. After 5 days of culture or following LPS stimulation as per previous experiments, cell culture supernatants were collected and analysed for TNF-α levels.

Differences in morphology between monocytes cultured under the various conditions outlined above were observed and photographed using a phase contrast inverted microscope [Olympus, Tokyo, Japan].

7.2.3.2 Analysis of Cytokine Production in Cell Culture Supernatants

Cell culture supernatants were analysed using an 8-analyte Bio-Plex magnetic bead-based human multiplex assay to quantify IL-2 IL-4, IL-6, IL-8, IL-10, TNF-α, GM-CSF and IFN-γ levels [Bio-Rad, California, USA] as per the manufacturer’s instructions. 50 µL of 1x group 1 conjugated magnetic-coupled beads and 50 µL of cell culture supernatants, standards or controls were added to a pre-wetted 96-well filter plate and incubated for 30 minutes at

133 room temperature on a microplate shaker at 300 rpm. The initial cell culture supernatants were analysed either undiluted or diluted 1:5 in RPMI to determine the optimal dilution, and subsequent cell culture supernatants were diluted 1:2 in RPMI with 10% heat inactivated human serum, with an extended standard curve using two extra 4-fold dilutions. The plate was washed 3 times using a vacuum manifold, and then 25 µL of 1x detection antibodies were added to each well. The samples were incubated for 30 mins at room temperature on the microplate shaker and then washed 3 times before 50 µL of streptavidin-PE was added to each well. After a further incubation for 10 minutes on the microplate shaker, the plate was again washed and the beads resuspended in 125 µL assay buffer per well. The plate was read on a Bio-Plex 200 Machine with Bio-Plex Manager Software, version 5.0 [Bio-Rad, California, USA].

TGF-β levels in the cell culture supernatants were measured using an ELISA DuoSet kit [R&D Systems, Minnesota, USA]. 96-well flat bottom Nunc Maxisorp plates [Thermo- Scientific, Illinois, USA] were coated with 2 µg/mL mouse anti-human TGF-β at room temperature overnight. The plates were washed three times with 0.05% Tween 20 in D- PBS using an EL406 microplate washer [BioTek, Vermont, USA], and then blocked with 5%

Tween 20 with 0.05% NaN3. Cell culture supernatants were diluted 1:4 or 1:5 in D-PBS with 1.4% delipidized bovine serum and 0.05% Tween 20, then activated with 1N HCl (20 µL per 100 µL of diluted cell culture supernatant). The acid was neutralised after 10 minutes with an equal volume of 1.2 N NaOH/0.5M HEPES, and 100 µL of activated supernatant or standard (recombinant human TGF-β) was added to each well. Samples were run in duplicate, along with a 7-point standard curve (31.25 to 2000 pg/mL). After 2 hours the plate was washed, then incubated for a further 2 hours with 300 ng/mL of biotinylated chicken anti-human TGF-β. A further wash step was performed prior to adding Streptavidin HRP for 20 minutes. This was followed by the addition of the chromogen, 3-3’,5-5’-tetramethylbenzidine [PanBio, Queensland, Australia]. The substrate reaction was stopped after 20 minutes with 1 M sulphuric acid. The optical density was measured at 450 nm with a wavelength correction at 555 nm on a SpectraMax M2 microplate reader [Molecular Devices, California, USA].

IL-1β and TNF-α levels were also measured in the cell culture supernatants using the ELISA DuoSet kits [R&D Systems, Minnesota, USA] as per the manufacturer’s instructions. Samples were run in duplicate, and 7-point (3.91 to 250 pg/mL) or 8-point (7.81 to 1000 pg/mL) standard curves were included, respectively. The optical density was again measured on a SpectraMax M2 microplate reader [Molecular Devices, California, USA].

134 TGF-β1 levels in the cell cultures were corrected for the amount of baseline TGF-β1 in the media (385.5 pg/mL), as per protocol. TGF-β1 levels were also measured in media containing different concentrations of SCM. There were no differences in TGF-β1 levels when compared to media without SCM, p=0.66.

IL-1β levels in the cell cultures were adjusted for the amount of baseline IL-1β in the media, with or without SCM. Similarly, cytokine levels for each of the cytokines in the 8- plex assay (IL-2, IL-4, IL-6, IL-8, IL-10, TNF-α, GM-CSF and IFN-γ) were adjusted for the corresponding baseline cytokine levels in cultured cell-free media.

7.2.3.3 Flow Cytometry

To assess cells cultured with SCM using flow cytometry, peripheral blood monocytes from one of the above donors were isolated by density gradient separation and then magnetic negative selection as described above. These cells were cultured in 24-well ultra-low attachment cell culture plates (8 x 105 monocytes per well) [Corning Costar,

Massachusetts, USA] with or without 1:100 SCM at 37°C with 5% CO2. The media contained RPMI, GM-CSF (25 ng/mL), 10% heat inactivated human serum, 2 mmol L- glutamine and penicillin-streptomycin. After 3 days, the wells were washed with cold D- PBS and the supernatant was transferred to 1.6 mL tubes, which were centrifuged at 300 g for 5 mins. The supernatant was discarded and the cells were then replated in 500 µL per well of fresh media without GM-CSF for a further 48 hours. The cells and supernatants were collected into FACS tubes either after 5 days culture or following a further 6 or 24 hours of 100 ng/mL LPS stimulation.

The cells were washed with cold D-PBS, blocked with 50% heat inactivated human AB serum for 30 mins at 4°C and then stained with CD45 FITC/CD14 PE and CD3 PerCP antibodies or the corresponding irrelevant mouse Ig controls [Becton Dickinson, New Jersey, USA] for 30 mins on ice in the dark. The cells were then washed three times with cold PAB buffer and then fixed in 1% paraformaldehyde until ready to be analysed on the BD FACS Calibur [Becton Dickinson, New Jersey, USA]. For comparison, 1 x 106 freshly isolated peripheral blood monocytes were stained with CD45/14/3 antibodies or their corresponding irrelevant mouse Ig controls in the same way and analysed via flow cytometry.

135 7.3 Results

7.3.1 Cytokine Production by Isolated Lamina Propria Mononuclear Cells

7.3.1.1 Response of Normal LPMCs to LPS Stimulation

The fit of the TNF-α ELISA standard curves were good (r2 = 1). At all time points lamina propria mononuclear cells produced negligible TNF-α (< 140 pg/mL) and were unresponsive to LPS, even at supra-physiological doses.

Unstimulated peripheral blood mononuclear cells produced negligible TNF-α (< 70 pg/mL). These cells produced high levels of TNF-α (> 4000 pg/mL) after stimulation with LPS for 12, 24 or 36 hours. There was no increase in TNF-α levels seen with increasing doses of LPS.

Figure 7.1 TNF-α levels in (a) cultured lamina propria mononuclear cells and (b) peripheral blood mononuclear cells

Similarly, for IL-8 and IL-1β, LPMCs produced very low or undetectable levels of these cytokines, even with LPS stimulation. Unstimulated PBMCs again produced low levels of IL-8 and absent IL-1β, but were able to be stimulated with LPS to produce high levels of IL- 8 (12020 to 19906 pg/mL) and IL-1β (1450 to 1638 pg/mL). LPMCs produced low levels of TGF-β (< 450 pg/mL) even with LPS stimulation. PBMCs produced 3 times as much TGF-β, however, levels were unchanged with LPS stimulation.

136

Figure 7.2 IL-8, IL-1β and TGF-β levels in cultured LPMCs and PBMCs

137 7.3.1.2 Cytokine Production in Response to a Range of Other Stimuli

IL-1β production from LPMCs was again very low (< 80 pg/mL). There was no difference between stimulated and unstimulated cells. IL-lβ levels were higher at the 6-hour stimulation time point, compared with latter time points.

IL-8 levels were moderately elevated with or without stimulation of the LPMCs. There was a trend towards a higher level of IL-8 after 6 hours of stimulation, which decreased at 12 hours then increased with longer stimulation up to 36 hours. PMA-stimulated LPMCs produced the highest IL-8 levels, and this was higher than unstimulated cells.

TGF-β levels were lower in stimulated than in unstimulated cells, and there were no differences between type of stimuli nor length of LPMC stimulation.

TNF-α production was low, without changes with longer stimulation. PMA-stimulation resulted in the highest TNF-α levels. There were no significant differences in cytokine response depending upon the dose of the stimuli used.

Figure 7.3 IL-1β, IL-8, TGF-β and TNF-α production by cultured LPMCs unstimulated or stimulated for up to 36 hours with 8 different stimuli (following 2 pages)

138

139

140 7.3.2 Stromal-Derived Conditioned Media

7.3.2.1 Subjects and Specimens

Of the 12 subjects included, 2 underwent colonic resection for large colonic polyps and 10 had colorectal cancer. The average age was 67.1 ± 3.3 (SEM) years. There was an equal gender distribution. Lamina propria was obtained at the time of the operation from macroscopically normal colon at least 2 cm away from the pathological lesion. The mean weight of lamina propria was 22.3 ± 1.9 (SEM) g per subject and the average viability of extracted lamina propria mononuclear cells was 92 ± 1.7% (SEM). A total volume of 234 mL of SCM was pooled for use in the cell culture experiments.

7.3.2.2 Endotoxin Level

There were no significant differences in endotoxin levels between individual and pooled SCM, p-value = 0.65. The endotoxin level in the pooled SCM was 0.94 ± 0.41 (SEM) EU/mL, and the average endotoxin level in the 12 individual SCM was 0.82 ± 0.09 (SEM) EU/mL. Endotoxin levels ranged from 0.53 to 1.36 EU/mL.

Figure 7.4 Endotoxin levels in pooled and individual SCM

141 7.3.2.3 Protein Quantification

The protein level in the pooled SCM was 6001 ± 230 (SEM) µg/mL. This was significantly higher than the protein level in RPMI with 10% human serum of 4725 ± 120 (SEM) µg/mL, p-value = 0.0002. The protein levels of individual SCM ranged from 3869 to 6917 µg/mL.

Figure 7.5 Total protein levels in individual and pooled SCM

7.3.2.4 Cytokine Analysis i. ELISA

IL-8 levels were high in the 12 individual SCM samples and ranged from 10346 to 28823 pg/mL with an average level of 14935 ± 1470 (SEM) pg/mL. IL-1β levels in the individual media ranged from 677 to 1878 pg/mL, with an average level of 1609 ± 98 (SEM) pg/mL. TGF-β levels were lower and ranged from 349 to 3103 pg/mL, with an average of 1329 ± 207 (SEM) pg/mL. In pooled SCM, the TGF-β level was 993 ± 119 (SEM) pg/mL. The lowest cytokine levels were TNF-α, with individual levels ranging from 22 to 769 pg/mL

142 and an average level of 158 ± 68 (SEM) pg/mL. The TNF-α level in pooled SCM was 132 ± 5 (SEM) pg/mL.

There was little variation in cytokine levels between subjects overall. Of the subjects with high or low cytokine values, there were no clinical patient characteristics that were thought to be contributing to these outlying results. There was also a lack of correlation between levels of different cytokines within individual subjects.

The mean levels in the individual SCM varied from the levels in the pooled SCM, because different volumes of SCM were generated per subject depending on the starting weight of lamina propria obtained.

Figure 7.6 IL-8, IL-1β, TGF-β and TNF-α levels in individual SCM

143 ii. Multiplex assay

Pooled SCM contained high levels of IL-6, IL-8/CXCL8, G-CSF, MCP-1/CCL2 and VEGF. There was some variability with cytokine levels measured with the multiplex assay, but IL- 8/CXCL8, IL-1β and TNF-α levels were 1.5 to 2 times higher when measured with the multiplex assay than with the ELISA assays. Whilst GM-CSF was present in the SCM, levels were at least 100 times lower than the amount used for standard in vitro differentiation of monocytes to macrophages.

Figure 7.7 Cytokines, chemokines and growth factors in the SCM as measured with the 27-analyte multiplex assay

144 Table 7.1 Cytokine, chemokines and growth factors in the SCM (mean ± SEM)

IL-1β 3257 ± 450 pg/mL

IL-1ra 1289 ± 345 pg/mL

IL-2 61.42 ± 15.98 pg/mL

IL-4 10.55 ± 2.83 pg/mL

IL-5 22.43 ± 7.35 pg/mL

IL-6 131746 ± 91116 pg/mL

IL-7 15.53 ± 4.63 pg/mL

IL-8/CXCL8 25000 ± 0 pg/mL

IL-9 109.3 ± 36.5 pg/mL

IL-10 345 ± 110 pg/mL

IL-12p70 548.6 ± 176.1 pg/mL

IL-13 113.1 ± 28.0 pg/mL

IL-15 68.88 ± 25.61 pg/mL

IL-17A 100.3 ± 39.6 pg/mL

Eotaxin 89.48 ± 30.20 pg/mL

Basic FGF 2940 ± 795 pg/mL

G-CSF 59052 ± 9384 pg/mL

GM-CSF 2291 ± 896 pg/mL

IFN-γ 1230 ± 255 pg/mL

IP-10 382.4 ± 37.3 pg/mL

MCP-1/CCL2 25069 ± 6626 pg/mL

MIP-1a 581.2 ± 93.7 pg/mL

MIP-1b 564.8 ± 98.6 pg/mL

PDGF-BB 149.6 ± 27.4 pg/mL

Rantes 114.4 ± 7.7 pg/mL

TNF-α 238.4 ± 86.1 pg/mL

VEGF 6508 ± 430 pg/mL

145 7.3.3 Effect of Stromal-Derived Conditioned Media on Cultured Peripheral Blood Mononuclear Cells

7.3.3.1 Cell Morphology

Cells cultured with GM-CSF differentiated into large adherent cells. Pleomorphism was seen in the presence of SCM. In addition to large adherent cells similar to those cultured with GM-CSF without SCM, smaller cells that were rounder in shaper were present. There was also some clumping of cells.

When monocytes were differentiated in the presence of SCM without GM-CSF, there was cell shrinkage and loss, with fewer cells per microscopy field. The cells that were present varied in size but overall were rounder and smaller than those cultured in the presence of GM-CSF.

a) GM-CSF b) SCM + GM-CSF c) SCM Figure 7.8 Cell morphology after 5 days of culture. Monocytes cultured for 60 hours with (a) GM-CSF, (b) SCM and GM-CSF, and (c) SCM. Photos at 200x magnification

When monocytes were differentiated in the presence of LPS and GM-CSF, they appeared large and elongated, similar to cells cultured in the presence of GM-CSF alone.

146 a) GM-CSF

b) SCM + GM-CSF

c) LPS + GM-CSF

d) SCM

Figure 7.9 Cell morphology in the presence or absence of SCM, LPS and GM-CSF. Representative photos from monocytes cultured for 60 hours with (a) GM-CSF, (b) SCM and GM-CSF, (c) LPS and GM-CSF, and (d) SCM. Photos at 100x (left) and 400x (right) magnification

147 When cultured in ultra-low attachment plates, the cells were rounder than those cultured in adherent standard cell culture plates.

a) Adherent plate b) Ultra-low attachment plate GM-CSF

SCM + GM-CSF

Figure 7.10 Differences in cell morphology between adherent standard or ultra-low attachment cell culture plates. These images compare monocytes cultured either with GM- CSF or SCM and GM-CSF in (a) standard adherent cell culture plates and (b) ultra-low attachment cell culture plates. Photos at 400x magnification

148 7.3.3.2 Phenotype Determined by Flow Cytometry

Table 7.2 Expression of CD45, CD14 and CD3 on cultured monocytes

CD45 CD14 CD3

Unstimulated 56% 9% < 1%

Unstimulated, with SCM 15% 13% < 1%

6 hr LPS stimulation 38% 10% < 1%

6 hr LPS stimulation, with SCM 16% 12% < 1%

24 hr LPS stimulation 44% 10% < 1%

24 hr LPS stimulation, with SCM 11% 10% < 1%

SCM = stromal-derived conditioned media

Peripheral blood monocytes isolated by magnetic negative selection were 91% CD14+CD45+. After culturing in ultra-low attachment cell culture plates for 6 days the cells were 38-56% CD45+. When SCM was present, only 11-16% of cells were CD45+. CD14 expression was markedly reduced to 9-13% of the cell population. There was no difference in the number of CD45+ or CD14+ cells in cultures following LPS stimulation. All cell cultures contained less than 1% CD3+ cells.

Figure 7.11 CD45/CD14 expression with corresponding isotype controls on MACS- isolated peripheral blood monocytes (left), monocytes cultured with GM-CSF (middle) and monocytes cultured with SCM and GM-CSF (right)

149 7.3.3.3 Cytokine Production i. Monocyte differentiation: GM-CSF vs. M-CSF

Monocytes selected by adherence to cell culture plates and differentiated with either GM- CSF or M-CSF produced low levels of TNF-α.

After LPS stimulation for six hours, cells produced higher levels of TNF-α. There was a higher level of TNF-α production by cells differentiated in M-CSF than in GM-CSF. With LPS stimulation for 24 hours, TNF-α levels had returned to baseline.

Interestingly, when different concentrations of SCM were present in the cell culture media, TNF-α production in response to LPS stimulation was decreased in a dose-dependent manner. The down-regulation of TNF-α was more marked with M-CSF than with GM-CSF.

Figure 7.12 TNF-α levels in cell culture supernatants from 2 subjects comparing GM- CSF vs. M-CSF differentiation of monocytes, with varying amounts of SCM and 6 or 24 hours of LPS stimulation

150 ii. Selection of monocytes: adherence vs. MACS

When monocytes were selected either by adherence to cell culture plates or by MACS negative selection and cultured for five days, they produced low levels of TNF-α. With 6 hours of LPS stimulation, monocytes selected by either method produced high levels of TNF-α, and this was again down-regulated with increasing amounts of SCM in the cell culture media. With 24 hours of LPS stimulation, cells selected via MACS, but not those selected by adherence, produced TNF-α, which was again down-regulated in the presence of SCM.

Figure 7.13 TNF-α levels in cell culture supernatants from 2 subjects comparing monocyte selection via adherence or MACS

151 iii. Response to varying amounts of SCM in the cell culture media

Monocytes isolated from PBMCs via MACS were cultured with varying amounts of SCM with or without GM-CSF. With increasing concentrations of SCM, there was more marked down-regulation of the TNF-α response to LPS stimulation.

Figure 7.14 TNF-α levels in cell culture supernatants from 2 subjects showing varying cytokine response of monocytes to different amounts of SCM

152 iv. Cytokine production following cell stimulation with LPS, IFN-γ or IL-10

Monocytes isolated via MACS, n=3, were cultured with GM-CSF ± SCM and stimulated with either LPS, IFN-γ or IL-10. Levels of 10 different cytokines were measured in the cell culture supernatants. a. TNF-α

As in previous experiments, unstimulated monocytes cultured with GM-CSF produced little TNF-α (22.7 pg/mL). Monocytes cultured with SCM for 5 days produced similar levels of TNF-α (36.4 pg/mL), but cells cultured with SCM for only 3 days produced very little TNF-α (< 5 pg/mL).

When stimulated with 6 hours of LPS, monocytes differentiated with GM-CSF produced very high levels of TNF-α. With a more prolonged exposure to LPS, this response was less marked, as measured by a lower TNF-α level.

The addition of SCM blunted the responsiveness of the cells to produce TNF-α, and this was dose- and time- dependent.

Monocytes differentiated with SCM in the absence of GM-CSF still responded to LPS stimulation in a similar fashion, but the increase in TNF-α production was less marked compared with cells cultured in the presence of GM-CSF.

153

Figure 7.15 TNF-α levels in cell culture supernatants of monocytes cultured (a) in the presence or absence of SCM for 5 days followed by stimulation with LPS for 6 or 24 hours or (b) with SCM for 3 or 5 days in the presence or absence of GM-CSF, followed by stimulation with LPS for 6 or 24 hours

154 Monocytes differentiated with GM-CSF alone or with 5 days of SCM in the presence or absence of GM- CSF produced low levels of TNF-α with no significant change following stimulation with

IFN-γ. Monocytes differentiated with 3 days of SCM expressed minimal TNF-α, which was enhanced following IFN-γ stimulation.

Following IL-10 stimulation, cells cultured with GM-CSF alone or with 5 days of SCM in the presence or absence of GM- CSF produced lower levels of TNF-α. Cells cultured for a shorter period of time with

SCM in the presence or absence of GM-CSF produced minimal or no TNF-α with or without IL-10 stimulation.

Figure 7.16 TNF-α levels in response to IFN-γ (top) and IL-10 (bottom) stimulation

155 b. IL-1β

Cells cultured with GM- CSF produced no IL-1β, even with LPS, IFN-γ or IL-10 stimulation.

Cells differentiated in the presence of SCM produced IL-1β in a dose-dependent manner.

This was down-regulated by all three of the stimuli: LPS, IFN-γ and IL-10.

Figure 7.17 IL-1β levels in response to (a) LPS, (b) IFN-γ and (c) IL-10 stimulation

156 c. TGF-β

Cells cultured with GM-CSF without SCM produced a low level of TGF-β, mean level 47.8 ± 17.3 (SEM) pg/mL.

When stimulated with LPS for a further 6 or 24 hours, the cells produced higher levels of TGF-β, but the difference did not reach statistical significance, p = 0.06.

Similarly, cells stimulated with IFN-γ or IL-10 for a further 6 or 24 hours produced an increased level of TGF-β.

Cells cultured with GM-CSF and SCM, or with SCM alone did not produce TGF-β, even when stimulated with LPS, IFN-γ or IL-10.

Figure 7.18 TGF-β levels in response to (a) LPS, (b) IFN- γ and (c) IL-10 stimulation

157 d. IL-6

GM-CSF differentiated macrophages and monocytes differentiated with SCM for 3 days with or without GM-CSF did not produce IL-6, even after stimulation.

Monocytes differentiated with 5 days of SCM produced high levels of IL-6 in a dose-dependent manner, without a significant difference in IL-6 levels following cell stimulation.

Figure 7.19 IL-6 levels in response to (a) LPS, (b) IFN-γ and (c) IL-10 stimulation

158 e. IL-2, IL-4 and IL-10

Cultured cells produced low levels of these three cytokines (see figures in Appendix 2a and 2b).

Overall, IL-2 production was low or absent from these cultured cells. Monocytes differentiated with GM-CSF alone or with SCM for 3 days produced no IL-2 and cells differentiated with SCM for 5 days produced low levels of IL-2 (1.4 pg/mL).

GM-CSF differentiated macrophages produced IL-2 in response to 6 hours of LPS and IFN- γ, but not IL-10, stimulation. By 24 hours of stimulation, the IL-2 levels had returned almost to baseline.

Cells differentiated for 3 days with a small amount of SCM (1:200) responded to 6 hours of LPS stimulation in a similar way. However, cells differentiated with higher amounts of SCM or for a longer time period did not produce IL-2 in response to LPS stimulation.

Unlike after LPS stimulation, IFN-γ stimulation for 6 hours resulted in IL-2 production. 24 hours of IFN-γ stimulation resulted in levels comparable to baseline IL-2 production.

Cultured cells did not change their IL-2 production levels in response to either 6 or 24 hours of IL-10 stimulation.

IL-4 production by these cells was also low, and mimicked IL-2 production in unstimulated cultured cells, as well as following LPS, IFN-γ or IL-10 production.

Unstimulated GM-CSF differentiated macrophages did not produce IL-10 and, with the addition of LPS or IFN-γ, a low level of IL-10 was generated. Cells cultured with SCM at a concentration of 1:50 or higher for 5 days produced higher levels of IL-10, which were down-regulated by 6 or 24 hours of LPS or IFN-γ stimulation. Cells cultured with a lower dose of SCM (1:100 or 1:200) for a shorter duration (3 days) produced low baseline levels of IL-10, but responded to 6 or 24 hours of LPS stimulation with a 5 to 40 fold increase in IL-10 production, and to 6 hours of IFN-γ stimulation with an 3 to 20 fold increase in IL-10 production.

159 f. IL-8 and IFN-γ

There was a wide variability of both IL-8 and IFN-γ production between experiments (see figures in Appendix 2c and 2d). Unstimulated GM-GSF-differentiated macrophages produced neither IL-8 nor IFN-γ. These cells produced IFN-γ following LPS stimulation. Cells cultured with SCM for 5 days also produced low levels of IFN-γ.

Cells differentiated with GM-CSF produced a low level of IL-8 in response to 24 hours of LPS stimulation. In the presence of SCM for 5 days, with or without GM-CSF, cells produced IL-8 but again showing wide variation in levels between different experiments. These cells also produced high levels of IL-8 after 6 or 24 hours of LPS, IFN-γ or IL-10 stimulation. Cells cultured with only 3 days of SCM in the presence or absence of GM-CSF, however, did not consistently produce high amounts of IL-8. g. GM-CSF

25 ng/mL GM-CSF was used as a differentiating factor in the media for the first 72 hours of the cell cultures. This was removed and the cells washed with D-PBS before fresh media without GM-CSF was added for a further 48 to 72 hours. Cells not cultured with GM-CSF produced minimal or no GM-CSF. Cells cultured with GM-CSF were stimulated to produce moderate levels of GM-CSF. This was enhanced by the presence of SCM and down- regulated by LPS, IFN-γ and IL-10 (see figures in Appendix 2e).

160 v. Effect of SCM vs. LPS on cell differentiation

Monocytes from 4 subjects were cultured in the presence of GM-CSF with either SCM or a similar amount of LPS to that contained in the conditioned media. TNF-α levels in the baseline cell-free media and in the SCM at that dilution were negligible (< 2 pg/mL).

As demonstrated in the earlier experiments, monocytes cultured with GM-CSF expressed low levels of TNF-α and, following stimulation with LPS for 6 or 24 hours but not IFN-γ or IL-10, high levels of TNF-α were expressed.

Differentiation of the cells in the presence of either SCM or LPS decreased the ability of these cells to produce TNF-α following LPS stimulation.

Figure 7.20 Comparison of TNF-α levels following cell differentiation with either SCM or LPS, in response to LPS stimulation

161 Cells differentiated with LPS expressed similar levels of TNF-α following IFN-γ and IL-10 stimulation compared with SCM-differentiated cells.

Figure 7.21 Comparison of TNF-α levels following cell differentiation with either SCM or LPS, in response to (b) IFN-γ and (c) IL-10 stimulation

162 7.4 Discussion i. Colonic Lamina Propria Macrophages

The intestine is a down-regulated environment in the normal healthy state and, as expected599, cytokine production by lamina propria macrophages was low. Isolated colonic macrophages cultured for up to 36 hours produced low levels of TNF-α, IL-1β, IL-8 and TGF-β. Stimulation with LPS, even at large doses, did not have any effect on cytokine production by lamina propria macrophages. PMA stimulation resulted in higher levels of IL-8 and TNF-α production by colonic macrophages, and all stimuli suppressed TGF-β levels by approximately 25 to 35%. In distinct contrast to lamina propria macrophages, peripheral blood mononuclear cells were highly responsive to cell stimulation. ii. The Extra-Cellular Matrix of the Gastrointestinal Tract

The colonic lamina propria is highly cellular, containing numerous lymphocytes, plasma cells, mast cells and macrophages that secrete cytokines and growth factors and provide immunological protection from the microorganisms and other pathogens in the lumen. In addition to the abundant haematopoietic cells, there is a myriad of non-haematopoietic stromal cells that both produce and maintain the framework of the lamina propria, and provide innate-like immune responses.

