Campylobacter Concisus and Its Possible Role in Inflammatory Bowel Disease

Campylobacter concisus and its possible role in inflammatory bowel disease

Yazan Ismail

A thesis submitted in fulfilment of the requirements for the degree of Doctor of Philosophy (Microbiology and Immunology)

Supervisor: Dr. Li Zhang

School of Biotechnology and Biomolecular Sciences

Faculty of Science

University of New South Wales

Australia

2012

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 ……………………………………………......

Acknowledgements

To Li, I thank god every day that I have chosen this research task and more importantly you as my supervisor. You were always there when I needed you and lifted me every time I felt down, I don’t regret any minute I spent under your supervision.

Viki and Siew my beloved sisters can you believe it, at last it is my turn to write you in my acknowledgement. We had beautiful days and days that we hate to remember, but that very dark days made our relation more than just a friendship to remember, or words to describe it is something for eternity, I will never fulfil your favours. From the bottom of my heart and from my family THANK YOU, THANK YOU.

To Leo, man I will never find a better mate to strip with him all night I mean the Western membranes off course, I am sure that you will be great researcher one day. Lisa, Lisa Lisa the lab is only not the same when your smile and beautiful sprit left our lab, I miss you girl. My in laws and brother Fakhry Al-edwan, thanks for everything you did.

Professor Hazel Mitchell, Assoc. Professor Andrew Collins, Dr. John Wilson, Dr. Tim Williams, Dr. Walied Al-momany and Dr. Qasem Abu-Shaqra thanks a lot for your support. Dr. Nadeem I will always remember our chats after all we will always have the same case, man thank you for raising my sprit when I really needed it. Natalia, Sophie, Connie, Nupur and Shirly what beautiful persons you are, you always had answers and the solutions for every problem I had. Natalia by the way I think I need another blood pressure check-up.

My high school friends, Moatz, Mohammed, Osama and Tk3yb (Note the alphabetic order). May god never let us apart. Jose I knew that you will reach what you aim for, since the first day I met you. Peter what an honest and Nobel person you are. Johnny where ever you are, I wish your life is filled with happiness you really deserve it. Ye Sing, Wilson, Azade, Lorance and Omar I knew you for a short period of time, but it was enough to know that you are honest and hardworking people. Wish you luck and the best future.

Mother, Father, Sisters, Foad and at last but not least my dearest brother captain Yazyd. I love you all MORE and MORE but NEVER LESS.

Sana’a my dearest, my life began when I met you. Kids the joy of my life, always aim high and don’t worry of missing the target, I will always be there FOR YOU (Inshallah).

Publications

Peer-reviewed Journal Articles

1. Zhang, L., V. Budiman, A. S. Day, H. Mitchell, D. A. Lemberg, S. M. Riordan, M. Grimm, S. T. Leach, and Y. Ismail. 2010. Isolation and detection of Campylobacter concisus from saliva of healthy individuals and patients with inflammatory bowel disease. J. Clin. Microbiol. 48:2965.

2. Ismail, Y.*, V. Mahendran, S*. Octavia, A. S. Day, S. M. Riordan, M. C. Grimm, R. Lan, D. Lemberg, T. A. T. Tran, and L. Zhang. 2012. Investigation of the Enteric Pathogenic Potential of Oral Campylobacter concisus Strains Isolated from Patients with Inflammatory Bowel Disease. PloS one. 7:e38217

3. Ismail, Y.*, H. Lee, S. Riordan, M. Grimm, L. Zhang. 2012. The effects of oral and enteric Campylobacter concisus strains on expression of TLR4, MD-2, TLR2, TLR5 and COX-2 in HT-29 cells. (Manuscript in review).

4. Ismail, Y.*, H. Lee, S. Riordan, M. Grimm, L. Zhang. 2012. Study of oral and enteric Campylobacter concisus strains proinflammatory properties on human intestinal epithelial cells in vitro. (Manuscript in preparation).

5. Ismail, Y.*, S. Riordan, M. Grimm, L. Zhang.2012. The lipopolysaccharide profiles of oral and enteric C. concisus strains isolated from patients with inflammatory bowel disease and healthy controls. (Manuscript in preparation).

Conference Proceedings

1. Ismail, Y.*, V. Mahendran, S*. Octavia, A. S. Day, S. M. Riordan, M. C. Grimm, R. Lan, D. Lemberg, T. A. T. Tran, and L. Zhang. 2012. Investigation of the enteric

iii

pathogenic potential of oral Campylobacter concisus strains isolated from patients with inflammatory bowel disease. Australian Society for Microbiology. Brisbane, Australia. Poster presentation.

2. Ismail, Y.*, H. Lee, S. Riordan, M. Grimm, L. Zhang. 2011. Upregulation of intestinal epithelial expression of TLR-4, MD-2, COX-2 and IL-8 by Campylobacter concisus: possible relevance to the pathophysiology of inflammatory bowel disease. The Second International Conference on Immune Tolerance. Amsterdam, The Netherlands. Oral presentation.

3. Ismail, Y.*, H. Lee, S. Riordan, M. Grimm, L. Zhang. 2011. Investigation of the interaction between Campylobacter concisus and the gut innate immune system. The 41st Australasian Society for Immunology Annual Meeting. Adelaide, Australia. Poster presentation.

4. Mahendran, V., Y. Ismail, A. Day, S. Riordan, R. Lan, D. Lemberg, L. zhang. 2011. Invistigation of the enteric pathogenic potential of oral Campylobacter concisus strains isolated from patients with inflammatory bowel disease. Australian Society for Microbiology. Tasmania, Australia. Poster presentation.

5. Mum, S., A. Day, D. Lemberg, S. Riordan, M. C. Grimm, V. Budiman, H. Mitchell, Y. Ismail, L. Zhang. 2010. Antibiotic susceptibility of Campylobacter concisus isolated from patients with inflammatory bowel diseases. Australian gastroenterology. Gold Coast, Australia. Poster presentation.

Table of Contents

ORIGINALITY STATEMENT ...... I ACKNOWLEDGEMENTS ...... II PUBLICATIONS ...... III TABLE OF CONTENTS ...... V LIST OF TABLES ...... XI LIST OF FIGURES ...... XII LIST OF ABBREVIATIONS ...... XIV ABSTRACT ...... XVII CHAPTER 1: INTRODUCTION ...... 1

1.1 INFLAMMATORY BOWEL DISEASE ...... 1 1.1.1 Overview ...... 1 1.1.2 Epidemiology ...... 2 1.1.3 Aetiology...... 9 1.1.3.1 Environmental factor ...... 10 1.1.3.2 Immunological response dysregulation ...... 12 1.1.3.3 Genetic Factors ...... 19 1.1.3.3.1 NOD2 (CARD15 ) ...... 20 1.1.3.3.2 SLC22A4 and SLC22A5 ...... 22 1.1.3.3.3 DLG5...... 23 1.1.3.3.4 PPARG ...... 24 1.1.3.3.5 MDR1 ...... 26 1.1.3.4 Bacterial factors ...... 27 1.1.3.4.1 Mycobacterium avium subspecies paratuberculosis and CD ...... 28 1.1.3.4.2 Escherichia coli and IBD ...... 30 1.1.3.4.3 Bacteroides vulgatus and UC...... 32 1.1.3.4.4 Campylobacter concisus ...... 34 1.2 CAMPYLOBACTER CONCISUS: AN EMERGING BACTERIA IN THE PATHOGENESIS OF IBD ...... 34 1.2.1 Campylobacter the Genera ...... 34 1.2.1.1 History...... 34 1.2.1.2 Taxonomy ...... 35 1.2.2 Campylobacter concisus; a general review ...... 38 1.2.3 Campylobacter concisus and Disease ...... 39 1.2.3.1 Campylobacter concisus and the human oral cavity ...... 39 1.2.3.2 Campylobacter concisus and the human gastroenteritis ...... 40

1.2.3.3 Campylobacter concisus and the IBD ...... 41 1.3 HYPOTHESIS AND AIMS ...... 42 CHAPTER 2: GENERAL METHODOLOGY ...... 43

2.1 MEDIA AND BUFFER ...... 43 2.1.1 Horse blood agar (HBA) ...... 43 2.1.2 Horse blood agar with vancomycin (HBA+V) ...... 43 2.1.3 Nutrient agar (NA) ...... 43 2.1.4 Phosphate buffer saline (PBS) ...... 43 2.2 PROTOCOLS ...... 44 2.2.1 Electrophoresis ...... 44 2.2.1.1 Aim: electrophoresis was used to separate proteins or lipopolysaccharides (LPS). 44 2.2.1.2 Instrument: The mini-PROTEAN tetra cell for mini precast gels (Bio-Rad, California, USA) was used, catalogue number: 165-8004...... 44 2.2.1.3 Reagent ...... 44 2.2.1.4 Method ...... 45 2.2.2 Silver Staining ...... 45 2.2.2.1 Aim: Silver staining was used to visualise the separated LPS profile bands, LPS separating was by PAGE...... 45 2.2.2.2 Reagents ...... 45 2.2.2.3 Method ...... 46 2.2.3 Coomassie brilliant blue staining ...... 46 2.2.3.1 Aim: Coomassie brilliant blue staining was used to visualise the separated whole protein profile bands after separating the proteins by SDS-PAGE...... 46 2.2.3.2 Reagents ...... 46 2.2.3.3 Method ...... 46 2.2.4 Western blotting ...... 47 2.2.4.1 Aim: Western blotting was used to transfer the separated protein bands by electrophoresis from the SDS-Polyacrylamide gels to polyvinylidene difluoride (PVDF) membrane sheath...... 47 2.2.4.2 Instrument: The mini trans-blot® electrophoretic transfer cell instrument (Bio- Rad, California, USA) was used, catalogue number: 170-3930...... 47 2.2.4.3 Reagents ...... 47 2.2.4.4 Method ...... 47 2.2.5 Mammalian cell culture technique ...... 48 2.2.5.1 Maintaining the Caco-2 human intestinal epithelial cell line...... 48 2.2.5.2 Maintaining the HT-29 human intestinal epithelial cell line...... 48 2.3 ETHICS APPROVAL ...... 49

CHAPTER 3: THE HIGH PREVALENCE OF MULTIPLE ORAL C. CONCISUS STRAINS DETECTED IN PATIENTS WITH INFLAMMATORY BOWEL DISEASE AND THEIR ENTERIC PATHOGENIC POTENTIAL ...... 50

3.1 INTRODUCTION ...... 50 3.2 MATERIALS AND METHODS ...... 52 3.2.1 Study subjects: ...... 52 3.2.2 Isolation and identification of oral C. concisus ...... 53 3.2.2.1 Specimen collection ...... 53 3.2.2.2 Isolation of multiple oral C. concisus isolates from saliva samples ...... 53 3.2.3 Confirmation of the identity of oral C. concisus by C. concisus specific PCR ...... 54 3.2.4 Determination of the number of oral C. concisus strains colonizing each individual (patients with IBD and healthy controls) ...... 55 3.2.4.1 The whole cell proteins preparation for the C. concisus isolates ...... 55 3.2.4.2 Whole cell protein profiles analysis by SDS-PAGE ...... 55 3.2.4.3 Whole cell protein profiles visualization ...... 56 3.2.5 Examination of adhesion and invasion of oral C. concisus strains to human intestinal epithelial cells, using a modified gentamicin protection assay ...... 56 3.2.6 Testing the susceptibility of C. concisus strains to gentamicin antibiotic ...... 57 3.2.6.1 Bacterial culture conditions ...... 57 3.2.6.2 Epsilometer test ...... 58 3.2.7 Statistical Analysis ...... 58 3.3 RESULTS ...... 58 3.3.1 Patient characteristics...... 58 3.3.2 Identifying the putative oral C. concisus isolates from patients with IBD and healthy controls ...... 59 3.3.3 Determining the number of oral C. concisus strains colonizing the oral cavity of each patient and control ………………………………………………………………………………………… ……………………………………………………………59 3.3.4 Comparison of the prevalence of multiple oral C. concisus strains between patients and controls as well as between children and adults ...... 64 3.3.5 Grouping C. concisus strains according to the whole protein profile ...... 64 3.3.6 The adherence and invasive ability of oral C. concisus strains to the human intestinal epithelial cells ...... 66 3.3.7 Gentamicin minimal inhibition concentration (MIC) ...... 71 3.4 DISCUSSION ...... 71 CHAPTER 4: THE LIPOPOLYSACCHARIDE PROFILES OF ORAL AND ENTERIC C. CONCISUS STRAINS ISOLATED FROM HEALTHY CONTROLS AND PATIENTS WITH INFLAMMATORY BOWEL DISEASE...... 76

vii

4.1 INTRODUCTION ...... 76 4.2 MATERIALS AND METHODS ...... 80 4.2.1 Study subjects: ...... 80 4.2.2 C. concisus strains used in this study ...... 80 4.2.3 C. concisus growth condition ...... 84 4.2.4 Extraction of LPS and separation on Polyacrylamide Gel Electrophoresis ...... 84 4.2.4.1 LPS extraction ...... 84 4.2.4.2 Separation of LPS by polyacrylamide gel electrophoresis ...... 84 4.2.4.3 Visualizing LPS patterns...... 85 4.2.5 Band intensity measurement: ...... 85 4.2.6 Statistical analysis ...... 85 4.3 RESULTS ...... 85 4.3.1 Patient characteristics...... 85 4.3.2 Oral C. concisus LPS profile ...... 86 4.3.3 LPS profile associated with IBD disease ...... 88 4.3.4 Comparing LPS profiles between Oral and enteric C. concisus Isolates ...... 89 4.4 DISCUSSION ...... 92 CHAPTER 5: THE EFFECTS OF C. CONCISUS ON INTESTINAL EPITHELIAL EXPRESSION OF MD-2, TLR2, 4 AND 5 ...... 95

5.1 INTRODUCTION ...... 95 5.2 MATERIAL AND METHODS ...... 98 5.2.1 C. concisus strains and cultivation conditions ...... 98 5.2.2 Cultivation of HT-29 cells ...... 99 5.2.3 Western blot ...... 100 5.2.3.1 Antibodies used for Western blotting ...... 100 5.2.3.2 Infection of HT-29 cells with C. concisus ...... 100 5.2.3.3 Preparation of HT-29 cells whole cell proteins ...... 100 5.2.3.4 Western blot ...... 101 5.2.4 Immunofluorescence staining and confocal microscopy ...... 101 5.2.4.1 Antibodies used for immunofluorescence staining ...... 101 5.2.4.2 Infection of HT-29 cells with C. concisus ...... 101 5.2.4.3 Immunostaining and visualization of TLR4, MD-2 and TLR5 by confocal microscopy ...... 101 5.2.5 Flow cytometry ...... 102 5.2.5.1 Antibodies used for immunofluorescence staining ...... 102 5.2.5.2 Infection of HT-29 cells with C. concisus ...... 102 5.2.5.3 Flow cytometry ...... 102 5.2.6 Statistical analysis ...... 103 5.3 RESULTS ...... 103

viii

5.3.1 Effects of C. concisus on TLR4 expression in the intestinal epithelial cells HT-29 103 5.3.2 Effects of C. concisus on MD-2 expression in HT-29 cells ...... 107 5.3.3 Effects of C. concisus on TLR5 expression in HT-29 cells ...... 111 5.3.4 Effects of C. concisus on TLR2 expression in HT-29 cells ...... 115 5.3.5 The effects of oral and enteric C. concisus strains isolated from individual patients with IBD on expression of TLR4 and MD-2 in HT-29 cells ...... 119 5.4 DISCUSSION ...... 120 CHAPTER 6: . STUDY OF C. CONCISUS PROINFLAMMATORY PROPERTIES USING HUMAN INTESTINAL EPITHELIAL CELL LINE HT-29 AND CACO-2 ...... 124

6.1 INTRODUCTION ...... 124 6.2 MATERIALS AND METHODS ...... 129 6.2.1 C. concisus strains and cultivation conditions ...... 129 6.2.2 Cultivation of HT-29 cells ...... 129 6.2.3 Cultivation of Caco-2 cells ...... 129 6.2.4 Western blot ...... 130 6.2.4.1 Antibodies used for Western blotting ...... 130 6.2.4.2 Infection of HT-29 cells with C. concisus ...... 130 6.2.4.3 Preparation of HT-29 cells whole cell proteins ...... 130 6.2.4.4 Western blot ...... 131 6.2.5 Immunofluorescence staining and confocal microscopy ...... 131 6.2.5.1 Antibodies used for immunofluorescence staining ...... 131 6.2.5.2 Infection of HT-29 cells with C. concisus ...... 131 6.2.5.3 Immunostaining and visualization of COX-2 and IκBα by confocal microscopy 131 6.2.6 Flow cytometry ...... 132 6.2.6.1 Antibodies used for immunofluorescence staining ...... 132 6.2.6.2 Infection of HT-29 cells with C. concisus ...... 132 6.2.6.3 Flow cytometry ...... 132 6.2.7 Measurement of IL-8 in HT-29 cell culture supernatant by ELISA ...... 132 6.2.8 The effect of TNF-α and IFN-γ on the adhesion and invasion ability of C. concisus strains to the Human intestinal epithelial cells ...... 133 6.2.9 Statistical analysis ...... 134 6.3 RESULTS ...... 134 6.3.1 Effects of C. concisus on COX-2 expression in HT-29 cells ...... 134 6.3.2 Effects of C. concisus on IL-8 production in HT-29 cells ...... 138 6.3.3 Effects of C. concisus on activation of NFκB in HT-29 cells ...... 141 6.3.4 Effects of C. concisus concentration on the expression of TLR4 in HT-29 cells ... 144 6.3.5 The effects of oral and enteric C. concisus strains isolated from individual patients with IBD on expression of COX-2 and IL-8 in HT-29 cells ...... 146

6.3.6 The effect of pre-existing inflammation on Campylobacter concisus adhering and invading HT-29 cells...... 147 6.4 DISCUSSION ...... 150 CHAPTER 7: GENERAL DISCUSSION AND FUTURE DIRECTIONS ...... 154

7.1 GENERAL DISCUSSION ...... 154 7.2 FUTURE DIRECTIONS...... 160 REFERENCES ...... 162

List of Tables

TABLE 1.1 IBD INCIDENCE IN DIFFERENT GEOGRAPHICAL REGIONS IN DIFFERENT TIME PERIODS, SHOWING THE INCREASED INCIDENCE OF IBD OVER TIME IN DIFFERENT GEOGRAPHICAL REGIONS...... 5 TABLE 1.2 INNATE IMMUNE RESPONSE CYTOKINES ASSOCIATED WITH IBD...... 13 TABLE 1.3 T-CELL RESPONSE LYMPHOKINES ASSOCIATED WITH IBD...... 17 TABLE 3.1 INFORMATION OF PATIENTS WITH IBD AND HEALTHY CONTROLS IN THIS STUDY INCLUDING THEIR GIVEN ID THROUGH THIS STUDY, SEX, AGE ON THE DAY OF DIAGNOSIS ..... 52 TABLE 3.2 ORAL C. CONCISUS STRAINS DETECTED IN INDIVIDUAL PATIENTS WITH IBD AND HEALTHY CONTROLS AND THEIR ASSIGNED GROUP* ...... 61 TABLE 3.3 INVASION AND ADHESION INDEX OF ORAL C. CONCISUS STRAINS ISOLATED FROM PATIENTS WITH CD AND CONTROLS TO CACO-2 CELLS...... 68 TABLE 4.1 C. CONCISUS STRAINS USED IN LPS EXTRACTION, THEIR LPS FORM AND LPS FORM GROUP IF APPLICABLE...... 81 TABLE 4.2 COMPARISON OF LPS PATTERN GROUPS PRESENTED IN PATIENTS WITH IBD AND IN HEALTHY CONTROLS...... 89 TABLE 5.1 C. CONCISUS STRAINS USED IN THIS STUDY...... 99 TABLE 6.1 THE EFFECT OF TNF- Α AT THREE DIFFERENT CONCENTRATIONS (20, 40, AND 80 NG/ML) ON THE ADHESION AND INVASION ABILITY OF C. CONCISUS STRAINS (P1CDO2, P1CDO3, AND P1CDB1(UNSWCD)) TO THE HUMAN INTESTINAL EPITHELIAL CELLS CACO-2...... 148 TABLE 6.2 THE EFFECT OF IFN-Γ AT THREE DIFFERENT CONCENTRATIONS (20, 40, AND 80 NG/ML) ON THE ADHESION AND INVASION ABILITY OF C. CONCISUS STRAINS (P1CDO2, P1CDO3, AND P1CDB1(UNSWCD) TO THE HUMAN INTESTINAL EPITHELIAL CELLS CACO-2...... 149

List of Figures

FIGURE 1.1 LEFT IMAGE IS A PHOTOMICROGRAPH OF HISTOLOGICAL SPECIMEN TAKEN FROM A PATIENT WITH CD, INFLAMMATION INVOLVE FULL THICKNESS OF THE INTESTINAL WALL FROM THE MUCOSA TO THE SEROSA. RIGHT IMAGE SHOWS MUCOSAL INFLAMMATION AND EROSION IN A PATIENT WITH UC ...... 2 FIGURE 1.2 INCIDENCE RATES (PER 100,000) OF CD AND UC IN RELATION TO AGE, GEELONG, VICTORIA, AUSTRALIA, BETWEEN APRIL 2007 AND MARCH 2008 (WILSON ET AL. 2010)...... 9 FIGURE 1.3 FACTORS THAT CONTRIBUTE TO THE DEVELOPMENT OF IBD. IT IS WIDE ACCEPTED THAT IBD IS AN OUTCOME OF THE INTERACTION OF THE FOLLOWING FACTORS: ENVIRONMENTAL FACTOR, IMMUNOLOGICAL DYSREGULATION, GENETIC SUSCEPTIBILITY AND BACTERIAL FACTOR. ADAPTED FROM...... 10 FIGURE 1.4 SUMMARY OF THE CYTOKINES AND CHEMOKINES PRODUCED BY THE INTESTINAL EPITHELIAL CELLS AND THE CHEMOKINES USUAL TARGET...... 14 FIGURE 1.5 THE SCIENTIFIC CLASSIFICATION OF THE GENUS CAMPYLOBACTER ACCORDING TO BERGEY’S MANUAL VOLUME TWO THE PROTEOBACTERIA...... 36 FIGURE 1.6 CAMPYLOBACTERACEAE FAMILY PHYLOGENETIC TREE AND ITS CLOSEST PHYLOGENETIC NEIGHBOURS, BASED ON THE 16S RRNA SEQUENCE SIMILARITY. THE SCALE REPRESENT A 10% SEQUENCE DISSIMILARITY...... 37 FIGURE 3.1 AN EXAMPLE SHOWING THE DETERMINATION OF THE NUMBER OF ORAL C. CONCISUS STRAINS COLONIZING A GIVEN INDIVIDUAL...... 59 FIGURE 3.2 WHOLE CELL PROTEIN PROFILES OF THE 52 ORAL C. CONCISUS STRAINS ISOLATED FROM 15 PATIENTS WITH IBD AND 17 CONTROLS...... 63 FIGURE 3.3 THE PREVALENCE OF MULTIPLE ORAL C. CONCISUS STRAINS IN PATIENTS WITH IBD AND HEALTHY CONTROLS THE TWO MAJOR FORMS OF IBD ARE CD AND UC...... 64 FIGURE 3.4 A REPRESENTATIVE WHOLE PROTEIN PROFILE FOR GROUP 1 AND GROUP 2 AS DETERMINED BY THE GROUPING METHOD REPORTED BY AABENHUS ET AL ...... 66 FIGURE 3.5 COMPARISON OF MEAN ADHESION INDEX AND MEAN INVASION INDEX BETWEEN C. CONCISUS STRAINS ISOLATED FROM PATIENT WITH CD GROUP AND C. CONCISUS STRAINS ISOLATED FROM HEALTHY GROUP...... 70 FIGURE 3.6 COMPARISON OF THE MEAN ADHESION INDEX AND THE MEAN INVASION INDEX BETWEEN THE TWO GROUPS (GROUP 1 AND GROUP 2) CATEGORIZED ACCORDING TO AABENHUS ET AL...... 71 FIGURE 4.1 INNER AND OUTER MEMBRANE OF GRAM NEGATIVE BACTERIA. THE OUTER LEAFLET OF THE OUTER MEMBRANE IS POSSESSED OF LPS WHICH HAS THREE COMPONENTS:...... 76 FIGURE 4.2 SCHEMATIC REPRESENTATION OF THE DETAILED STRUCTURE OF LPS...... 79 FIGURE 4.3 ORAL C. CONCISUS LPS PROFILES, LPS WAS EXTRACTED BY THE RAPID PHENOL MICRO METHOD, THEN ANALYSED BY 12% PAGE AND RESOLVED BY SILVER STAIN...... 87 FIGURE 4.4 LPS PROFILE OF THE TWO IBD CASES P1CD AND P3UC ...... 91

xii

FIGURE 4.5 COMPARING THE INTENSITY OF THE LOWER LPS BAND BETWEEN P3UCO1, P3UCB1 AND P3UCLW2...... 91 FIGURE 5.1 DETECTION OF TLR4 IN HT-29 CELLS INFECTED WITH C. CONCISUS STRAINS FOR 24 HOURS ...... 106 FIGURE 5.2 COMPARISON OF MEAN EXPRESSION (FOLD CHANGE) OF TLR4 IN HT-29 CELLS INFECTED WITH C. CONCISUS STRAINS ISOLATED FROM PATIENT WITH IBD AND HEALTHY CONTROLS...... 107 FIGURE 5.3 DETECTION OF MD-2 IN HT-29 CELLS INFECTED WITH C. CONCISUS STRAINS FOR 24 HOURS...... 110 FIGURE 5.4 COMPARISON OF MEAN EXPRESSION (FOLD CHANGE) OF MD-2 IN HT-29 CELLS INFECTED WITH C. CONCISUS STRAINS ISOLATED FROM PATIENT WITH IBD AND HEALTHY CONTROLS...... 111 FIGURE 5.5 DETECTION OF TLR5 IN HT-29 CELLS INFECTED WITH C. CONCISUS STRAINS FOR 24 HOURS BY: ………………………………114 FIGURE 5.6 COMPARISON OF MEAN EXPRESSION (FOLD CHANGE) OF TLR5 IN HT-29 CELLS INFECTED WITH C. CONCISUS STRAINS ISOLATED FROM PATIENT WITH IBD AND HEALTHY CONTROLS...... 115 FIGURE 5.7 DETECTION OF TLR2 IN HT-29 CELLS INFECTED WITH C. CONCISUS STRAINS FOR 24 HOURS ...... 118 FIGURE 5.8 COMPARISON OF MEAN EXPRESSION (FOLD CHANGE) OF TLR2 IN HT-29 CELLS INFECTED WITH C. CONCISUS STRAINS ISOLATED FROM PATIENT WITH IBD AND HEALTHY CONTROLS...... 119 FIGURE 6.1 TLR4 DOWNSTREAM SIGNALLING PATHWAY INDUCED BY LPS LEADS TO THE EXPRESSION OF TARGET GENES, INCLUDING IL-8 AND COX2...... 126 FIGURE 6.2 RECOGNISING BACTERIAL PAMPS BY PRRS (SUCH TLRS) CAUSES THE ACTIVATION OF THE TRANSCRIPTION FACTOR NFΚB THROUGH THE PHOSPHORYLATION AND THEN THE DEGRADATION OF THE NFΚB INHIBITOR IΚB (SUCH AS IΚBΑ)...... 127 FIGURE 6.3 DETECTION OF COX-2 IN HT-29 CELLS INFECTED WITH C. CONCISUS STRAINS FOR 24 HOURS ...... 136 FIGURE 6.4 COMPARISON OF MEAN EXPRESSION OF COX-2 IN INFECTED HT-29 CELLS BETWEEN C. CONCISUS STRAINS ISOLATED FROM PATIENT WITH IBD AND C. CONCISUS STRAINS ISOLATED FROM HEALTHY CONTROLS...... 137 FIGURE 6.5 LEVELS OF COX-2 IN HT-29 INDUCED BY THE C. CONCISUS STRAIN P1CDB1(UNSWCD) AT FIVE DIFFERENT MOIS (5, 12.5, 25, 50, 100)...... 138 FIGURE 6.6 PRODUCTION OF IL-8 BY HT-29 CELLS INDUCED BY C. CONCISUS STRAINS...... 140 FIGURE 6.7 EFFECTS OF C. CONCISUS ON DEGRADATION OF IΚBΑ IN HT-29 CELLS...... 142 FIGURE 6.8 LEVELS OF IΚBΑ IN HT-29 INDUCED BY C. CONCISUS STRAIN P1CDB1(UNSWCD) AT FIVE DIFFERENT MOIS (5, 12.5, 25, 50, 100) AFTER 15 MIN OF INFECTION...... 144 FIGURE 6.9 LEVELS OF GLY-TLR4 AND NON-GLYTLR4 IN HT-29 INDUCED BY THE C. CONCISUS STRAIN P1CDB1(UNSWCD) AT FIVE DIFFERENT MOIS (5, 12.5, 25, 50, 100)...... 146

xiii

List of Abbreviations

AFLP Ampliﬁed fragment length polymorphism

CD Crohn’s disease

COX-2 Cyclooxygenase-2

DLG5 Drosophila Discs Large Homolog 5 dNTP Deoxy-nucleotide-triphosphate

DPBS Dulbecco's Phosphate-Buffered Saline

ELISA Enzyme linked immunosorbent assay

FBS Foetal bovine serum

FC Flow cytometry

FISH Fluorescent in situ hybridisation

GIT Gastrointestinal tract

HBA Horse blood agar

HRP Horseradish peroxidase

IBD Inflammatory bowel disease

IECs Intestinal epithelial cells

IFN Interferon

Ig Immunoglobulin

IκB Inhibitors of nuclear factor Κb

xiv

IKKs Inhibitor κB kinases

IL Interleukin

LPSs Lipopolysaccharides

MAC M. avium complex

MAP Mycobacterium avium subspecies paratuberculosis

MAPK Mitogen-Activated Protein kinase

MD-2 Myeloid differentiation-2

MHC Major Histocompatibility Complex

MLST Multi locus sequence typing

MOI Multiplicity of infection

NFκB Nuclear Factor κB

NOD Nucleotide-binding oligomerisation domain

PAMPs Pathogen-associated molecular patterns

PBS Phosphate buffer saline

PCR Polymerase chain reaction

PPARs Peroxisome proliferator–activated receptors

PRRs Pattern-recognition receptors

PVDF Polyvinylidine difluoride

RAPD Random Amplified Polymorphic DNA

SNP Single nucleotide polymorphism

SDS Sodium dodecyl sulphate

Th1 T helper type-1

Th2 T helper type-2

TNF Tumour necrosis

TLRs Toll-like receptors

UC Ulcerative Colitis

WB Western blot

xvi

Abstract

Inflammatory bowel disease (IBD) is an idiopathic gastrointestinal disease that covers a group of disorders with two major types, Crohn’s disease (CD) and ulcerative Colitis (UC). The main characteristic of IBD is the uncontrolled inflammation of the intestines that can’t be down- regulated, resulting in chronic activation of the mucosal immune system and chronic intestinal inflammation. Epidemiological studies reported in the past 50 years showed that there is dramatic increase in the incidence of IBD worldwide. Furthermore, it was found that IBD is a risk factor of colorectal cancer (CRC) which accounts for 15% of total deaths in IBD patients.

The aetiology of IBD is unknown, but it is widely approved that the pathogenesis of IBD is an outcome of the interaction of multiple factors including environmental factors, immunological response dysregulation, genetic susceptibility and bacterial factors.

Studies have shown that IBD arises as a result of the host immune defence response to the enteric commensal bacteria. However, what triggered the gut immune system to attack the enteric commensal bacteria, a cohort of organisms that the gut immune system has co-evolved and lived peacefully with, is still unknown.

Recently, Campylobacter concisus has gained attention as an emerging human enteric pathogen and many recent studies have reported the association between C. concisus and IBD. Whereas, the role of C. concisus in the pathogenesis of IBD is still unknown, the major aim of this PhD thesis is to investigate the potential role that C. concisus plays in the pathogenesis of IBD.

C. concisus is a Gram-negative curved rod; motile by means of a single polar flagellum. C. concisus was first isolated from human gingival plaque.

Studies have shown that oral C. concisus is a source of intestinal colonizing C. concisus that is found in patients with IBD.

In chapter 3, the prevalence of multiple (≥2) oral C. concisus strains in patients with IBD and in healthy controls was examined. A total of 288 oral C. concisus isolates were isolated from saliva samples of 15 patients with IBD (9 CD and 6 UC) and 17 healthy controls using a filtration method, C. concisus was identified by C. concisus specific PCR. Out of the 288 oral C. concisus

xvii

Abstract

isolates 52 different oral C. concisus strains were identified using whole protein profile. It was found that 60% (9/15) of patients with IBD were colonized with multiple oral C. concisus strains (≥ two strains), which was significantly (P < 0.05, Fisher’s exact test) higher than that of the healthy controls (3/17, 18%). We then studied the adhesion and invasion properties of oral C. concisus strains to the human intestinal epithelial cell line Caco-2 using a modified gentamicin protection assay, 16 oral C. concisus strains isolated from the 9 patients with CD and 18 oral C. concisus strains isolated from 12 healthy controls were used in this study. It was found that the mean invasion index of oral C. concisus strains isolated from patients with CD (1.51±0.69) was significantly higher than that of the controls (0.07±0.02) (P < 0.05, unpaired t test). However the mean adhesion index of oral C. concisus strains isolated from patients with CD (1.09±0.39) was not significantly different (P > 0.05, unpaired t test) when compared to the controls (0.94±0.31). This is the first study that reported these two findings; results are shown as mean ± SE.

These findings suggest that patients with IBD are colonized with specific oral C. concisus strains that have the potential to cause enteric disease if colonizing the intestinal tract. The high prevalence of multiple oral C. concisus in patients with IBD may increase the chance of lateral (horizontal) gene transfer between these strains which may produce a new recombinant strain that is more virulent

Furthermore, it was also found that using whole protein profile is an easy, fast and cost effective way to define C. concisus strains. In addition to that, the C. concisus protein profiles revealed that oral C. concisus strains have heterogeneous whole protein profiles (strains isolated from each individual revealed a unique whole protein profile). Unfortunately, it was found that whole protein profile is not suitable for differentiating oral C. concisus strains isolated from patients with IBD from oral C. concisus strains isolated from healthy individuals.

In chapter 4, LPS patterns of C. concisus strains were examined. LPS was extracted from 56 C. concisus strains which included the 52 oral C. concisus strains isolated and shown in chapter 3, two C. concisus strains isolated from intestinal biopsies, and two C. concisus strains isolated from luminal-washout fluid. LPS was extracted by a modified rapid phenol micro method, LPS profiles were then obtained by separating the extracted LPS by polyacrylamide gel electrophoresis and visualised by silver staining. Studying the LPS profiles showed for the first

xviii

Abstract

time that all of the C. concisus strains have a smooth-form LPS profile. Furthermore, it was found that grouping C. concisus strains using the LPS profiles was more successful than using whole protein profile; Out of the 52 oral strains used in this study 33 different LPS profiles were recorded and 25 strains were assigned into 6 different groups (group A-F, LPS patterns 1-6). Group A and B contained 8 (32% of the grouped strains) and 9 (36% of the grouped strains) strains respectively, which were the largest two groups; 33% of the strains isolated were either in group A (pattern 1) or group B (pattern 2), LPS profile was not able to differentiate between C. concisus strains isolated from patients with IBD from C. concisus strains isolated from healthy individuals.

Furthermore, we compared the LPS profile of 3 C. concisus isolates taken from the saliva, intestinal biopsy and luminal-washout fluid of a patient with UC (P3UC), the three isolates had an identical whole protein profile. We found that, the LPS patterns between the isolates were also identical, except that the lower band of the oral strain had a significantly (P < 0.05, unpaired t test) lower intensity in comparison to the lower band of the enteric isolates. Knowing that the lower band of the LPS profile consisted of the core oligosaccharide and lipid A, it was suggested that oral C. concisus could increase the production or/and changes the composition of core/lipidA when introduced to the enteric environment.

In chapter 5, the aim was to investigated if C. concisus could act as a trigger to initiate the development of human IBD. Therefore, the expression of TLR2, TLR4 and its co-receptor MD- 2, and TLR5 in an in vitro human intestinal epithelial cells model (HT-29) induced by eight strains isolated from patients with IBD and three strains isolated from healthy controls were studied. The expression was measured by Western blot (WB) and flow cytometry (FC), and visualised by confocal microscopy. It was found for the first time that C. concisus strains have significantly (P < 0.05, unpaired t test) upregulated the expression of TLR4 and its co-receptor MD-2 but not TLR2 or TLR5 in IECs. Interestingly, the average level of Glycosylated-TLR4 (measured by WB), surface TLR4 (measured by FC), and the surface MD-2 (measured by FC) induced by the C. concisus strains isolated from patients with IBD was higher than that induced by the C. concisus strains isolated from healthy controls, (5.49 vs 2.24), (3.70 vs 1.93), and (2.10 vs 1.36) respectively. The results were significant in the case of surface TLR4 and Gly-TLR4 (P < 0.05, unpaired t test) and showed a borderline results (P = 0.06, unpaired t test) in the case of

xix

Abstract

surface MD-2. It was concluded that the upregulation of intestinal epithelial expression of TLR4 and its co-receptor MD-2 (especially the Gly-TLR4, surface TLR4, and the surface MD-2) may enhance the responses of the gut immune system to Gram-negative commensal bacterial species and lead to the breakdown in tolerance of the gut immune system to intestinal commensal bacteria.

In chapter 6, the proinflammatory properties of C. concisus strains on human IECs and the downstream signaling pathway involved were investigated. Therefore, we examined the in vitro effects of eight strains isolated from patients with IBD and three strains isolated from healthy controls on human intestinal epithelial secretion of the chemokine interleukin (IL)-8, the expression of the proinflammatory mediator cyclooxygenase-2 (COX-2) and the activation of nuclear factor kappa B (NFκB), HT-29 human IEC line was used. It was found that C. concisus strains significantly (P < 0.05, unpaired t test) induced the secretion of IL-8 by the IEC line HT- 29 which was measured by enzyme linked immunosorbent assay (ELISA). It was also found that C. concisus strains significantly (P < 0.05, unpaired t test) induced the expression of COX-2 by the IEC line HT-29 which was measured by WB and FC and visualised by confocal microscopy. Given that all C. concisus isolates induced the production of IL-8 and the expression of COX-2, the effect of C. concisus on activation of NFκB was assessed by examining the degradation of the inhibitors of nuclear factor κB (IκB)α in HT-29 cells following incubation with a representative C. concisus strain (P1CDB1(UNSWCD)), IκBα degradation was measured by WB and visualised by confocal microscopy. It was found for the first time that C. concisus strains activates the NFκB.

Furthermore, we showed that the increase in the C. concisus concentrations from multiple of infection (MOI) 5 up to MOI 100 did not affect the production of IL-8 and the activation of NFκB by the human IECs. However, at high C. concisus concentration (MOI 200) IL-8 production was significantly (P < 0.05, unpaired t test) increased compared to the lower bacterial concentrations (MOI 5-100). Interestingly by using WB, it was found that TLR4 especially Gly- TLR4 was also significantly (P < 0.05) upregulated on the human IECs at a low C. concisus dose (MOI 5). It was concluded from chapter 6 together with chapter 5 that C. concisus could modulate the gut innate immune system and act as a trigger of chronic intestinal inflammation such as IBD.

Abstract

Finally we have shown that the cytokines TNF and IFN-γ at three different concentration (20 ng/ml, 40 ng/ml and 80 ng/ml) significantly (P < 0.05, unpaired t test) increased the adhesion and invasion ability of C. concisus strains to Caco-2 cells, which was measured by a modified gentamicin protection assay. From this finding it was concluded that in IBD patients, the increased production of cytokines would increase the adhesion and invasion of C. concisus to the inflamed IECs, further exacerbating chronic intestinal inflammation.

Based on the above findings model to explain the role of C. concisus in triggering chronic intestinal inflammation such as IBD was proposed, which is shown in the general discussion section (chapter7).

xxi

Chapter 1

Chapter 1: Introduction

1.1 Inflammatory bowel disease

1.1.1 Overview

Inflammatory bowel disease (IBD) is an idiopathic gastrointestinal disease that covers a group of disorders with two major types, Crohn’s disease (CD) and Ulcerative Colitis (UC). IBD is a chronic inflammation of the intestines; this chronic inflammation is a result of chronic activation of the mucosal immune system which causes uncontrolled inflammation of the intestines that cannot be down-regulated 1. The main symptoms of IBD are abdominal pain, rectal bleeding, and severe diarrhoea which lead to malnutrition 2. Furthermore, IBD patients are likely to have arthralgias and could have other chronic inflammatory diseases, particularly primary sclerosing cholangitis, ankylosing spondylitis and psoriasis 3. IBD could develop into colorectal cancer (CRC) which accounts for 15% of total deaths in IBD patients 4.

CD and UC can be distinguished by pathogenesis, underlying inflammatory profiles, pathological diagnosis and treatment strategies 5. The main differences between CD and UC are that the inflammation in CD is transmural which involves full thickness of the intestinal wall from the mucosa to the serosa and are often discontinuous, while inflammation of UC is more involved in superficial mucosal and submucosal layers of the intestinal wall (Figure 1.1) 5. Secondly, the site of inflammation in CD could involve any region of the GIT but mainly in the ileum and colon, while UC always involves the rectum and may extend to the caecum 5. Thirdly, CD has a predominant T helper type 1 (Th1) immune response, which is mediated by the increased production of IL-12 6 and interferon (IFN)-γ 7 in the intestinal mucosa of patients with CD 8,9, while UC is characterized by an atypical T helper type 2 (Th2) immune response in which IL-5, IL-10 and IL-13 are shown to be increased in inflamed UC tissue while IL-4 is rarely detected in UC tissue 10. Lastly CD has a limited response to treatment and surgery, while surgery or colectomy in UC can be curative but could result in the creation of an ostomy 11.

Chapter 1

Figure 1.1 Left image is a photomicrograph of a histological specimen from a patient with CD. Inflammation involved full thickness of the intestinal wall from the mucosa to the serosa. Right image shows mucosal inflammation and erosion in a patient with UC 12.

1.1.2 Epidemiology

The incidence and prevalence of IBD are increasing worldwide, as shown in different epidemiological studies reported in the past 50 years 13,14. In general, North Europe, the United Kingdom (UK) and North America have the highest incidence rates and prevalence for both CD and UC. It is lower in other areas of the world such as southern and central Europe, Asia, Africa, Middle East and Latin America 1314115.

The highest incidence and prevalence of IBD in the world is reported from North America; United States of America (USA) and Canada. It is estimated that about 1.3 million person have IBD in these countries with the ratio of CD to UC being almost equal (1:1.15 respectively) 16. In the USA, an estimated 1 million individuals have IBD, with about 30 000 new cases reported each year 1. The incidence rate in USA has been shown to increase over the years. For example, a study conducted in Olmsted County (Minnesota, USA), has shown an increased incidence in CD from 2.6 cases per 105 person in 1950 - 1959 to 7.9 cases per 105 persons in 1990 - 2000. The incidence rate of UC in the same area was 4.2 cases per 105 persons in 1950 to 1959 which increased to 8.8 cases per 105 persons in 1990 to 2000 17. The study showed a higher incidence rate for CD and UC in women as compared to men during the period 1950-1959 which changed to be more males compared to females during the period 1990-200017 (Table 1.1).