The role of the extracellular matrix of the intestinal submucosa has been more extensively studied than that of the mucosa, but there are similarities between the two. The presence of more than 100 cross-linked proteins in the submucosa has been described and, besides providing a structural support for the stromal cells, the extracellular matrix has important roles in enabling intercellular communications and regulating cell proliferation, survival, differentiation and migration600, 601. Indeed, porcine small intestinal submucosa is used as a biological scaffold for tissue regeneration and formation of biological grafts [Cook Biotech, Indiana, USA].

One research group has used conditioned media from cell-depleted mucosa of the small intestine to characterise the intestinal stroma. They induced peripheral blood monocytes to differentiate into an intestinal macrophage phenotype using jejunal stromal-derived conditioned media. With prolonged culture of more than 24 to 48 hours, jejunal stromal-

163 derived conditioned media induced a dose and time-dependent reduction in expression of CD14, CD16 and CCR5, but not HLA-DR or CD13. When monocytes were cultured with conditioned media for 1 hour and then stimulated for 24 hours with either Helicobacter pylori urease, LPS or IFN-γ, there was a dose-dependent decline in the release of inducible IL-1, IL-6, TNF-α, IL-10 and RANTES and a reciprocal increase in TGF-β release212. The changes were induced through Smad-related nuclear factor kappa B (NF-κB) inactivation602. They also showed that epithelial cells and mast cells of the non-inflamed jejunal mucosa constitutively produced IL-8 and TGF-β and, largely due to the presence of these cytokines, conditioned media led to recruitment of peripheral blood monocytes212.

In a similar way, colonic stroma is likely to influence the phenotype and function of cells contained within the lamina propria, with the major difference that the colonic environment is likely to contain far greater mounts of bacterial products than the jejunal environment and thus differentially influence cellular maturation and function. The effect of the colonic mucosa on neuronal networks has been studied in patients with irritable bowel syndrome. Conditioned media generated from distal colonic biopsies from subjects with either diarrhoea- or constipation- predominant irritable bowel syndrome was shown to excite submucosal neurons from human and guinea pig distal ileum and colon. This did not occur using conditioned media from healthy controls, and it correlated with mast cell numbers in the human colonic mucosa603-605. A different research group showed that supernatants from colonic biopsies from patients with active ulcerative colitis excited nociceptive T9 to T13 dorsal root ganglia neurons in adult mice. These colonic supernatants contained high levels of TNF-α, and the neuronal effects were replicated with TNF-α and attenuated in TNF-receptor knockout mice606. Colonic stromal cells have also been implicated in the progression of colorectal cancer607-609.

In this study, conditioned media from twelve subjects was pooled to reduce the effect of inter-subject variability that occurs in experiments involving human subjects. A combined mechanical and enzymatic digestion technique was used to extract the lamina propria mononuclear cells from the colonic mucosa, as our research group has successfully generated viable colonic cells via this method610, 611. Lamina propria mononuclear cell viability was used as a surrogate marker of adequate stromal viability. The starting lamina propria weight, rather than the wet stromal weight, was used to determine the amount of media used and, in addition to RPMI, 10% human serum and antibiotics were included in the media, as the colonic environment is rich in microorganisms 612.

164 The measured endotoxin level was, as expected, moderately high in the stromal-derived conditioned media (0.94 EU/mL). The protein level was 1.27 times higher than baseline media, and levels of cytokines and growth factors involved in inflammation and angiogenesis were also elevated. In comparison, jejunal stromal-derived conditioned media has been shown to contain low or undetectable levels of regulatory cytokines including GM-CSF, M-CSF, IL-2, IL-4 and IL-10, with TGF-β levels less than 300 pg/mL and IL-8 levels less than 600 pg/mL212, 613. Similarly, terminal ileal stromal-derived conditioned media contained TGF-β levels of 210 pg/mL and low levels of Il-1, IL-2, IL-6, IL-10, IL-17, IFN-γ and TNF-α 310.

There were, in addition to these differences, a number of methodological differences between this present study and the studies using jejunal stromal-derived conditioned media. Jejunal samples were obtained from patients undergoing obesity surgery or from healthy organ donors, whereas the reasons for colonic resection were large colonic polyps or colon cancer. As suggested, the small intestinal microbial environment is vastly different to that of the colon612, 614. The jejunum harbours relatively low numbers of commensal organisms, which are predominantly acid-tolerant lactobacilli and streptococci. The colon contains the highest density of bacteria, ten million times more than the jejunum, with a great diversity of bacteria. Although there are more than 400 species present, these are predominantly obligate anaerobes. The recognition of these commensal organisms is essential for maintenance of normal immunological tolerance in the intestine through Toll-like receptor and NOD-like receptor mechanisms615, 616. The jejunal lamina propria macrophages were released from the small intestinal tissue using dispase rather than collagenase, and the isolated jejunal lamina propria macrophages did not adhere210. This is in contrast to other tissue macrophages, including colonic macrophages, where adherence is an established characteristic584. iii. In vitro Macrophage Cell Cultures

Primary human tissue macrophages do not proliferate in culture and are relatively difficult to isolate in large numbers. A widely accepted method to study macrophages and other related mononuclear phagocytes in vitro is through the use of human blood monocytes differentiated with GM-CSF or M-CSF617, 618. To generate colonic-like macrophages, peripheral blood monocytes isolated via density gradient separation and negative magnetic selection were used. Selection of cells via adherence to tissue culture

165 plates was considered, but this method produced a more heterogeneous population of cells.

GM-CSF and stromal-derived conditioned media were used to differentiate the monocytes into colonic-like macrophages. When the effect of GM-CSF and M-CSF on monocytes was compared, there were similar patterns of TNF-α production seen before and after LPS stimulation. The effect of stromal-derived conditioned media on monocytes was dose- dependent and, as the stromal-derived conditioned media was pooled, it was used at varying concentrations rather than based on amount of total protein per se. No further LPS was added to the cell culture in the differentiation phase, as endotoxin was already present in the stromal-derived conditioned media.

Cells were cultured for 5 days, with removal of GM-CSF and refreshing of the media after 3 days. This is the standard minimum time to differentiate monocytes to macrophages and, in studies using jejunal stromal-derived conditioned media, phenotypic changes were seen after exposure to the conditioned media for 36 to 48 hours212. In the colon, tissue macrophages are continuously regenerated from circulating monocytes which, following migration into the intestinal lamina propria, differentiate into an immune-tolerant resident macrophage phenotype.

LPS, IFN-γ and IL-10 were chosen as stimulants, as these are important factors that influence macrophages in the colon. LPS is a major component of gram-negative bacteria, common pathogens in the lower gastrointestinal tract, that can stimulate macrophages to produce pro-inflammatory mediators619-621, IFN-γ is a key activator of macrophages acting through induction of direct antimicrobial mechanisms as well as up-regulation of antigen processing and presenting pathways622 and IL-10 is an anti-inflammatory cytokine that regulates the inflammatory response by macrophages and other immune cells in the normal tolerogenic state623, 624 and its dysregulation plays an important role in inflammatory bowel disease625, 626.

Monocyte differentiation with stromal-derived conditioned media resulted in increased heterogeneity of cell size and shape compared to differentiation with GM-CSF alone. The heterogeneous nature of these cells, similar to isolated colonic lamina propria mononuclear cells, was also reflected in the flow cytometry forward- and side-scatter plots. In contrast, MACS-isolated peripheral blood monocytes comprised a distinct, homogeneous population of cells.

166 When LPS was included as a differentiating factor with GM-CSF, cells were closer in appearance to those cultured with GM-CSF alone than with stromal-derived conditioned media and GM-CSF. In contrast, following LPS stimulation, the pattern of TNF-α production by these LPS-differentiated cells mimicked cells differentiated with both stromal-derived conditioned media and GM-CSF, rather than with GM-CSF alone.

Both CD45 and CD14 expression were reduced following cell culture in low-adherent plates. All CD14 positive cells, which comprised approximately 10% of the cells, were CD45 positive, as expected. It has previously been established that macrophage expression of CD14 is reduced in vitro627 and that resident colonic macrophages have low CD14 expression levels206. Interestingly, colonic stromal-derived conditioned media had no additional effect on CD14 levels in these cultured cells and CD14 expression did not change in response to LPS stimulation. CD45-expressing leucocytic cells were predicted to remain positive627 and there are no data to support the reduction in CD45 expression following cell culture. In addition, CD45 levels were markedly reduced when stromal- derived conditioned media was included as a monocyte-differentiating factor. A component of the colonic stroma may have altered the expression of this cell marker or promoted the outgrowth of non-haematopoietic-type cells. Non-adherent plates were chosen for this experiment, and this may have influenced the cell phenotype. Clearly these preliminary findings need to be replicated, with a broader array of non-haematopoietic markers to characterise the phenotype of these cultured cells, particularly as the starting cell population was over 90% CD14/CD45 positive. iv. Cytokine Expression of the Cultured Monocytes

The pro- and anti-inflammatory cytokine profile was assessed in these cultured monocytes. TNF-α and IL-1β are both important acute phase response mediators. TNF-α levels were low in GM-CSF-, M-CSF- and stromal-derived conditioned media-differentiated macrophage-like cells. Following 6 and 24 hours of stimulation with LPS (lipopolysaccharide from Salmonella minnesota) mimicking in vitro a toxigenic gram- negative bacterial insult, monocytes that were classically differentiated to macrophages with GM-CSF in the absence of stromal-derived conditioned media produced high levels of TNF-α. This was an expected pro-inflammatory response of the macrophages to LPS. Interestingly, differentiation of monocytes in the presence of stromal-derived conditioned media resulted in a blunting of their responsiveness to LPS stimulation in a dose- and time- dependent manner.

167 In variance to TNF-α, classically differentiated macrophages did not produce IL-1β, even after stimulation with LPS. Differentiation with stromal-derived conditioned media resulted in increasing amounts of IL-1β production, which was partially lessened but not negated by stimulation of the cells with LPS. This suggests that, despite sharing a similar range of biological activities, transcriptional regulation of these cytokines differs in these cells.

TGF-β1 is a cytokine with predominantly anti-inflammatory properties, and its expression pattern was different to both that of TNF-α and IL-1β. Only cells differentiated with GM- CSF produced TGF-β1, and this was increased following 24 hours, but not 6 hours, of LPS stimulation. Stromal-derived conditioned media blunted the ability of these cells to produced TGF-β1.

IL-6 is associated with cell activation and acts as an anti-inflammatory or pro- inflammatory mediator, with an essential role in the transition from acute inflammation to either acquired immunity or chronic inflammation. IL-6 levels were elevated, irrespective of the presence or absence of GM-CSF, in monocytes differentiated for 5 days with stromal- derived conditioned media. There was no significant IL-6 production either with a shorter 3-day differentiation or in the absence of stromal-derived conditioned media.

Elevation of IL-10 was expected in colonic-like macrophages, as this is a key regulatory cytokine in the gastrointestinal tract. IL-10 levels in the colon are dependent upon the presence of commensal microbiota203 and it enhances the presence and tolerance-inducing functions of macrophage and regulatory T lymphocyte populations245, 263, 623, 628. In this study, IL-10 levels overall were low. Whilst unstimulated classically differentiated macrophages did not produce IL-10, its production was inducible by LPS and IFN-γ. Colonic-like macrophages that were differentiated with at least 2% stromal-derived conditioned media for 5 days produced higher levels of IL-10 at baseline and, in converse to classical macrophages, were down-regulated by both LPS and IFN-γ stimulation. The optimal concentration of stromal-derived conditioned media that correlates with the colonic environment in vivo has not been determined, and these findings suggest that colonic stromal factors result in higher basal IL-10 production and a lower responsiveness to stimulation, and that the regulatory effects of IL-10 in the intestine may not require high absolute concentrations of this cytokine.

168 In addition to LPS, the effects of IFN-γ and IL-10 stimulation on these cultured cells were also studied. There was only slight up– or down– regulation of TNF-α production following stimulation with IFN-γ or IL-10, respectively, and, intriguingly, the expression pattern of IL-1β, TGF-β1 and IL-6 by these cells was similar in response to all three stimuli.

Although a similar down-regulation of TNF-α in response to LPS stimulation was reported with jejunal stromal-derived conditioned media, in that study monocytes were pre- incubated with conditioned media for only an hour, an exposure time too short to differentiate the cells into an intestinal macrophage-like phenotype. This down-regulation of TNF-α was attributed to mast cell-derived TGF-β in the conditioned media, however the concentration of the conditioned media used was based upon protein levels not TGF-β concentration, and this is likely to have varied between individual subjects212.

In this study, the down-regulation of TNF-α production may partly be explained by the presence of LPS in the colonic stromal-derived conditioned media, as a similar down- regulation was also seen following monocyte differentiation with an equivalent dose of Lipopolysaccharide from Salmonella minnesota. TNF-α responses by LPS-differentiated cells following IFN-γ and IL-10 stimulation were also similar to that of stromal-derived conditioned media-differentiated cells. The morphological changes seen under phase contrast microscopy, however, did not appear to correlate with the cytokine production, and a broader spectrum of responses by LPS-differentiated monocytes needs to be evaluated. Lipopolysaccharides act primarily through the CD14/TLR4/MD2 receptor complex, although there is diversity particularly between different strains and subspecies629, 630. Humans are sensitive to low doses of LPS and in these experiments the amount of LPS used was below the acceptable endotoxin limit for medical devices of 0.06 EU/mL631 and 5.6 x 104 times lower than the amount used to stimulate the cells.

It is more likely that the combination of LPS with the multiple cytokines and growth factors present in the colonic stromal-derived conditioned media resulted in specific intestinal-like macrophage differentiation. Indeed, plasticity of gene expression has been well-characterised, where macrophages develop distinct pro-inflammatory or immunoregulatory functional phenotypes, depending upon cues from the local environment632, 633.

169 7.5 Summary

Using colonic stromal-derived conditioned media as a differentiating factor in monocyte cultures, a model of colonic-like intestinal macrophages has been developed. As colonic macrophages are finite in number and relatively difficult to culture in sufficient quantities, this in vitro model is an attractive alternative in characterising these important immune cells of the intestine. Colonic stromal-derived conditioned media can be more readily generated in volume and stored for later use and use of this model would enable further elucidation of the regulating factors and cellular responses of colonic macrophages and characterisation of the role of leucocyte immunoglobulin-like receptors in the gastrointestinal tract.

An interesting finding was the blunted TNF-α responsiveness to LPS stimulation of these in vitro colonic-like macrophages, in distinction to the highly LPS-responsive classically differentiated in vitro macrophages. This finding was seen at both mRNA (see Chapter 8) and cytokine expression levels and signifies that colonic stromal factors are likely to hold the key to the immune tolerant state of the intestine.

170 8 Leucocyte Immunoglobulin-Like Receptors and TNF-α Expression of in vitro Human Colonic-Like Macrophages

8.1 Introduction

In this final experimental chapter, the effect of colonic stromal-derived conditioned media on leucocyte immunoglobulin-like receptor expression in cultured colonic-like macrophages was studied by qRT-PCR. In addition to the 8 LILRs included in earlier immunohistochemistry studies, TNF-α mRNA expression was also analysed, as this cytokine was shown in the previous chapter to be down-regulated by colonic stromal- derived conditioned media.

8.2 Materials and Methods

8.2.1 Extraction of RNA from Monocytes and Cultured PBMCs

8.2.1.1 Preliminary Experiments

In order to determine the best method of extraction and expected yield of RNA from cells, a series of preliminary experiments was performed.

PBMCs from two healthy donors were washed and separated via MACS magnetic negative selection [Miltenyi Biotec, Cologne, Germany], as described in section 7.2.3.1. The isolated monocytes were cultured with 25 ng/mL GM-CSF in 96-well flat bottom cell culture plates [Corning Costar, Massachusetts, USA] at 1.8 and 2.1 x 105 cells per well for 3 days, and then re-cultured for a further 3 days without GM-CSF. The supernatants were removed and the cells lysed in Buffer RLT with β-mercaptoethanol, 350 µL per 5 wells [RNeasy Mini Kit, Qiagen, Hilden, Germany]. The lysed cells were transferred to a QIAshredder mini spin column and centrifuged to remove genomic DNA. The flow-through was transferred to an RNeasy Mini spin column, which was washed as per protocol. The extracted RNA was eluted in 30 µL DNase-free RNase-free water.

For comparison, cultured cells were lysed in Trizol [Ambion, Life Technologies, California, USA], 500 µL per 5 wells, and then 100 µL cold chloroform was added. After centrifugation,

171 the upper aqueous phase was removed, mixed with glycogen and isopropanol, and incubated at either room temperature for 30 minutes or at -20°C overnight. Following centrifugation, the supernatant was removed, the pellet washed with ethanol and then dried in a laminar flow hood. The RNA was resuspended in 12 µL DNase-free RNase free water and the RNA quality and yield was measured using a NanoDrop Spectophotometer ND-1000 [Thermo Fischer Scientific, Massachusetts, USA].

Cell cultures were repeated to test RNA extraction when stromal-derived conditioned media was present in the media. Peripheral blood mononuclear cells from the same donors were again separated via MACS, and 1.5 and 1.3 x 105 monocytes per well were cultured in 96-well flat bottom cell culture plates. The media contained either GM-CSF, SCM 1:12.5 or both GM-CSF and SCM 1:12.5 for the first 3 days of the 5-day culture. Cells from five wells were lysed in Buffer RLT or Trizol and pooled. A 23-gauge needle and syringe were used to homogenise the cells. RNA was extracted as per the protocols described above and the RNA was analysed on both the NanoDrop Spectophotometer ND- 1000 and Agilent 2100 Electrophoresis Bioanalyser using an RNA 6000 Nanochip [Agilent, California, USA] to check the RNA integrity and yield.

As the custom-designed qRT-PCR plates required the use of the RNEasy clean-up procedure for RNA extracted with Trizol, the yield and integrity of the RNA before and after this clean-up step was compared. PBMCs were isolated using a Lymphoprep 1.077 g/mL density gradient and washed with D-PBS. RNA was extracted from 1.2 × 106 PBMCs using Trizol as outlined above, followed by the RNEasy clean-up protocol [Qiagen, Hilden, Germany]. The RNA was resuspended in DNase-free RNase-free water [Gibco, New York, USA], Buffer RLT and ethanol were added and the sample was transferred to the RNEasy Mini spin column. This was washed with Buffer RPE and ethanol as per protocol, and the extracted RNA was eluted in 30 µL DNase-free RNase-free water. For comparison RNA was extracted from 1.2 × 106 PBMCs using the RNEasy Mini kit.

When comparing Trizol and RNEasy Mini Kit RNA extraction protocols, the second method was more rapid and resulted in purer RNA isolation (data not shown), therefore this extraction method was used to extract mRNA for subsequent qRT-PCR experiments.

172 8.2.1.2 Culturing Monocytes and Extraction of mRNA for qRT-PCR

The monocyte fractions of PBMCs from four Australian Red Cross Blood Service donors were cultured to determine the effect of stromal-derived conditioned media on LILR expression of these cells. PBMCs were washed and the monocyte fraction isolated by MACS magnetic negative selection. For comparison, 2 x 106 monocytes were lysed in 350 µL Buffer RLT with 10 µL/mL β-mercaptoethanol [RNeasy Mini Kit, Qiagen, Hilden, Germany] and then frozen at -20°C prior to extraction of RNA. The remaining monocytes were cultured in media with GM-CSF (25 ng/mL), SCM (1:100), or both GM-CSF and SCM at 37°C with 5% CO2. After 3 days, media were refreshed without GM-CSF. After a further 2 days, the cells were either collected, or stimulated for 6 or 24 hours with LPS (100 ng/mL), IFN-γ (25 ng/mL) or IL-10 (25 ng/mL) before collection (see section 7.2.3.1). The supernatants were removed, and 600 µL Buffer RLT with 10 µL/mL β-mercaptoethanol was added to each set of wells. Pipetting of the lysis buffer three times was performed to lyse the adherent cells, and the lysed cells were transferred to 1.6 mL tubes for storage at - 80°C prior to extraction of the RNA.

The lysed cells were rapidly thawed in a 37°C waterbath, an equal amount of 70% ethanol was added [Absolute ethanol, Ajax Finechem, Sydney, Australia, diluted with Gibco ultrapure distilled water, New York, USA], and mixed via pipetting. The sample was transferred to an RNeasy Mini Spin Column placed in a 2 mL collection tube and centrifuged for 1 minute at 10 000 g. The flow through was discarded and the column was washed with Buffer RW1. DNase 1 was added to the column to remove genomic DNA. After incubation for 15 minutes at room temperature, the column was again washed with Buffer RW1. The column was then washed three times with Buffer RPE or 80% ethanol and then centrifuged dry for 2 minutes at 10 000 g. The RNA was eluted in 30 µL DNase-free RNase- free water and collected into 1.6 mL tubes. The RNA was concentrated by spinning the samples at room temperature for 12-15 minutes in the SpeedVac SC110A [Thermo Scientific, Massachusetts, USA] prior to analysing the RNA concentration, purity and integrity using the NanoDrop Spectophotometer ND-1000 and the Agilent 2100 Electrophoresis Bioanalyzer.

173 8.2.2 Synthesis of cDNA

The RNA was converted to cDNA using the RT2 HT First Strand Kit [Qiagen, Hilden, Germany] as described in section 6.2.4. RNA, 1620 ng from the monocytes separated by negative selection and up to 1000 ng of RNA from the cell cultures, were used per reaction.

8.2.3 qRT-PCR Array

384-well custom qRT-PCR plates [Qiagen, Hilden, Germany] preloaded with primers to the eight LILRs, TNF-α, GAPDH and β-actin as well as the PCR controls, as previously, were used to study RNA expression of the isolated monocytes and cultured cells. The cDNA was diluted by a factor of 2 for the isolated monocytes and a factor of 1.5 for the cultured cells. The qRT-PCR reaction was performed on a Roche LightCycler 480 [Roche, Basel, Switzerland] and analysed with the RT2 Profiler PCR Array Data Analysis programme version 3.5 [Qiagen, Hilden, Germany] as described in section 6.2.5. Ct values were normalised to the housekeeping gene, GAPDH, and a fold regulation of ≥ 4 and a p-value of < 0.005 were deemed to be significant differences.

174 8.3 Results

8.3.1 PBMCs

PBMCs were obtained from 4 healthy Australian Red Cross Blood Service donors. These were 84.8 ± 2.3(SEM)% lymphocytes, 13.9 ± 2.1% monocytes and 1.3 ± 0.3% granulocytes. Viability was 93.9 ± 0.8% and significantly improved to 97.2 ± 1.0%, p = 0.04, following the MACS negative selection process. The monocyte-enriched fraction as assessed by flow cytometry was 98.2 ± 1.5% CD45+, 96.1 ± 1.5% CD14+ and 2.0 ± 0.3% CD3+.

Figure 8.1 Cell differential of PBMCs prior to MACS negative selection

175 8.3.2 LILR mRNA Expression of Culture Monocytes

8.3.2.1 Effect of Stromal-Derived Conditioned Media on LILR Expression of Unstimulated Cultured Monocytes i. Compared with monocytes differentiated with GM-CSF

Compared with monocytes cultured without SCM, there was a higher expression of LILRB1 (24 to 32 fold) and LILRB2 (47 to 68 fold) in cells cultured in the presence of SCM. LILRs were generally up-regulated by SCM.

Table 8.1 Genes over-expressed in cultured monocytes, in cells differentiated with SCM with or without GM-CSF cf. cells differentiated with GM-CSF without SCM

Unstimulated cells GM-CSF + SCM vs. GM-CSF SCM vs. GM-CSF

value value - - Position Gene Symbol Fold Regulation p Fold Regulation p

1 LILRB1 24.29 0.000001 *** 32.00 0.0002 ***

2 LILRB2 47.18 0.0051 ** 68.36 0.0002 ***

3 LILRB3 10.09 0.0203 * 9.92 0.1049 NS

4 LILRB4 9.90 0.7816 NS 7.93 0.9230 NS

5 LILRA1 28.34 0.0149 * 27.95 0.0230 *

6 LILRA2 3.71 0.2142 NS 3.90 0.2570 NS

7 LILRA3 7.93 0.4593 NS 6.61 0.8615 NS

8 LILRA5 3.03 0.4550 NS 3.12 0.4387 NS

* p < 0.05; **p 0.05 to < 0.005; ***p ≤ 0.005

176 ii. Compared with peripheral blood monocytes a. Comparing MACS-isolated peripheral blood monocytes with PBMCs

Overall, LILR mRNA levels were lower in MACS-isolated peripheral blood monocytes (up to 9-fold) than in unfractionated peripheral blood mononuclear cells. In particular, LILRB1 and LILRA5 mRNA levels were significantly reduced following MACS negative selection.

Table 8.2 Expression of LILRs in MACS-isolated monocytes (n=4) cf. PBMCS in control subjects (n=7)

MACS-monocytes vs. PBMCS Fold Regulation p-value

LILRB1 -9.16 0.0040 ***

LILRB2 -8.57 0.0066 **

LILRB3 -9.35 0.0084 **

LILRB4 -2.66 0.0679 NS

LILRA1 -3.65 0.0051 **

LILRA2 -1.81 0.0901 NS

LILRA3 -1.34 0.3108 NS

LILRA5 -2.84 0.0005 ***

177 b. Comparing cultured monocytes with MACS-isolated peripheral blood monocytes

When these MACS-isolated peripheral blood monocytes were cultured and classically differentiated into macrophages for 5 days, LILR expression was reduced further. Specifically, LILRA5 expression was significantly lower by 26-fold. SCM up-regulated LILR expression, and LILRB1 and LILRB4 levels were significantly higher in colonic-like macrophages compared with MACS-isolated monocytes.

Table 8.3 LILR mRNA expression in cultured monocytes cf. MACS-isolated monocytes, in same 4 subjects

GM-CSF vs. GM-CSF + SCM vs. SCM vs.

MACS-isolated MACS-isolated MACS-isolated monocytes monocytes monocytes

value value value - - - Position Gene Symbol Fold Regulation p Fold Regulation p Fold Regulation p

1 LILRB1 -4.99 0.1238 NS 4.87 0.0030 *** 6.41 0.0023 ***

2 LILRB2 -50.13 0.0944 NS -1.06 0.5442 NS 1.36 0.8876 NS

3 LILRB3 -3.92 0.6167 NS 2.57 0.2297 NS 2.53 0.2955 NS

4 LILRB4 1.07 0.0961 NS 10.63 0.0011 *** 8.52 0.0063 **

5 LILRA1 -121.10 0.0514 NS -4.27 0.1033 NS -4.33 0.1043 NS

6 LILRA2 -7.58 0.9404 NS -2.04 0.1166 NS -1.95 0.1516 NS

7 LILRA3 -11.14 0.1173 NS -1.40 0.2903 NS -1.68 0.0186 *

8 LILRA5 -26.72 0.0006 *** -8.82 0.0006 *** -8.56 0.0006 ***

178 iii. Compared with colonic lamina propria

Classically differentiated macrophages expressed slightly higher LILRB4, LILRA2 and LILRA3 mRNA levels than colonic lamina propria. When SCM was present in cell cultures, there were higher mRNA levels of all eight LILRs, with especially high levels of these same 3 LILRs, compared to levels in colonic lamina propria.

Table 8.4 LILR mRNA expression in cultured monocytes (n=4) cf. colonic lamina propria (n=11)

GM-CSF vs. lamina GM-CSF + SCM vs. SCM vs. lamina propria lamina propria propria

value value value - - - Position Gene Symbol Fold Regulation p Fold Regulation p Fold Regulation p

1 LILRB1 -6.84 0.1399 NS 3.55 0.0015 *** 4.68 0.0001 ***

2 LILRB2 -13.96 0.0061 ** 3.38 0.0004 *** 4.90 0.0000 ***

3 LILRB3 -3.20 0.9761 NS 3.16 0.0016 *** 3.10 0.0105 *

4 LILRB4 2.79 0.0020 *** 27.60 0.0000 *** 22.11 0.0000 ***

5 LILRA1 -3.86 0.1897 NS 7.34 0.0001 *** 7.24 0.0003 ***

6 LILRA2 4.60 0.0006 *** 17.09 0.0000 *** 17.94 0.0000 ***

7 LILRA3 3.60 0.0010 *** 28.56 0.0000 *** 23.81 0.0000 ***

8 LILRA5 1.78 0.0246 * 5.41 0.0001 *** 5.57 0.0000 ***

179 8.3.2.2 Effect of Cell Stimulation on LILR Expression of Cultured Monocytes i. Monocytes differentiated with GM-CSF

Classically differentiated macrophages differentiated with GM-CSF expressed 21-fold increased levels of LILRB2 mRNA after 24 hours of LPS stimulation compared with unstimulated cultured monocytes.

When stimulated with IFN-γ and IL-10, there was a 10- and 13- fold increase respectively in LILRB1 production compared with unstimulated classically differentiated macrophages. There was a trend towards a higher LILRB1 production following LPS stimulation.

Table 8.5 Genes over-expressed in monocytes differentiated with GM-CSF after 24 hours of stimulation with LPS, IFN-γ or IL-10 cf. unstimulated cultured monocytes

GM-CSF LPS stimulated vs. IFN-γ stimulated vs. IL-10 stimulated vs. differentia- unstimulated unstimulated unstimulated tion

value value value - - - Position Gene Symbol Fold Regulation p Fold Regulation p Fold Regulation p

1 LILRB1 15.16 0.0065 ** 10.06 0.0029 *** 13.55 0.0017 ***

2 LILRB2 21.19 0.0029 *** 10.04 0.0157 * 9.02 0.0677 NS

3 LILRB3 16.14 0.0312 * 7.99 0.1251 NS 14.37 0.0574 NS

4 LILRB4 21.00 0.0910 NS 13.88 0.3175 NS 17.18 0.1244 NS

5 LILRA1 5.93 0.0724 NS 3.64 0.4047 NS 8.06 0.0454 *

6 LILRA2 10.56 0.4651 NS 13.29 0.1833 NS 9.88 0.5639 NS

7 LILRA3 16.77 0.0206 * 24.25 0.0216 * 10.93 0.1149 NS

8 LILRA5 2.35 0.8912 NS 4.78 0.1237 NS 3.23 0.3878 NS

180

Figure 8.2 LILR gene expression in cell cultures, unstimulated or stimulated for 24 hours with LPS, IFN-γ or IL-10. The genes shown in the top right of the graph are significantly over-expressed in cell cultures (p ≤ 0.005) with greater than a 4-fold increase in expression, compared with the RNA expression of unstimulated cells cultured with GM- CSF

ii. Monocytes differentiated with SCM, with or without GM-CSF

There were no important differences in LILR expression after stimulation with LPS, IFN-γ or IL-10 in colonic-like macrophages. LILRB4 expression was slightly up-regulated in cells stimulated with IFN-γ.

Table 8.6 Genes over-expressed in monocytes differentiated with GM-CSF and SCM (top) or SCM (bottom) after 24 hours of stimulation with LPS, IFN-γ or IL-10 cf. unstimulated cultured monocytes (following page)

181 GM-CSF + LPS stimulated vs. IFN-γ stimulated vs. IL-10 stimulated vs. SCM unstimulated unstimulated unstimulated differentia- tion

value value value - - - Position Gene Symbol Fold Regulation p Fold Regulation p Fold Regulation p

1 LILRB1 1.94 0.1327 NS 4.06 0.0101 * 1.90 0.1367 NS

2 LILRB2 2.17 0.0701 NS 4.85 0.0219 * 2.67 0.0643 NS

3 LILRB3 1.68 0.2122 NS 3.72 0.0730 NS 1.64 0.0991 NS

4 LILRB4 1.14 0.5518 NS 2.14 0.0089 ** -1.20 0.4646 NS

5 LILRA1 1.19 0.6014 NS 1.38 0.3997 NS -1.61 0.8383 NS

6 LILRA2 1.15 0.5041 NS 2.20 0.0474 * -1.41 0.3913 NS

7 LILRA3 1.47 0.2830 NS 3.29 0.0054 ** -1.10 0.9192 NS

8 LILRA5 -1.85 0.5995 NS 6.67 0.0263 * 2.32 0.1437 NS

SCM LPS stimulated vs. IFN-γ stimulated vs. IL-10 stimulated vs. differentia- unstimulated unstimulated unstimulated tion

1 LILRB1 1.65 0.1725 NS 3.71 0.0135 * 2.30 0.0162 *

2 LILRB2 1.77 0.1841 NS 5.15 0.0388 * 2.16 0.0459 *

3 LILRB3 1.89 0.1592 NS 4.98 0.0555 NS 1.89 0.1575 NS

4 LILRB4 1.84 0.1197 NS 3.87 0.0039 *** 1.76 0.0868 NS

5 LILRA1 1.74 0.2472 NS 2.59 0.0584 NS 1.23 0.3953 NS

6 LILRA2 1.25 0.4267 NS 2.86 0.0115 * -1.09 0.5513 NS

7 LILRA3 1.34 0.3506 NS 6.06 0.0619 NS 1.76 0.1105 NS

8 LILRA5 -1.30 0.9604 NS 6.36 0.0114 * 2.78 0.0687 NS

182

Figure 8.3 LILR expression following stimulation (6 or 24 hs with LPS, IFN-γ or IL-10)

183 In figure 8.3, RNA fold changes following LPS, IFN-γ and IL-10 stimulation are plotted for each of the LILRs. All 3 stimulants resulted in higher LILR expression in classically differentiated macrophages, mostly requiring 24 hours of stimulation. There was earlier up-regulation of LILRB3 and LILRA3 following 6 hours of IFN-γ stimulation. Up-regulation of the LILRs was suppressed overall in colonic-like cells.

184 8.3.3 TNF-α mRNA Expression of Cultured Monocytes

8.3.3.1 Effect of Stromal-Derived Conditioned Media on TNF-α Expression of Unstimulated Cultured Monocytes

TNF-α mRNA levels were 4-fold higher in cells cultured in the presence of SCM compared to monocytes differentiated with GM-CSF alone, although this did not reach statistical significance.

Table 8.7 TNF-α expression in unstimulated cells

Fold Regulation p-value

GM-CSF + SCM vs. GM-CSF 4.09 0.0695 NS

SCM vs. GM-CSF 4.70 0.0344 *

There was no difference in TNF-α mRNA expression levels between MACS-isolated monocytes and unfractionated PBMCs (fold regulation -1.18, p-value = 0.544 NS). Cultured monocytes showed no significant difference in TNF-α mRNA levels compared with their starting population of MACS-isolated monocytes. Compared with lamina propria, however, there were significantly higher TNF-α mRNA levels in monocytes differentiated with SCM.

Table 8.8 TNF-α expression in unstimulated cells differentiated with GM-CSF, SCM or both cf. colonic lamina propria [LP] or MACS-isolated monocytes

GM-CSF GM-CSF + SCM SCM

TNF-α levels in

cultured cells value value value - - - Fold Regulation p Fold Regulation p Fold Regulation p vs. monocytes -1.98 0.6471 NS 2.06 0.0515 NS 2.37 0.0194 * vs. LP 2.82 0.0571 NS 11.52 0.0000 *** 13.23 0.0000 ***

185 8.3.3.2 Effect of Cell Stimulation on TNF-α Expression of Cultured Monocytes

TNF-α mRNA levels paralleled the cytokine levels, with a 33-fold over-expression of TNF-α mRNA following 6 hours of LPS stimulation, which decreased to 4-fold above baseline by 24 hours of LPS stimulation in classically-differentiated macrophages. The presence of SCM abrogated this over-expression, with a 3-fold increase in TNF-α mRNA expression levels after 6 hours of LPS stimulation and a return to baseline by 24 hours of LPS stimulation.

Figure 8.4 TNF-α expression following 6 and 24 hour stimulation with LPS

186 Following IFN-γ stimulation, there was a trend towards increased TNF-α mRNA levels by 2- to 5- fold in cultured cells, particularly in the presence of SCM.

IL-10 stimulation suppressed TNF-α mRNA levels by approximately 2-fold, especially with more prolonged stimulation in the presence of SCM.

Table 8.9 TNF-α expression following 6 and 24 hour stimulation

GM-CSF: GM-CSF + SCM: SCM:

Stimulated vs. Stimulated vs. Stimulated vs. Unstimulated Unstimulated Unstimulated

value value value - - - Stimulation Fold Regulation p Fold Regulation p Fold Regulation p

6 hrs LPS 33.30 0.0431 * 3.43 0.0088 ** 3.28 0.0142 *

24 hrs LPS 4.06 0.1750 NS 1.45 0.2977 NS 1.24 0.4415 NS

6 hrs IFN-γ 3.16 0.2131 NS 5.00 0.0261 * 3.62 0.0793 NS

24 hrs IFN-γ 2.32 0.4022 NS 3.42 0.0542 NS 3.57 0.0356 *

6 hrs IL-10 -1.47 0.4281 NS -1.96 0.0391 * -1.56 0.3156 NS

24 hrs IL-10 -1.75 0.5350 NS -1.43 0.3627 NS -2.17 0.0010 ***

187 8.4 Discussion i. Leucocyte Immunoglobulin-like Receptor Expression of Cultured Monocytes

Multiple leucocyte immunoglobulin-like receptors were expressed in the colon, predominantly on lamina propria macrophages, as demonstrated in Chapter 4. Using the colonic intestinal-like macrophage model developed in the previous chapter, expression of leucocyte immunoglobulin-like receptor mRNA in these cells was studied and compared with peripheral blood monocytes, colonic lamina propria and classically-differentiated macrophages in cell culture.

To account for biological variability, peripheral blood monocytes from 4 healthy donors were used for these cell cultures. Technical replicates and PCR controls were included on the customised qRT-PCR plates to ensure qRT-PCR quality, and the data were normalised to the housekeeping gene, GAPDH.

Monocytes differentiated in the presence of colonic stromal-derived conditioned media expressed an excess of LILRB1 and LILRB2 mRNA. There was greater than a 24–fold increase of these inhibitory LILRs, compared to monocytes differentiated with GM-CSF alone. LILRB1 mRNA expression correlated with protein expression in earlier immunohistochemistry experiments (see section 4.3.3), where it was shown to be highly expressed in colonic lamina propria macrophages. LILRA1 mRNA was also increased 28– fold when colonic stromal factors were present, although this did not reach statistical significance. LILRA5, the other LILR that was highly expressed in colonic macrophages in tissue sections, was increased 3–fold in these cell cultures although this, too, did not reach statistical significance.

There are differences in LILR expression between lymphocytes and monocytes, the former cell type accounting for the majority (84.8 ± 2.3%) of unfractionated PBMCs, whilst LILRs are predominantly expressed on the latter. LILR mRNA expression, however, was unexpectedly lower in MACS-isolated peripheral blood monocytes than in unfractionated peripheral blood mononuclear cells. The negative magnetic selection process may have altered LILR expression, either through cleavage of surface receptors, alteration of the conditions required for protein translation or, potentially, down-regulation of the expression of necessary but yet-unidentified co-molecules.

188 Following culture of MACS-isolated monocytes with GM-CSF, there was a marked decrease in LILRA5 mRNA in these classically-differentiated macrophages, in concordance with a previously demonstrated modulation of LILRA5 expression in vitro149. Additionally, the other LILRs were also down-regulated with culture. Colonic stromal-derived conditioned media led to a relative up-regulation of all 8 LILRs and, compared with levels in the pre- cultured blood monocytes, there was significant up-regulation of LILRB1 and LILRB4 in the colonic-like macrophages. LILRB1 has wide cellular expression, but studies specifically of in vitro differentiated peripheral blood monocytes are limited. There are two studies of peripheral blood mononuclear cells differentiated into dendritic cells with 50 ng/mL GM- CSF and 15 or 50 ng/mL IL-4 that assessed LILRB1 expression by flow cytometric analysis. One group analysed LILRB1, LILRB2 and LILRB4 and showed selective up-regulation of LILRB1 expression compared with PBMCs140. The other group showed low LILRB1 expression in circulating dendritic cells in subjects with systemic lupus erythematosis compared with controls, but almost universal expression of LILRB1 in cultured dendritic cells from both groups179. However, the colonic-like cells in this present study, whilst similarly showing up-regulation of LILRB1, were not differentiated to dendritic cells with IL-4; the colonic stromal-derived conditioned media used in these cell cultures contained negligible (0.1 pg/mL) IL-4.

This in vitro model of colonic-like macrophages showed significant up-regulation of all LILRs (3– to 28– fold) when compared with colonic lamina propria. The converse was seen when stromal-derived conditioned media was absent from the cell cultures; in classically differentiated macrophages there was only a minor (< 5–fold) up-regulation of LILRB4, LILRA2 and LILRA3 and a down-regulation of LILRB2 compared with lamina propria. Although the lower proportion of cells of monocytic lineage in the colonic mucosa may partly account for the relative differences in expression, factors in the colonic stromal- derived conditioned media are likely to up-regulate LILR mRNA expression. Colonic stromal-derived conditioned media contains a plethora of cytokines, chemokines and growth factors (see section 7.3.2.4) and, whilst the absolute concentration of these can be determined, the required concentration of these individual molecules and the synergism between them to effect gene transcription and translation are unknown. Furthermore, the equivalent amount of stromal-derived conditioned media that replicates the colonic environment in situ has not been determined.

Following cell stimulation, LILR mRNA expression of classically differentiated macrophages was responsive to stimulation with LPS, IFN-γ and IL-10. In particular,

189 LILRB1 was increased 10– to 15 – fold in response to all three stimuli, and LILRB2 was increased 21– fold following LPS-stimulation. IL-10 stimulation has been shown to induce LILRB2 and LILRB4 expression in monocytes128, 141 and human monocytic leukemic cell line THP-1 cells634. Expression of inhibitory LILRs in response to IL-10 stimulation has not previously been studied in macrophages in vitro. Although LILRB1 was inducible in classically differentiated macrophages, interestingly, this occurred with both activating and inhibitory stimuli and perhaps it is a non-specific cell response; this makes the changes difficult to interpret.

Colonic stromal-derived conditioned media suppressed the ability of in vitro macrophages to further up-regulate LILRB1 and LILRB2 expression in response to cell stimulation. Although there was a trend towards higher LILR expression after stimulation with IFN-γ in these colonic-like macrophages, the degree of up-regulation was in general less than half that seen in classically differentiated macrophages. ii. TNF-α Expression of Cultured Monocytes

TNF-α cytokine levels were low in cell culture supernatants from unstimulated classical macrophages, with marginally higher levels in the presence of colonic stromal-derived conditioned media, as demonstrated in the previous chapter. These cytokine levels were reflected in the corresponding mRNA levels from the cultured colonic-like macrophages, which showed a 4–fold increase in mRNA compared with classical macrophages.

Additionally, in response to cell stimulation, changes in TNF-α mRNA levels paralleled the changes seen in TNF-α cytokine levels. LPS stimulation of monocytes that were classically differentiated to macrophages with GM-CSF resulted in high early levels of TNF-α mRNA at 6 hours, which decreased with longer exposure to LPS. This induction of TNF-α mRNA expression following LPS stimulation was suppressed by colonic stromal-derived conditioned media, correlating with down-regulation of TNF-α cytokine levels in the supernatants.

In response to the activating stimulant, IFN-γ, up to 5–fold higher TNF-α mRNA levels were seen in cultured cells, in the presence or absence of colonic stromal-derived conditioned media and this suggests that these cultured cells may be relatively unresponsive.

190 Stimulation with the inhibitory regulator, IL-10, led to lower TNF-α mRNA and cytokine levels in the presence of colonic stromal-derived conditioned media. TNF-α was particularly suppressed following a longer stimulation of these colonic-like macrophages, and similarly, in vivo, IL-10 helps to maintain homeostasis by providing inhibitory signals to lamina propria macrophages623.

8.5 Summary

Intestinal-like macrophages generated from peripheral blood monocytes with colonic stromal-derived conditioned media as a differentiating factor had generally higher LILR expression than classically differentiated in vitro macrophages, with particularly high levels of the inhibitory LILRs, LILRB1, LILRB2 and LILRB4. Whilst stimulation of classically differentiated macrophages induced LILRB1, LILRB2 and TNF-α mRNA levels, colonic-like macrophages were unresponsive to these stimulatory factors, and this correlated with the effect of stromal-derived conditioned media on TNF-α cytokine levels. Although it would have been interesting to correlate LILR surface expression with mRNA levels in this in vitro model of colonic-like macrophages, this was limited by the availability of the non- commercial LILR antibodies.

Of particular importance is the agreement between colonic lamina propria macrophages and this colonic macrophage-like cell culture model of LILRB1 over-expression. Its high expression by colonic macrophages may serve as an important regulator of the intestinal immune response both in health and in chronic inflammation. A model of chronic inflammatory bowel disease-type colonic macrophages is vital in order to confirm the differential expression in LILRB1 expression in inflamed colon, and to determine what factors control this alteration in chronic inflammatory bowel disease.

191 9 Discussion and Future Directions

The human gastrointestinal tract is a crucial organ, which functions not only to digest food but, as the body’s largest lymphoid organ interfacing the luminal contents, it is central in maintaining immunological protection against microbial and other antigens. The normal gut immune system maintains a delicate balance between pro- and anti- inflammatory responses. Homeostasis is achieved through multiple mechanisms, including a robust epithelial barrier, the symbiotic relationship between luminal commensal microorganisms and innate immune cells lining the gut, and the immune contributions of the various haematopoietic and non-haematopoietic cells within the mucosa. Complex interactions between this intricate network of mucosal cells occur via mediators such as cytokines, chemokines and growth factors. Moreover, in addition to providing the supportive structure of the lamina propria, the gastrointestinal stroma actively influences the cells that reside within it.

In chronic inflammatory bowel disease dysregulation of this finely tuned balance results in the chronic inflammation and subsequent tissue damage seen in Crohn’s disease and ulcerative colitis. There are alterations at numerous levels, including increased permeability of the epithelial barrier allowing entry of pathogens into the mucosa, dysbiosis consisting of reduction in bacterial and increases in fungal and viral biodiversity and increases in subdominant commensal organisms, and changes in cells and cell mediators of both the innate and acquired immune system. The ~200 identified susceptibility genes encompass a broad spectrum of cell types, receptors and signalling molecules, although associations are strongest with the innate immune system.

Within this context are the potential roles of the leucocyte immunoglobulin-like receptors in the gastrointestinal tract, both in the healthy immune tolerant state and in inflammation. Their predicted involvement in immune regulation is based upon their structure, cellular distribution and signalling mechanisms. These activating and inhibitory receptors are expressed predominantly on cells of monocytic-macrophage lineage, which account for a large proportion of the cells of the lamina propria. They have been described in a range of other regulatory, inflammatory and autoimmune conditions and their known ligands apart from non-MHC class 1 are mainly microbial in origin, highlighting their likely importance in the intestinal environment.

192 The hypotheses relating to the presence of leucocyte immunoglobulin-like receptors in the human colon examined in this thesis are that: 1. immune tolerance in the colon is achieved through leucocyte immunoglobulin-like receptors; 2. imbalance in their number or function contributes to the pathogenesis of chronic inflammatory bowel disease; and 3. colonic stromal factors influence monocyte differentiation and can be used in vitro to alter the phenotype of monocytes to become immune-tolerant down-regulated intestinal- like macrophages, thus providing a model to enable further characterisation of these receptors in the colon.

With the paucity of information available about their presence and distribution in the gastrointestinal tract, initial studies in Chapters 4, 5 and 6 focused on leucocyte immunoglobulin-like receptor expression in colonic tissue and in isolated lamina propria mononuclear cells, at both transcriptional and translational levels.

Using a panel of mouse anti-human monoclonal antibodies to eight of the 11 leucocyte immunoglobulin-like receptors, these receptors are indeed present in the colonic lamina propria. Interestingly, one inhibitory (LILRB1) and one activating (LILRA5) receptor were universally and highly expressed in the colon, with median cell counts of 104.5 ± 15.8 and 132.6 ± 18.4 per high power microscopy field, respectively. Both LILRB1 and LILRA5 were expressed on CD68+ colonic macrophages in the lamina propria, as detected by immunohistochemistry serial sections and confirmed by immunofluorescence double staining. These leucocyte immunoglobulin-like receptors were also expressed on T and B lymphocytes. LILRA5 had a broader distribution that included endothelial cells. Smaller numbers of submucosa cells expressed these leucocyte immunoglobulin-like receptors. Whilst the strong, unanimous presence of an activating and an inhibitory leucocyte immunoglobulin-like receptor was not initially predicted, this highlights the important balance between active controlled inflammation and the suppressive immune responses that regulate immune tolerance in the intestine.

Of note, co-expression of LILRB1 on LILRA5+ lamina propria cells was also demonstrated. LILRB1 has 4 extracellular C2-type immunoglobulin-like domains that can bind MHC class 1 antigens and homologues and the calcium binding proteins, S100A8 and S100A9. In contrast, LILRA5 has 2 extracellular C2-type domains that have a different distribution of N-linked glycosylation sites, and it does not bind any of these ligands. Therefore, this co-

193 expressed pair of receptors may control the degree of cell activation or inhibition, depending on local ligands in the colon. This could be assessed further by the use of known ligands or neutralising antibodies to activate or inhibit either of these receptors on lamina propria macrophages, or through the use of LILRB1 or LILRA5 mRNA knockdown macrophage models.

All of the other leucocyte immunoglobulin-like receptors except LILRB4 were selectively present in non-inflamed colon, with LILRA1 and LILRA3 expressed on macrophages and mast cells, respectively, in the majority of subjects. This differential expression suggests that environmental, genetic or clinical factors may influence expression of these receptors. Whilst subjects provided demographic, smoking and medical details (see Chapter 3), this study was underpowered for any meaningful associations to be drawn, and a larger cohort would provide the opportunity to further explore these disparities and the circumstances that induce or inhibit leucocyte immunoglobulin-like receptor expression.

In inflammatory bowel disease, whilst LILRB1 was expressed in all subjects, both the mRNA expression level and density of receptors were significantly elevated, approximately 3 times higher in inflamed colonic lamina propria. At both a transcriptional and translational level, LILRB4 was also increased in inflammatory bowel disease, although this difference needs to be examined further as this receptor was only present in a few subjects with colitis. The role of these two inhibitory receptors in inflammatory bowel disease has not been determined, although they may be secondarily up-regulated to limit the degree of inflammation. Alternatively, their presumed inhibitory functions may be subverted, allowing these receptors to contribute to the active inflammation. As LILRB1 and LILRB4 are predominantly expressed on macrophages, and it has been established that newly-recruited CD14+ blood monocytes account for the inflammatory tissue macrophages in colitis, it would be important to establish whether they are increased due to infiltration of this fresh population of cells. Interestingly, higher LILRB1 and LILRB4 mRNA levels were not seen in circulating peripheral blood mononuclear cells in parallel with the colonic inflammation.

Activating, rather than inhibitory, leucocyte immunoglobulin-like receptors were predicted to be abundant in inflammatory bowel disease. LILRA5 was ubiquitously expressed in the colon and is therefore likely to have a significant immunomodulatory role in both health and inflammation, thus it was the focus of these three chapters. There were increases at both the mRNA (5-fold) and receptor-expression levels (1.5 times more

194 abundant on positive cell enumeration) in subjects with colitis. Wide variations between individuals may account for the statistical non-significance in this study, including differences in the degree of inflammation and the current and recent medical treatments. These could more readily be pursued through studies of colonic tissue obtained from endoscopic biopsies which, whilst not providing as large or representative specimens as from a colectomy, would enable a spectrum of disease activity and stages, the effect of various immunosuppressive and immunomodulatory medications, and differences between Crohn's disease and ulcerative colitis to be assessed. Furthermore, distinctions between the expression of leucocyte immunoglobulin-like receptors in healthy and inflamed ileum, where there is a significantly decreased microbial presence compared with the colon, could be determined.

Thus, within the immunoregulatory environment of the colon, both inhibitory and activating leucocyte immunoglobulin-like receptors are present and are likely to be involved in maintaining homeostasis, and an imbalance occurs in chronic inflammatory bowel disease. The predominance of these on lamina propria macrophages highlights the importance of the innate immune system in the colon, both in health and in inflammation.

In Chapters 7 and 8, a cell culture model of intestinal-like macrophages was developed using colonic stromal-derived conditioned media to differentiate peripheral blood monocytes in vitro. Colonic stroma has important immunoregulatory influences on the multiple cell types that reside in the lamina propria. The conditioned media contained an assortment of cytokines, chemokines, growth factors and proteins. Lipopolysaccharide was also present, as was anticipated from the abundance of commensal microorganisms in the colon. This is in contrast to other in vitro macrophage models where typically a colony stimulating factor and a variable number of other cytokines are used in the cell differentiation process. The modulatory effects of the stromal-derived conditioned media is likely due to the concerted effects of the plethora of mediators, and gaining an understanding of how the most crucial factors interact to influence cell differentiation will provide insights into the unique microenvironment of the colon. For example, pursuing these cultures in the presence of anti-TGF-β antibodies, or with neutralisation of LPS, would help to demonstrate the relative importance of these two critical mediators in the colonic milieu.

This cell culture model shares similarities with the colonic macrophages it is replicating. Similar to the down-regulated phenotype of colonic macrophages, cytokine production

195 was low in the intestinal-like macrophages. In particular, these cells produced minimal amounts of TNF-α, IL-1, IL-2, IL-4, IFN-γ, TGF-β and IL-10. Of importance, colonic stromal- derived conditioned media suppressed LPS-induced TNF-α production in a dose and time- dependent manner, and this was confirmed by qRT-PCR. It also inhibited the cells’ ability to produce TGF-β in response to stimulation. Comparable to colonic macrophages, LILRB1 was up-regulated on these cells in vitro and, to a lesser degree LILRB2 was also up- regulated. Although leucocyte immunoglobulin-like receptor mRNA levels were generally higher in the presence of colonic stromal-derived conditioned media when compared with classically-differentiated macrophages, the elevated LILRA5 on colonic macrophages was not paralleled to the same extent in these colonic-like cells. The up-regulation mainly of inhibitory receptors in this colonic macrophage-like model provides the opportunity to define their function and their roles in immunoregulation in the colon.