Chapter 1

In the Canadian population, the incidence during the period 1998-2000 in Alberta was 16.5 and 11 per 105 persons per year for CD and UC respectively 18. This is higher than the reported incidence during the period 1966 -1981 for the same area which was 6.56 and 3.31 cases per 105 persons for CD and UC respectively 19. The highest incidence rate recorded in the Canadian population during the period of 1998-2000 was in Nova Scotia where it was 20.2 and 19.5 for CD and UC respectively.

UK has a high incidence rate for IBD which has increased over the years. It was reported that in North Tees District the incidence of CD was 5.3 cases per 105 person, and for UC it was 10.4 cases per 105 person during the period 1971-1977 20. The IBD incidence of the same region has increased in CD reaching to 8.3 cases per 105 person while in UC 13.9 cases per 105 person during the period 1990 - 1994 21. Both studies showed similar incidence rate between males and females. The incidence rate in North Europe countries are similar to the results observed in UK 13. Such as in Sweden, Norway, The Netherlands and Denmark as shown in Table 1.1

The occurrence of UC has been elevated in areas including Japan, South Korea, Singapore, northern India and Latin America which were previously considered to have low incidence of UC, while CD in general remains rare 15. For example in Japan, UC is more common than CD. A report by Ouyang et al. showed a 7 fold increase in the number of UC patients from 10 000 cases in 1985 to above 72 000 patient in 2001 25.

The incidence rate of UC in Japan has also increased over time. During the period 1965- 1979, the incidence was reported to be 0.28 cases per 105 persons 26 which increased dramatically to reach 1.95 cases per 105 persons in 199127, as in CD the incidence rate was low during the period 1965 -1979 which was 0.4 cases per 105 persons 26, the CD incidence rate remained low and almost unchanged in 1991 scoring 0.51 cases per 105 persons 27. Similar trends were shown in Seoul (South Korea) and the middle East countries as shown in Table1.1.

As indicated previously, Asian countries have a lower IBD prevalence and incidence compared to western countries. But that is not always the case, in Northern India a study by Sood et al. 2003 reported an incidence of 6.02 cases per 105 and a prevalence of 44.3 cases per 105 for UC during the period 1999-2000 30. This is similar to that reported from North Europe and North America.

Chapter 1

In Latin America, the incidence of IBD is also low especially CD. A study by Linares et al in1999 reported that the mean annual incidence of UC in Panama was 1.2 cases per 105 persons per year while no CD case reported, and the mean annual incidence of UC in Argentina (Partido General Pueyrredón) was 2.2 cases per 105 inhabitants per year and only one CD case, during the period 1987-1993 31. It was also reported that the IBD incidence is also increasing in Puerto Rico as shown by Appleyard et al in 2004 32 as shown in Table 1.1.

The rising incidence of IBD in Asia and developing countries could be a result of different factors. These include changing the lifestyle to be more western and more industrialized which could affect the diet and possibly other environmental exposures. Other factors could be misdiagnosis of IBD, confusing it with other similar diseases such as infectious causes of diarrhoea, which falsely raised the incidence rate in these countries 15.

Chapter 1

Table 1.1 IBD incidence in different geographical regions in different time periods, showing the increased incidence of IBD over time in different geographical regions.

Country Study Incidence CD Incidence UC Reference period M to F ratio M to F ratio

UK 1971-1977 5.3 10.4 20

1 : 1.3 1 : 1.2

1990-1994 8.3 13.9 21

1.1 : 1 1 : 1.1

Northen Europe

Sweden 1955-1959 1.4 22

1 : 1.3

1975-1979 4.4 22

1 : 1.2

1990-2001 8.3 23

1 : 1

1945 2.0 24

1.1 : 1

2007 19.2 24

1.2 : 1

Norway 1964-1969 1.05 3.29 33

1.16 : 1 1.12 : 1

1990 5.1 10.6 34

Chapter 1

The Netherlands 1979-1983 3.9 6.8 35,36

1.1 : 1

1991-2003 6.21 7.72 37

1 : 1.6 1 : 1.2

Denmark 1962- 1978 1.9 8.1 38

1979-1987 4.1 39

1 : 1.4

2003-2005 8.6 13.4 40

1 : 1.1 1 : 1

North America

USA (Minnesota) 1950-1959 2.6 4.2 17

1:1.7 1:1.3

1990-2000 7.9 8.8 17

1.2 : 1 1.5 : 1

Canada (Alberta) 1966-1981 6.6 3.3 19

1 : 2 1.2:1

Chapter 1

1998-2000 16.5 11.0 18

1 : 1.3 1:1

Canada (Nova 1998-2000 20.2 19.2 18 Scotia) 1:1.44 1:1

Asia

Japan 1965-1979 0.4 0.28 26

0.51 1.95 27

South Korea 1986-1988 0.2 28

1986-1990 0.05 29

2001-2005 1.34 3.08 29

India (Punjab) 1999-2000 6.02 30

10:13

Sri Lanka 2007-2008 0.09 0.69 41

1:1 1:1.5

Middle East

Kuwait 1985 2.6 42

29 : 27

1999 2.4 42

26 : 24

1977-1982 0.45 43

Turkey 2001-2003 2.2 4.4 44

Chapter 1

37:38 45:40

Oman 1.35 45

1:1

Australia

Australia 1967-1977 0.66 46

(NSW, Hunter Valley) 1:1

(NSW, Hunter Valley) 1978-1988 2.1 46

1:1

(Victoria, 2007-2008 17.4 11.2 47 Geelong) 1:1.4 1:1

New Zealand 1969-1978 1.75 5.4 48

51:86 254:202

2004-2005 16.5 7.6 49

293:422 342:326

Latin America

Puerto Rico 1996 0.49 1.96 32

Puerto Rico 2000 1.96 3.32 32

Argentina 1987-1993 1 case only 2.2 31

Panama 1987-1993 No cases 1.2 31

M, male; F, female

The incidence rate of IBD here in Australia (Geelong, Victoria) is among the highest reported in the literature as shown by Wilson et al in 2010. The study showed that the incidence rates for CD and UC were 17.4 and 11.2 per 105 persons respectively in 2007-2008. Furthermore

Chapter 1 they demonstrated that the peak incidence rate was for the age group (15-24) reaching 40 per 105 person (Figure1.2) 47. This incidence rate is extensively higher than reported in 1967- 1977 (0.66 per105 persons) 46, showing more than a 26 fold increase. No significant difference was found between males and females in these studies.

The prevalence rate of IBD in Australia is also among the highest reported in the literature, which was in the range of 300-350 per 105 persons during the period 2007-2008 47.

The similarity of the incidence and prevalence rates of IBD between Australia, North America and North Europe could be of similar genetic and environmental risk factors 47. The author reported that the study population is demographically similar in terms of age structure, socioeconomic status, and ethnic makeup to the rest of the Australian population

Figure 1.2 Incidence rates (per 100,000) of CD and UC in relation to Age, Geelong, Victoria, Australia, between April 2007 and March 2008 47.

1.1.3 Aetiology

IBD is a complex disease in which scientists took different approaches to explain it, but it is widely approved that the pathogenesis of IBD is an outcome of the interaction of the following factors: environmental factors, immunological response dysregulation, genetic susceptibility and bacterial factors (Figure 1.3).

Chapter 1

Figure 1.3 Factors that contribute to the development of IBD. It is widely accepted that IBD is an outcome of the interaction of the following factors: environmental factors, immunological dysregulation, genetic susceptibility and bacterial factors. Adapted from: Sartor, R.B. 2006 50

1.1.3.1 Environmental factor

The high prevalence of IBD in western countries accompanied with increasing incidence in different countries that adopt the western lifestyle, suggests that environmental factors may play a role in the development of IBD 15.

The most studied environmental factor related to IBD is cigarette smoking. A meta-analysis by Mahid et al in 2006 found an association between current smoking and CD. Furthermore they found an association between former smoking and UC, on the other hand smoking had a protective effect on the development of UC when compared to controls 51. Researchers have studied the effect of cigarette smoking on immune function. A study by Sher et al 52 showed that in the colonic mucosa IL-8 concentration was significantly higher in healthy smokers compared to healthy non-smokers, while IL-8 was found to be significantly reduced in smokers with CD compared with non-smokers with CD. On the other hand, smokers with UC had a significantly reduced IL-8 when compared to non-smokers with UC 52. Other studies

Chapter 1 showed the relation of nicotine to IBD, and it was found that subjecting experimental rabbits to high doses of nicotine increase the thickness of adherent mucus on rectal mucosa while subjecting them to low dose decreased thickness 53. A study by Srivastava et al in 1991 found a significant reduction of immunoglobulin (Ig)A but not IgG and IgM in intestinal fluid of smokers with UC when compared to controls (Healthy and non-smokers) 54. The mechanism by which smoking affects IBD is unclear to date.

Another environmental risk factor that could contribute to the development of IBD is the diet. It was found that patients with IBD have a higher dietary intake of sucrose, refined carbohydrates, n−6 polyunsaturated fatty acids, total fats including monounsaturated and saturated fats, and reduced intake of fruit and vegetables prior to the development of IBD 55, 56 57. This type of diet is usually found in western countries which have high incidence rate of IBD.

The change to westernized food diet in Asian and developing countries was proposed to be a factor in explaining the increasing incidence in these countries in the last few decades 58. A study by Shoda et al in 1996 showed that an increased dietary intake of animal protein and n- 6 polyunsaturated fatty acids with less n-3 polyunsaturated fatty acids may contribute to the development of CD in the Japanese population 59. Taking in to consideration that the n-6 to n- 3 fatty acid ratio in typical USA diet is 10-25:1 60 and in European diet is 10-14:1 61, which is higher than the ratio found in the Mesolithic man 1–4:1 61.

Stress is another environmental factor that could be related to IBD. It was found that stress could induce ultra-structural damage in the rat intestinal mucosa causing increase permeability 62. Furthermore it was found that subjecting humans to cold pain stress increased the release of intestinal mast cell (MC) mediators in the jejunum and water secretion was also found to be increased in the subjects’ jejunum. The authors used a modified double-lumen, closed-segment perfusion technique 63. MC mediators can profoundly affect GIT physiology such as increasing intestinal permeability which could lead to mucosal inflammation 64,65. It was also found that stress can change the mucosal pro-inflammatory cytokines, a study by Mawdsley et al in 2006 found that subjecting humans to stress have increased significantly natural killer cell count, leukocyte count, platelet-leukocyte aggregate formation, mucosal TNF release and reactive oxygen metabolite production while rectal mucosal blood flow was found to be reduced 66.

Chapter 1

Another interesting environmental factor is the hygiene factor, it was found that poor sanitation has a protective effect in CD 67. Gent et al in1994 found an association between delayed exposure to enteric infection (caused by increased hygiene) and CD, the study was conducted on 133 patients with CD and matched with the same number of controls with the same age and sex, the results showed that CD was more common in subjects whose first houses had a hot-water tap and separate bathroom in infancy (before the age of 5), on the other hand the same study showed that UC had no clear relation to household amenities in infancy 68. Similar results were found by Duggan et al in 1998 69. It was also found that disrupting the pattern of bowel colonization in childhood by antibiotics increased the risk of IBD 70, 71.

Other interesting studies are those concerned with the luminal microenvironment, which will be discussed in section (1.2.3.4 Bacterial factor). Despite these associations, the exact mechanisms by which these environmental factors initiate the onset of IBD or reactivate quiescent IBD are not well understood 50.

1.1.3.2 Immunological response dysregulation

CD and UC patients have elevated innate and acquired immune responses, chronic inflammation and loss of tolerance to the enteric commensal bacteria 50.

Chronic inﬂammation of IBD occurs when the immune control mechanism is overcome by the production of proinflammatory cytokines, chemokines and the high activity of lymphocytes.

The innate immune system is highly activated in IBD, and produces many proinflammatory cytokines and chemokines. Table 1.2 illustrates the cytokines produced by the innate immune system in IBD. As Table 1.2 show, interleukin (IL)-12, IL-23 and IL-27 are only elevated in CD but not in UC. These cytokines are related in activating Th1 72 and Th17 immune response73.

Chapter 1

Table 1.2 Innate immune response cytokines associated with IBD.

Cytokine CD UC Reference

IL-1β Increase Increase 74

TNF Increase (more Increase 7 than UC)

IL-6 Increase Increase 75

IL-8 Increase Increase 76

IL-18 Increase Increase 77

IL-12 Increase Normal 6

IL-23 Increase Normal 78

IL-27 Increase Normal 78

Intestinal epithelial cells (IECs) are an important part of the innate immune system. They act as a selective barrier between the luminal environment and the rest of the body, activate the immune system by antigen presenting, and have the ability to produce chemokines and cytokines that could regulate the acute and the chronic inflammatory cells in the intestinal mucosa79,80. Cytokines such as tumour necrosis factor alpha (TNF), IL-1, and IL-6 81which are highly thought to have a direct effect on the pathogenesis of IBD. Any defect in the epithelial cells will contribute or even could be the primary defect in IBD 79,80, 82, 83. It was found that in a TNF ΔARE mouse model (a TNF-overexpressing mouse model 84) of ileitis, a selective chronic overproduction of TNF by IEC caused full development of Crohn’s-like pathology 85.

Summary of the cytokines and chemokines produced by IECs is illustrated in Figure1.4.

Chapter 1

Figure 1.4 Summary of the cytokines and chemokines produced by the intestinal epithelial cells and the chemokines usual target 86. Abbreviations: ENA, Epithelial- derived neurphil –activating peptide; IL, Interleukin; IP-10, Interferon-inducible protein-10; I-TAC, Interferon-inducible T cell chemoattractant; GM-CSF; Granulocyte macrophage –colony stimulating factor; GRO, Growth-related oncogene; MDC, Macrophage-derived chemokine; MEC, Mucosa-associated epithelial chemokine; MCP, Monocyte /macrophage chemotactic peptide; MEC, Mucosa associated epithelial chemokine; RANTES, Regular upo activation; SCF, Stem cell factor; TECK, Thymus- expressed chemokine; TGF, Transforming growth factor.

For the IECs to act as a selective barrier, tight junctions between the cells have to be maintained. In IBD alterations in gut permeability factors have been reported 87, and recently an association between genes important in mucosal transport and integrity; SLC22A4, SLC22A5 and DLG5 were found to be associated with IBD 58. Details of these genes will be discussed in the genetic factor section 1.1.3.3.

The innate immune system carries receptors that recognize conserved microbial motifs called pathogen-associated molecular patterns (PAMPs). These receptors are termed pattern recognition receptors (PRRs). PRRs are classified into four main families: Toll-like receptors (TLRs), nucleotide-binding oligomerisation domain, (NOD) receptors, retinoic acid-inducible gene I (RIG-I)-like receptors and the C-type lectin receptors 88. In the GIT, TLRs are expressed on the IECs, myofibroblasts, enteroendocrine cells, and on the immune cells within the lamina propria, such as T cells, macrophages and dendritic cells (DCs) 89.

Chapter 1

Recognition of PAMPs by TLRs leads to the activation of NFκB, MAPK, and interferon response factors through cellular signalling cascades 90, 91. TLR3 and TLR5 are normally expressed on human IECs, while TLR2 and TLR4 are minimally expressed on these cells. A study by Cario et al in 2000 showed that in IECs of patients with CD there is significant down-regulation of TLR3, an up-regulation in TLR4, while there were no changes in TLR2 or TLR5 expression. Further in IEC of UC patients it was found that TLR4 was strongly upregulated with no change in TLR2, TLR3 and TLR5. These results suggest that alterations in the innate response system may contribute to the pathogenesis of IBD 92.

NOD1 and 2 (also known as CARD4 and CARD15 respectively) have an important role in innate immunity by recognizing peptidoglycan, a component of the bacterial cell wall. NOD1 and 2 are expressed predominantly by antigen-presenting cells (APC) and epithelial cells 93. Polymorphisms in NOD1 94 and 2, result in a susceptibility to develop IBD, particularly CD in the case of NOD2 95, 96, 5. NOD2 polymorphisms will be discussed in further details in the genetic disregulation section (1.1.3.3.1).

IL-23 and IL-12 were found to be elevated in CD inflamed tissue. These cytokines are predominantly produced by activated dendritic cells and phagocytic cells 97. It was found that infecting IL-10-/- mice with heat-killed whole bacteria antigens will lead to the hyperproduction of IL-12 and IL-23 by abnormally differentiated subsets of intestinal macrophages, which cause Th1 activation resulting in the development of intestinal colitis 98. Further a review by Duerr et al in 2006 found a highly significant association between CD and the IL23R gene, in which uncommon coding variant in the gene conferred strong protection against CD 99.

TNF is also found to be elevated in IBD, non-lymphoid cells mostly macrophages are found to be responsible producing this cytokine 7, 97. It was found that treating IBD patients with monoclonal antibody to TNF has a curative effect. A study by Targan et al showed that giving chimeric monoclonal antibody cA2 (an antibody for TNF) has an effective impact in treating patients with moderate to severe CD 100, and it was shown by Seow et al that treating UC patients with infliximab (a monoclonal antibody against TNF) predicts clinical remission, endoscopic improvement and a lower risk of colectomy 101.

CD is associated with Th1 and Th17 cytokine profiles, whereas UC is associated with atypically Th2 response 102, see Table 1.3.

Chapter 1

Studies indicate a predominant expression of Th1 cytokines in CD inflamed tissue such as IFN-γ7103 and IL-2 74, and it was found that cytokines necessary for the development of a Th1 immune response such as IL-126,TNF (TNF is a cofactor for the mucosal Th1 response that has a costimulatory effect on Th1 cells104) 7 and IL-18 (involved in Th1 clone development)77 were also increased in CD inflamed tissue6. It was found that IL-12Rß2 (which is only expressed in Th1 but not Th2) expression was increased in active CD 105. In addition, T-cells isolated from inflamed CD lesions were found to have an up-regulated T-bet, T-bet is a Th1 transcription factor that plays an important role in Th1 specific cytokine production 106. All the previous studies indicate that Th1 cells are a main mediator of the pathogenesis of CD.

Th17 cells, a subset of T helper cells which are associated with inflammatory response, were recently shown to be associated with IBD. Th17 development is stimulated by the production of TGF-β, IL-6, and IL-23 by innate immune cells and APC, especially dendritic cells (DC) 50, 97,102,107. Th17 produces IL-17, IL-21, IL-6, TNF, IFN-γ, and IL-22 cytokines 108,109. It was found that IL-23 and IL-17 are increased in IBD tissues and in different animal models of IBD 110,111. A study by Yen et al in 2006 showed that colitis was prevented in IL-10 deficient mice in the absence of IL-23, and the target of IL-23 is a unique subset of tissue-homing memory T cells, which are specifically activated by IL-23 to produce the proinflammatory mediators IL-17 and IL-6 111.

UC is characterised by a Th2 atypical immune response. Typical Th2 response is characterised by the production of IL-4, IL-5, IL-9 and IL-13 112. While IL-5 and IL-13 112,113 are over expressed in UC, there is no evidence that T cells from UC produce increased amount of IL-4, adding to that, it was even found IL-4 production by lamina propria (LP) CD4+ T cells from inflamed UC mucosa is actually decreased when compared to healthy controls LP T cells 12,50,113. Fuss et al in 2004 suggested that UC has an atypical Th2 response mediated by non-classical natural killer T (NKT) cells that produces IL-13, these non-classical NKT cells are activated by APCs that express CD1d and were found to have a cytotoxic potential for epithelial cells 114.

Chapter 1

Table 1.3 T-cell response lymphokines associated with IBD.

Lymphokines CD UC Reference

IFN-γ Increased Normal 114

IL-2 Increase Increase 74

IL-4 Decrease Decrease 114

IL-5 Normal Increase 114

IL-13 Normal Increase 114

IL-17 Increase (significantly Increase 110 higher than UC)

IL-21 Increase Normal 115

The role of the other hand of the acquired immune system the B-cells in IBD is not extensively studied as T-cells.

It was found that in the intestinal mucosa of IBD patients (CD and UC) have a high production of IgG by mucosal lymphocytes 116 which was found to be almost 30 times more when compared to healthy controls 116. IgA and IgM antibodies also were increased 2.0 and 4.8 times respectively 116. A study by Macpherson et al in 1996 showed by western blots and by enzyme linked immunosorbent assay (ELISA) that the mucosal immune response in IBD patients is directed against cytoplasmic proteins of bacteria within the intestinal lumen, the study showed that the mucosal IgG from CD and UC inflamed area bound to proteins of a range of non-pathogenic commensal faecal bacteria such as Escherichia coli (E. coli), Clostridium perfringens and Bacteroides fragilis in a significantly greater manner when compared to healthy controls 117. A similar finding in colitic animal model was found by Brandwein et al in 1997 in which they reported by using enhanced chemiluminescence western blotting that sera from C3H/HeJBir mice (a spontaneously colitic mouse strain) had a reproducible banding pattern on Western blot to bacterial Ags, whereas sera from C3H/HeJ mice (noncolitic parental mice) did not 118. The serum Abs detected on immunoblot were

Chapter 1 primarily IgG2a. Bacteria were the mouse own cecum bacteria, Eubacterium species, two Lactobacillus species, E. coli, Clostridium ramosum, Bacteroides fragilis and others 118. Lodes et al in 2004 found that serum IgG levels to flagellins were elevated in patients with CD, but not in patients with UC when compared to controls119.

The serum level for anaerobic gram-positive coccoid rods such as Euhacterium, Peptostreptococcus and Coprococcus spp were found to be higher in patients with IBD comparing to healthy controls, and higher in CD when compared to UC120,121. One of the early studies by Matthews et al in 1980 found that by using agglutination test with the genera Eubacterium and Peptostreptococcus, 54% and 11% of sera from patients with CD and UC were positive respectively, while none of the sera from healthy controls showed positive results122. Further, the prevalence of serum antibody to many strains of E. coli appears to be increased in CD and UC 123.