In colonic lamina propria mononuclear cells and in this novel cell culture model, phagocytosis was not examined. This key function of intestinal macrophages that protects the body from foreign antigens and the impact of leucocyte immunoglobulin-like receptors on this would be able to be explored using this colonic macrophage-like model. Additionally, the immunological changes in chronic inflammatory bowel disease were not determined. Generation of stromal-derived conditioned media from both subjects with Crohn’s disease and ulcerative colitis would be critical in assessing the key changes in macrophages in these conditions, particularly in regards to cytokine production, phagocytosis and leucocyte immunoglobulin-like receptor function.

From the work presented in this thesis, the role of innate immune receptors on intestinal macrophages is highlighted. What is known about their ligands is scanty, but they can bind microbial antigens. The gastrointestinal tract is in constant contact with a diverse and abundant population of microorganisms, and these receptors are likely to be central in responding to both the commensals and pathogenic microbes to induce immunological tolerance and control inflammation. The imbalance in the microbial environment in chronic inflammatory bowel disease may be a reason for the imbalance in leucocyte immunoglobulin-like receptors seen in these conditions.

Interactions of the leucocyte immunoglobulin-like receptors with colonic commensal microorganisms and pathogens and the subsequent macrophage responses could be assessed in vitro in colonic lamina propria macrophages and colonic-like and classically- differentiated macrophages. The in vivo and in vitro effects of antibiotics and current anti-

196 inflammatory, immunoregulatory or immunosuppressive medications used in chronic inflammatory bowel disease on leucocyte immunoglobulin-like receptors in the colon is yet to be explored. In future, there may be a potential role for targeted manipulation of these receptors by microbial or microbial-related components to correct the effects of the dysbiosis in chronic inflammatory bowel disease.

197 10 Appendices

10.1 Structured Medical Questionnaire

Leucocyte Immunoglobulin-Like Receptors in the GUT Study

Patient Code:

1st 2 letters of 1st 2 letters of Patient number given name surname

DOB:

Day Month Year

Sex: Female Male

Hospital:

Group: Control Colitis

PREOPERATIVE INTERVIEW - DATE: ______

198 Country of birth: o Australia o Overseas (Please specify: )

Aboriginal or Torres Strait Islander: o No o Yes

First language: o English o Other (Please specify: )

Mother’s country of birth: o Australia o Overseas (Please specify: )

Father’s country of birth: o Australia o Overseas (Please specify: )

Highest level of education achieved:

< 6 years 7-10 years 11-12 years Certificate or Degree Postgraduate at school at school at school diploma diploma, masters or doctorate

Smoking history: o Never smoked 7 or more cigarettes per week o Ex-smoker: formerly smoked 7 or more cigarettes per week o Year started: o Year stopped: o Currently smoking 7 or more cigarettes per week o Year started:

199 MEDICAL HISTORY Have you ever had any of the following medical illnesses?

No Yes Details: Angina or a myocardial infarction (heart attack)? Rheumatic fever? Valvular heart disease? Pericarditis? Hypertension? Asthma? Chronic bronchitis or emphysema? Tuberculosis? Indigestion? Heartburn? Reflux? Peptic ulcer disease? Crohn’s or ulcerative colitis? Fistulae? Irritable bowel syndrome? Gallstones? Primary sclerosing cholangitis? Viral hepatitis? Type: Any other liver disease? Migraines? Seizures / fits / epilepsy? Strokes, mini-strokes or TIAs? Multiple sclerosis or optic neuritis? Iritis, uveitis or episcleritis? Thyroid disease? Diabetes mellitus? Age at onset: Requiring insulin: High cholesterol? Iron, vitamin B12 or folate deficiency? Anaemia? Osteoporosis or osteopaenia? Osteoarthritis?

200 Rheumatoid arthritis? Ankylosing spondylitis? Psoriasis? Eczema or hay fever? Erythema nodosum? Pyoderma gangrenosum? Thromboembolic events (DVTs, pulmonary emboli)? Kidney stones? Chronic renal failure? Gout? Cancer? Type:

SURGICAL HISTORY Have you had an appendicectomy? o No o Yes If yes, what year? Have you had any other operations for which you required a general or spinal anaesthetic (not including caesarean sections)?

Operation Reason Hospital Year

Are there any other medical problems that you think are important that you haven’t been asked about?

201 FAMILY HISTORY Please give details of any medical conditions in your first-degree relatives. Mother: Father: Siblings: Children: Are there any other gastrointestinal diseases or cancers in your extended family (e.g. grandparents, aunts, uncles, cousins)?

MEDICATIONS What medications or drugs are you currently taking, or have taken in the past 30 days? Please include over-the-counter medications that are sold without a prescription, as well as prescription medications. Name of medication Dose Indication Approx. duration (months, years)

In the past 30 days, have you taken any aspirin (including Astrix, Cardiprin, Cartia, Disprin, Solprin)? o No o Yes: Please provide details (dose, quantity, duration)

In the past 30 days, have you taken any NSAIDs / COX2 inhibitors? (e.g. Brufen, Feldene, Indocid, Mobic, Mobilis, Naprosyn, Nurofen, Orudis, Voltaren, Celebrex) o No o Yes: Please provide details (drug, dose, frequency, duration)

202 Have you ever taken any of the following medications? 1. 5-ASA compounds? (e.g. Salazopyrin, Pyralin, Mesasal, Salofalk, Pentasa, Dipentum, Colazide) o No o Yes: Please provide details (which one, route, start date, stop date)

2. Steroids? (e.g. Prednisone, Prednisolone, Panafcort, Predsol, Colifoam, Entocort) o No o Yes: Please provide details (which one, route, start date, stop date)

3. Immunomodulatory agents? (e.g. Azathioprine, Imuran, 6-Mercaptopurine, Puri-Nethol, Methotrexate, Methoblastine, Cyclosporin, Neoral, Infliximab, Remicade) o No o Yes: Please provide details (which ones, route, start and stop date)

4. Have you ever participated in any clinical trials of gastrointestinal medications? o No o Yes: Please provide details (study / drug, trial dates)

5. Have you ever taken any food supplements, probiotics or been on parenteral nutrition? o No o Yes: Please provide details (what and when)

6. Have you ever been on the oral contraceptive pill? o No o Yes: Please provide details (start and stop date)

203 CLINICAL QUESTIONS What is the indication for the colonic resection? (Tick one) o Inflammatory bowel disease – active disease o Inflammatory bowel disease – high risk of colorectal cancer o Colorectal cancer o Severe diverticular disease o Other, please specify:

How would you describe your general health or wellbeing (choose 1 of the following)?

Very well Slightly below Poor Very poor Terrible par

Do you have any of the following? o Arthritis o Erythema nodosum o Pyoderma gangrenosum o Uveitis or iritis

How many times do you open your bowels during the day?

How many times do you open your bowels during the night?

Do you experience an urgency to defaecate?

None Hurry to the Need to go Incontinence toilet immediately

Is there blood in your motions?

None Trace of blood Occasionally Usually frank frank blood blood

204 Do you suffer from abdominal pain?

None Mild Moderate Severe

In general, would you say your health is:

Excellent Very good Good Fair Poor

Compared to one year ago, how would you rate your health in general now?

Much better Somewhat About the same Somewhat Much worse now than 1 year better now than as 1 year ago worse now than now than 1 year ago 1 year ago 1 year ago ago

Weight in kilograms: Height in centimetres:

BMI: .

THANK YOU FOR COMPLETING THESE QUESTIONS

205 10.2 Additional Cytokine Graphs from the Cell Culture Model

Additional graphs of (a) IL-2 and IL-4, (b) IL-10, (c) IL-8, (d) IFN-γ and (e) GM-CSF levels in cultured cells following LPS, IFN-γ and IL-10 stimulation (section 7.3.3.3). a) IL-2 and IL-4 following LPS stimulation

206 IL-2 and IL-4 following IFN-γ stimulation

207 IL-2 and IL-4 following IL-10 stimulation

208 b) IL-10 following LPS and IFN-γ stimulation

209 e) IL-8 following LPS and IFN-γ stimulation

210 IL-8 following IL-10 stimulation

211 d) IFN-γ following LPS and IL-10 stimulation

212 e) GM-CSF following LPS and IFN-γ stimulation

213 GM-CSF following IL-10 stimulation

214 11 References

1. Crohn BB, Ginzburg L Fau - Oppenheimer GD, Oppenheimer GD. Regional ileitis: a pathologic and clinical entity. 1932. 2. White WH. On simple ulcerative colitis and other rare intestinal ulcers. Guy's Hosp. Rep. 1888;45:131-162. 3. Kirsner JB. The Historical Basis of the Idiopathic Inflammatory Bowel Diseases. Inflammatory Bowel Diseases 1995;1:2-26. 4. Lanier LL. Face off--the interplay between activating and inhibitory immune receptors. Current Opinion in Immunology 2001;13:326-31. 5. Wende H, Colonna M, Ziegler A, et al. Organization of the leukocyte receptor cluster (LRC) on human chromosome 19q13.4. Mammalian Genome 1999;10:154-60. 6. Wende H, Volz A, Ziegler A. Extensive gene duplications and a large inversion characterize the human leukocyte receptor cluster.[Erratum appears in Immunogenetics 2001;52(3-4):308]. Immunogenetics 2000;51:703-13. 7. Wilson MJ, Torkar M, Haude A, et al. Plasticity in the organization and sequences of human KIR/ILT gene families. Proceedings of the National Academy of Sciences of the United States of America 2000;97:4778-83. 8. Tsuchiya N, Kyogoku C, Miyashita R, et al. Diversity of human immune system multigene families and its implication in the genetic background of rheumatic diseases. Current Medicinal Chemistry 2007;14:431-9. 9. Barrow AD, Trowsdale J. The extended human leukocyte receptor complex: diverse ways of modulating immune responses. Immunological Reviews 2008;224:98-123. 10. Canavez F, Young NT, Guethlein LA, et al. Comparison of chimpanzee and human leukocyte Ig-like receptor genes reveals framework and rapidly evolving genes. Journal of Immunology 2001;167:5786-94. 11. Nikolaidis N, Klein J, Nei M. Origin and evolution of the Ig-like domains present in mammalian leukocyte receptors: insights from chicken, frog, and fish homologues. Immunogenetics 2005;57:151-7. 12. Straub C, Neulen ML, Sperling B, et al. Chicken NK cell receptors. Developmental & Comparative Immunology 2013;41:324-33. 13. Viertlboeck BC, Hanczaruk MA, Amann B, et al. Chicken immunoregulatory Ig-like receptor families: an overview and expression details on ggTREM-A1. Developmental & Comparative Immunology 2013;41:403-12. 14. Kubagawa H, Burrows PD, Cooper MD. A novel pair of immunoglobulin-like receptors expressed by B cells and myeloid cells. Proceedings of the National Academy of Sciences of the United States of America 1997;94:5261-6. 15. Castells MC, Wu X, Arm JP, et al. Cloning of the gp49B gene of the immunoglobulin superfamily and demonstration that one of its two products is an early-expressed mast cell surface protein originally described as gp49. Journal of Biological Chemistry 1994;269:8393-401. 16. Katz HR, Vivier E, Castells MC, et al. Mouse mast cell gp49B1 contains two immunoreceptor tyrosine-based inhibition motifs and suppresses mast cell activation when coligated with the high-affinity Fc receptor for IgE.

215 Proceedings of the National Academy of Sciences of the United States of America 1996;93:10809-14. 17. Samaridis J, Colonna M. Cloning of novel immunoglobulin superfamily receptors expressed on human myeloid and lymphoid cells: structural evidence for new stimulatory and inhibitory pathways. European Journal of Immunology 1997;27:660-5. 18. Cella M, Dohring C, Samaridis J, et al. A novel inhibitory receptor (ILT3) expressed on monocytes, macrophages, and dendritic cells involved in antigen processing. Journal of Experimental Medicine 1997;185:1743-51. 19. Colonna M, Navarro F, Bellon T, et al. A common inhibitory receptor for major histocompatibility complex class I molecules on human lymphoid and myelomonocytic cells.[see comment]. Journal of Experimental Medicine 1997;186:1809-18. 20. Cosman D, Fanger N, Borges L, et al. A novel immunoglobulin superfamily receptor for cellular and viral MHC class I molecules. Immunity 1997;7:273- 82. 21. Borges L, Hsu ML, Fanger N, et al. A family of human lymphoid and myeloid Ig-like receptors, some of which bind to MHC class I molecules. Journal of Immunology 1997;159:5192-6. 22. Arm JP, Nwankwo C, Austen KF. Molecular identification of a novel family of human Ig superfamily members that possess immunoreceptor tyrosine- based inhibition motifs and homology to the mouse gp49B1 inhibitory receptor. Journal of Immunology 1997;159:2342-9. 23. Wagtmann N, Rojo S, Eichler E, et al. A new human gene complex encoding the killer cell inhibitory receptors and related monocyte/macrophage receptors. Current Biology 1997;7:615-8. 24. Torkar M, Norgate Z, Colonna M, et al. Isotypic variation of novel immunoglobulin-like transcript/killer cell inhibitory receptor loci in the leukocyte receptor complex. European Journal of Immunology 1998;28:3959-67. 25. Young NT, Canavez F, Uhrberg M, et al. Conserved organization of the ILT/LIR gene family within the polymorphic human leukocyte receptor complex. Immunogenetics 2001;53:270-8. 26. Rissoan M-C, Duhen T, Bridon J-M, et al. Subtractive hybridization reveals the expression of immunoglobulin-like transcript 7, Eph-B1, granzyme B, and 3 novel transcripts in human plasmacytoid dendritic cells. Blood 2002;100:3295-303. 27. Engel P, Tedder TF. New CD from the B cell section of the Fifth International Workshop on Human Leukocyte Differentiation Antigens. Leukemia & Lymphoma 1994;13 Suppl 1:61-4. 28. Banham AH, Colonna M, Cella M, et al. Identification of the CD85 antigen as ILT2, an inhibitory MHC class I receptor of the immunoglobulin superfamily. Journal of Leukocyte Biology 1999;65:841-5. 29. Andre P, Biassoni R, Colonna M, et al. New nomenclature for MHC receptors. Nature Immunology 2001;2:661. 30. ncbi.nlm.nih.gov/gene. 2014. 31. Chapman TL, Heikeman AP, Bjorkman PJ. The inhibitory receptor LIR-1 uses a common binding interaction to recognize class I MHC molecules and the viral homolog UL18. Immunity 1999;11:603-13.

216 32. Allen RL, Raine T, Haude A, et al. Leukocyte receptor complex-encoded immunomodulatory receptors show differing specificity for alternative HLA-B27 structures. Journal of Immunology 2001;167:5543-7. 33. Jones DC, Kosmoliaptsis V, Apps R, et al. HLA class I allelic sequence and conformation regulate leukocyte Ig-like receptor binding. Journal of Immunology 2011;186:2990-7. 34. Vitale M, Castriconi R, Parolini S, et al. The leukocyte Ig-like receptor (LIR)- 1 for the cytomegalovirus UL18 protein displays a broad specificity for different HLA class I alleles: analysis of LIR-1 + NK cell clones. International Immunology 1999;11:29-35. 35. Fanger NA, Cosman D, Peterson L, et al. The MHC class I binding proteins LIR-1 and LIR-2 inhibit Fc receptor-mediated signaling in monocytes. European Journal of Immunology 1998;28:3423-34. 36. Fanger NA, Borges L, Cosman D. The leukocyte immunoglobulin-like receptors (LIRs): a new family of immune regulators. Journal of Leukocyte Biology 1999;66:231-6. 37. Shiroishi M, Kajikawa M, Kuroki K, et al. Crystal structure of the human monocyte-activating receptor, "Group 2" leukocyte Ig-like receptor A5 (LILRA5/LIR9/ILT11). Journal of Biological Chemistry 2006;281:19536-44. 38. Nakajima H, Samaridis J, Angman L, et al. Human myeloid cells express an activating ILT receptor (ILT1) that associates with Fc receptor gamma- chain. Journal of Immunology 1999;162:5-8. 39. Wormald MR, Dwek RA. Glycoproteins: glycan presentation and protein- fold stability. Structure 1999;7:R155-60. 40. Janik ME, Litynska A, Vereecken P. Cell migration-the role of integrin glycosylation. Biochim Biophys Acta 2010;1800:545-55. 41. Lee TH, Mitchell A, Liu Lau S, et al. Glycosylation in a mammalian expression system is critical for the production of functionally active leukocyte immunoglobulin-like receptor A3 protein. Journal of Biological Chemistry 2013;288:32873-85. 42. Borges L, Cosman D. LIRs/ILTs/MIRs, inhibitory and stimulatory Ig- superfamily receptors expressed in myeloid and lymphoid cells. Cytokine & Growth Factor Reviews 2000;11:209-17. 43. Brown D, Trowsdale J, Allen R. The LILR family: modulators of innate and adaptive immune pathways in health and disease. Tissue Antigens 2004;64:215-25. 44. Cambier JC. Antigen and Fc receptor signaling. The awesome power of the immunoreceptor tyrosine-based activation motif (ITAM). Journal of Immunology 1995;155:3281-5. 45. Cambier JC. New nomenclature for the Reth motif (or ARH1/TAM/ARAM/YXXL). Immunology Today 1995;16:110. 46. Reth M. Antigen receptor tail clue. Nature 1989;338:383-4. 47. Billadeau DD, Leibson PJ. ITAMs versus ITIMs: striking a balance during cell regulation. Journal of Clinical Investigation 2002;109:161-8. 48. Blank U, Launay P, Benhamou M, et al. Inhibitory ITAMs as novel regulators of immunity. Immunological Reviews 2009;232:59-71. 49. Waterman PM, Cambier JC. The conundrum of inhibitory signaling by ITAM- containing immunoreceptors: potential molecular mechanisms. FEBS Letters 2010;584:4878-82.

217 50. Abram CL, Lowell CA. The expanding role for ITAM-based signaling pathways in immune cells. Science's Stke [Electronic Resource]: Signal Transduction Knowledge Environment 2007;2007:re2. 51. Underhill DM, Goodridge HS. The many faces of ITAMs. Trends in Immunology 2007;28:66-73. 52. Vivier ED, M. Immunoreceptor tyrosine-based inhibition motifs. Immunol today 1997;18:286-291. 53. Bolland S, Ravetch JV. Inhibitory pathways triggered by ITIM-containing receptors. Advances in Immunology 1999;72:149-77. 54. Tamir I, Dal Porto JM, Cambier JC. Cytoplasmic protein tyrosine phosphatases SHP-1 and SHP-2: regulators of B cell signal transduction. Current Opinion in Immunology 2000;12:307-15. 55. Rohrschneider LR, Fuller JF, Wolf I, et al. Structure, function, and biology of SHIP proteins. Genes & Development 2000;14:505-20. 56. Hatherley D, Cherwinski HM, Moshref M, et al. Recombinant CD200 protein does not bind activating proteins closely related to CD200 receptor. Journal of Immunology 2005;175:2469-74. 57. Kaiser BK, Barahmand-Pour F, Paulsene W, et al. Interactions between NKG2x immunoreceptors and HLA-E ligands display overlapping affinities and thermodynamics. Journal of Immunology 2005;174:2878-84. 58. Hatherley D, Graham SC, Turner J, et al. Paired receptor specificity explained by structures of signal regulatory proteins alone and complexed with CD47. Molecular Cell 2008;31:266-77. 59. Kaiser BK, Pizarro JC, Kerns J, et al. Structural basis for NKG2A/CD94 recognition of HLA-E. Proceedings of the National Academy of Sciences of the United States of America 2008;105:6696-701. 60. Yamada E, McVicar DW. Paired receptor systems of the innate immune system. Current Protocols in Immunology 2008;Chapter 1:Appendix 1X. 61. Ravetch JV, Lanier LL. Immune inhibitory receptors. Science 2000;290:84-9. 62. Jones DC, Roghanian A, Brown DP, et al. Alternative mRNA splicing creates transcripts encoding soluble proteins from most LILR genes. European Journal of Immunology 2009;39:3195-206. 63. Borges L, Kubin M, Kuhlman T. LIR9, an immunoglobulin-superfamily- activating receptor, is expressed as a transmembrane and as a secreted molecule. Blood 2003;101:1484-6. 64. Beinhauer BG, McBride JM, Graf P, et al. Interleukin 10 regulates cell surface and soluble LIR-2 (CD85d) expression on dendritic cells resulting in T cell hyporesponsiveness in vitro. European Journal of Immunology 2004;34:74- 80. 65. Suciu-Foca N, Feirt N, Zhang Q-Y, et al. Soluble Ig-like transcript 3 inhibits tumor allograft rejection in humanized SCID mice and T cell responses in cancer patients. Journal of Immunology 2007;178:7432-41. 66. Lopez-Alvarez MR, Jones DC, Jiang W, et al. Copy number and nucleotide variation of the LILR family of myelomonocytic cell activating and inhibitory receptors. Immunogenetics 2014;66:73-83. 67. Sudmant PH, Kitzman JO, Antonacci F, et al. Diversity of human copy number variation and multicopy genes. Science 2010;330:641-6.

218 68. Torkar M, Haude A, Milne S, et al. Arrangement of the ILT gene cluster: a common null allele of the ILT6 gene results from a 6.7-kbp deletion. European Journal of Immunology 2000;30:3655-62. 69. Norman PJ, Carey BS, Stephens HAF, et al. DNA sequence variation and molecular genotyping of natural killer leukocyte immunoglobulin-like receptor, LILRA3. Immunogenetics 2003;55:165-71. 70. Hirayasu K, Ohashi J, Kashiwase K, et al. Long-term persistence of both functional and non-functional alleles at the leukocyte immunoglobulin-like receptor A3 (LILRA3) locus suggests balancing selection. Human Genetics 2006;119:436-43. 71. Hirayasu K, Ohashi J, Tanaka H, et al. Evidence for natural selection on leukocyte immunoglobulin-like receptors for HLA class I in Northeast Asians. American Journal of Human Genetics 2008;82:1075-83. 72. Norman PJ, Cook MA, Carey BS, et al. SNP haplotypes and allele frequencies show evidence for disruptive and balancing selection in the human leukocyte receptor complex. Immunogenetics 2004;56:225-37. 73. Bashirova AA, Apps R, Vince N, et al. Diversity of the human LILRB3/A6 locus encoding a myeloid inhibitory and activating receptor pair. Immunogenetics 2014;66:1-8. 74. Cella M, Jarrossay D, Facchetti F, et al. Plasmacytoid monocytes migrate to inflamed lymph nodes and produce large amounts of type I interferon. Nature Medicine 1999;5:919-23. 75. Saverino D, Fabbi M, Ghiotto F, et al. The CD85/LIR-1/ILT2 inhibitory receptor is expressed by all human T lymphocytes and down-regulates their functions. Journal of Immunology 2000;165:3742-55. 76. Tedla N, Gibson K, McNeil HP, et al. The co-expression of activating and inhibitory leukocyte immunoglobulin-like receptors in rheumatoid synovium. American Journal of Pathology 2002;160:425-31. 77. Tedla N, Bandeira-Melo C, Tassinari P, et al. Activation of human eosinophils through leukocyte immunoglobulin-like receptor 7. Proceedings of the National Academy of Sciences of the United States of America 2003;100:1174-9. 78. Sloane DE, Tedla N, Awoniyi M, et al. Leukocyte immunoglobulin-like receptors: novel innate receptors for human basophil activation and inhibition. Blood 2004;104:2832-9. 79. Ju X-S, Hacker C, Scherer B, et al. Immunoglobulin-like transcripts ILT2, ILT3 and ILT7 are expressed by human dendritic cells and down-regulated following activation. Gene 2004;331:159-64. 80. Cao W, Rosen DB, Ito T, et al. Plasmacytoid dendritic cell-specific receptor ILT7-Fc epsilonRI gamma inhibits Toll-like receptor-induced interferon production. Journal of Experimental Medicine 2006;203:1399-405. 81. Cho M, Ishida K, Chen J, et al. SAGE library screening reveals ILT7 as a specific plasmacytoid dendritic cell marker that regulates type I IFN production. International Immunology 2008;20:155-64. 82. Tedla N, Lee C-W, Borges L, et al. Differential expression of leukocyte immunoglobulin-like receptors on cord-blood-derived human mast cell progenitors and mature mast cells. Journal of Leukocyte Biology 2008;83:334-43.

219 83. Anderson KJ, Allen RL. Regulation of T-cell immunity by leucocyte immunoglobulin-like receptors: innate immune receptors for self on antigen-presenting cells. Immunology 2009;127:8-17. 84. An H, Chandra V, Piraino B, et al. Soluble LILRA3, a potential natural antiinflammatory protein, is increased in patients with rheumatoid arthritis and is tightly regulated by interleukin 10, tumor necrosis factor-alpha, and interferon-gamma. Journal of Rheumatology 2010;37:1596-606. 85. Bankey PE, Banerjee S, Zucchiatti A, et al. Cytokine induced expression of programmed death ligands in human neutrophils. Immunology Letters 2010;129:100-7. 86. Baudhuin J, Migraine J, Faivre V, et al. Exocytosis acts as a modulator of the ILT4-mediated inhibition of neutrophil functions. Proceedings of the National Academy of Sciences of the United States of America 2013;110:17957-62. 87. Davidson CL, Cameron LE, Burshtyn DN. The AP-1 transcription factor JunD activates the leukocyte immunoglobulin-like receptor 1 distal promoter. International Immunology 2014;26:21-33. 88. Schutz F, Hackstein H. Identification of novel dendritic cell subset markers in human blood. Biochemical & Biophysical Research Communications 2014;443:453-7. 89. Chen Y, Chu F, Gao F, et al. Stability engineering, biophysical, and biological characterization of the myeloid activating receptor immunoglobulin-like transcript 1 (ILT1/LIR-7/LILRA2). Protein Expression & Purification 2007;56:253-60. 90. Garner LI, Salim M, Mohammed F, et al. Expression, purification, and refolding of the myeloid inhibitory receptor leukocyte immunoglobulin-like receptor-5 for structural and ligand identification studies. Protein Expression & Purification 2006;47:490-7. 91. Horton R, Wilming L, Rand V, et al. Gene map of the extended human MHC. Nature Reviews Genetics 2004;5:889-99. 92. Chapman TL, Heikema AP, West AP, Jr., et al. Crystal structure and ligand binding properties of the D1D2 region of the inhibitory receptor LIR-1 (ILT2). Immunity 2000;13:727-36. 93. Allan DS, Colonna M, Lanier LL, et al. Tetrameric complexes of human histocompatibility leukocyte antigen (HLA)-G bind to peripheral blood myelomonocytic cells. Journal of Experimental Medicine 1999;189:1149-56. 94. Colonna M, Samaridis J, Cella M, et al. Human myelomonocytic cells express an inhibitory receptor for classical and nonclassical MHC class I molecules. Journal of Immunology 1998;160:3096-100. 95. Gonen-Gross T, Achdout H, Gazit R, et al. Complexes of HLA-G protein on the cell surface are important for leukocyte Ig-like receptor-1 function. Journal of Immunology 2003;171:1343-51. 96. Lepin EJ, Bastin JM, Allan DS, et al. Functional characterization of HLA-F and binding of HLA-F tetramers to ILT2 and ILT4 receptors. European Journal of Immunology 2000;30:3552-61. 97. Navarro F, Llano M, Bellon T, et al. The ILT2(LIR1) and CD94/NKG2A NK cell receptors respectively recognize HLA-G1 and HLA-E molecules co- expressed on target cells. European Journal of Immunology 1999;29:277-83.

220 98. Shiroishi M, Tsumoto K, Amano K, et al. Human inhibitory receptors Ig-like transcript 2 (ILT2) and ILT4 compete with CD8 for MHC class I binding and bind preferentially to HLA-G. Proceedings of the National Academy of Sciences of the United States of America 2003;100:8856-61. 99. Shiroishi M, Kuroki K, Rasubala L, et al. Structural basis for recognition of the nonclassical MHC molecule HLA-G by the leukocyte Ig-like receptor B2 (LILRB2/LIR2/ILT4/CD85d). Proceedings of the National Academy of Sciences of the United States of America 2006;103:16412-7. 100. Carosella ED, Moreau P, Lemaoult J, et al. HLA-G: from biology to clinical benefits. Trends in Immunology 2008;29:125-32. 101. Kovats S, Main EK, Librach C, et al. A class I antigen, HLA-G, expressed in human trophoblasts. Science 1990;248:220-3. 102. Houlihan JM, Biro PA, Harper HM, et al. The human amnion is a site of MHC class Ib expression: evidence for the expression of HLA-E and HLA-G. Journal of Immunology 1995;154:5665-74. 103. Crisa L, McMaster MT, Ishii JK, et al. Identification of a thymic epithelial cell subset sharing expression of the class Ib HLA-G molecule with fetal trophoblasts. Journal of Experimental Medicine 1997;186:289-98. 104. Le Discorde M, Moreau P, Sabatier P, et al. Expression of HLA-G in human cornea, an immune-privileged tissue. Human Immunology 2003;64:1039- 44. 105. Cirulli V, Zalatan J, McMaster M, et al. The class I HLA repertoire of pancreatic islets comprises the nonclassical class Ib antigen HLA-G. Diabetes 2006;55:1214-22. 106. Ito T, Ito N, Saathoff M, et al. Immunology of the human nail apparatus: the nail matrix is a site of relative immune privilege. Journal of Investigative Dermatology 2005;125:1139-48. 107. Ryu M, Chen Y, Qi J, et al. LILRA3 binds both classical and non-classical HLA class I molecules but with reduced affinities compared to LILRB1/LILRB2: structural evidence. PLoS ONE [Electronic Resource] 2011;6:e19245. 108. Nakamura A, Kobayashi E, Takai T. Exacerbated graft-versus-host disease in Pirb-/- mice. Nature Immunology 2004;5:623-9. 109. Arnold V, Cummings JS, Moreno-Nieves UY, et al. S100A9 protein is a novel ligand for the CD85j receptor and its interaction is implicated in the control of HIV-1 replication by NK cells. Retrovirology 2013;10:122. 110. Atwal JK, Pinkston-Gosse J, Syken J, et al. PirB is a functional receptor for myelin inhibitors of axonal regeneration. Science 2008;322:967-70. 111. Kim T, Vidal GS, Djurisic M, et al. Human LilrB2 is a beta-amyloid receptor and its murine homolog PirB regulates synaptic plasticity in an Alzheimer's model. Science 2013;341:1399-404. 112. Nakamura Y, Fujita Y, Ueno M, et al. Paired immunoglobulin-like receptor B knockout does not enhance axonal regeneration or locomotor recovery after spinal cord injury. Journal of Biological Chemistry 2011;286:1876-83. 113. Omoto S, Ueno M, Mochio S, et al. Genetic deletion of paired immunoglobulin-like receptor B does not promote axonal plasticity or functional recovery after traumatic brain injury. Journal of Neuroscience 2010;30:13045-52. 114. Zheng J, Umikawa M, Cui C, et al. Inhibitory receptors bind ANGPTLs and support blood stem cells and leukaemia development.[Erratum appears in

221 Nature. 2012 Aug 30;488(7413):684 Note: Hyunh, Hoangdinh [corrected to Huynh, HoangDinh]]. Nature 2012;485:656-60. 115. Cao W, Bover L. Signaling and ligand interaction of ILT7: receptor-mediated regulatory mechanisms for plasmacytoid dendritic cells. Immunological Reviews 2010;234:163-76. 116. Cao W, Bover L, Cho M, et al. Regulation of TLR7/9 responses in plasmacytoid dendritic cells by BST2 and ILT7 receptor interaction. Journal of Experimental Medicine 2009;206:1603-14. 117. Palma G, De Laurenzi V, De Marco M, et al. Plasmacytoids dendritic cells are a therapeutic target in anticancer immunity. Biochimica et Biophysica Acta 2012;1826:407-14. 118. Tavano B, Galao RP, Graham DR, et al. Ig-like transcript 7, but not bone marrow stromal cell antigen 2 (also known as HM1.24, tetherin, or CD317), modulates plasmacytoid dendritic cell function in primary human blood leukocytes. Journal of Immunology 2013;190:2622-30. 119. Ince MN, Harnisch B, Xu Z, et al. Increased expression of the natural killer cell inhibitory receptor CD85j/ILT2 on antigen-specific effector CD8 T cells and its impact on CD8 T-cell function. Immunology 2004;112:531-42. 120. Anfossi N, Doisne J-M, Peyrat M-A, et al. Coordinated expression of Ig-like inhibitory MHC class I receptors and acquisition of cytotoxic function in human CD8+ T cells. Journal of Immunology 2004;173:7223-9. 121. Guma M, Angulo A, Vilches C, et al. Imprint of human cytomegalovirus infection on the NK cell receptor repertoire. Blood 2004;104:3664-71. 122. Monsivais-Urenda A, Noyola-Cherpitel D, Hernandez-Salinas A, et al. Influence of human cytomegalovirus infection on the NK cell receptor repertoire in children.[Erratum appears in Eur J Immunol. 2011 Jan;41(1):258]. European Journal of Immunology 2010;40:1418-27. 123. Antrobus RD, Khan N, Hislop AD, et al. Virus-specific cytotoxic T lymphocytes differentially express cell-surface leukocyte immunoglobulin- like receptor-1, an inhibitory receptor for class I major histocompatibility complex molecules. Journal of Infectious Diseases 2005;191:1842-53. 124. Northfield J, Lucas M, Jones H, et al. Does memory improve with age? CD85j (ILT-2/LIR-1) expression on CD8 T cells correlates with 'memory inflation' in human cytomegalovirus infection. Immunology & Cell Biology 2005;83:182-8. 125. Poon K, Montamat-Sicotte D, Cumberbatch N, et al. Expression of leukocyte immunoglobulin-like receptors and natural killer receptors on virus- specific CD8+ T cells during the evolution of Epstein-Barr virus-specific immune responses in vivo. Viral Immunology 2005;18:513-22. 126. Scott-Algara D, Arnold V, Didier C, et al. The CD85j+ NK cell subset potently controls HIV-1 replication in autologous dendritic cells. PLoS ONE [Electronic Resource] 2008;3:e1975. 127. Lichterfeld M, Kavanagh DG, Williams KL, et al. A viral CTL escape mutation leading to immunoglobulin-like transcript 4-mediated functional inhibition of myelomonocytic cells. Journal of Experimental Medicine 2007;204:2813- 24. 128. Vlad G, Piazza F, Colovai A, et al. Interleukin-10 induces the upregulation of the inhibitory receptor ILT4 in monocytes from HIV positive individuals. Human Immunology 2003;64:483-9.

222 129. Huang J, Burke PS, Cung TDH, et al. Leukocyte immunoglobulin-like receptors maintain unique antigen-presenting properties of circulating myeloid dendritic cells in HIV-1-infected elite controllers. Journal of Virology 2010;84:9463-71. 130. Huang J, Goedert JJ, Sundberg EJ, et al. HLA-B 35-Px-mediated acceleration of HIV-1 infection by increased inhibitory immunoregulatory impulses. Journal of Experimental Medicine 2009;206:2959-66. 131. Huang J, Burke P, Yang Y, et al. Soluble HLA-G inhibits myeloid dendritic cell function in HIV-1 infection by interacting with leukocyte immunoglobulin- like receptor B2. Journal of Virology 2010;84:10784-91. 132. Nakayama M, Underhill DM, Petersen TW, et al. Paired Ig-like receptors bind to bacteria and shape TLR-mediated cytokine production. Journal of Immunology 2007;178:4250-9. 133. Chan KR, Ong EZ, Tan HC, et al. Leukocyte immunoglobulin-like receptor B1 is critical for antibody-dependent dengue. Proceedings of the National Academy of Sciences of the United States of America 2014;111:2722-7. 134. Brown DP, Jones DC, Anderson KJ, et al. The inhibitory receptor LILRB4 (ILT3) modulates antigen presenting cell phenotype and, along with LILRB2 (ILT4), is upregulated in response to Salmonella infection. BMC Immunology 2009;10:56. 135. Dietrich J, Cella M, Colonna M. Ig-like transcript 2 (ILT2)/leukocyte Ig-like receptor 1 (LIR1) inhibits TCR signaling and actin cytoskeleton reorganization. Journal of Immunology 2001;166:2514-21. 136. Merlo A, Saverino D, Tenca C, et al. CD85/LIR-1/ILT2 and CD152 (cytotoxic T lymphocyte antigen 4) inhibitory molecules down-regulate the cytolytic activity of human CD4+ T-cell clones specific for Mycobacterium tuberculosis. Infection & Immunity 2001;69:6022-9. 137. Saverino D, Merlo A, Bruno S, et al. Dual effect of CD85/leukocyte Ig-like receptor-1/Ig-like transcript 2 and CD152 (CTLA-4) on cytokine production by antigen-stimulated human T cells. Journal of Immunology 2002;168:207- 15. 138. Naji A, Menier C, Morandi F, et al. Binding of HLA-G to ITIM-bearing Ig-like transcript 2 receptor suppresses B cell responses. Journal of Immunology 2014;192:1536-46. 139. Wagner CS, Walther-Jallow L, Buentke E, et al. Human cytomegalovirus- derived protein UL18 alters the phenotype and function of monocyte- derived dendritic cells. Journal of Leukocyte Biology 2008;83:56-63. 140. Young NT, Waller ECP, Patel R, et al. The inhibitory receptor LILRB1 modulates the differentiation and regulatory potential of human dendritic cells. Blood 2008;111:3090-6. 141. Chang CC, Ciubotariu R, Manavalan JS, et al. Tolerization of dendritic cells by T(S) cells: the crucial role of inhibitory receptors ILT3 and ILT4. Nature Immunology 2002;3:237-43. 142. Ristich V, Liang S, Zhang W, et al. Tolerization of dendritic cells by HLA-G. European Journal of Immunology 2005;35:1133-42. 143. Svajger U, Obermajer N, Jeras M. IFN--rich environment programs dendritic cells toward silencing of cytotoxic immune responses. Journal of Leukocyte Biology 2014;95:33-46.

223 144. Chen L, Xu Z, Chang C, et al. Allospecific CD8 T suppressor cells induced by multiple MLC stimulation or priming in the presence of ILT3.Fc have similar gene expression profiles. Human Immunology 2014;75:190-6. 145. Chang C-C, Liu Z, Vlad G, et al. Ig-like transcript 3 regulates expression of proinflammatory cytokines and migration of activated T cells. Journal of Immunology 2009;182:5208-16. 146. Lee DJ, Sieling PA, Ochoa MT, et al. LILRA2 activation inhibits dendritic cell differentiation and antigen presentation to T cells. Journal of Immunology 2007;179:8128-36. 147. Lu HK, Mitchell A, Endoh Y, et al. LILRA2 selectively modulates LPS- mediated cytokine production and inhibits phagocytosis by monocytes.[Erratum appears in PLoS One. 2012;7(8). doi: 10.1371/annotation/f4475cbd-3518-46cd-bb6b-a326008af388]. PLoS ONE [Electronic Resource] 2012;7:e33478. 148. Bleharski JR, Li H, Meinken C, et al. Use of genetic profiling in leprosy to discriminate clinical forms of the disease. Science 2003;301:1527-30. 149. Mitchell A, Rentero C, Endoh Y, et al. LILRA5 is expressed by synovial tissue macrophages in rheumatoid arthritis, selectively induces pro-inflammatory cytokines and IL-10 and is regulated by TNF-alpha, IL-10 and IFN-gamma. European Journal of Immunology 2008;38:3459-73. 150. Low HZ, Reuter S, Topperwien M, et al. Association of the LILRA3 deletion with B-NHL and functional characterization of the immunostimulatory molecule. PLoS ONE [Electronic Resource] 2013;8:e81360. 151. Munitz A, Cole ET, Beichler A, et al. Paired immunoglobulin-like receptor B (PIR-B) negatively regulates macrophage activation in experimental colitis. Gastroenterology 2010;139:530-41. 152. Montero-Melendez T, Llor X, Garcia-Planella E, et al. Identification of novel predictor classifiers for inflammatory bowel disease by gene expression profiling. PLoS ONE [Electronic Resource] 2013;8:e76235. 153. Sun Y, Liu J, Gao P, et al. Expression of Ig-like transcript 4 inhibitory receptor in human non-small cell lung cancer. Chest 2008;134:783-8. 154. Rouas-Freiss N, Moreau P, LeMaoult J, et al. The dual role of HLA-G in cancer. J Immunol Res 2014;2014:359748. 155. Naji A, Menier C, Maki G, et al. Neoplastic B-cell growth is impaired by HLA- G/ILT2 interaction. Leukemia 2012;26:1889-92. 156. Leleu X, Le Friec G, Facon T, et al. Total soluble HLA class I and soluble HLA- G in multiple myeloma and monoclonal gammopathy of undetermined significance. Clinical Cancer Research 2005;11:7297-303. 157. Sebti Y, Le Maux A, Gros F, et al. Expression of functional soluble human leucocyte antigen-G molecules in lymphoproliferative disorders. British Journal of Haematology 2007;138:202-12. 158. Wlasiuk P, Tomczak W, Zajac M, et al. Total expression of HLA-G and TLR-9 in chronic lymphocytic leukemia patients. Human Immunology 2013;74:1592-7. 159. Colovai AI, Tsao L, Wang S, et al. Expression of inhibitory receptor ILT3 on neoplastic B cells is associated with lymphoid tissue involvement in chronic lymphocytic leukemia. Cytometry Part B, Clinical Cytometry 2007;72:354- 62.

224 160. Casado LF, Granados E, Algara P, et al. High expression of the ILT2 (LIR-1) inhibitory receptor for major histocompatibility complex class I molecules on clonal expansions of T large granular lymphocytes in asymptomatic patients. Haematologica 2001;86:457-63. 161. Kamarashev J, Burg G, Mingari MC, et al. Differential expression of cytotoxic molecules and killer cell inhibitory receptors in CD8+ and CD56+ cutaneous lymphomas. American Journal of Pathology 2001;158:1593-8. 162. Nikolova M, Musette P, Bagot M, et al. Engagement of ILT2/CD85j in Sezary syndrome cells inhibits their CD3/TCR signaling. Blood 2002;100:1019-25. 163. Urosevic M, Kamarashev J, Burg G, et al. Primary cutaneous CD8+ and CD56+ T-cell lymphomas express HLA-G and killer-cell inhibitory ligand, ILT2. Blood 2004;103:1796-8. 164. Hunt JS, Petroff MG, McIntire RH, et al. HLA-G and immune tolerance in pregnancy. FASEB Journal 2005;19:681-93. 165. Guo Y, Lee CL, So KH, et al. Soluble human leukocyte antigen-g5 activates extracellular signal-regulated protein kinase signaling and stimulates trophoblast invasion. PLoS ONE [Electronic Resource] 2013;8:e76023. 166. Hsu P, Santner-Nanan B, Joung S, et al. Expansion of CD4(+) HLA-G(+) T Cell in human pregnancy is impaired in pre-eclampsia. American Journal of Reproductive Immunology 2014;71:217-28. 167. Liu Y, Zhang L, Gao M, et al. Changes of inhibitory receptors on NK-92 cells and HLA-G on BeWo cells with Toxoplasma gondii infection. Inflammation 2013;36:1440-7. 168. Manavalan JS, Kim-Schulze S, Scotto L, et al. Alloantigen specific CD8+CD28- FOXP3+ T suppressor cells induce ILT3+ ILT4+ tolerogenic endothelial cells, inhibiting alloreactivity. International Immunology 2004;16:1055-68. 169. Benito MJ, Lopez-Hoyos M, Fernandez-Fresnedo G, et al. Changes in the expression of the immunoglobulin-like transcript 3 (ILT3) and ILT4 receptors in renal allograft recipients: effect of donor and recipient aging. Transplantation Proceedings 2008;40:2894-6. 170. Wu J, Zhang W, Hernandez-Lopez P, et al. Isoforms of human leukocyte antigen-G and their inhibitory receptors in human kidney allograft acceptance. Human Immunology 2009;70:988-94. 171. Le Rond S, Azema C, Krawice-Radanne I, et al. Evidence to support the role of HLA-G5 in allograft acceptance through induction of immunosuppressive/ regulatory T cells. Journal of Immunology 2006;176:3266-76. 172. LeMaoult J, Daouya M, Wu J, et al. Synthetic HLA-G proteins for therapeutic use in transplantation. FASEB Journal 2013;27:3643-51. 173. Pfistershammer K, Lawitschka A, Klauser C, et al. Allogeneic disparities in immunoglobulin-like transcript 5 induce potent antibody responses in hematopoietic stem cell transplant recipients. Blood 2009;114:2323-32. 174. Baffari E, Fiume D, Caiazzo G, et al. Upregulation of the inhibitory receptor ILT4 in monocytes from septic patients. Human Immunology 2013;74:1244-50. 175. Tedla N, An H, Borges L, et al. Expression of activating and inhibitory leukocyte immunoglobulin-like receptors in rheumatoid synovium: correlations to disease activity. Tissue Antigens 2011;77:305-16.

225 176. Huynh OA, Hampartzoumian T, Arm JP, et al. Down-regulation of leucocyte immunoglobulin-like receptor expression in the synovium of rheumatoid arthritis patients after treatment with disease-modifying anti-rheumatic drugs. Rheumatology 2007;46:742-51. 177. Kuroki K, Hirose K, Okabe Y, et al. The long-term immunosuppressive effects of disulfide-linked HLA-G dimer in mice with collagen-induced arthritis. Human Immunology 2013;74:433-8. 178. Monsivais-Urenda A, Nino-Moreno P, Abud-Mendoza C, et al. Analysis of expression and function of the inhibitory receptor ILT2 (CD85j/LILRB1/LIR-1) in peripheral blood mononuclear cells from patients with systemic lupus erythematosus (SLE). Journal of Autoimmunity 2007;29:97-105. 179. Monsivais-Urenda A, Gomez-Martin D, Santana-de-Anda K, et al. Defective expression and function of the ILT2/CD85j regulatory receptor in dendritic cells from patients with systemic lupus erythematosus. Human Immunology 2013;74:1088-96. 180. Jensen MA, Yanowitch RN, Reder AT, et al. Immunoglobulin-like transcript 3, an inhibitor of T cell activation, is reduced on blood monocytes during multiple sclerosis relapses and is induced by interferon beta-1b. Multiple Sclerosis 2010;16:30-8. 181. Wiendl H, Feger U, Mittelbronn M, et al. Expression of the immune- tolerogenic major histocompatibility molecule HLA-G in multiple sclerosis: implications for CNS immunity. Brain 2005;128:2689-704. 182. Aractingi S, Briand N, Le Danff C, et al. HLA-G and NK receptor are expressed in psoriatic skin: a possible pathway for regulating infiltrating T cells? American Journal of Pathology 2001;159:71-7. 183. Schleinitz N, Cognet C, Guia S, et al. Expression of the CD85j (leukocyte Ig- like receptor 1, Ig-like transcript 2) receptor for class I major histocompatibility complex molecules in idiopathic inflammatory myopathies. Arthritis & Rheumatism 2008;58:3216-23. 184. Du Y, Cui Y, Liu X, et al. Contribution of functional LILRA3, but not nonfunctional LILRA3, to sex bias in susceptibility and severity of anti- citrullinated protein antibody-positive rheumatoid arthritis. Arthritis & Rheumatology 2014;66:822-30. 185. Kabalak G, Dobberstein SB, Matthias T, et al. Association of immunoglobulin-like transcript 6 deficiency with Sjogren's syndrome. Arthritis & Rheumatism 2009;60:2923-5. 186. Wisniewski A, Luszczek W, Manczak M, et al. Distribution of LILRA3 (ILT6/LIR4) deletion in psoriatic patients and healthy controls. Human Immunology 2003;64:458-61. 187. Wisniewski A, Wagner M, Nowak I, et al. 6.7-kbp deletion in LILRA3 (ILT6) gene is associated with later onset of the multiple sclerosis in a Polish population. Human Immunology 2013;74:353-7. 188. Koch S, Goedde R, Nigmatova V, et al. Association of multiple sclerosis with ILT6 deficiency. Genes & Immunity 2005;6:445-7. 189. Ordonez D, Sanchez AJ, Martinez-Rodriguez JE, et al. Multiple sclerosis associates with LILRA3 deletion in Spanish patients. Genes & Immunity 2009;10:579-85.

226 190. Mamegano K, Kuroki K, Miyashita R, et al. Association of LILRA2 (ILT1, LIR7) splice site polymorphism with systemic lupus erythematosus and microscopic polyangiitis.[Erratum appears in Genes Immun. 2008 Oct;9(7):650]. Genes & Immunity 2008;9:214-23. 191. Jensen MA, Patterson KC, Kumar AA, et al. Functional genetic polymorphisms in ILT3 are associated with decreased surface expression on dendritic cells and increased serum cytokines in lupus patients. Annals of the Rheumatic Diseases 2013;72:596-601. 192. Kuroki K, Tsuchiya N, Shiroishi M, et al. Extensive polymorphisms of LILRB1 (ILT2, LIR1) and their association with HLA-DRB1 shared epitope negative rheumatoid arthritis. Human Molecular Genetics 2005;14:2469-80. 193. Savage DC. Microbial ecology of the gastrointestinal tract. Annual Review of Microbiology 1977;31:107-33. 194. Madara JL, Nash S, Moore R, et al. Structure and function of the intestinal epithelial barrier in health and disease. Monographs in Pathology 1990:306-24. 195. Raleigh DR, Marchiando AM, Zhang Y, et al. Tight junction-associated MARVEL proteins marveld3, tricellulin, and occludin have distinct but overlapping functions. Molecular Biology of the Cell 2010;21:1200-13. 196. Amasheh S, Fromm M, Gunzel D. Claudins of intestine and nephron - a correlation of molecular tight junction structure and barrier function. Acta Physiologica 2011;201:133-40. 197. Shen L, Weber CR, Raleigh DR, et al. Tight junction pore and leak pathways: a dynamic duo. Annual Review of Physiology 2011;73:283-309. 198. Suzuki T. Regulation of intestinal epithelial permeability by tight junctions. Cellular & Molecular Life Sciences 2013;70:631-59. 199. Turner JR. Intestinal mucosal barrier function in health and disease. Nature Reviews. Immunology 2009;9:799-809. 200. Kunisawa J, Takahashi I, Kiyono H. Intraepithelial lymphocytes: their shared and divergent immunological behaviors in the small and large intestine. Immunological Reviews 2007;215:136-53. 201. Cheroutre H, Lambolez F, Mucida D. The light and dark sides of intestinal intraepithelial lymphocytes. Nature Reviews. Immunology 2011;11:445-56. 202. Lee E, Schiller LR, Fordtran JS. Quantification of colonic lamina propria cells by means of a morphometric point-counting method. Gastroenterology 1988;94:409-18. 203. Ueda Y, Kayama H, Jeon SG, et al. Commensal microbiota induce LPS hyporesponsiveness in colonic macrophages via the production of IL-10. 2010. 204. Mahida YR, Patel S, Gionchetti P, et al. Macrophage subpopulations in lamina propria of normal and inflamed colon and terminal ileum. Gut 1989;30:826-34. 205. Malizia G, Calabrese A, Cottone M, et al. Expression of leukocyte adhesion molecules by mucosal mononuclear phagocytes in inflammatory bowel disease. Gastroenterology 1991;100:150-9. 206. Grimm MC, Pavli P, Van de Pol E, et al. Evidence for a CD14+ population of monocytes in inflammatory bowel disease mucosa--implications for pathogenesis. Clinical & Experimental Immunology 1995;100:291-7.

227 207. Rogler G, Hausmann M, Vogl D, et al. Isolation and phenotypic characterization of colonic macrophages. Clinical & Experimental Immunology 1998;112:205-15. 208. Kamada N, Hisamatsu T, Okamoto S, et al. Unique CD14 intestinal macrophages contribute to the pathogenesis of Crohn disease via IL- 23/IFN-gamma axis. Journal of Clinical Investigation 2008;118:2269-80. 209. Mowat AM, Bain CC. Mucosal macrophages in intestinal homeostasis and inflammation. Journal of Innate Immunity 2011;3:550-64. 210. Smith PD, Janoff EN, Mosteller-Barnum M, et al. Isolation and purification of CD14-negative mucosal macrophages from normal human small intestine. Journal of Immunological Methods 1997;202:1-11. 211. Smith PD, Smythies LE, Mosteller-Barnum M, et al. Intestinal macrophages lack CD14 and CD89 and consequently are down-regulated for LPS- and IgA-mediated activities. Journal of Immunology 2001;167:2651-6. 212. Smythies LE, Sellers M, Clements RH, et al. Human intestinal macrophages display profound inflammatory anergy despite avid phagocytic and bacteriocidal activity. Journal of Clinical Investigation 2005;115:66-75. 213. Rescigno M, Rotta G, Valzasina B, et al. Dendritic cells shuttle microbes across gut epithelial monolayers. Immunobiology 2001;204:572-81. 214. Chang SY, Song JH, Guleng B, et al. Circulatory antigen processing by mucosal dendritic cells controls CD8(+) T cell activation. Immunity 2013;38:153-65. 215. Chang SY, Ko HJ, Kweon MN. Mucosal dendritic cells shape mucosal immunity. Experimental & Molecular Medicine 2014;46:e84. 216. Groschwitz KR, Ahrens R, Osterfeld H, et al. Mast cells regulate homeostatic intestinal epithelial migration and barrier function by a chymase/Mcpt4- dependent mechanism. Proceedings of the National Academy of Sciences of the United States of America 2009;106:22381-6. 217. Groschwitz KR, Wu D, Osterfeld H, et al. Chymase-mediated intestinal epithelial permeability is regulated by a protease-activating receptor/matrix metalloproteinase-2-dependent mechanism. American Journal of Physiology - Gastrointestinal & Liver Physiology 2013;304:G479- 89. 218. Gurish MF, Austen KF. Developmental origin and functional specialization of mast cell subsets. Immunity 2012;37:25-33. 219. Kurashima Y, Kiyono H. New era for mucosal mast cells: their roles in inflammation, allergic immune responses and adjuvant development. Experimental & Molecular Medicine 2014;46:e83. 220. Ramsay DB, Stephen S, Borum M, et al. Mast Cells in Gastrointestinal Disease. Gastroenterology & Hepatology 2010;6:772-777. 221. Pascal RR, Gramlich TL, Parker KM, et al. Geographic variations in eosinophil concentration in normal colonic mucosa. Modern Pathology 1997;10:363-5. 222. DeBrosse CW, Case JW, Putnam PE, et al. Quantity and distribution of eosinophils in the gastrointestinal tract of children. Pediatric & Developmental Pathology 2006;9:210-8. 223. Lowichik A, Weinberg AG. A quantitative evaluation of mucosal eosinophils in the pediatric gastrointestinal tract. Modern Pathology 1996;9:110-4.