Autoantibodies in patients with IBD sera for antigens in the cytoplasm of neutrophils (ANCA), exocrine pancreas and cytoskeletal proteins was reported. Perinuclear antineutrophil cytoplasmic antibodies (P-ANCA) have been reported in sera of patients with IBD. Saxon et al in 1989 found P-ANCA in 84% of patients with UC, and 20% of patients with CD 124. Further study by Gigase et al in 1997 found IgA P-ANCA in 52% of patients with UC and in 9% of CD patients, and the presence of IgA P-ANCA was significantly associated with the occurrence of blood in the faeces of UC, while IgG P-ANCA was found in 56% of patients with UC and in 7% of patients with CD 125. Serum autoantibodies against exocrine pancreas had a prevalence of 39% in patients with CD and only 4.5% in patient with UC 126. Autoantibody against cytoskeletal proteins of human intestinal cells were detected in Sera from patients with CD using western blotting in a study by Mayet et al in 1990., The same study also demonstrated the following antibodies: cytokeratin 18 autoantibodies (IgG 20.0%; IgM 6.7%; IgA 13.3%), actin antibodies (IgG 36.7%; IgM 48.3%, IgA 26.7%), desmin antibodies (IgG 6.7%; IgM 15.08%; IgA 50%), vimentin antibodies (IgG 3.3%; IgM 16.7%; IgA 10.0%) and tropomyosin antibodies (IgG 3.3%; IgM 3.3%, IgA 5.0%). significant correlation was only found on two cases; between cytokeratin 18 antibodies (IgM- type) level and the BEST index of activity, and between desmin antibodies (IgM-type) level and the van HEES index of activity (BEST and van HEES are index used to evaluate the activity and severity of CD) 127.

Chapter 1

1.1.3.3 Genetic Factors

Genetic factors play an important role in the pathogenesis of IBD. Genes associated with IBD are involved in several pathways that are crucial for intestinal homeostasis and immune- regulation such as regulation of adaptive immunity, microbial clearance, microbial defence, reactive oxygen species (ROS) generation, and mucosal barrier integrity 50,128.

Genetic contributions to IBD have been confirmed by studies on twins, and the contribution was found to be stronger in CD as compared to UC 129. A study on Danish twins by Orholm et al in 2000 found a strong genetic influence on IBD occurrence. The authors showed that the concordance rate in monozygotic (MZ) twins was 58.3% and 18.2% in CD and UC, respectively, while it was only 0% and 4.5% in the case of dizygotic (DZ) twins, respectively 130. Further, a study on Swedish twins by Tysk et al in 1988 showed a higher rate of concordance in MZ twin pairs with CD 58.3% compared with twins with UC 6.3%. The study concluded that heredity as an aetiological factor is stronger in CD than in UC 131.

Although CD and UC share some susceptibility genes, it seems that the level of severity in CD and UC is caused by separate genetic factors 132,133. Epidemiological studies have suggested that while a group of genetic alterations are common to both CD and UC, other gene alterations are specific to CD or UC 3. The genes associated with both CD and UC are MDR1 (multidrug resistance protein , encodes P-glycoprotein 170), PPARG (encodes peroxisome-proliferator-activated receptor), IL23R (encodes a crucial subunit for the IL- 23 receptor), IL12B (encodes the p40 subunit of IL-12 and IL-23), STAT3 (encodes signal transducer and activator of transcription3), and DLG5 (Disks large homolog 5, encodes a scaffolding protein involved in epithelial integrity) 3,50.

The genes which are only associated with CD are NOD2 (encodes nucleotide-binding oligomerization domain protein 2), SLC22A4 (solute carrier family 22 member 4, encodes OCTN1; organic cation transporter novel type 1), SLC22A5 (solute carrier family 22 member 5, encodes OCTN2; organic ction transporter novel type 2), ATG16L1 (encodes autophagy related 16-like protein 1) and IRGM (encodes immunity-related GTPase family M) 3, 50.

The following is a brief insight of some of the dysregulated genes in IBD.

Chapter 1

1.1.3.3.1 NOD2 (CARD15 )

Nucleotide-binding oligomerization domain-containing protein 2 (NOD2), also known as caspase recruitment domain-containing protein 15 (CARD15) is a pattern recognition receptor that functions as an intracellular sensor for bacterial peptidoglycan. The gene responsible of encoding NOD2 is Card15 134. NOD2 gene is located at the IBD1 locus on chromosome 16 (16q12) 135.

Activation of NOD2 leads to the activation of nuclear factor κB (NFκB) and mitogen- activated protein (MAP) kinase signalling pathways 136. An increased activation of NFκB molecules has been observed in the lamina propria of individuals with CD and UC 137.

Studies have shown the importance of NOD2 in the host immune defences, especially on the intestinal epithelial cells 138. A study by Hisamatsu et al in 2003 found that intestinal cell lines with a NOD2 frame shift mutation (Leu1007fsinsC) cause a weak control against the intracellular growth of Salmonella typhimurium. 139. In addition a study by Kim et al in 2008 showed that NOD2 plays a role in mediating antibacterial responses in vivo using NOD2 deficient mice that were orally inoculated with Listeria monocytogenes. After the incubation period, L. monocytogenes was found colonizing the liver and spleen 3,140.

Hugot et al in 1996 identified a putative CD-susceptibility locus on chromosome 16 which is centred around loci D16S409 and D16S419, IBD1. The identification of the locus was after a non-parametric two-point sibling-pair linkage method study, performed on two independent panels of families with multiple affected members with CD. The first panels of families had 25 Caucasian families while the other panel had 53 families, each family of both panels had at least two affected siblings with CD. This early study 135 supported with other studies 141,142 points to a gene in the pericentromeric region of chromosome 16, which contribute to susceptibility to CD.

In 2001, using a positional-cloning strategy, based on linkage analysis followed by linkage disequilibrium mapping at a 235 CD families and 100 multiplex and 59 simplex UC families, Hugot et al found three independent associations for CD, a frameshift variant and two missense variants in NOD2 143. Furthermore, in the same year using transmission disequilibrium test and case-control analysis, Ogura et al found an association between a frameshift mutation caused by a cytosine insertion, 3020insC in NOD2 and CD 144.

Chapter 1

The major three variants identified in NOD2 are Arg702Trp (2104 C>T), Gly908Arg (2722 G>C), and Leu1007fsinsC (cytosine insertion causing a frameshift at Leu1007 145,146 Yazdanyar et al in 2009 reported a meta-analysis of 75 articles to assess the impact of NOD2 variants on the CD and UC risk across diverse populations found that the odds ratios per allele for CD were 2.2 for Arg702Trp, 2.6 for Gly908Arg, and 3.8 for Leu1007fsinsC. Further, the study showed an odds ratio for simple NOD2 heterozygotes of 2.4 (confidence interval 2.0-2.8), 6.7 for compound heterozygotes (confidence interval 6.0-13) and for NOD2 homozygous of 6.7 (confidence interval 4.1-10.9). Control individuals were free of IBD 147.

The prevalence of these three major NOD2 variants differed between populations. At least one of these variants is presented in 25.3% of German and British population with CD 148, 25.7% of American white children with CD 149, and 18% of Italian population with CD, but are rare in African Americans children and Hispanic American children with CD (1.6% and 2.2% respectively) 149 and not present in Turkish 150, Japanese 151, Chinese (Hong Kong) 152 and Chinese (Zhejiang) 153 patients with IBD. No association between UC and NOD2 variants was shown in any of the previous studies.

These three major variants of NOD2 were found to affect the leucine-rich repeat of the CARD15, which is the domain that senses the bacterial peptidoglycan and lipopolysaccharide 144,154 . These variants lead to a decreased activation in NFκB in human mononuclear cells following stimulation of the CARD15 disease variants, which contrasts the reality of what happens in CD 155. However, it was hypothesized that an impaired inﬂammatory response to microbes, shown by the lack of defensins and cytokines production, rather than an overly aggressive inﬂammatory response by a defective intestinal innate immune system may underlay the initial phase of IBD 83,156. A decrease in alpha defensin expression by the affected ileum and colon of patients with CD that have NOD2 mutation was reported by Wehkamp et al in 2004 157. Another hypothesis is that mutation in NOD2 will affect the antigen presenting cells response to recognize microbial peptides, which will affect the response by the regulatory T-cells and effector T-cells, leading to disruption in the mucosal homeostasis 156.

Nevertheless, dysregulation of the CARD15 pathway alone is insufficient to induce CD, which is supported by the fact that CARD15 deficient mice do not present intestinal inflammation after bacterial infection 158 and also CARD15 mutations are absent in patients with IBD from some geographical regions such as China, Korea and Japan 138.

Chapter 1

1.1.3.3.2 SLC22A4 and SLC22A5

SLC22A4 and SLC22A5 are located on the IBD5 locus at chromosome 5 (5q31). SLC22A4 encodes for organic cation transporter 1 (OCTN1) and SLC22A5 encodes for organic cation transporter 2 (OCTN2) 159.

Peltekova et al in 2004 have identified a mutation on SLC22A4 in which a C is substituted into a T in the gene exon 9 (1672C>T), and another mutation on the SLC22A5 in which a G is transverse into a C in the gene promoter (207 G->C). The polymorphism on SLC22A4 causes a Leu503Phe amino-acid substitution in OCTN1, while the polymorphism on SLC22A5 causes a 5' untranslated region variant. These two mutations form a haplotype which was associated with CD 138,160.

A meta-analysis study by Xuan et al in 2011 found a significant association between the polymorphisms on (SLC22A4/SLC22A5) and susceptibility of CD in Caucasian population, while significant associations were not found in the East Asian population161. Countries included in the Caucasian population are161: Belgium 162, Italy163,164, England 165, Germany 166, New Zealand 167, Spain 168, Sweden 169, Netherlands 170 and Poland 171. The East Asian countries are: Japan 172,173 and China 174. Further, a recent study in 2011 by Chua et al did not observe any significant correlation of SLC22A5 polymorphisms with CD in the Malaysian population175 which supports Xuan et al findings161.

Polymorphisms on SLC22A4/SLC22A5 could contribute to the development of particularly refractory CD, as it was found that refractory CD in Slovenian patients who do not respond to standard therapy, including patients who develop fistulas are significantly associated with these polymorphisms. The same study also found that the main contributor to the IBD pathogenesis in the IBD5 region is SLC22A5, after finding that the expression of the SLC22A5 is lower in blood lymphocytes from IBD patients compared to control group, and colon tissue biopsies showed a decreased expression of SLC22A5 in inflamed tissue biopsies compared to non-inflamed colon176.

The polymorphism on SLC22A4 and SLC22A5 could contribute to CD susceptibility by altering the transcription and the transporting functions of OCTN and interacting with variants in another gene associated with CD such as CARD15 as shown by a functional study done by Peltekova et al 177,178. It has been reported that the combination of SLC22A-TC

Chapter 1 homozygosity and one or more of the common CARD15 disease susceptibility alleles engendered a 7.5-fold increase in risk for CD 179.

The mechanism by which the SLC22A4 and SLC22A5 variants contribute to CD is not clear to date.

1.1.3.3.3 DLG5

Drosophila Discs Large Homolog 5 (DLG5) gene is located at chromosome 10 (q22-23). By using positional cloning to identify genetic variants in DLG5 associated with IBD, Stoll et al in 2004 found that there are four common haplotypes (A, B, C and D) at DLG5. Two haplotypes were associated with IBD and CD. One of these two haplotypes (haplotype D) had a non-synonymous single nucleotide polymorphism (SNP) (G > A) at position 113(113G > A) and it has been found to be significantly over-transmitted in patients with IBD and CD in particular, this polymorphism causes an amino acid substitution in which an arginine is substituted with glutamine (R30Q) in the DUF622 domain of DLG5. The second haplotype (haplotype A) distinguished by the SNP DLG5_e26 (insertion or deletion of adenine in exon 26), was significantly undertransmitted in IBD and CD. The group also identified a third SNP in DLG5 in which a transversion (C > A) at axon 23 (4136C > A )which also had a significant association with CD, this polymorphism causes an amino acid substitution of proline by glutamine (P1371Q) 166,180. Further, it was suggested by Stoll et al that there could be a genetic interaction between 113 G > A polymorphism and CD associated CARD15 mutations 180.

The 113 G > A polymorphism in DLG5 was found to be associated with patients with IBD in German 180, American (Central Pennsylvania) 181, Canadian (Quebec) and Italian 182 populations, as well as with patients with CD in German population180, and American (Central Pennsylvania) 181.

The DLG5_e26 polymorphism was found to be associated with patients with IBD and patients with CD in Canadian population 183 and Malaysian population 175.

The 4136 C > A polymorphism in DLG5 was found to be associated with IBD and CD in, Canadian population (Non-Jewish)183, Central Pennsylvania-USA 184, and Malaysian population 175.

Chapter 1

Török et al in 2005 did not find any association between any of the three polymorphisms in DLG5 with patients with CD or UC from Southern Germany 166

A study by Friedrichs et al in 2006 found that The 113 G > A polymorphisms in DLG5 is a susceptibility factor for CD in men (OR = 2.49, P < 0.001) but not women (OR = 1.01, P = 0.979) in a study sample from Germany, Italy and Canada (Quebec), using multivariate logistic regression analyses 185. Further a study by Lin et al in 2011 found that the association between the 4136 C > A polymorphism (A allele) in DLG5 and IBD was female specific (OR = 3.77, P = 0.01) but not male (OR = 1.1, P = 0.81) using a gender distribution analysis on patients with IBD and healthy controls from Central Pennsylvania-USA 184.

It has been found that DLG5 plays a role in regulating cell growth and maintaining cell shape and polarity 159,186. A study by Wakabayashi et al in 2003 suggested an epithelial function for DLG5 as a binding partner of vinexin at sites of cell-cell contact 187,188.Preliminary results from Stoll et al in 2004 on the expression of DLG5 mRNA in a variety of tissues confirmed the presence of the transcript in the colon, the intestine and isolated intestinal epithelial cells. Thus, it can be hypothesised that genetic variants in DLG5 interfere with epithelial barrier function in the colon and has a role in maintaining epithelial structure 50,188.

Other population based studies need to be done to confirm the association between these polymorphisms and IBD, and to confirm the mechanism by which the DLG5 polymorphisms causes IBD.

1.1.3.3.4 PPARG

Peroxisome proliferator–activated receptors (PPARs) are nuclear transcription factors which belong to the steroid receptor super family. Fatty acids activate these receptors, when activated they are involved in the transduction of metabolic and nutritional signals into transcriptional responses 189. One of these transcription factors is PPAR-γ which plays an important role in the maintenance of mucosal integrity in the intestine. PPARG is the gene responsible of encoding PPAR-γ 190, 191.

After a Linkage disequilibrium mapping, followed by sequencing, expression analysis and immunohistochemistry was performed on SAMP1/YitFc (SAMP1/Fc) mouse strain (a mouse model of spontaneous chronic ileitis that has many features of CD in humans), it was found that PPARG is a novel susceptible gene in this mouse model. Further, in the same study rare 3 SNPs in PPARG1 were found to be significantly associated with patients with CD in

Chapter 1

Virginia- USA (almost 95% of patients and controls are Caucasian). The 3 SNPs were given the name SNP 1-3. SNP1 and SNP2 had the strongest association showing a P value of <10-5 and <0.005 respectively, SNP3 showed a weak evidence of association with patients with CD P = 0.053. SNP1 and SNP2 are located in the intron between exon A2 and coding exon 1, while SNP3 encodes a silent base change in coding exon 6 of the gene 192.

SNP1 is a G > A SNP at position 12350898 (12350898G>A), SNP2 is a G > A SNP at position 12359887 (12359887G>A), and SNP3 is a C>T SNP at position 12467406 (12467406C>T) 192.

A genotyping of a large IBD cohort study, 2261 individuals (616 CD, 365 UC and 1280 healthy) from Germany, for the three SNIPs :12350898G>A, 12359887G>A, 12467406C>T 192 using polymerase chain reaction (PCR), restriction fragment length polymorphism (RFLP) and subsequent agarose gel electrophoresis. Showed no significant differences in the minor allele frequencies of the three SNPs in CD or UC patients compared to healthy controls. Further the genotype–phenotype analysis in the same study did not reveal any association of the three SNPs with an ileal disease involvement or any other specific anatomic location 193. This contrasts the results shown by Sugawara et al 192.

The Pro12Ala polymorphism, a polymorphism in PPARG that influences insulin sensitivity and risk of type II diabetes in various ethnic populations, was examined in Turkish patients with IBD. The results showed no significant differences in the frequency of the Pro12Ala polymorphism in the PPAR-γ gene among subjects with CD, UC and controls (P > 0.05) 194.

It has been shown that PPAR-γ is expressed at high levels in the colonic epithelium. PPAR-γ and its high-affinity synthetic ligands such as thiazolidinediones are involved in the regulation of colon inflammation by different mechanisms, such as inhibiting NFκB activity which leads to the down regulation production of inflammatory cytokines, chemokines and other inflammatory signals 191,192. A study by Lytle et al in 2006 showed that treating IL-10 -/- mice (IBD mouse model) with PPAR-γ ligand rosiglitazone has slowed the onset of spontaneous IBD, rosiglitazone has shown an anti-inflammatory activity in the mouse model 195.

Further studies need to be done to investigate the PPARG variants in other populations, and clarify their role in IBD.

Chapter 1

1.1.3.3.5 MDR1

MDR1 is a multidrug resistance gene that encodes the membrane transport protein P- glycoprotein-170 (pgp-170). P-glycoprotein is an ATP-binding cassette (ABC) transporter which was initially identified as a protein responsible for multidrug resistance in cancer cells 196 197 198. P-glycoprotein 170 functions as an ATP-dependent efflux transporter pump and plays an important role in drug disposition and response 198.

MDR1 has been proposed as a candidate gene for IBD for several reasons 199. First, it was found that mdr1a-/- mice could develop a severe, spontaneous intestinal inflammation when maintained under specific pathogen-free conditions. The pathology of the intestinal inflammation seen in mdr1a-/- mice was similar to human IBD; dysregulated epithelial cell growth and leukocytic infiltration into the lamina propria of the large intestine 200. Second, MDR1 is located at chromosome 7q22 which was identified by Satsangi et al in 1996 as a chromosome that could carry a susceptible loci for both CD and UC, based on a genome search for susceptibility genes in IBD involving 186 affected sibling pairs from 160 nuclear families in UK (all Caucasian) 201. Third, Langmann et al in 2004 found a significant reduced expression of the messenger RNA of MDR1 in the colon of patients with UC but the expression was unaffected in patients with CD.

The most studied SNPs that correlate with the activity and expression of pgp-170 are; 2677G > T/A in exon 21 (Ala893Ser/Thr) and 3435C> T in exon 26 (exchanging C with T will not affect the amino acid sequence of PGP, but it was found that the T allele appears to be associated with markedly lower MDR1 expression compared with the C allele 202).

The 2677G>T/A SNP was found to be significantly associated with IBD in American (Jewish and non-Jewish) population (the 2677T allele) 203, and British population (the 2677T allele with UC not CD) 204. The 2677G>T/A SNP was not significantly associated with IBD in Scottish 205, Italian 206, white Spanish (CD not UC) 207 and Hungarian 208 populations.

The 3435C >T SNP was found to be significantly associated with IBD in German population (UC, not CD) 209, Scottish population (UC, not CD) 205, and white Spanish population (CD not UC) 207. The 3435C >T SNP was not significantly associated with IBD in American population (Jewish and non-Jewish) 203, Italian population 206, British population 204 and Hungarian population 208.

Chapter 1

A recent study by Huebner et al in 2009 aimed to correlate 9 different SNPs in the MDR1 to patients with IBD. Patients and healthy controls used in this study were from New Zealand and were Caucasian from European ancestry, a total of 373–388 CD participants, 386–401 UC participants and 195–201 healthy controls were used in this study. SNPs studied were: 2677G > T/A, 3435C >T, rs1045642 (SNP in exon 26), rs1128503 (SNP in exon 12_1), 1236C >T, rs1186745 (SNP in intron 27_5), rs2235046 (SNP in intron 16), rs3789243 (SNP in intron 3), rs2032582 (SNP in exon21). The results showed that heterozygous carriers for the variants 1236C > T, rs2235046 and 2677G>T/A had a lower risk of developing ulcerative colitis compared to homozygotes. A significant association was found between MDR1 C3435T and disease behaviour in CD, in which the T allele increased the probability of ileal/stricturing CD. SNP rs3789243 was found to be associated with pancolitis in UC patients, and none of the analysed markers were associated with CD or IBD 210. The results also showed an increased frequency of heterozygosity in the healthy controls which suggests a heterozygous advantage 210.

The results of these studies imply that MDR1 is a complex gene, and the contribution of MDR1 variants to IBD is unclear to date.

1.1.3.4 Bacterial factors

Many observations and experimental trials have shown strong evidence for the role of bacterial factors in the pathogenesis of IBD. Evidence such as: prevention of further inflammation in patients with CD by faecal stream diversion but the recurrence of inflammation upon restoration of the faecal stream 211,212 and washing out luminal content (whole gut irrigation) had a beneficial effect in active CD 213. Secondly, the positive effect of antibiotics on the clinical progress in some cases of both CD 214 and UC 215. Thirdly, probiotic combinations were found to be effective in preventing the onset of acute pouchitis in human IBD patients 216 and improved clinical scores of human CD 217. Fourth, the association between genetic factor variants that affects intracellular pathways concerned with recognition of microbial molecular patterns and IBD, such as NOD2 variants (receptor for bacterial muramyl dipeptide) as explained in the genetic factor section (1.1.3.3). Fifth, triggering of intestinal inflammation in most IBD animal models (B27 transgenic rats and IL- 10 -/- mice) depends on the presence of bacterial flora 218, 219. Furthermore, experimental colitis is attenuated in IBD animal models when treated with broad spectrum antibiotics 132.