228 224. Soehnlein O, Lindbom L. Phagocyte partnership during the onset and resolution of inflammation. Nature Reviews. Immunology 2010;10:427-39. 225. Fournier BM, Parkos CA. The role of neutrophils during intestinal inflammation. Mucosal immunology 2012;5:354-66. 226. Amulic B, Cazalet C, Hayes GL, et al. Neutrophil Function: From Mechanisms to Disease. Annual Review of Immunology 2012;30:459-489. 227. Kolaczkowska E, Kubes P. Neutrophil recruitment and function in health and inflammation. Nature Reviews. Immunology 2013;13:159-75. 228. Spits H, Cupedo T. Innate lymphoid cells: emerging insights in development, lineage relationships, and function. Annual Review of Immunology 2012;30:647-75. 229. Chinen H, Matsuoka K, Sato T, et al. Lamina propria c-kit+ immune precursors reside in human adult intestine and differentiate into natural killer cells. Gastroenterology 2007;133:559-73. 230. Hogan PG, Hapel AJ, Doe WF. Lymphokine-activated and natural killer cell activity in human intestinal mucosa. Journal of Immunology 1985;135:1731-8. 231. Pang G, Buret A, Batey RT, et al. Morphological, phenotypic and functional characteristics of a pure population of CD56+ CD16- CD3- large granular lymphocytes generated from human duodenal mucosa. Immunology 1993;79:498-505. 232. Neill DR, Wong SH, Bellosi A, et al. Nuocytes represent a new innate effector leukocyte that mediates type-2 immunity. Nature 2010;464:1367-70. 233. Cherrier M, Eberl G. The development of LTi cells. Current Opinion in Immunology 2012;24:178-83. 234. Cherrier M, Sawa S, Eberl G. Notch, Id2, and RORt sequentially orchestrate the fetal development of lymphoid tissue inducer cells. Journal of Experimental Medicine 2012;209:729-40. 235. Spits H, Artis D, Colonna M, et al. Innate lymphoid cells--a proposal for uniform nomenclature. Nature Reviews. Immunology 2013;13:145-9. 236. Mortha A, Chudnovskiy A, Hashimoto D, et al. Microbiota-dependent crosstalk between macrophages and ILC3 promotes intestinal homeostasis. Science 2014;343:1249288. 237. Cherrier M, Ohnmacht C, Cording S, et al. Development and function of intestinal innate lymphoid cells. Current Opinion in Immunology 2012;24:277-83. 238. Diefenbach A, Colonna M, Koyasu S. Development, differentiation, and diversity of innate lymphoid cells. Immunity 2014;41:354-65. 239. Hazenberg MD, Spits H. Human innate lymphoid cells. Blood 2014;124:700- 9. 240. Cella M, Fuchs A, Vermi W, et al. A human natural killer cell subset provides an innate source of IL-22 for mucosal immunity. Nature 2009;457:722-5. 241. Sawa S, Lochner M, Satoh-Takayama N, et al. RORt+ innate lymphoid cells regulate intestinal homeostasis by integrating negative signals from the symbiotic microbiota. Nature Immunology 2011;12:320-6. 242. Sakaguchi S, Sakaguchi N, Asano M, et al. 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 1995;155:1151-64.

229 243. Fontenot JD, Gavin MA, Rudensky AY. Foxp3 programs the development and function of CD4+CD25+ regulatory T cells. Nat Immunol 2003;4:330-6. 244. Weiner HL. Induction and mechanism of action of transforming growth factor-beta-secreting Th3 regulatory cells. Immunol Rev 2001;182:207-14. 245. Groux H, O'Garra A, Bigler M, et al. A CD4+ T-cell subset inhibits antigen- specific T-cell responses and prevents colitis. Nature 1997;389:737-42. 246. Noble A, Giorgini A, Leggat JA. Cytokine-induced IL-10-secreting CD8 T cells represent a phenotypically distinct suppressor T-cell lineage. Blood 2006;107:4475-83. 247. Makita S, Kanai T, Oshima S, et al. CD4+CD25bright T cells in human intestinal lamina propria as regulatory cells. J Immunol 2004;173:3119-30. 248. Makita S, Kanai T, Nemoto Y, et al. Intestinal lamina propria retaining CD4+CD25+ regulatory T cells is a suppressive site of intestinal inflammation. J Immunol 2007;178:4937-46. 249. Hori S, Nomura T, Sakaguchi S. Control of regulatory T cell development by the transcription factor Foxp3. Science 2003;299:1057-61. 250. Weiss JM, Bilate AM, Gobert M, et al. Neuropilin 1 is expressed on thymus- derived natural regulatory T cells, but not mucosa-generated induced Foxp3+ T reg cells. J Exp Med 2012;209:1723-42, S1. 251. Yadav M, Louvet C, Davini D, et al. Neuropilin-1 distinguishes natural and inducible regulatory T cells among regulatory T cell subsets in vivo. J Exp Med 2012;209:1713-22, S1-19. 252. Cebula A, Seweryn M, Rempala GA, et al. Thymus-derived regulatory T cells contribute to tolerance to commensal microbiota. Nature 2013;497:258-62. 253. Lathrop SK, Bloom SM, Rao SM, et al. Peripheral education of the immune system by colonic commensal microbiota. Nature 2011;478:250-4. 254. Fontenot JD, Rasmussen JP, Gavin MA, et al. A function for interleukin 2 in Foxp3-expressing regulatory T cells.[Erratum appears in Nat Immunol. 2006 Apr;7(4):427]. Nature Immunology 2005;6:1142-51. 255. Beyer M, Schultze JL. Plasticity of T(reg) cells: is reprogramming of T(reg) cells possible in the presence of FOXP3? International Immunopharmacology 2011;11:555-60. 256. Allan SE, Song-Zhao GX, Abraham T, et al. Inducible reprogramming of human T cells into Treg cells by a conditionally active form of FOXP3. Eur J Immunol 2008;38:3282-9. 257. Li MO, Flavell RA. Contextual regulation of inflammation: a duet by transforming growth factor-beta and interleukin-10. Immunity 2008;28:468-76. 258. Fahlen L, Read S, Gorelik L, et al. T cells that cannot respond to TGF-beta escape control by CD4(+)CD25(+) regulatory T cells. Journal of Experimental Medicine 2005;201:737-46. 259. Kinoshita M, Takeda K. Microbial and dietary factors modulating intestinal regulatory T cell homeostasis. FEBS Letters 2014;588:4182-7. 260. Collison LW, Workman CJ, Kuo TT, et al. The inhibitory cytokine IL-35 contributes to regulatory T-cell function. Nature 2007;450:566-9. 261. Bardel E, Larousserie F, Charlot-Rabiega P, et al. Human CD4+ CD25+ Foxp3+ regulatory T cells do not constitutively express IL-35. J Immunol 2008;181:6898-905.

230 262. Bollrath J, Powrie FM. Controlling the frontier: regulatory T-cells and intestinal homeostasis. Semin Immunol 2013;25:352-7. 263. Vignali DA, Collison LW, Workman CJ. How regulatory T cells work. Nature Reviews. Immunology 2008;8:523-32. 264. Atarashi K, Tanoue T, Shima T, et al. Induction of colonic regulatory T cells by indigenous Clostridium species. Science 2011;331:337-41. 265. Furusawa Y, Obata Y, Fukuda S, et al. Commensal microbe-derived butyrate induces the differentiation of colonic regulatory T cells.[Erratum appears in Nature. 2014 Feb 13;506(7487):254]. Nature 2013;504:446-50. 266. Zhu J, Yamane H, Paul WE. Differentiation of effector CD4 T cell populations. Annu Rev Immunol 2010;28:445-89. 267. Korn T, Bettelli E, Oukka M, et al. IL-17 and Th17 Cells. Annu Rev Immunol 2009;27:485-517. 268. Annunziato F, Cosmi L, Liotta F, et al. Main features of human T helper 17 cells. Annals of the New York Academy of Sciences 2013;1284:66-70. 269. Lee YK, Turner H, Maynard CL, et al. Late developmental plasticity in the T helper 17 lineage. Immunity 2009;30:92-107. 270. Lexberg MH, Taubner A, Albrecht I, et al. IFN-gamma and IL-12 synergize to convert in vivo generated Th17 into Th1/Th17 cells. Eur J Immunol 2010;40:3017-27. 271. Cosmi L, Maggi L, Santarlasci V, et al. T helper cells plasticity in inflammation. Cytometry Part A: The Journal of the International Society for Analytical Cytology 2014;85:36-42. 272. Garrido-Mesa N, Algieri F, Rodriguez Nogales A, et al. Functional plasticity of Th17 cells: implications in gastrointestinal tract function. International Reviews of Immunology 2013;32:493-510. 273. Geginat J, Paroni M, Maglie S, et al. Plasticity of human CD4 T cell subsets. Front Immunol 2014;5:630. 274. Wong P, Pamer EG. CD8 T cell responses to infectious pathogens. Annu Rev Immunol 2003;21:29-70. 275. Russell JH, Ley TJ. Lymphocyte-mediated cytotoxicity. Annu Rev Immunol 2002;20:323-70. 276. Barber DL, Wherry EJ, Ahmed R. Cutting Edge: Rapid In Vivo Killing by Memory CD8 T Cells. The Journal of Immunology 2003;171:27-31. 277. Middendorp S, Nieuwenhuis EE. NKT cells in mucosal immunity. Mucosal immunology 2009;2:393-402. 278. Wei B, Wingender G, Fujiwara D, et al. Commensal microbiota and CD8+ T cells shape the formation of invariant NKT cells. Journal of Immunology 2010;184:1218-26. 279. Zeissig S, Blumberg RS. Commensal microbiota and NKT cells in the control of inflammatory diseases at mucosal surfaces. Current Opinion in Immunology 2013;25:690-6. 280. Olszak T, Neves JF, Dowds CM, et al. Protective mucosal immunity mediated by epithelial CD1d and IL-10. Nature 2014;509:497-502. 281. Dorn I, Schlenke P, Mascher B, et al. Lamina propria plasma cells in inflammatory bowel disease: intracellular detection of immunoglobulins using flow cytometry. Immunobiology 2002;206:546-57.

231 282. Medina F, Segundo C, Campos-Caro A, et al. Isolation, maturational level, and functional capacity of human colon lamina propria plasma cells. Gut 2003;52:383-9. 283. Fagarasan S, Kawamoto S, Kanagawa O, et al. Adaptive immune regulation in the gut: T cell-dependent and T cell-independent IgA synthesis. Annu Rev Immunol 2010;28:243-73. 284. Corthesy B. Roundtrip ticket for secretory IgA: role in mucosal homeostasis? Journal of Immunology 2007;178:27-32. 285. Macpherson AJ, McCoy KD, Johansen FE, et al. The immune geography of IgA induction and function. Mucosal immunology 2008;1:11-22. 286. Slack E, Balmer ML, Macpherson AJ. B cells as a critical node in the microbiota-host immune system network. Immunological Reviews 2014;260:50-66. 287. Fritz JH, Rojas OL, Simard N, et al. Acquisition of a multifunctional IgA+ plasma cell phenotype in the gut. Nature 2012;481:199-203. 288. Oka A, Ishihara S, Mishima Y, et al. Role of regulatory B cells in chronic intestinal inflammation: association with pathogenesis of Crohn's disease. Inflammatory Bowel Diseases 2014;20:315-28. 289. Sattler S, Ling GS, Xu D, et al. IL-10-producing regulatory B cells induced by IL-33 (Breg(IL-33)) effectively attenuate mucosal inflammatory responses in the gut. Journal of Autoimmunity 2014;50:107-22. 290. Mizoguchi A, Mizoguchi E, Takedatsu H, et al. Chronic intestinal inflammatory condition generates IL-10-producing regulatory B cell subset characterized by CD1d upregulation. Immunity 2002;16:219-30. 291. Mauri C, Bosma A. Immune regulatory function of B cells. Annu Rev Immunol 2012;30:221-41. 292. Frantz C, Stewart KM, Weaver VM. The extracellular matrix at a glance. J Cell Sci 2010;123:4195-200. 293. Adegboyega PA, Mifflin RC, DiMari JF, et al. Immunohistochemical study of myofibroblasts in normal colonic mucosa, hyperplastic polyps, and adenomatous colorectal polyps. Archives of Pathology & Laboratory Medicine 2002;126:829-36. 294. Rege TA, Hagood JS. Thy-1 as a regulator of cell-cell and cell-matrix interactions in axon regeneration, apoptosis, adhesion, migration, cancer, and fibrosis. FASEB Journal 2006;20:1045-54. 295. Vannucchi M-G, Traini C, Manetti M, et al. Telocytes express PDGFRα in the human gastrointestinal tract. Journal of Cellular and Molecular Medicine 2013;17:1099-1108. 296. Owens BM, Simmons A. Intestinal stromal cells in mucosal immunity and homeostasis. Mucosal immunology 2013;6:224-34. 297. Powell DW, Pinchuk IV, Saada JI, et al. Mesenchymal cells of the intestinal lamina propria. Annual Review of Physiology 2011;73:213-37. 298. Pinchuk IV, Mifflin RC, Saada JI, et al. Intestinal mesenchymal cells. Current Gastroenterology Reports 2010;12:310-8. 299. Okuno T, Andoh A, Bamba S, et al. Interleukin-1beta and tumor necrosis factor-alpha induce chemokine and matrix metalloproteinase gene expression in human colonic subepithelial myofibroblasts. Scand J Gastroenterol 2002;37:317-24.

232 300. Rogler G, Gelbmann CM, Vogl D, et al. Differential activation of cytokine secretion in primary human colonic fibroblast/myofibroblast cultures. Scand J Gastroenterol 2001;36:389-98. 301. Andoh A, Shioya M, Nishida A, et al. Expression of IL-24, an activator of the JAK1/STAT3/SOCS3 cascade, is enhanced in inflammatory bowel disease. J Immunol 2009;183:687-95. 302. Lahar N, Lei NY, Wang J, et al. Intestinal Subepithelial Myofibroblasts Support in vitro and in vivo Growth of Human Small Intestinal Epithelium. PLoS ONE 2011;6:e26898. 303. Samuel S, Walsh R, Webb J, et al. Characterization of putative stem cells in isolated human colonic crypt epithelial cells and their interactions with myofibroblasts. American Journal of Physiology - Cell Physiology 2009;296:C296-C305. 304. Kosinski C, Stange DE, Xu C, et al. Indian Hedgehog Regulates Intestinal Stem Cell Fate through Epithelial–Mesenchymal Interactions During Development. Gastroenterology 2010;139:893-903. 305. Büller NVJA, Rosekrans SL, Westerlund J, et al. Hedgehog Signaling and Maintenance of Homeostasis in the Intestinal Epithelium. Physiology 2012;27:148-155. 306. Zacharias WJ, Li X, Madison BB, et al. Hedgehog is an anti-inflammatory epithelial signal for the intestinal lamina propria. Gastroenterology 2010;138:2368-77, 2377.e1-4. 307. Pinchuk IV, Saada JI, Beswick EJ, et al. PD-1 ligand expression by human colonic myofibroblasts/fibroblasts regulates CD4+ T-cell activity. Gastroenterology 2008;135:1228-1237, 1237.e1-2. 308. Pinchuk IV, Beswick EJ, Saada JI, et al. Human colonic myofibroblasts promote expansion of CD4+ CD25high Foxp3+ regulatory T cells. Gastroenterology 2011;140:2019-30. 309. Yates J, Rovis F, Mitchell P, et al. The maintenance of human CD4+ CD25+ regulatory T cell function: IL-2, IL-4, IL-7 and IL-15 preserve optimal suppressive potency in vitro. Int Immunol 2007;19:785-99. 310. Huff KR, Akhtar LN, Fox AL, et al. Extracellular matrix-associated cytokines regulate CD4+ effector T-cell responses in the human intestinal mucosa. Mucosal immunology 2011;4:420-7. 311. Owens BM, Kaye PM. Stromal cell induction of regulatory dendritic cells. Front Immunol 2012;3:262. 312. Pang G, Couch L, Batey R, et al. GM-CSF, IL-1 alpha, IL-1 beta, IL-6, IL-8, IL- 10, ICAM-1 and VCAM-1 gene expression and cytokine production in human duodenal fibroblasts stimulated with lipopolysaccharide, IL-1 alpha and TNF-alpha. Clin Exp Immunol 1994;96:437-43. 313. Otte JM, Rosenberg IM, Podolsky DK. Intestinal myofibroblasts in innate immune responses of the intestine. Gastroenterology 2003;124:1866-78. 314. Owens BM, Steevels TA, Dudek M, et al. CD90(+) Stromal Cells are Non- Professional Innate Immune Effectors of the Human Colonic Mucosa. Front Immunol 2013;4:307. 315. https://www.crohnsandcolitis.com.au: Crohns and Colitis Australia, 2015. 316. Wilson J, Hair C, Knight R, et al. High incidence of inflammatory bowel disease in Australia: a prospective population-based Australian incidence study. Inflamm Bowel Dis 2010;16:1550-6.

233 317. Ng SC, Tang W, Ching JY, et al. Incidence and phenotype of inflammatory bowel disease based on results from the Asia-pacific Crohn's and colitis epidemiology study. Gastroenterology 2013;145:158-165.e2. 318. Rocchi A, Benchimol EI, Bernstein CN, et al. Inflammatory bowel disease: a Canadian burden of illness review. Canadian Journal of Gastroenterology 2012;26:811-7. 319. Bernstein CN, Wajda A, Svenson LW, et al. The epidemiology of inflammatory bowel disease in Canada: a population-based study.[Erratum appears in Am J Gastroenterol. 2006 Aug;101(8):1945]. American Journal of Gastroenterology 2006;101:1559-68. 320. Gearry RB, Richardson A, Frampton CM, et al. High incidence of Crohn's disease in Canterbury, New Zealand: results of an epidemiologic study. Inflammatory Bowel Diseases 2006;12:936-43. 321. Vind I, Riis L, Jess T, et al. Increasing incidences of inflammatory bowel disease and decreasing surgery rates in Copenhagen City and County, 2003- 2005: a population-based study from the Danish Crohn colitis database. American Journal of Gastroenterology 2006;101:1274-82. 322. Rubin GP, Hungin AP, Kelly PJ, et al. Inflammatory bowel disease: epidemiology and management in an English general practice population. Alimentary Pharmacology & Therapeutics 2000;14:1553-9. 323. Loftus CG, Loftus EV, Jr., Harmsen WS, et al. Update on the incidence and prevalence of Crohn's disease and ulcerative colitis in Olmsted County, Minnesota, 1940-2000. Inflammatory Bowel Diseases 2007;13:254-61. 324. Hiatt RA, Kaufman L. Epidemiology of inflammatory bowel disease in a defined northern California population. Western Journal of Medicine 1988;149:541-546. 325. The Health and Welfare of Australia's Aboriginal and Torres Strait Islander Peoples: Australian Bureau of Statistics, 2008. 326. Iyngkaran GHJT, T.; Bottolfsen, M; Tse, E.; Sivanesan, S.; Liew, D. P669. Inflammatory Bowel Disease is prevalent in Australia but rare in Indigenous Australians. Journal of Crohns and Colitis 2015;9:S415-S416. 327. Nguyen GC, Chong CA, Chong RY. National estimates of the burden of inflammatory bowel disease among racial and ethnic groups in the United States. Journal of Crohn's & colitis 2014;8:288-95. 328. Shivananda S, Lennard-Jones J, Logan R, et al. Incidence of inflammatory bowel disease across Europe: is there a difference between north and south? Results of the European Collaborative Study on Inflammatory Bowel Disease (EC-IBD). Gut 1996;39:690-7. 329. Nerich V, Monnet E, Etienne A, et al. Geographical variations of inflammatory bowel disease in France: a study based on national health insurance data. Inflamm Bowel Dis 2006;12:218-26. 330. Burisch J. Crohn's disease and ulcerative colitis. Occurrence, course and prognosis during the first year of disease in a European population-based inception cohort. Danish Medical Journal 2014;61:B4778. 331. Sonnenberg A, Wasserman IH. Epidemiology of inflammatory bowel disease among U.S. military veterans. Gastroenterology 1991;101:122-30. 332. Cosnes J, Gower-Rousseau C, Seksik P, et al. Epidemiology and natural history of inflammatory bowel diseases. Gastroenterology 2011;140:1785- 94.

234 333. Prideaux L, Kamm MA, De Cruz PP, et al. Inflammatory bowel disease in Asia: a systematic review. J Gastroenterol Hepatol 2012;27:1266-80. 334. Pezerovic D, Klarin I, Zulj M, et al. Incidence and prevalence of inflammatory bowel disease in Vukovarsko-Srijemska County, Croatia, 1991-2000 and 2001-2010: a population-based study. Collegium Antropologicum 2014;38:115-23. 335. Mahid SS, Minor KS, Soto RE, et al. Smoking and inflammatory bowel disease: a meta-analysis. Mayo Clin Proc 2006;81:1462-71. 336. Lakatos PL, Szalay F, Tulassay Z, et al. Clinical presentation of Crohn's disease. association between familial disease, smoking, disease phenotype, extraintestinal manifestations and need for surgery. Hepatogastroenterology 2005;52:817-22. 337. De Cruz P, Kamm MA, Hamilton AL, et al. Crohn's disease management after intestinal resection: a randomised trial. Lancet 2014. 338. Vester-Andersen MK, Vind I, Prosberg MV, et al. Hospitalisation, surgical and medical recurrence rates in inflammatory bowel disease 2003-2011-a Danish population-based cohort study. J Crohns Colitis 2014;8:1675-83. 339. Ott C, Takses A, Obermeier F, et al. Smoking increases the risk of extraintestinal manifestations in Crohn's disease. World J Gastroenterol 2014;20:12269-76. 340. Cosnes J, Nion-Larmurier I, Afchain P, et al. Gender differences in the response of colitis to smoking. Clinical Gastroenterology & Hepatology 2004;2:41-8. 341. Birrenbach T, Bocker U. Inflammatory bowel disease and smoking: a review of epidemiology, pathophysiology, and therapeutic implications. Inflamm Bowel Dis 2004;10:848-59. 342. Cosnes J. Tobacco and IBD: relevance in the understanding of disease mechanisms and clinical practice. Best Pract Res Clin Gastroenterol 2004;18:481-96. 343. Biedermann L, Brulisauer K, Zeitz J, et al. Smoking cessation alters intestinal microbiota: insights from quantitative investigations on human fecal samples using FISH. Inflamm Bowel Dis 2014;20:1496-501. 344. Monick MM, Powers LS, Walters K, et al. Identification of an autophagy defect in smokers' alveolar macrophages. J Immunol 2010;185:5425-35. 345. Cosnes J, Beaugerie L, Carbonnel F, et al. Smoking cessation and the course of Crohn's disease: an intervention study. Gastroenterology 2001;120:1093-9. 346. Franke A, McGovern DPB, Barrett JC, et al. Meta-Analysis Increases to 71 the Tally of Confirmed Crohn’s Disease Susceptibility Loci. Nature genetics 2010;42:1118-1125. 347. Khor B, Gardet A, Xavier RJ. Genetics and pathogenesis of inflammatory bowel disease. Nature 2011;474:307-317. 348. Jostins L, Ripke S, Weersma RK, et al. Host-microbe interactions have shaped the genetic architecture of inflammatory bowel disease. Nature 2012;491:119-24. 349. Anderson CA, Boucher G, Lees CW, et al. Meta-analysis identifies 29 additional ulcerative colitis risk loci, increasing the number of confirmed associations to 47. Nat Genet 2011;43:246-52.

235 350. Halme L, Paavola-Sakki P, Turunen U, et al. Family and twin studies in inflammatory bowel disease. World J Gastroenterol 2006;12:3668-72. 351. Spehlmann ME, Begun AZ, Burghardt J, et al. Epidemiology of inflammatory bowel disease in a German twin cohort: results of a nationwide study. Inflamm Bowel Dis 2008;14:968-76. 352. Hugot JP, Laurent-Puig P, Gower-Rousseau C, et al. Mapping of a susceptibility locus for Crohn's disease on chromosome 16. Nature 1996;379:821-3. 353. Hugot JP, Chamaillard M, Zouali H, et al. Association of NOD2 leucine-rich repeat variants with susceptibility to Crohn's disease. Nature 2001;411:599-603. 354. Ogura Y, Bonen DK, Inohara N, et al. A frameshift mutation in NOD2 associated with susceptibility to Crohn's disease. Nature 2001;411:603-6. 355. Inohara N, Ogura Y, Nunez G. Nods: a family of cytosolic proteins that regulate the host response to pathogens. Curr Opin Microbiol 2002;5:76-80. 356. Kobayashi KS, Chamaillard M, Ogura Y, et al. Nod2-dependent regulation of innate and adaptive immunity in the intestinal tract. Science 2005;307:731- 4. 357. Gutierrez O, Pipaon C, Inohara N, et al. Induction of Nod2 in myelomonocytic and intestinal epithelial cells via nuclear factor-kappa B activation. J Biol Chem 2002;277:41701-5. 358. Rosenstiel P, Fantini M, Brautigam K, et al. TNF-alpha and IFN-gamma regulate the expression of the NOD2 (CARD15) gene in human intestinal epithelial cells. Gastroenterology 2003;124:1001-9. 359. Lala S, Ogura Y, Osborne C, et al. Crohn's disease and the NOD2 gene: a role for paneth cells. Gastroenterology 2003;125:47-57. 360. Lesage S, Zouali H, Cezard JP, et al. CARD15/NOD2 mutational analysis and genotype-phenotype correlation in 612 patients with inflammatory bowel disease. Am J Hum Genet 2002;70:845-57. 361. Bhullar M, Macrae F, Brown G, et al. Prediction of Crohn's disease aggression through NOD2/CARD15 gene sequencing in an Australian cohort. World J Gastroenterol 2014;20:5008-16. 362. Solon JG, Burke JP, Walsh SR, et al. The effect of NOD2 polymorphism on postsurgical recurrence in Crohn's disease: a systematic review and meta- analysis of available literature. Inflamm Bowel Dis 2013;19:1099-105. 363. Abreu MT, Taylor KD, Lin YC, et al. Mutations in NOD2 are associated with fibrostenosing disease in patients with Crohn's disease. Gastroenterology 2002;123:679-88. 364. Hampe J, Franke A, Rosenstiel P, et al. A genome-wide association scan of nonsynonymous SNPs identifies a susceptibility variant for Crohn disease in ATG16L1. Nature Genetics 2007;39:207-11. 365. Rioux JD, Xavier RJ, Taylor KD, et al. Genome-wide association study identifies new susceptibility loci for Crohn disease and implicates autophagy in disease pathogenesis. Nature Genetics 2007;39:596-604. 366. Ellinghaus D, Zhang H, Zeissig S, et al. Association Between Variants of PRDM1 and NDP52 and Crohn’s Disease, Based on Exome Sequencing and Functional Studies. Gastroenterology 2013;145:339-347.