Chapter 1

Serological 117 and T-cell responses 220 to enteric bacteria have been displayed in CD and UC patients. it was found that mucosal IgG from CD and UC inflamed areas bound to proteins of a range of non-pathogenic commensal faecal bacteria such as Escherichia coli, Clostridium perfiringens and Bacteroides fragilis in a significantly greater manner compared to healthy controls 117, In UC it was found that almost 60% of patients exhibited antibodies that react with enteric bacteria. It is reported that CD is associated with an elevated Th1 and Th17 cytokine response shown by the increased production of IFN-γ and IL-17, while UC is associated with atypical Th2 cytokine response shown by increased production of IL-5 and IL-13 but not IL-4 towards bacteria within the intestinal lumen, as explained earlier in the immunological response dysregulation section (1.1.3.2).

Several bacteria have been related to IBD, including Bacteroides vulgatus for UC, specific invasive strain of Escherichia coli and Mycobacterium avium subspecies paratuberculosis for CD, and most recently, Campylobacter concisus which is gaining more attention as a potential causative agent of IBD.

The following is a brief overview showing the association between these bacteria and IBD.

1.1.3.4.1 Mycobacterium avium subspecies paratuberculosis and CD

Mycobacteria are Gram-positive, acid-fast, pleomorphic, non-motile rods belonging to the order Actinomycetales 221.

Mycobacterium avium subspecies paratuberculosis (MAP) is one of the M. avium complex (MAC) 222. MAC is a ubiquitous environmental microorganism which compromises two species M. Avium and M intracellular 222. MAC could colonize animal and human intestines and usually don’t cause disease unless the host is immune-compromised 223. On the other hand MAP is a specific causative agent of chronic inflammation in many different animals such as monogastric animals such as dogs and pigs, small and large ruminants such as cows, sheep and goats, and primates such as baboons and gibbons 224 causing what is known as Johne’s disease (JD).

In 1895, MAP which was called Johne’s bacillus at that time was identified causing JD in a German cow 224. A report in 1913 by Dalziel 225 described for the first time the similarities between CD in human and JD in cattle. The anatomic and pathologic distribution as well as the signs and symptoms of both CD and JD are largely shared, because of the similarities between CD and JD it has been argued that MAP might be also a cause of CD 226.

Chapter 1

In 1984, Chiodini et al successfully isolated spheroplast from two CD patient’s intestinal tissue samples, which was identified to be most closely related to Mycobacterium paratuberculosis (MAP) 227. The organism was found to be extremely fastidious (grown on Herrold egg yolk medium containing mycobactin) and require up to 18 months of incubation for primary isolation. After inoculating the primary isolated spheroblasts into 7H9 broth and incubating 8-9 weeks a strongly acid-fast bacilli form Mycobacterium species was observed 227. The same year the same group showed that by orally inoculating 50 mg of the previously isolated Mycobacterium paratuberculosis into a seven days old goat, a humoral and cell- mediated immunologic responses was developed and granulomatous disease of the distal small intestine, with noncaseating tuberculoid granulomas was observed after five months of inoculating228. In 1986 Chiodini et al added further two strains of an unclassified Mycobacterium species (closely related to Mycobacterium paratuberculosis) isolated from resected intestinal tissues from patients with CD, the isolation process took 18 and 30 months of incubation 229.

Isolating through culture methods is time consuming because of the fastidious nature and slow growth of MAP. Recently, improved detection techniques are established such as PCR, fluorescent in situ hybridisation (FISH) and ELISA 230. The in vitro culture techniques have also improved significantly in the last decade, supported with the improvements in culture media such as using Mycobacterial Growth Indicator Tube (Becton Dickinson, Franklin Lake, NJ, USA) and BACTEC culture media (Beckton Dickinson, Franklin Lake, NJ, USA), which reduced the primary isolation process from 30 months as shown by Chiodini et al in 1986 to 10-12 weeks231,232. These new techniques have increased the detection and isolation of MAP from patients with IBD.

A highly specific marker for the precise identification of MAP was identified by Green et al in 1989 called IS900 (a multicopy DNA insertion element) 233. Since then IS900 was used in detecting MAP from samples of patients with IBD and healthy controls using the two main methods PCR and in situ hybridization (ISH) assays. Many researchers found more frequent detection of IS900 in patients with CD intestinal tissue 234,235 and more recently by 236-239, and blood samples 231 compared to healthy. Increased serological reactivity to MAP was also reported (especially when IgA was analysed) in serum of patients with CD compared to controls 240-243 244.

Chapter 1

Adding to the previous findings, MAP spheroplasts were only found in CD patients as shown by a recent study by Mendoza et al in 2010. The study showed that the viable MAP in spheroplast form was detected from 30 out of 30 patients with CD, 1 out of 29 patients with UC, and 0 out of 10 healthy (non-IBD) controls. 232.

In contrast to the previous reports, a study by Ellingson et al in 2003 compared the detection of MAP from intestinal samples of bovine JD, human CD and UC using PCR. It was found that IS900 sequence was demonstrable in all samples of confirmed positive Johne's disease tissue but the sequence was not identified in the 35 CD and 36 UC samples 245. Other researchers also have not detected MAP in patients with CD 246,247. Some researchers didn’t find any significant differences of isolating MAP between CD and healthy controls, such as a study by Suenaga et al in 1995 in which they found by using colonic mucosa samples from CD, UC and healthy controls that, IS900 sequences was detected in all (100%) of 10 patients with CD, in 11 (61.1%) of 18 patients with ulcerative colitis, and in 14 (87.5%) of 16 control patients with non-inflammatory bowel disease. Indicating no significant difference between CD and controls 248. In addition, some studies found that the antibody levels against MAP antigen in the serum of CD were not significantly elevated over controls 249,250.

A study by Jones et al in 2006 aimed to determine whether exposure to clinical cases of bovine JD was a risk factor for CD. In UK, it was found that prevalence of CD among dairy farmers (in UK) is similar to that reported in other studies in the UK. The group concluded that contact with bovine infected with JD does not increase risk of having CD in human 251.

Using antituberculous chemotherapy in treating CD showed controversial results among research groups. Many reports showed negative results in attempting to treat patients with CD using antituberculous chemotherapy252-258, while others showed positive results 259-262.

According to the previous facts, which shows a controversial reports among researchers, the pathogenic role of MAP in CD is unclear to date.

1.1.3.4.2 Escherichia coli and IBD

Escherichia coli (E. coli) is a facultative anaerobic Gram-negative bacteria species, and one of the intestinal normal flora in which it plays a role in maintaining normal intestinal homeostat and promotes the stability of the luminal microbial flora. It does not cause disease in the GIT except in immunocompromised hosts or in a host having a breached gastrointestinal barrier 263.

Chapter 1

Mucosa-associated E coli have been isolated from two different parts of the intestinal tract, the ileal mucosa and colonic mucosa.

CD patient’s microbiota differs from healthy controls and UC patients as shown in different studies on faecal and mucosa-associated bacterial communities 264-266, and it was shown that the number of Proteobacteria, in particular E coli, are generally increased in CD patients ileal biopsies and faecal samples 267,268.

A study by Darfeuille-Michaud et al in 1998 showed that 84.6% of E. coli strains isolated from the chronic ileal mucosa of patients with CD adhere to differentiated intestinal cells (Caco-2), which was significantly higher than the adherence of the strains isolated from healthy ileal mucosa of the healthy controls 33.3%269. One of the strains which was isolated from the ileal biopsy of a patient with CD called LF82 by Darfeuille-Michaud et al269, was further studied by Boudeau et al in 1999 in which the invasive ability of LF82 was compared with that of reference enteroinvasive, enteropathogenic, enterotoxigenic , enteraggregative, enterohemorrhagic, and diffusely adhering E. coli strains. The findings showed that LF82 is a new potentially pathogenic group of E. coli that showed a true invasive ability, the group gave it the name of adherent-invasive E. coli (AIEC), the invasion ability was test on the cell line HEp-2, Caco-2, Intestine-407, and HCT-8 cells 270.

Darfeuille-Michaud et al in 2004 studied the prevalence of AIEC with intestinal mucosa (ileal and colonic specimens) of patients with CD and of controls. Results showed that in ileal specimens; 21.7% of the CD chronic lesions, 36.4% of CD early lesions (from neoterminal ileal specimen), 22.2% of healthy mucosa of CD patients and 6.2% of healthy controls found to have AIEC strains. Significant difference between the prevalence of AIEC in CD early lesions and healthy controls was reported. In colonic specimens; AIEC strains were found in 3.7% of CD patients, and 1.9% of healthy controls. The group concluded that AIEC strains are associated specifically with ileal mucosa in CD 271.

It is known that the surface of the mucus layer mostly supports the growth of anaerobic bacteria, while the sub-mucus (relatively oxygenated) supports the growth of microaerophilic bacteria. Martin et al in 2004 found that the aerobic culture of colonoscopic sigmoid colon biopsies, after the removal of the mucus layer with dithiothreitol treatment and extensive washing is often sterile (58%) in control colons (n=24), whereas the colon in CD (n=14) and colon cancer (n=21) contains increased bacterial number in this sub-mucus niche (79% of CD cases has bacteria and 21% was found to be sterile). More than half of the bacteria in the sub-

Chapter 1 mucus niche were found to be E coli (identification was carried out by gram staining and API20E bacterial identification kits) 272.

Many studies have reported that mucosa-associated E coli isolated from colonic mucosal samples are increased in CD272-277, all of these studies support a central role for mucosally adherent bacteria specially E. coli in the pathogenesis of CD, and 272,276,277 showed there invading abilities.

It was found that AIEC could survive and replicate extensively in large vacuoles within macrophages without triggering host cell death, there by producing large amounts of TNF, which could be related to some features of CD and particularly to granuloma formation (one of the hallmarks of CD lesions) 278.

To date, the involvement of mucosa-associated E coli in the pathogenesis of CD is still unknown.

1.1.3.4.3 Bacteroides vulgatus and UC

Bacteroides species are gram-negative rods that grow in anaerobic conditions. Non-spore forming bacteria, and some species are motile and others are non-motile 279,280. They are part of the normal gastrointestinal flora. They form almost 25% of the total human colon anaerobic bacteria 281282.

Bacteroides species could cause infections in humans at different sites, such as in necrotizing soft tissue, bone and joint infections in children, brain abscess and meningitis, intra- abdominal sepsis, and its isolates are most predominant in infections that have an intra- abdominal origin 281.

In 1995 Bamba et al demonstrated an elevation of the level of serum IgG antibody in patients with UC to Bacteroides vulgatus. The antibody reacted to a specific protein in the outer membrane of B. vulgatus which was recognized as a 26-kDa protein 283. In 2000 Matsuda et al found a higher agglutination titre against B. vulgatus (16 strains were used isolated from affected colonic mucosa of 16 patients with UC) in most UC patients compared to healthy controls, and the percentage of positive immunoreactivity was much higher in UC patients than in healthy controls. They also found that the bacterial counts per gram rectal tissue (affected rectal mucosa of the patients with UC and the normal rectal mucosa of the healthy controls) for both aerobes and anaerobes increased in UC patients as compared to healthy

Chapter 1 controls, and the highest bacterial counts was for B. vulgatus. Further testing using immunoblotting, found a higher serum IgG immunoreactivity against the 26-kDa protein in patients with UC (53.8%, 14 out of 26 serum samples) compared to healthy controls (9.1%, 2 out of 22 serum samples) 284.

A study by Fujita et al in 2001 detected B. vulgatus more frequently and in greater numbers in intestinal samples taken from patients with IBD (UC more than CD) than in samples from controls; detection of B. vulgatus was in all UC and CD cases but only in 50% of controls, the highest median number for B. vulgatus genome was for UC (147) then CD (128) which were both higher than controls (7). The group used real-time quantitative PCR in detecting and estimating the numbers of bacterial genomes in the intestinal samples from 16 patients with CD, 11 patients with UC, and 18 colon cancer were used as controls. Fujita and his colleagues concluded that B. vulgatus are not a direct cause of IBD, although they may contribute to the diseases by preventing or delaying remission 285.

In animal studies, using HLA-B27 transgenic rats it was found in 1996 by Rath et al that transgenic rats colonized with DESEP bacterial cocktails did not develop colitis and gastritis, but colonizing the transgenic rats with the bacterial cocktails DESEP-B (has the same bacterial component as DESEP except they added B. vulgatus) which contained B. vulgatus did developed colitis and gastritis. DESEP bacterial cocktails contained; Streptococcus faecum (group D), E. coli, Streptococcus avium, Eubacterium contortum, and Peptostreptococcus productus 218. Another study by Kim et al in 2005 found that HLA-B27 transgenic rats did not develop colitis when monoassociated with either E. coli or E. faecalis, but colitis was developed accompanied with T-cell responses in the transgenic mice when mono-associated with B. vulgatus 286. Adding to that a study by Sellon et al in 1998 found that B. vulgatus induced colitis when administrated to IL-10-/- mice 287.,

In contrast to the previous findings, a study by Waidmann et al in 2003 found that the gnotobiotic IL-2 -/- mice colonized with E. coli developed colitis, while colonizing the transgenic mice with B. vulgatus did not, and more surprisingly co-colonization the transgenic mice with E. coli and B. vulgatus did not induce colitis. By using FISH and culture methods it was found that the anti-colitogenic effect of B. vulgatus on E. coli cannot be explained by a significant reduction in numbers of E. coli in the colon. The group concluded that B. vulgatus protects against E. coli -triggered colitis in IL-2 -/- mice by an unknown mechanism 288. Furthermore a study by Bohn et al in 2006 showed that the prevention of E.

Chapter 1 coli triggered colitis in IL-2 -/- mice by co-colonization the transgenic mice with B. vulgatus could be because of increased expression of anti-inflammatory RegIII family genes, pancreatitis associated protein (PAP), and peroxisome proliferator-activated receptor- [gamma] regulated genes 289. PAP mRNA expression was found to have an anti- inflammatory negative regulatory feedback mechanism 290.

A recent study on mice by Mazmanian et al in 2008 showed that Bacteroides fragilis could induce tolerance in the gut by producing polysaccharide A, which suppresses IL-17 production and promotes the activity of IL-10-producing CD4+ T cells, demonstrating that molecules of the bacterial microbiota can mediate the critical balance between health and disease 291.

These controversial results indicate that the role of B. vulgatus in UC pathogenesis is unclear.

1.1.3.4.4 Campylobacter concisus

C. concisus and its relation to IBD will be discussed in more detail in section 1.2.

1.2 Campylobacter concisus: an emerging bacteria in the pathogenesis of IBD

1.2.1 Campylobacter the Genera

1.2.1.1 History

In 1909 McFadyean and Stockman reported the frequent isolation of Vibrio-like bacteria from epizootic abortion in ewes. This Vibrio-like bacterium was given the name Vibrio fetus in 1919 after it was confirmed by Smith and Taylor292. In 1931 Vibrio jejuni was the first Vibrio-like bacterium to be associated with disease as it was attributed to winter dysentery in calves 293.

The first documented association between Campylobacter infection and human disease was in 1938, when a milk-borne disease infected 355 inmates causing diarrhoea. The successful growth of microorganisms resembling V. jejuni in beef heart broth from blood samples of 13 out of the 39 patients was reported. However, transferring the microorganism from the broth culture to solid media such as Loeffler’s medium or blood agar resulted in minimal or no growth, respectively. 294.

V. fetus was isolated in 1947 by Vinzent et al from the blood of three pregnant women who suffered from fever of unknown origin 295,296. A bacteria called ‘related vibrio’ was isolated

Chapter 1 from blood by King in 1957. This new bacterium was related to V. fetus but had different serological reaction and optimum temperature for growth 297. ‘Related vibrio’ was reclassified into the Campylobacter by Sebald and Veron 298,296.

Changing some species from the Vibrio genus to the Campylobacter genus began in 1963 by Sebald and Veron. V. fetus and Vibrio bubulus were transferred to Campylobacter fetus and Campylobacter bubulus, respectively. The reasons for the change was that these species have non fermentative metabolism, require microaerobic conditions for growth, their DNA have low cytosine (C) and guanine (G) base composition, and they have a higher optimum growth temperature 299. Sebald and Veron published a more extensive study on the taxonomy of Campylobacter in 1973 in which they showed four distinct species of the genus Campylobacter: C. fetus, Campylobacter coli, Campylobacter jejuni (previously called V. jejuni) and Campylobacter sputorum which is subdivided to C. sputorum subspecies sputorum and C. sputorum subspecies bubulus 300.

One of the problems that played an important role in the delay of discovering Campylobacter was the difficulty of its isolation. An important technique in isolating Campylobacter was first described in 1972 by Dekeyser et al 301. The study introduced the filtration technique, and showed the isolation of ‘related vibrio’ from the blood and stool of one patient and from stool alone of another patient, both of which suffered from acute enteritis. Another important discovery was in 1977 when a selective media for Campylobacter was described by Skirrow. The media was composed of basal medium with a mixture of vancomycin, polymyxin B and trimethoprim 302. Discovering the filtration technique and Campylobacter selective media increased the chance of isolating Campylobacter. This opened the way to investigate and explore this microorganism and assess its clinical importance in more detail.

From 1973 onward various species of Campylobacter were isolated from a wide range of sources, such as the isolation of Campylobacter concisus from the oral cavity 303 and from biopsy samples of patients with inflammatory bowel disease 304

1.2.1.2 Taxonomy

The genus Campylobacter is one of the three genera that make up the family Campylobacteraceae. Campylobacteraceae is from the order Campylobacterales which is the only order that makes up the Epsilonproteobacteria Class (Figure1.5) 305,306.

Chapter 1

Figure 1.5 The scientific classification of the Genus Campylobacter according to Bergey’s Manual Volume two the Proteobacteria 305,306. According to Bergey’s Manual, the Class Epsilonproteobacteria is circumscribed to the Phylum Proteobacteria, the authors depended on the phylogenetic analysis of the 16S ribosomal RNA (rRNA) sequences in this classification 305. The members of this class are Gram negative, slender, and rod shape that could be helical curved or straight bacteria 307. Epsilonproteobacteria has only one order according to Bergey’s Manual which is Campylobacterales, the classification depended on the phylogenetic analysis of the 16S rRNA sequences 305. Campylobacterales order consists of two families Campylobacteraceae and Helicobacteraceae

The family Campylobacteraceae are Gram negative, microaerophilic with low G + C content. These bacteria are non-spore formers, 0.2-0.8 µm wide and 0.5 -5 µm long, could be spiral rod curved or S-shaped, most are motile with a single polar unsheathed flagellum that could be at one or both ends of the bacterium and they have a corkscrew-like motion 308. The Campylobacteraceae family consists of three genera which includes Campylobacter. The closest related phylogenetic neighbours to the Campylobacter genera are the two genera; Sulfurospirillum, and Arcobacter, demonstrated in Figure 1.6 309

Chapter 1

Figure 1.6 Campylobacteraceae family phylogenetic tree and its closest phylogenetic neighbours, based on the 16S rRNA sequence similarity. The scale represent a 10% sequence dissimilarity 309. Some strains are missing as new strains were discovered, the image has specie termed Bacteroides ureolyticus, this species has been reassigned to be termed campylobacter ureolyticus310. The species under the Campylobacter genus has the phenotypic characteristics of the Campylobacteraceae family, with some exceptions, such as; Campylobacter showae which has multiple flagella and Campylobacter gracilis which is non-motile. Most Campylobacter species grow under microaerobic conditions, while several species grow in anaerobic conditions with the addition of fumarate or nitrate as electron acceptors. The anaerobic species cannot grow in microaerobic conditions unless hydrogen, formate or succinate is supplemented as an electron source 306,308,311-313.

Campylobacter optimum growth temperature is from 30 to 37 0C, they can’t hydrolyze gelatin, tyrosin, starch and casein. Campylobacter species are positive for the Oxidase test except for C. gracilis, Campylobacter gives negative results for methyl red reaction, acetions and indole production308,312,313.

Several potential virulence factors have been proposed for Campylobacter species such as adhering to and invading the intestinal mucosa and epithelial cells, the ability to translocate, the production of toxins and the flagella-mediated motility 314,315. However, the definitive virulence mechanisms of Campylobacter species are not known.

Chapter 1

To date there are 22 species recognized in the Campylobacter genera including: Campylobacter concisus, Campylobacter hyointestinalis, Campylobacter showae, Campylobacter helveticus, Campylobacter Canadensis, Campylobacter rectus, Campylobacter fetus, Campylobacter peloridis, Campylobacter upsaliensis, Campylobacter curvus, Campylobacter lari, Campylobacter sputorum, Campylobacter jejuni, Campylobacter coli, Campylobacter gracilis, Campylobacter avium, Campylobacter cuniculorum, Campylobacter insulaenigrae, Campylobacter lanienae, Campylobacter mucosalis, Campylobacter hominis, and Campylobacter ureolyticus 310.

Campylobacter ureolyticus was recently allocated to the genus Campylobacter, after reassessing the taxonomy of this bacterial species from the genus Bacteroides. That was after protein profiles, genomic amplified fragment length polymorphism patterns and 16S rRNA and cpn60 gene sequences of a diverse collection of 26 B. ureolyticus strains 310.

1.2.2 Campylobacter concisus; a general review

C. concisus is a gram negative non-pigmented, non-spore former, slow growing rods that has curved or S-shaped forms with rounded ends, small 0.5- 4 µm, motile with unsheathed polar flagella and requires hydrogen for growing in microaerobic conditions, but do not grow at air 306,316 containing 10% CO2 . C. concisus was first isolated in1981 from the oral cavity of humans with periodontal disease 303,317.

Campylobacter concisus is heterogeneous specie with many genotypic subgroups, as shown by protein electrophoresis and DNA-DNA hybridization 318. An amplified fragment length polymorphism analysis (AFLP) study by Aabenhus et al 319 confirmed that the genetic diversity of the species and revealed at least four distinct genomospecies. The heterogeneity was also supported in other methods such as PCR amplification of 23S rDNA and Random Amplified Polymorphic DNA (RAPD) analysis 320.

Campylobacter concisus was found in different human samples, such as in the gingival crevices, faeces, saliva, blood, stomach and oesophagus specimens 306, and recently from biopsy samples of CD patients 321,322.