236 367. Parkes M, Barrett JC, Prescott NJ, et al. Sequence variants in the autophagy gene IRGM and multiple other replicating loci contribute to Crohn's disease susceptibility. Nature Genetics 2007;39:830-2. 368. Wellcome Trust Case Control C. Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature 2007;447:661-78. 369. Duerr RH, Taylor KD, Brant SR, et al. A genome-wide association study identifies IL23R as an inflammatory bowel disease gene. Science 2006;314:1461-3. 370. Di Meglio P, Di Cesare A, Laggner U, et al. The IL23R R381Q Gene Variant Protects against Immune-Mediated Diseases by Impairing IL-23-Induced Th17 Effector Response in Humans. PLoS ONE 2011;6:e17160. 371. Dubinsky MC, Wang D, Picornell Y, et al. IL-23 Receptor (IL-23R) Gene Protects Against Pediatric Crohn’s Disease. Inflammatory bowel diseases 2007;13:511-515. 372. Zwiers A, Kraal L, van de Pouw Kraan TC, et al. Cutting edge: a variant of the IL-23R gene associated with inflammatory bowel disease induces loss of microRNA regulation and enhanced protein production. Journal of Immunology 2012;188:1573-7. 373. Taurog JD, Richardson JA, Croft JT, et al. The germfree state prevents development of gut and joint inflammatory disease in HLA-B27 transgenic rats. J Exp Med 1994;180:2359-64. 374. Rath HC, Herfarth HH, Ikeda JS, et al. Normal luminal bacteria, especially Bacteroides species, mediate chronic colitis, gastritis, and arthritis in HLA- B27/human beta2 microglobulin transgenic rats. J Clin Invest 1996;98:945- 53. 375. Schultz M, Tonkonogy SL, Sellon RK, et al. IL-2-deficient mice raised under germfree conditions develop delayed mild focal intestinal inflammation. Am J Physiol 1999;276:G1461-72. 376. Zelas P, Jagelman DG. Loop illeostomy in the management of Crohn's colitis in the debilitated patient. Ann Surg 1980;191:164-8. 377. Harper PH, Truelove SC, Lee EC, et al. Split ileostomy and ileocolostomy for Crohn's disease of the colon and ulcerative colitis: a 20 year survey. Gut 1983;24:106-13. 378. Rutgeerts P, Goboes K, Peeters M, et al. Effect of faecal stream diversion on recurrence of Crohn's disease in the neoterminal ileum. Lancet 1991;338:771-4. 379. D'Haens GR, Geboes K, Peeters M, et al. Early lesions of recurrent Crohn's disease caused by infusion of intestinal contents in excluded ileum. Gastroenterology 1998;114:262-7. 380. Eckburg PB, Bik EM, Bernstein CN, et al. Diversity of the human intestinal microbial flora. Science 2005;308:1635-8. 381. Qin J, Li R, Raes J, et al. A human gut microbial gene catalogue established by metagenomic sequencing. Nature 2010;464:59-65. 382. Caporaso JG, Lauber CL, Costello EK, et al. Moving pictures of the human microbiome. Genome Biol 2011;12:R50. 383. Tamboli CP, Neut C, Desreumaux P, et al. Dysbiosis in inflammatory bowel disease. Gut 2004;53:1-4.

237 384. Dalal SR, Chang EB. The microbial basis of inflammatory bowel diseases. Journal of Clinical Investigation 2014;124:4190-6. 385. Bibiloni R, Mangold M, Madsen KL, et al. The bacteriology of biopsies differs between newly diagnosed, untreated, Crohn's disease and ulcerative colitis patients. J Med Microbiol 2006;55:1141-9. 386. Gophna U, Sommerfeld K, Gophna S, et al. Differences between tissue- associated intestinal microfloras of patients with Crohn's disease and ulcerative colitis. J Clin Microbiol 2006;44:4136-41. 387. Ott SJ, Musfeldt M, Wenderoth DF, et al. Reduction in diversity of the colonic mucosa associated bacterial microflora in patients with active inflammatory bowel disease. Gut 2004;53:685-93. 388. Sokol H, Seksik P, Furet JP, et al. Low counts of Faecalibacterium prausnitzii in colitis microbiota. Inflamm Bowel Dis 2009;15:1183-9. 389. Frank DN, St. Amand AL, Feldman RA, et al. Molecular-phylogenetic characterization of microbial community imbalances in human inflammatory bowel diseases. Proceedings of the National Academy of Sciences of the United States of America 2007;104:13780-13785. 390. Rajca S, Grondin V, Louis E, et al. Alterations in the intestinal microbiome (dysbiosis) as a predictor of relapse after infliximab withdrawal in Crohn's disease. Inflammatory Bowel Diseases 2014;20:978-86. 391. Gevers D, Kugathasan S, Denson LA, et al. The treatment-naive microbiome in new-onset Crohn's disease. Cell Host & Microbe 2014;15:382-92. 392. Sokol H, Pigneur B, Watterlot L, et al. Faecalibacterium prausnitzii is an anti-inflammatory commensal bacterium identified by gut microbiota analysis of Crohn disease patients. Proc Natl Acad Sci U S A 2008;105:16731-6. 393. Qiu X, Zhang M, Yang X, et al. Faecalibacterium prausnitzii upregulates regulatory T cells and anti-inflammatory cytokines in treating TNBS- induced colitis. J Crohns Colitis 2013;7:e558-68. 394. Zhang M, Qiu X, Zhang H, et al. Faecalibacterium prausnitzii inhibits interleukin-17 to ameliorate colorectal colitis in rats. PLoS One 2014;9:e109146. 395. Seksik P, Rigottier-Gois L, Gramet G, et al. Alterations of the dominant faecal bacterial groups in patients with Crohn's disease of the colon. Gut 2003;52:237-42. 396. Baumgart M, Dogan B, Rishniw M, et al. Culture independent analysis of ileal mucosa reveals a selective increase in invasive Escherichia coli of novel phylogeny relative to depletion of Clostridiales in Crohn's disease involving the ileum. Isme j 2007;1:403-18. 397. Willing B, Halfvarson J, Dicksved J, et al. Twin studies reveal specific imbalances in the mucosa-associated microbiota of patients with ileal Crohn's disease. Inflamm Bowel Dis 2009;15:653-60. 398. Agus A, Massier S, Darfeuille-Michaud A, et al. Understanding host- adherent-invasive Escherichia coli interaction in Crohn's disease: opening up new therapeutic strategies. Biomed Res Int 2014;2014:567929. 399. Darfeuille-Michaud A, Boudeau J, Bulois P, et al. High prevalence of adherent-invasive Escherichia coli associated with ileal mucosa in Crohn's disease. Gastroenterology 2004;127:412-21.

238 400. Nguyen HT, Dalmasso G, Muller S, et al. Crohn's disease-associated adherent invasive Escherichia coli modulate levels of microRNAs in intestinal epithelial cells to reduce autophagy. Gastroenterology 2014;146:508-19. 401. Walker AW, Sanderson JD, Churcher C, et al. High-throughput clone library analysis of the mucosa-associated microbiota reveals dysbiosis and differences between inflamed and non-inflamed regions of the intestine in inflammatory bowel disease. BMC Microbiol 2011;11:7. 402. Zitomersky NL, Atkinson BJ, Franklin SW, et al. Characterization of adherent bacteroidales from intestinal biopsies of children and young adults with inflammatory bowel disease. PLoS One 2013;8:e63686. 403. Swidsinski A, Weber J, Loening-Baucke V, et al. Spatial organization and composition of the mucosal flora in patients with inflammatory bowel disease. J Clin Microbiol 2005;43:3380-9. 404. Wang W, Chen L, Zhou R, et al. Increased proportions of Bifidobacterium and the Lactobacillus group and loss of butyrate-producing bacteria in inflammatory bowel disease. Journal of Clinical Microbiology 2014;52:398- 406. 405. Zhang L, Lee H, Grimm MC, et al. Campylobacter concisus and inflammatory bowel disease. World Journal of Gastroenterology 2014;20:1259-67. 406. Ott SJ, Kuhbacher T, Musfeldt M, et al. Fungi and inflammatory bowel diseases: Alterations of composition and diversity. Scand J Gastroenterol 2008;43:831-41. 407. Li Q, Wang C, Tang C, et al. Dysbiosis of gut fungal microbiota is associated with mucosal inflammation in Crohn's disease. Journal of Clinical Gastroenterology 2014;48:513-23. 408. Norman JM, Handley SA, Baldridge MT, et al. Disease-specific alterations in the enteric virome in inflammatory bowel disease. Cell 2015;160:447-60. 409. De Cruz P, Prideaux L, Wagner J, et al. Characterization of the gastrointestinal microbiota in health and inflammatory bowel disease. Inflamm Bowel Dis 2012;18:372-90. 410. Prideaux L, Kang S, Wagner J, et al. Impact of ethnicity, geography, and disease on the microbiota in health and inflammatory bowel disease. Inflammatory Bowel Diseases 2013;19:2906-18. 411. Biedermann L, Zeitz J, Mwinyi J, et al. Smoking cessation induces profound changes in the composition of the intestinal microbiota in humans. PLoS One 2013;8:e59260. 412. Ananthakrishnan AN, Khalili H, Konijeti GG, et al. A prospective study of long-term intake of dietary fiber and risk of Crohn's disease and ulcerative colitis. Gastroenterology 2013;145:970-7. 413. Ghouri YA, Richards DM, Rahimi EF, et al. Systematic review of randomized controlled trials of probiotics, prebiotics, and synbiotics in inflammatory bowel disease. Clin Exp Gastroenterol 2014;7:473-87. 414. Jonkers D, Penders J, Masclee A, et al. Probiotics in the management of inflammatory bowel disease: a systematic review of intervention studies in adult patients. Drugs 2012;72:803-23. 415. Ianiro G, Bibbo S, Scaldaferri F, et al. Fecal microbiota transplantation in inflammatory bowel disease: beyond the excitement. Medicine (Baltimore) 2014;93:e97.

239 416. Colman RJ, Rubin DT. Fecal microbiota transplantation as therapy for inflammatory bowel disease: a systematic review and meta-analysis. J Crohns Colitis 2014;8:1569-81. 417. Anderson JL, Edney RJ, Whelan K. Systematic review: faecal microbiota transplantation in the management of inflammatory bowel disease. Aliment Pharmacol Ther 2012;36:503-16. 418. Ukabam SO, Clamp JR, Cooper BT. Abnormal small intestinal permeability to sugars in patients with Crohn's disease of the terminal ileum and colon. Digestion 1983;27:70-4. 419. Pearson AD, Eastham EJ, Laker MF, et al. Intestinal permeability in children with Crohn's disease and coeliac disease. Br Med J (Clin Res Ed) 1982;285:20-1. 420. Hollander D, Vadheim CM, Brettholz E, et al. Increased intestinal permeability in patients with Crohn's disease and their relatives. A possible etiologic factor. Ann Intern Med 1986;105:883-5. 421. May GR, Sutherland LR, Meddings JB. Is small intestinal permeability really increased in relatives of patients with Crohn's disease? Gastroenterology 1993;104:1627-32. 422. Yacyshyn BR, Meddings JB. CD45RO expression on circulating CD19+ B cells in Crohn's disease correlates with intestinal permeability. Gastroenterology 1995;108:132-7. 423. Breslin NP, Nash C, Hilsden RJ, et al. Intestinal permeability is increased in a proportion of spouses of patients with Crohn's disease. Am J Gastroenterol 2001;96:2934-8. 424. Soderholm JD, Olaison G, Lindberg E, et al. Different intestinal permeability patterns in relatives and spouses of patients with Crohn's disease: an inherited defect in mucosal defence? Gut 1999;44:96-100. 425. Peeters M, Geypens B, Claus D, et al. Clustering of increased small intestinal permeability in families with Crohn's disease. Gastroenterology 1997;113:802-7. 426. Buhner S, Buning C, Genschel J, et al. Genetic basis for increased intestinal permeability in families with Crohn's disease: role of CARD15 3020insC mutation? Gut 2006;55:342-7. 427. D'Inca R, Annese V, di Leo V, et al. Increased intestinal permeability and NOD2 variants in familial and sporadic Crohn's disease. Aliment Pharmacol Ther 2006;23:1455-61. 428. Wehkamp J, Harder J, Weichenthal M, et al. NOD2 (CARD15) mutations in Crohn's disease are associated with diminished mucosal alpha-defensin expression. Gut 2004;53:1658-64. 429. Perminow G, Beisner J, Koslowski M, et al. Defective paneth cell-mediated host defense in pediatric ileal Crohn's disease. Am J Gastroenterol 2010;105:452-9. 430. Zeissig S, Burgel N, Gunzel D, et al. Changes in expression and distribution of claudin 2, 5 and 8 lead to discontinuous tight junctions and barrier dysfunction in active Crohn's disease. Gut 2007;56:61-72. 431. Soderholm JD, Olaison G, Peterson KH, et al. Augmented increase in tight junction permeability by luminal stimuli in the non-inflamed ileum of Crohn's disease. Gut 2002;50:307-13.

240 432. Wyatt J, Vogelsang H, Hubl W, et al. Intestinal permeability and the prediction of relapse in Crohn's disease. Lancet 1993;341:1437-9. 433. D'Inca R, Di Leo V, Corrao G, et al. Intestinal permeability test as a predictor of clinical course in Crohn's disease. Am J Gastroenterol 1999;94:2956-60. 434. Arnott ID, Kingstone K, Ghosh S. Abnormal intestinal permeability predicts relapse in inactive Crohn disease. Scand J Gastroenterol 2000;35:1163-9. 435. Tibble JA, Sigthorsson G, Bridger S, et al. Surrogate markers of intestinal inflammation are predictive of relapse in patients with inflammatory bowel disease. Gastroenterology 2000;119:15-22. 436. Zeissig S, Bojarski C, Buergel N, et al. Downregulation of epithelial apoptosis and barrier repair in active Crohn's disease by tumour necrosis factor alpha antibody treatment. Gut 2004;53:1295-302. 437. Suenaert P, Bulteel V, Lemmens L, et al. Anti-tumor necrosis factor treatment restores the gut barrier in Crohn's disease. Am J Gastroenterol 2002;97:2000-4. 438. Suenaert P, Maerten P, Van Assche G, et al. Effects of T cell-induced colonic inflammation on epithelial barrier function. Inflamm Bowel Dis 2010;16:1322-31. 439. Welcker K, Martin A, Kolle P, et al. Increased intestinal permeability in patients with inflammatory bowel disease. Eur J Med Res 2004;9:456-60. 440. Arslan G, Atasever T, Cindoruk M, et al. (51)CrEDTA colonic permeability and therapy response in patients with ulcerative colitis. Nucl Med Commun 2001;22:997-1001. 441. Heller F, Florian P, Bojarski C, et al. Interleukin-13 is the key effector Th2 cytokine in ulcerative colitis that affects epithelial tight junctions, apoptosis, and cell restitution. Gastroenterology 2005;129:550-64. 442. Heller F, Fromm A, Gitter AH, et al. Epithelial apoptosis is a prominent feature of the epithelial barrier disturbance in intestinal inflammation: effect of pro-inflammatory interleukin-13 on epithelial cell function. Mucosal Immunol 2008;1 Suppl 1:S58-61. 443. Fuss IJ, Heller F, Boirivant M, et al. Nonclassical CD1d-restricted NK T cells that produce IL-13 characterize an atypical Th2 response in ulcerative colitis. J Clin Invest 2004;113:1490-7. 444. Gilat T, Hacohen D, Lilos P, et al. Childhood factors in ulcerative colitis and Crohn's disease. An international cooperative study. Scandinavian Journal of Gastroenterology 1987;22:1009-24. 445. Rutgeerts P, D'Haens G, Hiele M, et al. Appendectomy protects against ulcerative colitis. Gastroenterology 1994;106:1251-3. 446. Koutroubakis IE, Vlachonikolis IG, Kouroumalis EA. Role of appendicitis and appendectomy in the pathogenesis of ulcerative colitis: a critical review. Inflammatory Bowel Diseases 2002;8:277-86. 447. Frisch M, Pedersen BV, Andersson RE. Appendicitis, mesenteric lymphadenitis, and subsequent risk of ulcerative colitis: cohort studies in Sweden and Denmark. Bmj 2009;338:b716. 448. Andersson RE, Olaison G, Tysk C, et al. Appendectomy and protection against ulcerative colitis. N Engl J Med 2001;344:808-14. 449. Gardenbroek TJ, Eshuis EJ, Ponsioen CI, et al. The effect of appendectomy on the course of ulcerative colitis: a systematic review. Colorectal Disease 2012;14:545-53.

241 450. Okazaki K, Onodera‡ H, Watanabe N, et al. A patient with improvement of ulcerative colitis after appendectomy. Gastroenterology 2000;119:502-506. 451. Järnerot G, Andersson M, Franzén L. Laparoscopic appendectomy in patients with refractory ulcerative colitis. Gastroenterology 2001;120:1562-1563. 452. Bolin TD, Wong S, Crouch R, et al. Appendicectomy as a Therapy for Ulcerative Proctitis. Am J Gastroenterol 2009;104:2476-2482. 453. Kaplan GG, Jackson T, Sands BE, et al. The risk of developing Crohn's disease after an appendectomy: a meta-analysis. Am J Gastroenterol 2008;103:2925-31. 454. Kaplan GG, Pedersen BV, Andersson RE, et al. The risk of developing Crohn's disease after an appendectomy: a population-based cohort study in Sweden and Denmark. Gut 2007;56:1387-92. 455. Cosnes J, Seksik P, Nion-Larmurier I, et al. Prior appendectomy and the phenotype and course of Crohn's disease. World J Gastroenterol 2006;12:1235-42. 456. Eroschenko VP. Di Fiore's atlas of histology with functional correlations. Baltimore: Williams & Wilkins, 2012. 457. Spencer J, Finn T, Isaacson PG. Gut associated lymphoid tissue: a morphological and immunocytochemical study of the human appendix. Gut 1985;26:672-679. 458. Mizumoto T. B and T cells in lymphoid tissues of human appendix. Int Arch Allergy Appl Immunol 1976;51:80-93. 459. Mizoguchi A, Mizoguchi E, Chiba C, et al. Role of appendix in the development of inflammatory bowel disease in TCR-alpha mutant mice. J Exp Med 1996;184:707-15. 460. Krieglstein CF, Cerwinka WH, Laroux FS, et al. Role of appendix and spleen in experimental colitis. J Surg Res 2001;101:166-75. 461. Farkas SA, Hornung M, Sattler C, et al. Preferential migration of CD62L cells into the appendix in mice with experimental chronic colitis. Eur Surg Res 2005;37:115-22. 462. Watson Ng WS, Hampartzoumian T, Lloyd AR, et al. A murine model of appendicitis and the impact of inflammation on appendiceal lymphocyte constituents. Clinical & Experimental Immunology 2007;150:169-78. 463. Cheluvappa R, Luo AS, Grimm MC. Autophagy suppression by appendicitis and appendectomy protects against colitis. Inflammatory Bowel Diseases 2014;20:847-55. 464. Cheluvappa R, Luo AS, Grimm MC. T helper type 17 pathway suppression by appendicitis and appendectomy protects against colitis. Clinical & Experimental Immunology 2014;175:316-22. 465. Cohen T, Pfeffer RB, Valensi Q. "Ulcerative appendicitis" occurring as a skip lesion in chronic ulcerative colitis; report of a case. Am J Gastroenterol 1974;62:151-5. 466. Park SH, Loftus EV, Jr., Yang SK. Appendiceal skip inflammation and ulcerative colitis. Dig Dis Sci 2014;59:2050-7. 467. D'Haens G, Geboes K, Peeters M, et al. Patchy cecal inflammation associated with distal ulcerative colitis: a prospective endoscopic study. Am J Gastroenterol 1997;92:1275-9.

242 468. Yang SK, Jung HY, Kang GH, et al. Appendiceal orifice inflammation as a skip lesion in ulcerative colitis: an analysis in relation to medical therapy and disease extent. Gastrointest Endosc 1999;49:743-7. 469. Groisman GM, George J, Harpaz N. Ulcerative appendicitis in universal and nonuniversal ulcerative colitis. Mod Pathol 1994;7:322-5. 470. Byeon JS, Yang SK, Myung SJ, et al. Clinical course of distal ulcerative colitis in relation to appendiceal orifice inflammation status. Inflamm Bowel Dis 2005;11:366-71. 471. Duerr RH, Targan SR, Landers CJ, et al. Anti-neutrophil cytoplasmic antibodies in ulcerative colitis. Comparison with other colitides/diarrheal illnesses. Gastroenterology 1991;100:1590-6. 472. Rump JA, Scholmerich J, Gross V, et al. A new type of perinuclear anti- neutrophil cytoplasmic antibody (p-ANCA) in active ulcerative colitis but not in Crohn's disease. Immunobiology 1990;181:406-13. 473. Eggena M, Cohavy O, Parseghian MH, et al. Identification of histone H1 as a cognate antigen of the ulcerative colitis-associated marker antibody pANCA. J Autoimmun 2000;14:83-97. 474. McKenzie H, Main J, Pennington CR, et al. Antibody to selected strains of Saccharomyces cerevisiae (baker's and brewer's yeast) and Candida albicans in Crohn's disease. Gut 1990;31:536-8. 475. Main J, McKenzie H, Yeaman GR, et al. Antibody to Saccharomyces cerevisiae (bakers' yeast) in Crohn's disease. Bmj 1988;297:1105-6. 476. Joossens S, Reinisch W, Vermeire S, et al. The value of serologic markers in indeterminate colitis: a prospective follow-up study. Gastroenterology 2002;122:1242-7. 477. Prideaux L, De Cruz P, Ng SC, et al. Serological antibodies in inflammatory bowel disease: a systematic review. Inflammatory Bowel Diseases 2012;18:1340-55. 478. Cohavy O, Bruckner D, Gordon LK, et al. Colonic bacteria express an ulcerative colitis pANCA-related protein epitope. Infect Immun 2000;68:1542-8. 479. Lodes MJ, Cong Y, Elson CO, et al. Bacterial flagellin is a dominant antigen in Crohn disease. J Clin Invest 2004;113:1296-306. 480. Sutton CL, Kim J, Yamane A, et al. Identification of a novel bacterial sequence associated with Crohn's disease. Gastroenterology 2000;119:23- 31. 481. Landers CJ, Cohavy O, Misra R, et al. Selected loss of tolerance evidenced by Crohn's disease-associated immune responses to auto- and microbial antigens. Gastroenterology 2002;123:689-99. 482. Targan SR, Landers CJ, Yang H, et al. Antibodies to CBir1 flagellin define a unique response that is associated independently with complicated Crohn's disease. Gastroenterology 2005;128:2020-8. 483. Mow WS, Vasiliauskas EA, Lin YC, et al. Association of antibody responses to microbial antigens and complications of small bowel Crohn's disease. Gastroenterology 2004;126:414-24. 484. Forcione DG, Rosen MJ, Kisiel JB, et al. Anti-Saccharomyces cerevisiae antibody (ASCA) positivity is associated with increased risk for early surgery in Crohn's disease. Gut 2004;53:1117-22.

243 485. Dubinsky MC, Kugathasan S, Mei L, et al. Increased immune reactivity predicts aggressive complicating Crohn's disease in children. Clin Gastroenterol Hepatol 2008;6:1105-11. 486. Vester-Andersen MK, Prosberg MV, Jess T, et al. Disease Course and Surgery Rates in Inflammatory Bowel Disease: A Population-Based, 7-Year Follow- Up Study in the Era of Immunomodulating Therapy. Am J Gastroenterol 2014;109:705-714. 487. Peyrin-Biroulet L, Ferrante M, Magro F, et al. Results from the 2nd Scientific Workshop of the ECCO. I: Impact of mucosal healing on the course of inflammatory bowel disease. J Crohns Colitis 2011;5:477-83. 488. Rungoe C, Langholz E, Andersson M, et al. Changes in medical treatment and surgery rates in inflammatory bowel disease: a nationwide cohort study 1979-2011. Gut 2014;63:1607-16. 489. Kruis W, Katalinic A, Klugmann T, et al. Predictive factors for an uncomplicated long-term course of Crohn's disease: a retrospective analysis. Journal of Crohn's & colitis 2013;7:e263-70. 490. MacDermott RP, Schloemann SR, Bertovich MJ, et al. Inhibition of antibody secretion by 5-aminosalicylic acid. Gastroenterology 1989;96:442-8. 491. Bantel H, Berg C, Vieth M, et al. Mesalazine inhibits activation of transcription factor NF-kappaB in inflamed mucosa of patients with ulcerative colitis. Am J Gastroenterol 2000;95:3452-7. 492. Sharon P, Ligumsky M, Rachmilewitz D, et al. Role of prostaglandins in ulcerative colitis. Enhanced production during active disease and inhibition by sulfasalazine. Gastroenterology 1978;75:638-40. 493. Stenson WF, Lobos E. Sulfasalazine inhibits the synthesis of chemotactic lipids by neutrophils. J Clin Invest 1982;69:494-7. 494. Neal TM, Winterbourn CC, Vissers MC. Inhibition of neutrophil degranulation and superoxide production by sulfasalazine. Comparison with 5-aminosalicylic acid, sulfapyridine and olsalazine. Biochem Pharmacol 1987;36:2765-8. 495. Craven PA, Pfanstiel J, Saito R, et al. Actions of sulfasalazine and 5- aminosalicylic acid as reactive oxygen scavengers in the suppression of bile acid-induced increases in colonic epithelial cell loss and proliferative activity. Gastroenterology 1987;92:1998-2008. 496. Stevens C, Lipman M, Fabry S, et al. 5-Aminosalicylic acid abrogates T-cell proliferation by blocking interleukin-2 production in peripheral blood mononuclear cells. J Pharmacol Exp Ther 1995;272:399-406. 497. Targownik LE, Nugent Z, Singh H, et al. Prevalence of and outcomes associated with corticosteroid prescription in inflammatory bowel disease. Inflammatory Bowel Diseases 2014;20:622-30. 498. Travis SP, Danese S, Kupcinskas L, et al. Once-daily budesonide MMX in active, mild-to-moderate ulcerative colitis: results from the randomised CORE II study. Gut 2014;63:433-41. 499. McDonald JW, Wang Y, Tsoulis DJ, et al. Methotrexate for induction of remission in refractory Crohn's disease. Cochrane Database of Systematic Reviews 2014;8:CD003459. 500. Targan SR, Feagan BG, Fedorak RN, et al. Natalizumab for the treatment of active Crohn's disease: results of the ENCORE Trial. Gastroenterology 2007;132:1672-83.

244 501. Sandborn WJ, Colombel JF, Enns R, et al. Natalizumab induction and maintenance therapy for Crohn's disease. New England Journal of Medicine 2005;353:1912-25. 502. Feagan BG, Rutgeerts P, Sands BE, et al. Vedolizumab as induction and maintenance therapy for ulcerative colitis. N Engl J Med 2013;369:699-710. 503. Sands BE, Feagan BG, Rutgeerts P, et al. Effects of vedolizumab induction therapy for patients with Crohn's disease in whom tumor necrosis factor antagonist treatment failed. Gastroenterology 2014;147:618-627 e3. 504. Van Assche G, Van Ranst M, Sciot R, et al. Progressive multifocal leukoencephalopathy after natalizumab therapy for Crohn's disease. New England Journal of Medicine 2005;353:362-8. 505. Danese S. New therapies for inflammatory bowel disease: from the bench to the bedside. Gut 2012;61:918-32. 506. Monteleone G, Neurath MF, Ardizzone S, et al. Mongersen, an Oral SMAD7 Antisense Oligonucleotide, and Crohn's Disease. N Engl J Med 2015;372:1104-1113. 507. Zorzi F, Calabrese E, Monteleone I, et al. A phase 1 open-label trial shows that smad7 antisense oligonucleotide (GED0301) does not increase the risk of small bowel strictures in Crohn's disease. Aliment Pharmacol Ther 2012;36:850-7. 508. Monteleone G, Fantini MC, Onali S, et al. Phase I clinical trial of Smad7 knockdown using antisense oligonucleotide in patients with active Crohn's disease. Mol Ther 2012;20:870-6. 509. Sandborn WJ, Gasink C, Gao L-L, et al. Ustekinumab Induction and Maintenance Therapy in Refractory Crohn's Disease. New England Journal of Medicine 2012;367:1519-1528. 510. Gaffen SL, Jain R, Garg AV, et al. The IL-23-IL-17 immune axis: from mechanisms to therapeutic testing. Nat Rev Immunol 2014;14:585-600. 511. Sandborn WJ, Ghosh S, Panes J, et al. A Phase 2 Study of Tofacitinib, an Oral Janus Kinase Inhibitor, in Patients With Crohn’s Disease. Clinical Gastroenterology and Hepatology 2014;12:1485-1493.e2. 512. Sandborn WJ, Ghosh S, Panes J, et al. Tofacitinib, an oral Janus kinase inhibitor, in active ulcerative colitis. N Engl J Med 2012;367:616-24. 513. Brubaker SW, Bonham KS, Zanoni I, et al. Innate Immune Pattern Recognition: A Cell Biological Perspective. Annu Rev Immunol 2015. 514. Peyrin-Biroulet L, Chamaillard M. NOD2 and defensins: translating innate to adaptive immunity in Crohn's disease. J Endotoxin Res 2007;13:135-9. 515. Veldhoen M, Hocking RJ, Atkins CJ, et al. TGFbeta in the context of an inflammatory cytokine milieu supports de novo differentiation of IL-17- producing T cells. Immunity 2006;24:179-89. 516. Dvorak AM, Monahan RA, Osage JE, et al. Mast-cell degranulation in Crohn's disease. Lancet 1978;1:498. 517. Smith AM, Rahman FZ, Hayee B, et al. Disordered macrophage cytokine secretion underlies impaired acute inflammation and bacterial clearance in Crohn's disease.[Erratum appears in J Exp Med. 2009 Sep 28;206(10):2301]. Journal of Experimental Medicine 2009;206:1883-97. 518. Casanova JL, Abel L. Revisiting Crohn's disease as a primary immunodeficiency of macrophages. J Exp Med 2009;206:1839-43.

245 519. Cook MG, Dixon MF. An analysis of the reliability of detection and diagnostic value of various pathological features in Crohn's disease and ulcerative colitis. Gut 1973;14:255-62. 520. Le Berre N, Heresbach D, Kerbaol M, et al. Histological discrimination of idiopathic inflammatory bowel disease from other types of colitis. J Clin Pathol 1995;48:749-53. 521. Rugtveit J, Brandtzaeg P, Halstensen TS, et al. Increased macrophage subset in inflammatory bowel disease: apparent recruitment from peripheral blood monocytes. Gut 1994;35:669-74. 522. Grimm MC, Pullman WE, Bennett GM, et al. Direct evidence of monocyte recruitment to inflammatory bowel disease mucosa. Journal of Gastroenterology & Hepatology 1995;10:387-95. 523. Rugtveit J, Haraldsen G, Hogasen AK, et al. Respiratory burst of intestinal macrophages in inflammatory bowel disease is mainly caused by CD14+L1+ monocyte derived cells. Gut 1995;37:367-73. 524. Hausmann M, Spottl T, Andus T, et al. Subtractive screening reveals up- regulation of NADPH oxidase expression in Crohn's disease intestinal macrophages. Clinical & Experimental Immunology 2001;125:48-55. 525. Mahida YR, Wu KC, Jewell DP. Characterization of antigen-presenting activity of intestinal mononuclear cells isolated from normal and inflammatory bowel disease colon and ileum. Immunology 1988;65:543-9. 526. Rugtveit J, Nilsen EM, Bakka A, et al. Cytokine profiles differ in newly recruited and resident subsets of mucosal macrophages from inflammatory bowel disease. Gastroenterology 1997;112:1493-505. 527. Steel AW, Mela CM, Lindsay JO, et al. Increased proportion of CD16(+) NK cells in the colonic lamina propria of inflammatory bowel disease patients, but not after azathioprine treatment. Alimentary Pharmacology & Therapeutics 2011;33:115-26. 528. Buonocore S, Ahern PP, Uhlig HH, et al. Innate lymphoid cells drive IL-23 dependent innate intestinal pathology. Nature 2010;464:1371-1375. 529. Fuchs A, Vermi W, Lee JS, et al. Intraepithelial type 1 innate lymphoid cells are a unique subset of IL-12- and IL-15-responsive IFN--producing cells. Immunity 2013;38:769-81. 530. Schiering C, Krausgruber T, Chomka A, et al. The alarmin IL-33 promotes regulatory T-cell function in the intestine. Nature 2014;513:564-8. 531. Martin B, Banz A, Bienvenu B, et al. Suppression of CD4+ T lymphocyte effector functions by CD4+CD25+ cells in vivo. J Immunol 2004;172:3391-8. 532. Langrish CL, Chen Y, Blumenschein WM, et al. IL-23 drives a pathogenic T cell population that induces autoimmune inflammation. J Exp Med 2005;201:233-40. 533. Reinecker HC, Steffen M, Witthoeft T, et al. Enhanced secretion of tumour necrosis factor-alpha, IL-6, and IL-1 beta by isolated lamina propria mononuclear cells from patients with ulcerative colitis and Crohn's disease. Clinical and Experimental Immunology 1993;94:174-181. 534. Reimund JM, Wittersheim C, Dumont S, et al. Mucosal inflammatory cytokine production by intestinal biopsies in patients with ulcerative colitis and Crohn's disease. J Clin Immunol 1996;16:144-50.

246 535. Plevy SE, Landers CJ, Prehn J, et al. A role for TNF-alpha and mucosal T helper-1 cytokines in the pathogenesis of Crohn's disease. J Immunol 1997;159:6276-82. 536. Fujino S, Andoh A, Bamba S, et al. Increased expression of interleukin 17 in inflammatory bowel disease. Gut 2003;52:65-70. 537. Troncone E, Marafini I, Pallone F, et al. Th17 cytokines in inflammatory bowel diseases: discerning the good from the bad. International Reviews of Immunology 2013;32:526-33. 538. Kleinschek MA, Boniface K, Sadekova S, et al. Circulating and gut-resident human Th17 cells express CD161 and promote intestinal inflammation. J Exp Med 2009;206:525-34. 539. Brand S, Beigel F, Olszak T, et al. IL-22 is increased in active Crohn's disease and promotes proinflammatory gene expression and intestinal epithelial cell migration. American Journal of Physiology - Gastrointestinal & Liver Physiology 2006;290:G827-38. 540. Dambacher J, Beigel F, Zitzmann K, et al. The role of the novel Th17 cytokine IL-26 in intestinal inflammation. Gut 2009;58:1207-17. 541. Seiderer J, Elben I, Diegelmann J, et al. Role of the novel Th17 cytokine IL- 17F in inflammatory bowel disease (IBD): upregulated colonic IL-17F expression in active Crohn's disease and analysis of the IL17F p.His161Arg polymorphism in IBD. Inflamm Bowel Dis 2008;14:437-45. 542. Ina K, Itoh J, Fukushima K, et al. Resistance of Crohn's disease T cells to multiple apoptotic signals is associated with a Bcl-2/Bax mucosal imbalance. J Immunol 1999;163:1081-90. 543. Di Sabatino A, Ciccocioppo R, Cinque B, et al. Defective mucosal T cell death is sustainably reverted by infliximab in a caspase dependent pathway in Crohn's disease. Gut 2004;53:70-7. 544. Van den Brande JM, Braat H, van den Brink GR, et al. Infliximab but not etanercept induces apoptosis in lamina propria T-lymphocytes from patients with Crohn's disease. Gastroenterology 2003;124:1774-85. 545. Fuss IJ, Neurath M, Boirivant M, et al. Disparate CD4+ lamina propria (LP) lymphokine secretion profiles in inflammatory bowel disease. Crohn's disease LP cells manifest increased secretion of IFN-gamma, whereas ulcerative colitis LP cells manifest increased secretion of IL-5. J Immunol 1996;157:1261-70. 546. Targan SR, Karp LC. Defects in mucosal immunity leading to ulcerative colitis. Immunological Reviews 2005;206:296-305. 547. Kucharzik T, Lugering N, Weigelt H, et al. Immunoregulatory properties of IL-13 in patients with inflammatory bowel disease; comparison with IL-4 and IL-10. Clin Exp Immunol 1996;104:483-90. 548. Kobori A, Yagi Y, Imaeda H, et al. Interleukin-33 expression is specifically enhanced in inflamed mucosa of ulcerative colitis. Journal of Gastroenterology 2010;45:999-1007. 549. Sponheim J, Pollheimer J, Olsen T, et al. Inflammatory bowel disease- associated interleukin-33 is preferentially expressed in ulceration- associated myofibroblasts. Am J Pathol 2010;177:2804-15. 550. Roulis M, Armaka M, Manoloukos M, et al. Intestinal epithelial cells as producers but not targets of chronic TNF suffice to cause murine Crohn-like

247 pathology. Proceedings of the National Academy of Sciences of the United States of America 2011;108:5396-5401. 551. Di Sabatino A, Pender SL, Jackson CL, et al. Functional modulation of Crohn's disease myofibroblasts by anti-tumor necrosis factor antibodies. Gastroenterology 2007;133:137-49. 552. Di Mari JF, Mifflin RC, Adegboyega PA, et al. IL-1alpha-induced COX-2 expression in human intestinal myofibroblasts is dependent on a PKCzeta- ROS pathway. Gastroenterology 2003;124:1855-65. 553. Hata K, Andoh A, Shimada M, et al. IL-17 stimulates inflammatory responses via NF-kappaB and MAP kinase pathways in human colonic myofibroblasts. Am J Physiol Gastrointest Liver Physiol 2002;282:G1035-44. 554. Greenstein AJ, Janowitz HD, Sachar DB. The extra-intestinal complications of Crohn's disease and ulcerative colitis: a study of 700 patients. Medicine (Baltimore) 1976;55:401-12. 555. Rankin GB, Watts HD, Melnyk CS, et al. National Cooperative Crohn's Disease Study: extraintestinal manifestations and perianal complications. Gastroenterology 1979;77:914-20. 556. Bernstein CN, Blanchard JF, Rawsthorne P, et al. The prevalence of extraintestinal diseases in inflammatory bowel disease: a population-based study. Am J Gastroenterol 2001;96:1116-22. 557. Hoffmann RM, Kruis W. Rare extraintestinal manifestations of inflammatory bowel disease. Inflamm Bowel Dis 2004;10:140-7. 558. Gupta G, Gelfand JM, Lewis JD. Increased risk for demyelinating diseases in patients with inflammatory bowel disease. Gastroenterology 2005;129:819- 26. 559. Bernstein CN, Wajda A, Blanchard JF. The clustering of other chronic inflammatory diseases in inflammatory bowel disease: a population-based study. Gastroenterology 2005;129:827-36. 560. Peyrin-Biroulet L, Loftus EV, Jr., Colombel JF, et al. Long-term complications, extraintestinal manifestations, and mortality in adult Crohn's disease in population-based cohorts. Inflamm Bowel Dis 2011;17:471-8. 561. Ott C, Scholmerich J. Extraintestinal manifestations and complications in IBD. Nat Rev Gastroenterol Hepatol 2013;10:585-95. 562. Walmsley RS, Ayres RCS, Pounder RE, et al. A simple clinical colitis activity index. Gut 1998;43:29-32. 563. Powell-Tuck J, Buckell NA, Lennard-Jones JE. A controlled comparison of corticotropin and hydrocortisone in the treatment of severe proctocolitis. Scand J Gastroenterol 1977;12:971-5. 564. Seo M, Okada M, Yao T, et al. An index of disease activity in patients with ulcerative colitis. Am J Gastroenterol 1992;87:971-6. 565. Higgins PDR, Schwartz M, Mapili J, et al. Patient defined dichotomous end points for remission and clinical improvement in ulcerative colitis. Gut 2005;54:782-788. 566. Jowett SL, Seal CJ, Phillips E, et al. Defining relapse of ulcerative colitis using a symptom-based activity index. Scand J Gastroenterol 2003;38:164-71. 567. Targownik LE, Singh H, Nugent Z, et al. The epidemiology of colectomy in ulcerative colitis: results from a population-based cohort. Am J Gastroenterol 2012;107:1228-35.

248 568. Riley SA, Mani V, Goodman MJ, et al. Microscopic activity in ulcerative colitis: what does it mean? Gut 1991;32:174-8. 569. Tedla N, Palladinetti P, Kelly M, et al. Chemokines and T lymphocyte recruitment to lymph nodes in HIV infection. American Journal of Pathology 1996;148:1367-73. 570. Papanikolaou NA, Vasilescu ER, Suciu-Foca N. Novel single nucleotide polymorphisms in the human immune inhibitory immunoglobulin-like T cell receptor type 4. Human Immunology 2004;65:700-5. 571. Chang C-C, Silvia EA, Ho EK, et al. Polymorphism and linkage disequilibrium of immunoglobulin-like transcript 3 gene. Human Immunology 2008;69:284-90. 572. Davidson CL, Li NL, Burshtyn DN. LILRB1 polymorphism and surface phenotypes of natural killer cells. Human Immunology 2010;71:942-9. 573. Camilleri M, Lee JS, Viramontes B, et al. Insights into the pathophysiology and mechanisms of constipation, irritable bowel syndrome, and diverticulosis in older people. Journal of the American Geriatrics Society 2000;48:1142-50. 574. Coleman MP, Forman D, Bryant H, et al. Cancer survival in Australia, Canada, Denmark, Norway, Sweden, and the UK, 1995-2007 (the International Cancer Benchmarking Partnership): an analysis of population-based cancer registry data. Lancet 2011;377:127-38. 575. Colonna M, Nakajima H, Navarro F, et al. A novel family of Ig-like receptors for HLA class I molecules that modulate function of lymphoid and myeloid cells. Journal of Leukocyte Biology 1999;66:375-81. 576. Apps R, Gardner L, Sharkey AM, et al. A homodimeric complex of HLA-G on normal trophoblast cells modulates antigen-presenting cells via LILRB1. European Journal of Immunology 2007;37:1924-37. 577. McIntire RH, Ganacias KG, Hunt JS. Programming of human monocytes by the uteroplacental environment. Reproductive Sciences 2008;15:437-47. 578. Li C, Houser BL, Nicotra ML, et al. HLA-G homodimer-induced cytokine secretion through HLA-G receptors on human decidual macrophages and natural killer cells. Proceedings of the National Academy of Sciences of the United States of America 2009;106:5767-72. 579. Barclay AN, Hatherley D. The counterbalance theory for evolution and function of paired receptors. Immunity 2008;29:675-8. 580. Diaz-Lagares A, Alegre E, Arroyo A, et al. Tyrosine nitration in the human leucocyte antigen-G-binding domain of the Ig-like transcript 2 protein. FEBS Journal 2009;276:4233-43. 581. Fava F, Danese S. Intestinal microbiota in inflammatory bowel disease: friend of foe? World Journal of Gastroenterology 2011;17:557-66. 582. Hold GL, Smith M, Grange C, et al. Role of the gut microbiota in inflammatory bowel disease pathogenesis: what have we learnt in the past 10 years? World Journal of Gastroenterology 2014;20:1192-210. 583. Manichanh C, Borruel N, Casellas F, et al. The gut microbiota in IBD. Nature Reviews Gastroenterology & Hepatology 2012;9:599-608. 584. Golder JP, Doe WF. Isolation and preliminary characterization of human intestinal macrophages. Gastroenterology 1983;84:795-802. 585. Lu HK, Rentero C, Raftery MJ, et al. Leukocyte Ig-like receptor B4 (LILRB4) is a potent inhibitor of FcgammaRI-mediated monocyte activation via

249 dephosphorylation of multiple kinases. Journal of Biological Chemistry 2009;284:34839-48. 586. Shackelford DA, Kaufman JF, Korman AJ, et al. HLA-DR antigens: structure, separation of subpopulations, gene cloning and function. Immunological Reviews 1982;66:133-87. 587. Bland PW, Richens ER, Britton DC, et al. Isolation and purification of human large bowel mucosal lymphoid cells: effect of separation technique on functional characteristics. Gut 1979;20:1037-46. 588. Bull DM, Bookman MA. Isolation and functional characterization of human intestinal mucosal lymphoid cells. J Clin Invest 1977;59:966-74. 589. Selby WS, Janossy G, Bofill M, et al. Intestinal lymphocyte subpopulations in inflammatory bowel disease: an analysis by immunohistological and cell isolation techniques. Gut 1984;25:32-40. 590. Simmons D, Seed B. Isolation of a cDNA encoding CD33, a differentiation antigen of myeloid progenitor cells. J Immunol 1988;141:2797-800. 591. Peiper SC, Ashmun RA, Look AT. Molecular cloning, expression, and chromosomal localization of a human gene encoding the CD33 myeloid differentiation antigen. Blood 1988;72:314-21. 592. Olsen T, Goll R, Cui G, et al. Tissue levels of tumor necrosis factor-alpha correlates with grade of inflammation in untreated ulcerative colitis. Scandinavian Journal of Gastroenterology 2007;42:1312-20. 593. Matsuda R, Koide T, Tokoro C, et al. Quantitive cytokine mRNA expression profiles in the colonic mucosa of patients with steroid naive ulcerative colitis during active and quiescent disease. Inflammatory Bowel Diseases 2009;15:328-34. 594. Konturek PC, Brzozowski T, Engel M, et al. Ghrelin ameliorates colonic inflammation. Role of nitric oxide and sensory nerves. Journal of Physiology & Pharmacology 2009;60:41-7. 595. Sanchez-Munoz F, Fonseca-Camarillo G, Villeda-Ramirez MA, et al. Transcript levels of Toll-Like Receptors 5, 8 and 9 correlate with inflammatory activity in Ulcerative Colitis. BMC Gastroenterology 2011;11:138. 596. Hozumi H, Hokari R, Kurihara C, et al. Endoscopic finding of spontaneous hemorrhage correlates with tumor necrosis factor alpha expression in colonic mucosa of patients with ulcerative colitis. International Journal of Colorectal Disease 2013;28:1049-55. 597. Verma R, Verma N, Paul J. Expression of inflammatory genes in the colon of ulcerative colitis patients varies with activity both at the mRNA and protein level. European Cytokine Network 2013;24:130-8. 598. Raddatz D, Bockemuhl M, Ramadori G. Quantitative measurement of cytokine mRNA in inflammatory bowel disease: relation to clinical and endoscopic activity and outcome. European Journal of Gastroenterology & Hepatology 2005;17:547-57. 599. Weber B, Saurer L, Mueller C. Intestinal macrophages: differentiation and involvement in intestinal immunopathologies. Seminars In Immunopathology 2009;31:171-84. 600. Naba A, Clauser KR, Hoersch S, et al. The Matrisome: In Silico Definition and In Vivo Characterization by Proteomics of Normal and Tumor Extracellular Matrices. Molecular & Cellular Proteomics : MCP 2012;11:M111.014647.

250 601. Hynes RO, Naba A. Overview of the matrisome--an inventory of extracellular matrix constituents and functions. Cold Spring Harb Perspect Biol 2012;4:a004903. 602. Smythies LE, Shen R, Bimczok D, et al. Inflammation anergy in human intestinal macrophages is due to Smad-induced IkappaBalpha expression and NF-kappaB inactivation. J Biol Chem 2010;285:19593-604. 603. Balestra B, Vicini R, Cremon C, et al. Colonic mucosal mediators from patients with irritable bowel syndrome excite enteric cholinergic motor neurons. Neurogastroenterol Motil 2012;24:1118-e570. 604. Buhner S, Li Q, Berger T, et al. Submucous rather than myenteric neurons are activated by mucosal biopsy supernatants from irritable bowel syndrome patients. Neurogastroenterol Motil 2012;24:1134-e572. 605. Buhner S, Li Q, Vignali S, et al. Activation of human enteric neurons by supernatants of colonic biopsy specimens from patients with irritable bowel syndrome. Gastroenterology 2009;137:1425-34. 606. Ibeakanma C, Vanner S. TNFalpha is a key mediator of the pronociceptive effects of mucosal supernatant from human ulcerative colitis on colonic DRG neurons. Gut 2010;59:612-21. 607. Mukaratirwa S, Koninkx JF, Gruys E, et al. Mutual paracrine effects of colorectal tumour cells and stromal cells: modulation of tumour and stromal cell differentiation and extracellular matrix component production in culture. International Journal of Experimental Pathology 2005;86:219- 229. 608. Conti J, Thomas G. The role of tumour stroma in colorectal cancer invasion and metastasis. Cancers (Basel) 2011;3:2160-8. 609. Dolznig H, Rupp C, Puri C, et al. Modeling colon adenocarcinomas in vitro a 3D co-culture system induces cancer-relevant pathways upon tumor cell and stromal fibroblast interaction. Am J Pathol 2011;179:487-501. 610. Connor SJ, Paraskevopoulos N, Newman R, et al. CCR2 expressing CD4+ T lymphocytes are preferentially recruited to the ileum in Crohn's disease.[Erratum appears in Gut. 2004 Nov;53(11):1722]. Gut 2004;53:1287-94. 611. Grimm MC, Elsbury SK, Pavli P, et al. Interleukin 8: cells of origin in inflammatory bowel disease. Gut 1996;38:90-8. 612. Hao WL, Lee YK. Microflora of the gastrointestinal tract: a review. Methods Mol Biol 2004;268:491-502. 613. Smythies LE, Maheshwari A, Clements R, et al. Mucosal IL-8 and TGF-beta recruit blood monocytes: evidence for cross-talk between the lamina propria stroma and myeloid cells. Journal of Leukocyte Biology 2006;80:492-9. 614. Guarner F, Malagelada JR. Gut flora in health and disease. Lancet 2003;361:512-9. 615. Macpherson AJ, Harris NL. Interactions between commensal intestinal bacteria and the immune system. Nat Rev Immunol 2004;4:478-485. 616. Honda K, Takeda K. Regulatory mechanisms of immune responses to intestinal bacteria. Mucosal Immunol 2009;2:187-196. 617. Akagawa KS, Komuro I, Kanazawa H, et al. Functional heterogeneity of colony-stimulating factor-induced human monocyte-derived macrophages. Respirology 2006;11 Suppl:S32-6.

251 618. Lacey DC, Achuthan A, Fleetwood AJ, et al. Defining GM-CSF- and macrophage-CSF-dependent macrophage responses by in vitro models. J Immunol 2012;188:5752-65. 619. Im E, Riegler FM, Pothoulakis C, et al. Elevated lipopolysaccharide in the colon evokes intestinal inflammation, aggravated in immune modulator- impaired mice. Am J Physiol Gastrointest Liver Physiol 2012;303:G490-7. 620. Kiener PA, Marek F, Rodgers G, et al. Induction of tumor necrosis factor, IFN-gamma, and acute lethality in mice by toxic and non-toxic forms of lipid A. 1988. 621. Fenton MJ, Golenbock DT. LPS-binding proteins and receptors. J Leukoc Biol 1998;64:25-32. 622. Schroder K, Hertzog PJ, Ravasi T, et al. Interferon-gamma: an overview of signals, mechanisms and functions. J Leukoc Biol 2004;75:163-89. 623. Sabat R, Grutz G, Warszawska K, et al. Biology of interleukin-10. Cytokine Growth Factor Rev 2010;21:331-44. 624. Veenbergen S, Samsom JN. Maintenance of small intestinal and colonic tolerance by IL-10-producing regulatory T cell subsets. Current Opinion in Immunology 2012;24:269-76. 625. Paul G, Khare V, Gasche C. Inflamed gut mucosa: downstream of interleukin- 10. Eur J Clin Invest 2012;42:95-109. 626. Shah N, Kammermeier J, Elawad M, et al. Interleukin-10 and interleukin-10- receptor defects in inflammatory bowel disease. Current Allergy & Asthma Reports 2012;12:373-9. 627. Gantner F, Kupferschmidt R, Schudt C, et al. In vitro differentiation of human monocytes to macrophages: change of PDE profile and its relationship to suppression of tumour necrosis factor-α release by PDE inhibitors. British Journal of Pharmacology 1997;121:221-231. 628. Groux H, Bigler M, de Vries JE, et al. Interleukin-10 induces a long-term antigen-specific anergic state in human CD4+ T cells. J Exp Med. 1996;184:19-29. 629. Means TK, Golenbock DT, Fenton MJ. The biology of Toll-like receptors. Cytokine Growth Factor Rev 2000;11:219-32. 630. Golenbock DT, Fenton MJ. Extolling the diversity of bacterial endotoxins. Nat Immunol 2001;2:286-288. 631. FDA. Guidance for Industry Pyrogen and Endotoxins Testing: Questions and Answers. In: Services UDoHaH, ed. Bethesda, 2012. 632. Sica A, Mantovani A. Macrophage plasticity and polarization: in vivo veritas. J Clin Invest 2012;122:787-95. 633. Murray PJ, Wynn TA. Obstacles and opportunities for understanding macrophage polarization. J Leukoc Biol 2011;89:557-63. 634. Xu X, Zou P, Chen L, et al. IL-10 enhances promoter activity of ILT4 gene and up-regulates its expression in THP-1 cells. Journal of Huazhong University of Science and Technology. Medical Sciences 2010;30:594-8.

252