Chapter 1

1.2.3 Campylobacter concisus and Disease

1.2.3.1 Campylobacter concisus and the human oral cavity

The earliest description of C. concisus was done in 1981 by Tanner et al in which they isolated Gram negative rods, with small curved unbranched cells, that could be helical or curved in shape, and motile by means of single polar flagella from the gingival crevices of humans with gingivitis and periodontitis. Tanner et al classified this bacterium in the genus Campylobacter, and the name C. concisus was proposed 303. In 1986 the chemotactic response by C. concisus to formate was described by Paster et al which could have a role in the gingival colonization 323.

The composition of the subgingival microbiota of primary incisors, canines and molars in healthy children aged 4–5 years was investigated by Kamma et al in 2000. The study showed that C. concisus was found in low numbers in these areas, as it was found that the relative proportions of C. concisus in children incisors was 0.2%, canines was 1.3% and molars was 1.7% (molars had a significantly higher numbers than incisors), in addition the frequency of isolation of C. concisus in children incisors was 2.5%, canines was 7.5% and molars was 5%. Interestingly, although C. concisus was found in low numbers they were found more frequently in bleeding sites 324. Later in the same year the same group also showed that, C. concisus was also less frequently detected in children oral cavity compared to other oral microbiota that was after examining the subgingival plaque from permanent teeth and deciduous teeth from subjects 7-8 years old. Interestingly, C. concisus was found to be associated with bleeding permanent and deciduous teeth, has also been associated with the progression of periodontitis, and was significantly more frequently detected in the permanent teeth plaques 325.

It was found that the serum antibody level in adults (16-61 years) with periodontal disease to C. concisus was significantly higher than healthy subjects 326. In 1999 Kamma et al showed that C. concisus is found in significantly higher numbers and more frequently in adult (22-35 years) smokers with early onset periodontitis. The association between C. concisus and the progress of periodontal disease in adults was also shown by Kamma et al in 2001 327.

Recently in 2010 Zhang et al detected 100% and 97% of C. concisus in the saliva sample taken from patients with IBD and healthy controls, respectively using bacterial culture and PCR. This indicates that the human cavity is a reservoir of C. concisus. Also, it was found

Chapter 1 that young children group (3-5 years) had a significant lower culture positive rate when compared to other age groups 328. It is hypothesised by Zhang group that C. concisus colonizing the gut could be originated from the oral cavity.

To date, the pathogenic role of C. concisus in oral cavity infections is not known.

1.2.3.2 Campylobacter concisus and the human gastroenteritis

It was reported that 1-10% and 10-25% of gastroenteritis is caused by Campylobacter infections in developed countries and in developing countries respectively 329. Even though C. jejuni and C. coli were reported as the main causative of Campylobacter enteritis in humans 308. C. concisus is gaining more interest as a major player in gastroenteritis.

The isolating of Campylobacter species (including C.concisus) other than C. jejuni/coli from stool samples of patients with gastroenteritis began in the mid-1990s, and one of the reasons that helped in isolating these Campylobacter species was addition of the filtration technique before growing it on blood agar plates 330-333. All the gastroenteritis patients in these reports were children.

Since 2000 many reports from Australia 334, South Africa 335 and Denmark 336, began to identify C. concisus in particular as a potential etiological agent of gastroenteritis in children, after isolation of C. concisus from a large number of children with diarrheal cases. The study by Engberg et al reported that out of the 1376 cases of gastroenteritis they tested, C. concisus was isolated from 39 clinical cases (3%) and it was found that 15 (39%) of these 39 clinical cases were children under 9 years old, and 10 (26%) of these 39 clinical cases were elderly people over 60 years 336.

Isolation of C. concisus from stool samples in many countries such as Australia, South Africa and Europe could indicate that C. concisus infection could be a global problem.

Recently in 2012 a study from Denmark showed, a high incidence of isolating C. concisus from patients with gastroenteritis (their diarrhoeic stool samples were used), recording an annual incidence of 35 cases per 105 persons, which was similar to the incidence of isolation of C. jejuni and C. coli. The study showed that C. concisus was isolated from 400 patients with gastroenteritis out of 8302 patients with gastroenteritis recruited for this study, as compared to 489 patients with gastroenteritis in which C. jejuni/coli was isolated. C.

Chapter 1 concisus was again found more frequently among small children (<1 year) and the elderly (≥65 years) 337.

A study by Aabenhus et al in 2002 showed that approximately half (49%) of Campylobacter species isolated from diarrhoea samples of immunocompromised patients were identified as C. concisus. The study isolated 224 Campylobacter, 110 of which were identified as C. concisus strains, the 98 immunocompromised patients included 13 IBD cases 329.

1.2.3.3 Campylobacter concisus and the IBD

The first association report between C. concisus and IBD was published recently in 2009 by Zhang et al in which they found that biopsy samples taken from children with CD had 51% PCR specific-positive rate for C. concisus, which was significantly higher than the controls (2%) 321. Later in 2010 Man et al showed that 65% of faecal samples from child patients with CD were positive for DNA detection of C. concisus which was significantly higher than healthy controls (33%), the group used PCR assay that targets the 16S rRNA gene in detection 338. Recently in 2011 Mahendran et al 339and Mukhopadhya et al 340 reported a significant high prevalence of C. concisus in intestinal biopsies of adult patients with UC and CD as compared with the controls by using PCR method. Furthermore, Mahendran et al reported that the increased prevalence of C. concisus in patients with IBD was mainly detected in the descending colon and rectum.

The first reports of isolating C. concisus from samples taken from patients with IBD was in 2002 by Aabenhus et al in which it was reported the isolation of C. concisus from stool samples taken from 13 IBD patients 329 but the association between C. concisus and IBD was not taken in consideration in this report 329. Then in 2009 Zhang et al reported for the first time the isolation of a C. concisus strain from an inflamed intestinal sample of a child patient with CD, C. concisus was identified by Oxoid biochemical identification system and sequencing of the 16S rRNA gene. Zhang et al also reported that the level of IgG antibody specific to C. concisus was significantly higher in children with CD as compared to healthy controls 321. Later in the same year Lastovica also reported a successful isolation of C. concisus from a gastric biopsy specimen from a 10-year-old boy suffering from IBD. C. concisus was identified by phenotypic and DNA-DNA hybridization studies. Lastovica also reported the isolation of C. concisus from stool samples of 6 patients (4 adults and 2 children) with IBD 322.

Chapter 1

C. concisus and its role in IBD will be discussed further in this thesis.

1.3 Hypothesis and aims

The hypothesis of this PhD project is that specific oral C. concisus strains have the potential to trigger the development of inflammatory bowel disease.

Specific aims:

1. To examine the prevalence of multiple oral C. concisus strains in patients with IBD and controls and the interaction of these C. concisus strains with human intestinal epithelial cells.

2. To examine the profiles of lipopolysaccharide (LPS) of C. concisus strains and the potential use of LPS profiles to group C. concisus

3. To examine the effects of C. concisus on intestinal epithelial expression of specific Toll- like receptors.

4. To examine the role of C. concisus in activation of NFkB and induction of cyclooxygenase-2 (COX-2) and IL-8

Chapter 2

Chapter 2: General Methodology

Methods described in this chapter are the common methods used in studies carried out in chapters 3-6. Methods specifically related to studies in individual chapters were described separately in the corresponding chapters.

2.1 Media and buffer

2.1.1 Horse blood agar (HBA)

Blood agar base No.2(40 g) (Oxoid, New Hampshire, USA) was suspended in 1 litre of distilled water (DW), dissolved then sterilized by autoclaving for 15 minutes. After cooling the mixture to about 45-50 0C 60 ml of defibrinated horse blood (Oxoid) was added.

2.1.2 Horse blood agar with vancomycin (HBA+V)

HBA supplemented with 10 µg/ml of vancomycin.

2.1.3 Nutrient agar (NA)

Nutrient agar powder (75 g) (Oxoid) was suspended in 1 litre of DW, dissolved then sterilized by autoclaving for 15 minutes.

2.1.4 Phosphate buffer saline (PBS)

NaCl (8 g) (Ajax Finechem, Massachusetts, USA), 0.2 g KCl (Ajax Finechem), 1.44g

Na2HPO4 (Ajax Finechem), and 0.24 g of KH2PO4 (Ajax Finechem) were dissolved in 1 litre of DW, pH of the mixture was then adjusted to 7.4, and sterilized by autoclaving for 20 minutes.

Chapter 2

2.2 Protocols

2.2.1 Electrophoresis

2.2.1.1 Aim: electrophoresis was used to separate proteins or lipopolysaccharides (LPS).

2.2.1.2 Instrument: The mini-PROTEAN tetra cell for mini precast gels (Bio-Rad, California, USA) was used, catalogue number: 165-8004.

2.2.1.3 Reagent

5X SDS/electrophoresis buffer: 15.1 g of Tris base (Sigma-Aldrich, Missouri, USA), 72.0 glycine (Ajax Finechem), and 5.0 g sodium dodecyl sulfate (SDS) (Bio-Rad), DW was then added to 1000ml.

2X sample buffer: 0.5 M tris base (25 ml) (Merck, New Jersey, USA), 1 ml of 10% SDS (Bio-Rad), 20 ml glycerol (AjaxFinechem), 4 g SDS (Bio-Rad), 2-mercaptoethanol (2 ml) (Sigma-Aldrich), and 1 mg of bromophenol blue (Sigma-Aldrich), mixture was then filled up to 100ml with DW. Note: store at -20 0C.

SDS-Polyacrylamide separating gel 12%: 40% acrylamide (2.25 ml) (Bio-Rad), 1.9 ml of 1.5 M Tris pH 8.8 (Merck), 75 µl of 10% SDS, 50 µl 10% ammonium persulfate (APS, Sigma-Aldrich), and 5 µl of N,N,N,N-Tetramethylenediamine (TEMED) (Sigma-Aldrich) were added to 2.45 ml DW . Mixture was then directly poured onto the glass plate. Note: this gel was used for protein separation.

Polyacrylamide separating gel 12%: same as 12% SDS-PAGE separating gel preparation except omitting the 75 µl 10% SDS and replacing it with DW. Note: this gel was used for LPS separation.

SDS-Polyacrylamide stacking gel 5%: 310 µl ml of 40% acrylamide (Bio-Rad), 473 µl of 0.5 M Tris (pH 6.8, Merck), 25 µl of 10% SDS (Bio-Rad), 25 µl 10% APS (Sigma-Aldrich), and 2.5 µl of TEMED (Sigma-Aldrich) were added to 1.7 ml DW. Mixture was then directly poured onto the glass plate. Note: this gel was used for protein separation

Chapter 2

Polyacrylamide staking gel 5%: same as 5% SDS-PAGE stacking gel preparation except omitting the 25 µl 10% SDS and replacing it with DW. Note: this gel was used for LPS separation.

2.2.1.4 Method

Polyacrylamide gels were assembled in the electrophoresis apparatus and filled with 1X SDS/electrophoresis buffer according to manufactory instructions. Samples (proteins or LPS) were mixed with equal amount (v/v) of 2X sample buffer and heated to 1000 C for the required time. The sample mixture was then loaded to the gel wells. Electrophoresis was then performed at a constant voltage of 80 V using the PowerPacTM universal power supply (Bio- Rad) for the required time (until the bromophenol blue had reached the bottom of the gel).

2.2.2 Silver Staining

2.2.2.1 Aim: Silver staining was used to visualise the separated LPS profile bands, LPS separating was by PAGE.

2.2.2.2 Reagents

Fixing solution: 25 ml of propan-2-ol (Ajax Finechem) and 7 ml acetic acid (Ajax Finechem) were mixed and filled up to 100 ml with DW.

Oxidizing solution: 0.7 g of periodic acid (Sigma-Aldrich) and 2.8 ml of the fixing solution were mixed and filled up to 100 ml with DW.

Ammonical silver nitrate stain: 1.4 ml ammonia (Ajax Finechem), 21 ml of 0.09 M NaOH

(Ajax Finechem), and 4 ml of 1.14 M AgNO3 (Sigma-Aldrich) were mixed and filled up to 100 ml with DW.

Developing solution: 0.005 g of citric acid (Sigma-Aldrich) was added to 100 ml of 0.019% (v/v) formaldehyde (Ajax Finechem).

Chapter 2

2.2.2.3 Method

Polyacrylamide Gels were fixed in fixing solution overnight, then oxidized for 7 minutes in freshly prepared oxidized solution. Oxidised gels were then washed by DW for 4 hours (water was change each 1 hour). Gels were then stained for 15 minutes in ammonical silver nitrate stain, and washed by DW for 40 minutes (water was changed each 10 minutes). Finally, gels were transferred to freshly prepare developing solution until LPS bands were visualised. Staining was stopped by immersing the gels in DW.

2.2.3 Coomassie brilliant blue staining

2.2.3.1 Aim: Coomassie brilliant blue staining was used to visualise the separated whole protein profile bands after separating the proteins by SDS-PAGE.

2.2.3.2 Reagents

Coomassie brilliant blue stain: 0.25 g of coomassie brilliant blue R250 (Sigma-Aldrich) was dissolved in 90 ml of 50% methanol (Ajax Finechem) and 10 ml glacial acetic acid (Ajax Finechem) was then added to the mixture. Destaining solution: 30 ml methanol (Ajax Finechem) were mixed with 10ml glacial acetic acid (Ajax Finechem), the mixture was filled up to 100 ml with DW.

2.2.3.3 Method SDS-Polyacrylamide Gels were immersed in at least 5 volumes of the coomassie brilliant blue stain overnight on a slow rotating platform, gels were then destained using the destaining solution for 3-7 hours on a slow rotating platform changing the destaining solution 3-4 times. When we see thoroughly the destained gel while the bands being sharp as possible destaining was stopped by immersing the gels in DW.

Chapter 2

2.2.4 Western blotting

2.2.4.1 Aim: Western blotting was used to transfer the separated protein bands by electrophoresis from the SDS-Polyacrylamide gels to polyvinylidene difluoride (PVDF) membrane sheath.

2.2.4.2 Instrument: The mini trans-blot® electrophoretic transfer cell instrument (Bio-Rad, California, USA) was used, catalogue number: 170-3930.

2.2.4.3 Reagents

Washing buffer: 0.05 ml of Tween20 (Sigma-Aldrich) was added to 100 ml PBS.

Blocking Solution: 0.05 ml of Tween20 (Sigma-Aldrich) and 5 g of skimmed milk powder were added to 100 ml PBS

Transfer buffer: 3.03g Tris-Base and 14.4 g glycine were dissolved into 200 ml methanol, the mixture was then filled up to 1000 ml with DW.

2.2.4.4 Method

After the proteins were separated by SDS-PAGE, the SDS-polyacrylamide gels were equilibrated in the transfer buffer for 20-30 minutes, while waiting; the PVDF membranes (Bio-Rad) were immersed in absolute methanol for few seconds then equilibrated in the transfer buffer for 10-20 minutes.

Proteins were transferred from the SDS-polyacrylamide gel to PVDF membrane at 100 volt for 2 hours in 4 0C room using the Mini Trans-Blot electrophoresis transfer cell (Bio-Rad, catalogue number 170-3930) according to the manufacturer manual. Care was taken at all times for the transfer buffer not to reach high temperatures.

After the proteins were transferred to the PVDF membranes, membranes were blocked in blocking solution for 1 hour at room temperature, and then washed 3 times with washing buffer 5-10 minutes each wash at room temperature. Membranes were then treated with the primary antibody for the required time (each target protein had different incubation time) at room temperature, then washed as previous. Membranes were then treated with the secondary antibody conjugated with horseradish peroxidase for the required time (each target protein had different incubation time) at room temperature and washed again as previous. All antibodies were diluted in blocking solution.

Chapter 2

Finally the HRP-labelled antibodies were detected using Immun-Star™ WesterC™ Chemiluminescence Kits (Bio-Rad) and LAS-3000 imaging system (Fujifilm, Tokyo, Japan).

2.2.5 Mammalian cell culture technique

2.2.5.1 Maintaining the Caco-2 human intestinal epithelial cell line.

Culture media were prepared as follows: minimum Essential Medium (MEM) (Invitrogen, California, USA) was supplemented with 10% heat-inactivated foetal bovine serum (FBS) (Bovogen Biologicals, Melbourne, Australia), 1 mM Sodium pyruvate (Invitrogen), 0.1 mM non-essential amino acids (Invitrogen), 100 Unit/ml Penicillin, 100 µg/ml streptomycin (Invitrogen) and 2.25 mg/l Sodium Bicarbonate (Invitrogen).

Caco-2 cells were maintained in 25 cm2 or 75 cm2 sterile tissue culture flasks (Greiner Bio- o One, Kremsmünster, Austria) at 37 C and 5% CO2 incubation conditions in the culture media.

Caco-2 cells were routinely examined microscopically to insure normal growth and to check the culture confluence.

The tissue culture flasks were passaged when the confluence reaches 85-95%; usually every 7 days if the initial concentration was 1 × 105 cell/ml, keeping in track that each cell line is not passaged more than 10 times. Passage was by using 0.25% trypsin-EDTA (Gibco) at 37 0C until the cells detaches from the flask surface (not more than 7 minutes). Trypsin was then neutralized by adding 10 or 30 ml of culture media to the 25 cm2 or 75 cm2 tissue flasks respectively. Cells were then pelleted by centrifugation at 300 RCF for 5 minutes and seeded at an initial concentration of 1 × 105 cell/ml in a new sterile 25 cm2 or 75 cm2 tissue culture flasks or the required initial concentration. Culture media was changed every 3-4 days.

2.2.5.2 Maintaining the HT-29 human intestinal epithelial cell line.

Culture media were prepared as follow: McCoy’s 5A medium (Invitrogen) was supplemented with 10% FBS, 100 Unit/ml Penicillin and 100 µg/ml streptomycin.

HT-29 cells were maintained in 25 cm2 or 75 cm2 sterile tissue culture flasks (Greiner) at 37 0 C and 5% CO2 incubation conditions in the culture media.

HT-29 cells were routinely examined microscopically to insure normal growth and to check the culture confluence.

Chapter 2

The tissue culture flasks were passaged when the confluence reaches 85-95%; usually every 5 days if the initial concentration was 1 × 105 cell/ml, keeping in track that each cell line is not passaged more than 10 times. Passage was by using 0.25% trypsin-EDTA (Gibco) at 37 0C until the cells detaches from the flask surface (not more than 8 minutes). Trypsin was then neutralized by adding 10 or 30 ml of culture media to the 25 cm2 or 75 cm2 tissue flasks respectively. Cells were then pelleted by centrifugation at 300 RCF for 5 minutes and seeded at an initial concentration of 1 × 105 cell/ml in a new sterile 25 cm2 or 75 cm2 tissue culture flasks or the required initial concentration. Culture media were changed every 3-4 days.

2.3 Ethics Approval

Ethics approvals for this study were granted by the human ethics committee of the University of New South Wales and the South East Sydney Area Healthy Service (Ethics Nos: HREC 09237/SESIAHS 09/078, HREC08335/SESIAHS(CHN)07/48 and HREC 06233/SESAHS (ES) 06/164). Written informed consent was obtained from all subjects in this study.

Chapter 3

Chapter 3: The high prevalence of multiple oral C. concisus strains detected in patients with inflammatory bowel disease and their enteric pathogenic potential

3.1 Introduction Campylobacter concisus is a rod to spiral-shaped, small in size 0.5 to 1 by 4 µm, gram negative, motile and microaerophilic bacterium that colonizes the human oral cavity 306,316. The organism was first described in 1981 by Tanner et al after isolation from the periodontal pockets of patients with gingivitis and periodontitis 303. In 1986 the chemotactic response by C. concisus to formate was observed by Paster et al and this phenomenon was found to play a significant role in the gingival colonization 323. C. concisus strains were also isolated from the gingival crevices of patients with gingivitis, periodontitis and advancing bone loss 341,342. In 2010, Zhang et al detected C. concisus in the saliva of 100% of patients with inflammatory bowel disease (IBD) and 97% of healthy individuals. The same authors were able to isolate C. concisus from saliva collected from 85% of patients with IBD and 75% of healthy controls 328. Zhang et al also found that the prevalence of C. concisus in the oral cavity of patients with IBD and healthy controls was not statistically different. The role of C. concisus as an oral commensal bacterium or oral pathogen bacterium is unclear 343, and whether the oral C. concisus colonizing patients with IBD are different from the oral C. concisus colonizing healthy controls still needs to be investigated.

A significantly higher prevalence of C. concisus in enteric (faecal and biopsy) specimens from patients with Crohn’s disease (CD) and ulcerative colitis (UC) as compared to healthy individuals was reported in a number of studies 304,315,339,340. Furthermore, Mahendran et al reported that the increased prevalence of C. concisus in patients with IBD was mainly in the descending colon and rectum 339. A recent study from our research group examined the genetic relationship between oral C. concisus and C. concisus colonizing the intestinal tissues using multi locus sequence typing (MLST), the study found that the intestinal-associated C. concisus in patients with IBD was related to their own oral C. concisus or to a different oral C. concisus from another patient with IBD. Furthermore, the study also found that C. concisus strains could undergo natural recombination in individual’s oral cavity 344.

Chapter 3

C. concisus genomic heterogeneity was confirmed by DNA: DNA hybridization 345, pulsed field gel electrophoresis 346 and ampliﬁed fragment length polymorphism (AFLP) 347 . The heterogeneity of C. concisus was also described in a study by Aabenhus et al in 2005 using protein profiling, biochemical testing and antibiotic susceptibility 319. In addition Aabenhus et al have identified two groups of C. concisus strains by using protein profiles; Group 1 had a characteristic double band in the molecular weight region of 170 kDa which was common to all group 1 strains whereas, group 2 isolates were distinguished from group 1 strains by having one or two additional protein bands in the region of 200 kDa 319. To date, no study has shown the heterogeneity between C. concisus strains isolated from the oral cavity.

In 2010 Man et al isolated a C. concisus strain from a biopsy sample taken from a child diagnosed with CD, the strain was termed C. concisus UNSWCD. Invasion ability of UNSWCD strain to the human intestinal cell line Caco-2 was investigated by Man et al It was found that the percentage invasion of the strain UNSWCD (named in this thesis as P1CDB1(UNSWCD)) is significantly higher than the C. concisus strains UNSWCS and ATCC 51562 which are two strains isolated from the faecal sample of patients with acute gastroenteritis 315. Furthermore, Man et al also found that the C. concisus ATCC 5156 which was isolated from a healthy individual did not have invasion ability to Caco-2 cells 315. A more recent study by Kaakoush et al in 2011 found that four C. concisus strains isolated from biopsy samples taken from patients with chronic intestinal disease have higher invasion ability to Caco-2 cells compared to three C. concisus strains isolated from faecal samples of patients with acute intestinal disease, the same observation was also made to another isolate of C. concisus strain recovered from a faecal sample of a healthy individual 348. All of the previous adhesion and invasion studies were performed on a small number of C. concisus strains, and none of these studies used strains isolated from the oral cavity.

A study by Zhang et al in 2010 detected multiple (≥2) oral C. concisus strains in a child with CD 328. However, the prevalence of oral C. concisus in patients with IBD and healthy controls is to date, unknown.

The aim of this study was to:

1. Examine the prevalence of multiple (≥2) oral C. concisus in patients with IBD and healthy controls. 2. Examine the enteric pathogenic potential of oral C. concisus strains. 3. Subgroup the oral C. concisus strains based on whole protein profiling.

Chapter 3

3.2 Materials and Methods

3.2.1 Study subjects:

C. concisus were isolated from 32 study subjects; 15 were patients with IBD (9 CD and 6 UC) and 17 were healthy individuals, the detailed information is given in Table 3.1.

The patients were recruited from the Prince of Wales Hospital, the St George Hospital and Sydney Children’s Hospital at Sydney, Australia. The healthy individuals were recruited from University of New South Wales (Sydney, Australia) staff and their family members. None of the healthy individuals suffered from chronic diseases or been treated with medications that could influence the oral or gut microbiota.

Table 3.1 Information of patients with IBD and healthy controls in this study including their given ID through this study, sex, age on the day of diagnosis, patient diagnosis and disease activity at time of sample collection.

Individuals ID Sex-Age at Diagnosis Disease activity at diagnosis the time of sample collection Patient No. 1 F-5y CD New case, active Patient No. 2 M-19y CD Relapse, active Patient No. 3 M-23y UC New case, active Patient No. 4 F-16y CD Remission Patient No. 5 M-13y CD Remission Patient No. 6 M-13y CD Remission Patient No. 7 M-65y UC New case, active Patient No. 10 M-18y CD New case, active Patient No. 11 M-34y CD New case, active Patient No. 12 F-55y CD New case, active Patient No. 13 F-73y CD New case, active Patient No. 14 M-22y UC New case, active Patient No. 15 M-34y UC New case, active Patient No. 16 F-39y UC New case, active Patient No. 17 M-67y UC New case, active Healthy No. 1 F-23y Healthy N/A Healthy No. 2 F-23y Healthy N/A Healthy No. 3 F-27y Healthy N/A

Chapter 3

Healthy No. 4 F-9y Healthy N/A Healthy No. 5 M 16y Healthy N/A Healthy No. 6 M-5y Healthy N/A Healthy No. 7 M-4y Healthy N/A Healthy No. 8 F-67y Healthy N/A Healthy No. 9 M-18y Healthy N/A Healthy No. 10 F-18y Healthy N/A Healthy No. 11 M-58y Healthy N/A Healthy No. 12 M-18y Healthy N/A Healthy No. 13 M-13y Healthy N/A Healthy No. 14 F-41y Healthy N/A Healthy No. 15 F-62y Healthy N/A Healthy No. 16 F-21y Healthy N/A Healthy No. 17 M-18y Healthy N/A M: Male; F: Female; N/A: Not applicable; CD: Crohn’s disease; UC: Ulcerative colitis

3.2.2 Isolation and identification of oral C. concisus

3.2.2.1 Specimen collection

Saliva samples (0.5ml) were collected in a sterile container from patients and healthy controls. Samples were processed for bacterial recovery within three hours of collection.

3.2.2.2 Isolation of multiple oral C. concisus isolates from saliva samples

Oral C. concisus was isolated from saliva samples using the filtration method described by Lastovica et al 349, with some modifications 328. Briefly, 6 µl of the saliva sample was inoculated on horse blood agar supplemented with 10µg/ml vancomycin (HBA +V) (refer chapter 2) and incubated at 37 0C for 3 days under microaerophilic conditions generated by the BR0056A gas generating kit- Campylobacter system (Oxoid, Hants, United Kingdom).

Following incubation, the mixed bacterial cultures were collected and suspended into 200µl of sterile phosphate buffer saline (PBS) (refer chapter 2), the suspension was then filtered through a 0.65-µm nitrocellulose ﬁlter onto a fresh HBA+V agar plate. Plates were then incubated at 37 0C for 2 days under microaerophilic conditions generated by BR0056A gas generating kit- Campylobacter system.

Chapter 3

Following incubation, 12 putative oral C. concisus colonies were collected based on a number of independent criteria: first, colony shape in which colonies with convex, translucent and approximately 1 mm in diameter characteristics were chosen. Second, bacterial morphology and motility as determined by wet mount. Spiral curved rod and motile bacteria were isolated. Third, Gram reaction; Gram negative bacteria were chosen.

The 12 putative oral C. concisus isolated colonies were collected from each individual. The collected colonies were subcultured onto fresh HBA+V agar plates to obtain a pure culture and were incubated as described previously for 2 days. The identity of the putative C. concisus isolates was further confirmed by C. concisus specific polymerase chain reaction (PCR).

3.2.3 Confirmation of the identity of oral C. concisus by C. concisus specific PCR

The identity of 384 putative C. concisus isolates (12 colonies from each case) was confirmed by as described earlier using C. concisus-specific PCR 338. Briefly, DNA was released from each of the candidate C. concisus isolate by dissolving each isolate into 200 µl of nuclease- free water (Ambion, California, USA) and heated to 96ºC for 10 minutes in order to release the DNA.

Twenty five microliter of the PCR master mix was prepared as follows: 10 pmol of each primer (forward and reverse primer, Sigma-Aldrich, Missouri, USA) was mixed with 1x PCR buffer (Fisher Biotech, Subiaco, Australia), 200 nM of deoxy-nucleotide-triphosphate (dNTP)

(Fisher Biotech), and 2.5 mM MgCl2 (Fisher Biotech). 200 ng of DNA was added into PCR master mix, when the temperature reached 95ºC (Hot–start method), the PCR was initiated by adding 5.5 U of Taq polymerase (Fisher Biotech) into the mixture to reduce nonspecific amplification. The thermal cycling conditions consisted of 35 cycles of 95ºC (10s), 65ºC (10s) and 72ºC (45s). Sequence of each primer was as follow:

Concisus F 5’-CTT-GTG-AAA-TCC-TAT-GGC-TTA-3’

Concisus R 5’-CTC-ATT-AGA-GTG-CTC-AGC-C-3’

Each of the PCR product (5µl) was separated by electrophoresis on 1.2% agarose gel at 87V for 45 minutes in TAE buffer [40 mM Tris base (Sigma-Aldrich), 40 mM of glacial acetic acid (Ajax chemicals), 0.1 mM of ethylenediamine tetra-acetic acid (EDTA) pH 8.0 (Sigma- Aldrich) and 1L sterile water]. FN-1 marker (Fisher Biotech) was used to determine the size

Chapter 3 of the PCR products. The gel was then stained with GelRedTM Nucleic Acid Gel Stain (Biotium, Hayward, USA) for 15 minutes and visualized by UV trans-illumination using Gel- DocX 2000 (Bio-Rad, California, USA).

For each PCR assay, a negative control (without DNA template) and a positive control (a known C. concisus strain) was included.

3.2.4 Determination of the number of oral C. concisus strains colonizing each individual (patients with IBD and healthy controls)

Of the 12 putative oral C. concisus isolates collected from each individual, a minimum of nine isolates were positive for C. concisus specific PCR. Given this, nine C. concisus isolates that were confirmed by C. concisus PCR from each individual were subjected to sodium dodecyl sulphate polyacrylamide gel electrophoresis (SDS-PAGE), to examine the number of strains colonizing the oral cavity of each individual.

3.2.4.1 Whole cell protein preparation for C. concisus isolates

Each of C. concisus isolates were grown on HBA+V agar plates for two days, then were harvested and washed three times with PBS. The pellet was frozen-thawed three times using liquid nitrogen and finally suspended in 600 µl of PBS. The bacterial mixture was sonicated on ice for 3 minutes with 0.5 seconds intervals at 40% amplitudes using Branson digital sonifier (Branson Ultrasonics, Connecticut, USA) fitted with a 3mm micro tapered tip (Consonic, Sydney, Australia). The protein concentrations were determined using the Pierce® BCA protein assay kit (Thermo scientific, Illinois, USA). The assay was performed according to manufacturer’s instruction.

3.2.4.2 Whole cell protein profile analysis by SDS-PAGE

Aliquots of 17 µg bacterial whole cell proteins prepared above (section 3.2.4.1) were separated by electrophoresis as follows. Equal volume of the whole cell protein was mixed with equal volume of the 2X sample buffer (preparation chapter 2) then heated at 100 0C for 5 minutes. The mixture was loaded onto 1mm thick 12% SDS-PAGE separating gel (stacking gel: 5% PAGE gel), then separated by electrophoresis (80 Volt) for 2 hours in a 1X electrophoresis buffer.

Chapter 3

3.2.4.3 Whole cell protein profiles visualization

Following electrophoresis, gels were stained using Coomassie brilliant blue (refer chapter 2). The stained gels were then scanned using UMAX powerlook 1000 (MIAF, UNSW, Sydney, Australia).

3.2.5 Examination of adhesion and invasion of oral C. concisus strains to human intestinal epithelial cells, using a modified gentamicin protection assay

The adherence and invasion abilities of oral C. concisus strains to human intestinal epithelial cells were investigated. Sixteen oral C. concisus strains isolated from the nine patients with CD (Patient No. 1, No. 2, No. 4, No. 5, No. 6, No. 10, No. 11, No. 12 and No. 13), In addition to 18 oral C. concisus strains isolated from twelve healthy controls (Healthy No. 1, No. 2, No. 3, No. 4, No. 5, No. 6, No. 8, No. 9, No. 10, No. 11, No. 12 and No. 14) were tested. The intestinal biopsy C. concisus strain P1CDB1(UNSWCD) of patient No. 1, which was previously shown to be invasive to Caco-2 cells 315, was used as the positive control.

The enteric C. concisus isolate of patient 1 (P1CDB1(UNSWCD)) was isolated by Zhang et al 321, which was named as UNSWCD in a following study by Man et al 338. To maintain the consistency with the previous publications and the naming system in this study, we used P1CDB1 (UNSWCD) to label this strain in this study.

The adherence assay and invasion assay procedures are as follows: 5 × 105 of the intestinal epithelial cells Caco-2 (refer chapter 2) were seeded onto 24-well plates (nunc, Roskilde, Denmark), plates were pre-coated with 0.05 mg/ml of rat tail collagene type 1 (BD, New Jersey, USA) for 45 minutes at room temperature, cell were then incubated at 370 C and 5%

CO2 for 4 days to form a monolayer.

The monolayer was then washed 4 times using phosphate buffer saline (PBS) and incubated with C. concisus strains at a multiplicity of infection (MOI) of 100 for 2 hours in MEM media containing no antibiotics. Centrifugation was achieved at 320 RCF for 5 minutes to promote the adherence to the surface of the epithelial cells.

Six wells of Caco-2 cells were infected with each of C. concisus strain. Following the 2 hour incubation, the wells were washed 5 times with PBS. The Caco-2 monolayer of 3 wells was lysed with 1% Triton X-100 (Gibco, California, USA) for 5 minutes. Serial dilutions of cell lysates were inoculated onto HBA plates and incubated at 37 0C for 48 hours in a

Chapter 3 microaerobic condition. Colony-forming units (CFU) of C. concisus were recorded, which were regarded as the numbers of C. concisus that were associated with the Caco-2 cells (the sum of adhering bacteria and invading bacteria).

The remaining 3 wells of Caco-2 cells infected with C. concisus were incubated with 1 ml of MEM containing 200 µg/ml of gentamicin to kill the extracellular bacteria. Following washing of the Caco-2 monolayer for 5 times using PBS, the cells were lysed with 1% Triton X-100 for 5 minutes. Serial dilutions of cell lysates were inoculated onto HBA plates and incubated for two days. CFU of C. concisus were recorded. Generated figures were regarded as the numbers of C. concisus that had invaded (internalized) Caco-2 cells. The assay was repeated three times. Prior to cell lysis, the number of viable extracellular bacteria was determined by culturing the supernatant on HBA. This was to insure that gentamicin had killed all extracellular bacteria which had not invaded the Caco-2 cells.

C. concisus adhesion index was calculated as follow:

The invasion index was calculated using the formula described by Larson et al in 2008

350, which is as follow:

3.2.6 Testing the susceptibility of C. concisus strains to gentamicin antibiotic

3.2.6.1 Bacterial culture conditions All C. concisus strains identified by the whole protein profile were examined for their susceptibility to gentamicin antibiotic.

Strains were grown on modified Iso-Sensitest agar to suit the growth of C. concisus. The agar was prepared by using 4.32% (w/v) Iso-SensitestTM agar (Oxoid) supplemented with 6% (v/v) defibrinated horse blood (Oxoid)

Chapter 3

3.2.6.2 Epsilometer test The C. concisus susceptibility to gentamicin was determined by Epsilometer test strips (E- test) (AB-BIODISK, Solna, Sweden). In brief; C. concisus strains were grown on Horse blood agar (HBA) for 2 days in microaerobic condition at 37 0C. A loopful of the bacteria was inoculated into 1 ml PBS. The mixture was spread on the modified Iso-Sensitest agar and the excess inoculum was aspirated. Agar plates were then left to dry at room temperature. The gentamicin-containing strip was applied at the centre of the dry agar plate which was then incubated for 48 hours at 370 C at microaerobic condition.

After the incubation period, the minimum inhibition concentration (MIC) was measured by the reading on the E-test strip at which the inhibition zone intercepted.

3.2.7 Statistical Analysis Student’s t test and the Fisher’s exact test were used in this study. Statistical analysis was performed using Graph Pad software (San Diego, California, USA). P values of less than (< 0.05) were considered significant for all statistical tests.

3.3 Results

3.3.1 Patient characteristics

C. concisus was isolated from 32 study subjects, 15 of which were patients (11 adults and 4 children) with IBD (9 CD and 6 UC) and 17 were healthy controls (12 adults and 5 children), children are defined as under 18 years old (Table 3.1).

The 15 patients with IBD (11 males and 4 females) ranged between 5-73 years old (mean±SD, 33±5.7), and the 17 healthy controls (8 males and 9 females) were of 4-67 years old (26±4.7). The age of patients and controls was not statistically different (unpaired t test, P > 0.05).

Out of the 15 patients; 14 patients had active IBD (13 newly diagnosed with IBD and one relapsing case) while one was in remission. None of the patients had received antibiotic treatment for their IBD except for the relapsed patient (patient number 2, Table 3.1) who received metronidazole and ciprofloxacin two years ago at the time of initial diagnosis.

All patients and controls had not received antibiotic treatment during the six months prior to sample collection.

Chapter 3

3.3.2 Identifying the putative oral C. concisus isolates from patients with IBD and healthy controls

Twelve putative oral C. concisus were isolated from each patient and control, making a total of 384 putative oral C. concisus isolates.

C. concisus specific PCR 338 was used to confirm the identity of the putative oral C. concisus isolates. It was found that at least 9 out of the 12 putative C. concisus isolates collected from each individual were positive for C. concisus specific PCR. According to that, a total of 288 C. concisus isolates (9 from each individual) were analysed by SDS-PAGE analysis to determine the number of oral C. concisus strains colonizing the oral cavity of each individual.

3.3.3 Determining the number of oral C. concisus strains colonizing the oral cavity of each patient and control

The number of oral C. concisus strains colonizing each individual was determined by the whole cell protein profile. Nine C. concisus isolates collected from each individual were subjected to SDS-PAGE analysis, C. concisus isolates with identical protein profiles (presence or absence of protein bands) were designated as the same strain and C. concisus isolates with different protein profiles were designated as different strains (Figure 3.1). We previously showed that different protein profiles of C. concisus isolates correlated with the difference in the housekeeping genes, thus representing different C. concisus strains 351.

Figure 3.1 An example showing the determination of the number of oral C. concisus strains colonizing a given individual. Nine oral C. concisus isolates collected from each individual were subjected to SDS-PAGE analysis. Among the nine C. concisus isolates collected from each individual; five isolates showed pattern 1 and four isolates showed pattern 2 on SDS-PAGE. Thus, the relevant individual was colonized with two different C. concisus strains. M: Molecular weight marker (kDa).

Of the 15 patients with IBD; five patients (Patient No. 1, No. 2, No. 10, No. 12 and No. 17) were colonized with two oral C. concisus strains, three patients (Patient No. 14, No.

Chapter 3

15, No. 16) were colonized with three C. concisus strains and one patient (Patient No. 4) was colonized with four C. concisus strains (Table 3.2 and Figure 3.2). The remaining six patients with IBD were colonized with a single oral C. concisus strain (Table 3.2 and Figure3.2). Of the 17 healthy controls, three individuals (Healthy No. 8, No. 9 and No. 11) were colonized with three oral C. concisus strains and the remaining 14 individuals were colonized with a single oral C. concisus strain. A total of 52 oral C. concisus strains were identified as shown in Table 3.2 and Figure 3.2, the same Figure shows the whole cell protein profiles of the 52 oral C. concisus strains.

Chapter 3

Table 3.2 Oral C. concisus strains detected in individual patients with IBD and healthy controls and their assigned group*.

Individuals ID Number Strains ID Group according to Aabenhus of Strains et al 319 Patient No. 1 2 P1CDO2, P1CDO3 Group 2 Patient No. 2 2 P2CDO1, P2CDO2 Group 2 Patient No. 3 1 P3UCO1 Group 2 Patient No. 4 4 P4CDO1, P4CDO2, Group1:P4CDO2,P4CDO3 P4CDO3, P4CDO4 Group2:P4CDO1,P4CDO4 Patient No. 5 1 P5CDO1 Group 2 Patient No. 6 1 P6CDO1 Group 2 Patient No. 7 1 P7UCO1 Group1 Patient No. 10 2 P10CDO1, P10CDO2 Group 2 Patient No. 11 1 P11CDO1 Group 2 Patient No. 12 2 P12CDO1, P12CDO2 Group 2 Patient No. 13 1 P13CDO1 Group 2 Patient No. 14 3 P14UCO1,P14UCO2, Group 2 P14UCO3 Patient No. 15 3 P15UCO1, P15UCO2, Group 1: P15UCO3 P15UCO3 Group 2: P15UCO1, P15UCO3 Patient No. 16 3 P16UCO1, P16UCO2, Group 2 P16UCO3 Patient No. 17 2 P17UCO1, 17UCO2 Group 2 Healthy No. 1 1 H1O1 Group 2 Healthy No. 2 1 H2O1 Group 2 Healthy No. 3 1 H3O1 Group 2 Healthy No. 4 1 H4O1 Group 2 Healthy No. 5 1 H5O1 Group 2 Healthy No. 6 1 H6O1 Group 2 Healthy No. 7 1 H7O1 Group 2 Healthy No. 8 3 H8O1, H8O2, H8O4 Group 2 Healthy No. 9 3 H9O1, H9O2, H9O3 Group 2

Chapter 3

Healthy No. 10 1 H10O1 Group 2 Healthy No. 11 3 H11O1, H11O2, Group 2 H11O3 Healthy No. 12 1 H12O1 Group 2 Healthy No. 13 1 H13O1 Group 2 Healthy No. 14 1 H14O1 Group 2 Healthy No. 15 1 H15O1 Group 2 Healthy No. 16 1 H16O1 Group 2 Healthy No. 17 1 H17O1 Group 2 *C. concisus was grouped using the method described by Aabenhus et al 319 , a total of 52 strains were examined. Note: The strain ID contains information including diagnosis, source of the strain and the number of strains isolated from a given individual. For example, P1CDO1 means that this strain was isolated from patient No. 1 (P1) who was diagnosed with CD (CD), it is an oral strain (O) and it is the first oral isolate from this patient (1). P1CDO2 is the second oral C. concisus isolate from patient No. 1. P1CDB1 is the first isolate from intestinal biopsies (B) of patient No. 1 with CD. H1O-1 means that this strain is the first oral isolate from healthy control No. 1

A. B.

Figure 3.2 Whole cell protein profiles of the 52 oral C. concisus strains isolated from 15 patients with IBD and 17 controls. The whole cell proteins of C. concisus strains were analyzed using 12% SDS-PAGE and stained using 0.25% of Coomassie Brilliant Blue. M: molecular marker. A. Oral C. concisus isolated from patients with CD; B. Oral C. concisus isolated from patients with UC; C. Oral C. concisus strains isolated from healthy controls. M: Molecular weight marker (kDa).

Chapter 3

3.3.4 Comparison of the prevalence of multiple oral C. concisus strains between patients and controls as well as between children and adults

Sixty percent (9/15) of patients with IBD were colonized with multiple oral C. concisus strains (≥ two strains), which was significantly higher than that of the healthy controls (3/17, 18%) (Fisher’s exact test, P < 0.05) (Figure 3.3). The prevalence of multiple oral C. concisus strains in patients with CD and UC were 56% (5/9) and 67% (4/6) respectively, which was not statistically different from that of the healthy individuals (P > 0.05) (Figure 3.3).

Fifty percent (2/4) of children (under 17 years of age) with IBD were colonized with multiple oral C. concisus strains, which was not significantly different from those depicted in adult patients (7/11, 64%).

Figure 3.3 The prevalence of multiple oral C. concisus strains in patients with IBD and healthy controls The two major forms of IBD are CD and UC (see text for detail). n.s, not significant; * P < 0.05, Fisher’s exact test.

3.3.5 Grouping C. concisus strains according to whole protein profile The oral C. concisus strains were initially categorised into IBD group (strains that were isolated from patients with IBD) and control group (strains that were isolated from healthy

Chapter 3 individuals). Attempts were made to identify specific protein bands that may appear only in strains of IBD group. As shown in Figure 3.2, each C. concisus strain revealed a unique protein profile and specific protein bands that may be associated with IBD strains. These were not detectable by visual inspection.

Oral strains were grouped using the method described by Aabenhus et al 319, This involved the grouping of C. concisus strains into two groups depending on the whole protein profiles. Group 1 had a characteristic double band in the molecular weight region of 170 kDa which was common to all group 1 strains. However, Group 2 was separated from group 1 strains by having one or two additional protein bands in the region of 200 kDa 319. According to this way of grouping, we found that the prevalence of group 2 among the oral C. concisus strains isolated from patients with IBD to be 86% (25 out of 29); among the strains from CD to be 87.5% (14 out of 16) and among strains from UC to be 84.5% (11 out of 13). The prevalence of group 2 in the oral C. concisus strains isolated from healthy controls was found to be 100% (23 out of 23 strains). The isolated strains were grouped as shown in Table 3.2. By using Fisher’s exact t-test, it was found that no significant difference (P > 0.05) between the prevalence of group 2 (or group 1) in the oral C. concisus strains isolated from patients with IBD and the prevalence of group 2 (or group 1) in oral C. concisus strains isolated from healthy controls. A representative whole protein profile for the two groups is illustrated in Figure 3.4. Figure 3.2 shows the whole cell protein profiles of the 52 oral C. concisus strains used in this study.

Group 2 had a prevalence of 92% (48 out of 52 strains) of the total oral C. concisus strains.

Chapter 3

Figure 3.4 A representative whole protein profile for group 1 and group 2 as determined by the grouping method reported by Aabenhus et al 347. Group 1 had a characteristic double band in the molecular weight region of 170 kDa which was common to all group 1 strains while Group 2 was separated from group 1 strains by having one or two additional protein bands in the region of 200 kDa (Blue arrow). M: Molecular weight marker (kDa).

3.3.6 The adherence and invasive ability of oral C. concisus strains to human intestinal epithelial cells The adherence and invasive abilities of 16 oral C. concisus strains isolated from the nine patients with CD (Patient No. 1, No. 2, No. 4, No. 5, No. 6, No. 10, No. 11, No. 12 and No. 13), as well as 18 oral C. concisus strains isolated from twelve healthy controls (Healthy No. 1, No. 2, No. 3, No. 4, No. 5, No. 6, No. 8, No. 9, No. 10, No. 11, No. 12 and No. 14) were evaluated (Table 3.3). C. concisus P1CDB1(UNSWCD) was used as a positive control.

A variation in the adherence ability between the 34 oral C. concisus strains was found. Some strains did not have any adhesion ability (P4CDO1 and H3O1), but some other oral C. concisus strains had a high adhesion index compared to the positive control P1CDB1(UNSWCD), such as P13CDO1 and H8O3 (6.06 and 5.25 respectively). The positive control P1CDB1(UNSWCD) had an adhesion index of 0.35 (Table 3.3)

The adhesion index (Table 3.3) shows that, 6 out of 9 patients with CD (67%) had at least one oral C. concisus strain with equal or even higher adhesion index as compared to the positive control P1CDB1(UNSWCD), while it was found that 7 out of 12 healthy controls (58%) were colonized with oral C. concisus strains that had an equal or higher adhesion index compared to the positive control P1CDB1(UNSWCD). No significant difference

Chapter 3

(P > 0.05) was found between the number of patients with CD and the number of healthy controls that had at least one oral C. concisus strain with equal or even higher adhesion index compared to the positive control P1CDB1(UNSWCD).

The mean adhesion index for the 16 oral C. concisus strains isolated from patients with CD was 1.09, while it was found to be 0.94 for the 18 oral C. concisus strains isolated from the healthy controls. The mean adhesion index between strains isolated from patients and controls was not significant (P > 0.05) (Figure 3.5, A).

The invasion ability of the 34 oral C. concisus strains tested in this study also varied. Some strains did not have any invasion ability (P11CDO1, H3O1, H8O1, H11O2, H11O3 and H12O1), while the highest invasion index was for the strain P2CDO1, followed by P2CDO2 and P5CDO1 with an invasion index of 9.5, 6.5 and 4.1, respectively. The positive control (P1CDB1(UNSWCD)) had an invasion index of 1.32, as shown in Table 3.3.

The invasion index (Table 3.3) shows that, 3 patients out of 9 patients with CD (33%) had at least one oral C. concisus strain with equal or even higher invasion index as compared to the positive control P1CDB1(UNSWCD). On the other hand, none of the healthy controls were colonized with oral C. concisus strains that had an equal or higher invasion index compared to the positive control P1CDB1(UNSWCD). In statistical terms, comparison of the two groups together provided borderline results (P = 0.06).

The highest invasion index found in the healthy controls was 0.39 (H9O2 strain). The number of patients with CD that had at least one oral C. concisus strain with higher invasion index compared to H9O2 (0.39) were 4 out of 9 patients with CD (44%) which was significantly higher than the healthy controls (P = 0.02).

The mean invasion index for the 16 oral C. concisus strains isolated from patients with CD was 1.51, while it was found to be 0.07 for the 18 oral C. concisus strains isolated from the healthy controls. The mean invasion index for strains isolated from patients was significantly higher than that from the controls (P < 0.05) (Figure 3.5, B).

Chapter 3

Table 3.3 Invasion and adhesion index of oral C. concisus strains isolated from patients with CD and controls to Caco-2 cells.

Bacteria *Invasion index *Adhesion index mean ± SEM mean ± SEM P1CDB1(UNSWCD) 1.32 ± 0.2 0.35 ± 0.02

P1CDO2 0.199 ± 0.019 1.2 ± 0.12

P1CDO3 1.954 ± 0.90 0.01 ± 0.00071

P2CDO1 9.5±0.9 0.01±0.001

P2CDO2 6.5±1.4 0.15±0.01

P4CDO1 0.13 ± 0.014 0

P4CDO2 0.045 ± 0.0071 0.44 ± 0.078

P4CDO3 0.17 ± 0.028 2.49 ± 0.43

P4CD04 0.09 ± 0.035 0.12 ± 0.018

P5CD01 4.1 ± 1.1 0.015 ±0.003

P6CDO1 0.07 ± 0.014 0.665 ± 0.1

P10CDO1 0.48 ± 0.071 0.45 ± 0.071

P10CDO2 0.46 ± 0.11 1.82 ± 0.11

P11CDO1 0 0.009 ± 0.0035

P12CDO1 0.08 ± 0 2.42 ± 0.38

P12CDO2 0.375 ± 0.064 1.62 ± 0.30

P13CDO1 0.01 ± 0.00 6.06 ± 1.04

H1O1 0.025±0.004 1.29±0.039

H2O1 0.045 ± 0.0014 2 ± 0.09

H3O1 0 0

H4O1 0.154±0.022 1.159±0.021

H501 0.18±0.035 1.23±0.157

Chapter 3

H6O1 0.1 ± 0.043 0.06 ± 0.0071

H8O1 0 0.02 ± 0.0071

H8O3 0.065 ± 0.0071 5.25 ± 0.53

H8O4 0.05 ± 0.0071 0.63 ± 0.10

H9O1 0.01 ± 0.00 1.37 ± 0.44

H9O2 0.39 ± 0.071 0.027 ± 0.011

H9O3 0.035 ± 0.0071 2.56 ± 0.34

H10O1 0.11 ± 0.031 0.18 ± 0.018 H11O1 0.02 ± 0.0071 0.025 ± 0.0.0071

H11O2 0 0.057 ± 0.0035

H11O3 0 0.006 ± 0.0

H12O1 0 0.0001 ± 0.00

H14O1 0.11 ± 0.014 1.11 ± 0.16 *The invasion and adhesion index were the average of duplicate experiments

Chapter 3

A. B. Figure 3.5 Comparison of mean adhesion index and mean invasion index between C. concisus strains isolated from patient with CD group and C. concisus strains isolated from healthy group. A. The adhesion index (Mean ± SEM) of oral C. concisus strains from patients with CD and healthy controls; B. The invasion index (Mean ± SEM) of oral C. concisus strains from patients with CD and healthy controls. n.s, not significant; *, Significant (P <0.05, unpaired t test).

The adhesion index of the two oral C. concisus groups that were classified previously (section 3.3.5) according to the published method of Aabenhus et al 319 was compared. It was found that the mean adhesion index for group1 was 1.47, while it was 0.97 for group 2. The mean adhesion index between strains in group 1 and group 2 was not significant (P > 0.05) (Figure 3.6 A).

The invasion index of the two oral C. concisus groups that we have classified previously (section 3.3.5) according to the published method of Aabenhus et al was compared. It was found that the mean invasion index for group1 was 0.11, while it was found to be 0.79 for group 2. The mean adhesion index between strains in group1 and group 2 was not significant (P > 0.05) (Figure 3.6 B).

Chapter 3

A. B. Figure 3.6 Comparison of the mean adhesion index and the mean invasion index between the two groups (group 1 and group 2) categorized according to Aabenhus et al 319 grouping method. A. The adhesion index (Mean ± SEM) of oral C. concisus strains from Group1 and group 2; B. The invasion index (Mean ± SEM) of oral C. concisus strains from group 1 and group 2. n.s, not significant.

3.3.7 Gentamicin minimal inhibition concentration (MIC)

The MIC to gentamicin for the C. concisus strains were found to be between 0.25 and 0.38 µg/ml and this is lower than the gentamicin concentration used in the invasion assay by almost 500 fold.

3.4 Discussion

A total 288 oral C. concisus isolates were purified from the oral cavity of 15 patients with IBD (nine CD and six UC) and 17 Healthy controls. C. concisus were identified by Microscopic morphology and then confirmed by 16S rRNA gene PCR. Out of the 288 C. concisus isolates 54 different strains were identified using SDS-PAGE whole cell protein profile analysis.

The multiple oral C. concisus strains that were isolated from the same individual’s oral cavity were subjected to the same environmental influences, This simulation indicated that the differences between the whole protein profiles of the multiple oral C. concisus strains isolated from an individual oral cavity were attributed to the differences in the genes rather than being caused by the effect of the environment. Therefore, using whole protein profile analysis it was concluded that the method was a fast and economical way for defining multiple oral C. concisus strains colonizing an individual’s oral cavity.

Chapter 3

The heterogeneity of the C. concisus was shown in different studies using genetic 345-347 and phenotypic studies 319,329. The heterogeneity of oral C. concisus strains by using whole protein profile was also shown in this study. As presented in Figure 3.2, C. concisus strains isolated from each individual revealed a unique whole protein profile. The effect of the environmental factors on the bacterial protein expression was minimised by growing C. concisus strains under the same growth conditions. This is the first study that shows the heterogeneity of oral C. concisus strains using a large sample size.

We compared the whole protein profiles between the oral C. concisus strains isolated from patients with IBD with the oral C. concisus strains isolated from healthy individuals, in order to find a specific protein bands that are uniquely found in oral C. concisus strains isolated from patients with IBD. However, such bands were not found.

In 2005 Aabenhus et al sub-grouped C. concisus strains isolated from faecal samples according to the whole protein profile into two groups, group 1 and group 2 as explained earlier in the results section 3.3.5. Aabenhus et al 319 found that group 2 comprised 86% of the total C. concisus strains they isolated, and significant differences were found between the 319 two groups particularly in nitrate reduction, esterase and H2S production . An earlier publication by Aabenhus et al in 2002 showed that group 2 is the major group found in immunocompromised patients (which includes IBD) and that only group 2 were recovered from immunocompetent individuals (which includes healthy individuals) 329.

In this study, we subgrouped the oral C. concisus strains into the two groups assigned by Aabenhus et al 319, and we found that group 2 encompassed most of the oral C. concisus strains with a prevalence of 92% (48 out of 52 oral C. concisus strains) of the total oral C. concisus strains isolated from patients and controls, our findings are consistent with those of Aabenhus et al 319. It was also found in this study that all (100%) of the oral C. concisus strains isolated from healthy controls were from group 2, while 86% of the oral C. concisus strains isolated from patients with IBD were from group 2. Although it is herein shown that group 2 was more related to healthy controls rather than patients, no significant difference in the prevalence of group 2 between patients and controls was found.

After defining strains by using whole protein profile. The study showed a significantly higher number of patients with IBD to be colonized with multiple oral C. concisus strains as compared to the healthy controls. The reason why patients with IBD were more often colonized with multiple oral C. concisus strains is not clear. Data obtained from our lab

Chapter 3

(unpublished) showed that by using MLST, the oral C. concisus strains had undergone natural recombination 344. One possibility is that C. concisus strains colonizing the oral cavity of patients with IBD are more capable of natural recombination as compared to the C. concisus strains colonizing healthy individuals; this could have resulted in the development of new C. concisus strains. Alternatively, patients with IBD may have acquired additional oral C. concisus strains from different sources.

The clinical significance of the high prevalence of multiple C. concisus strains in the oral cavity of patients with IBD remains to be determined. A recent study by Habashneh et al in 2011 showed that patients with IBD have a significantly higher prevalence of periodontitis as compared to individuals without IBD 352. Taking in consideration that C. concisus was previously shown to be associated with early periodontitis327,343, we speculate that the high prevalence of multiple oral C. concisus strains in patients with IBD could have contributed to the reported increased incidence of periodontitis in patients with IBD 352.

It is known that recombination could confer benefits to the bacteria by providing an opportunity for acquiring novel metabolic functions, antibiotics resistance and virulence genes 353. According to that, the possibility that oral C. concisus strains in patients with IBD are capable of natural recombination, could result in generating more invasive and virulent C. concisus strains that could have the potential of causing inflammatory manifestations when introduced to the intestinal tract.

The presence of multiple oral C. concisus strains in an individual may increase the chance of lateral (horizontal) gene transfer between these strains. Recently, A Master’s thesis from our group showed by using MLST that the biopsy strain P1CDB1(UNSWCD) is a recombinant of the oral strains P1CDO2 and P1CDO3 and it was found that P1CDO2 is more closely related to the biopsy strain P1CDB1(UNSWCD) 344. Furthermore, a study by Kaakoush et al in 2011 found that enteric C. concisus strains isolated from patients with chronic intestinal diseases have significantly higher invasive potential than those isolated from acute intestinal diseases, and by further investigation it was found that the reason of this high invasiveness was attributed to carrying a plasmid 348. Thus, these findings suggest that, having multiple C. concisus strains in the oral cavity could increase the chance of the lateral (horizontal) gene transfer in which plasmids could play a role. This recombination could lead to producing a new recombinant strain that is more virulent in adhering and invading the intestinal epithelial cells which could consequently lead to chronic inflammation.

Chapter 3

The assumption discussed above could be applicable to the adhesion invasion results. One could see that P1CDO2 strain has a high adhesion capability (1.2) but a very low invasion capability (0.19), on the other hand P1CDO3 has a low adhesion capability (0.01) but a very high invasion capability (1.954). Interestingly, P1CDB1(UNSWCD) has both high adhesion (0.35) and invasion (1.32) capability. Taking in consideration that P1CDB1(UNSWCD) is closely related to P1CDO2, it seems that having these two oral strains in the same niche may promote recombination, in which P1CDO2 has acquired the gene responsible for invasion by lateral gene transfer from P1CDO3. With reference to the finding of Kaakoush et al 348, we assume that the gene responsible for invasion could be carried on a plasmid that has transferred through conjugation from the donor P1CDO3 to the recipient P1CDO2.

The MIC to gentamicin by C. concisus strains used in this study was found to be between 0.25 and 0.38 µg/ml, this finding indicates that C. concisus strains had similar susceptibility to gentamicin. In the invasion assay we used gentamicin in a concentration of 200µg/ml which was greatly higher than the MIC been found. Suggesting that, the concentration used (200µg/ml) in the invasion assay is efficient in killing all the C. concisus that have not invaded the intestinal cells.

Adhesion and invasion play an important role in the pathogenesis of enteric pathogenic bacteria such as Salmonella species and Shigella species. It was found that host cell adhesion 354-356 and invasion 357-360 play an important role in the Campylobacter jejuni pathogenesis. The adhesion348 and invasion315,348 ability of C. concisus strains isolated from enteric (biopsy or faecal) samples has been reported, but it has to be noted that most of these studies used a comparatively small number of C. concisus strains. To date, the adhesion and invasion ability of C. concisus strains isolated from the oral cavity has not been studied. According to that, we investigated the adhesion and invasion ability of oral C. concisus strains isolated from 9 patients with CD (16 strain) and 12 healthy controls (18 strain) to human intestinal epithelial Caco-2 cells.

In this work it was found that, the mean adhesion index of oral C. concisus strains isolated from patients with CD was not significantly (P > 0.05) different from those of the controls. It was also found that there was no significant difference (P > 0.05) between the number of patients and controls colonized with oral C. concisus with equal or higher adhesion index compared to the positive control P1CDB1(UNSWCD). This indicates that patients and

Chapter 3 healthy individuals have similar possibility of being colonized with oral C. concisus strains that have the ability to adhere to the intestinal epithelial cells.

Interestingly, the mean invasion index of oral C. concisus strains isolated from patients with CD was significantly higher than that of the controls. The highest invasion index for C. concisus strain isolated from patients was 9.5 while, it was only 0.39 for healthy controls. Furthermore, it was found that the number of patients with CD who have an oral C. concisus strain with higher invasion index than 0.39 (the highest invasion index for healthy) is significantly higher than the healthy control (P = 0.02). These findings, supported with the finding by Nielsen et al in 2011 which shows that oral C. concisus strains are able to induce apoptosis in Caco-2 cells 361, suggested that patients with CD are colonized with specific oral C. concisus strains which may have the potential to cause intestinal inflammation if they get to colonise the intestinal tract.

The method of sub-grouping described by Aabenhus et al 319 was not suitable for discriminating between the oral C. concisus according to their adhesion and invasion properties. As it was found that, there was no significant difference in the mean adhesion index or the mean invasion index between the oral C. concisus strains in group1 and group 2.

In conclusion, to our knowledge, this is the first report showing the significant higher prevalence of multiple (≥2) oral C. concisus strains colonizing the oral cavity of patients with IBD compared to controls, and that the mean invasion index of oral C. concisus strains isolated from patients with CD was significantly higher than that of the controls. These findings suggest that patients with CD are colonized with specific oral C. concisus strains that have the potential to cause enteric disease if colonizing the intestinal tract.

This study shows that oral C. concisus strains have heterogeneous whole protein profile, consistent with previous studies on enteric C. concisus strains. Another important conclusion that can be made is the inadequacy of whole protein profile in differentiating between oral C. concisus strains isolated from patients with IBD and oral C. concisus strains isolated from healthy individuals.

Chapter 4

Chapter 4: The lipopolysaccharide profiles of oral and enteric C. concisus strains isolated from healthy controls and patients with inflammatory bowel disease

4.1 Introduction The cell wall of Gram negative bacteria has an outer bilayer envelope with an asymmetric organized membrane called, the outer membrane. The inner layer of the outer membrane is composed mainly of phospholipids and proteins, while the outer layer consists of proteins and lipopolysaccharides (LPSs) which are the dominant constituents362,363 , Figure 4.1.

LPS has three components lipid A, the core, and the O-antigen. As shown in Figure 4.1.

Figure 4.1 Inner and outer membrane of gram negative bacteria. The outer layer of the outer membrane is possessed of LPS which has three components: lipid A, core oligosaccharide (inner and outer core) and O-antigen 364.Kdo, 3-deoxy-D-manno- oct-2-ulosonic acid; PPEtn,Pyrophosphorylethanolamine.

Chapter 4

Lipid A is composed of sugars and fatty acids, which is attached to LPS in the outer leaflet of the outer membrane. Lipid A is the bioactive component of LPS as shown by numerous studies 365-368. A study by Loppnowet al in 1989 found that lipid A is responsible for the induction of IL-1 and immune-regulation in human mononuclear cells within the pg/ml range365.

Lipid A is considered a highly conserved component of the LPS, however recent studies have shown that heterogeneity could arise in this region which can be caused by hydroxylation, secondary substitution and length of fatty acid chains in this region369. It was found that environmental conditions can regulate the numbers and types of lipid A 369, these modifications was found to cause different affect on the innate immune responses370-373. Guoet al studied the Salmonella typhimuriumLipid A modifications by the regulatory system PhoP-PhoQ, and suggested that bacteria could combat the immune system and cause chronic illness by: LPS modifications which is in part a result of modifications of lipid A)which promotes resistance to cationic antimicrobial peptides (such as polymyxin and defensins) and host-adapted lipid A which promotes bacterial survival by causing a lower cytokine and chemokine production370. The regulatory system PhoP-PhoQsense host microenvironments by sensing low level Mg2+ 374 and induces transcription of genes essential to intracellular survival within macrophages375, and resistance to cationic antimicrobial peptides376, 377

Environmental changes were also reported by kawaharaet al to modify the structure of lipid A in Yersinia pestis a pathogenic bacteria that causes plague378. The study showed that the growth of Yersinia pestisat two different temperatures, 27oC and 37 oC, changed the lipid A structure. Interestingly the LPS and the lipid A from bacteria grown at 27 oC had induced more TNF production from the mouse and human macrophage cell lines, indicating that Yersinia pestishad modified the structure of their lipid A at 37 oC in order to reduce the activation of human macrophages 372.similar findings were shown for the enteropathogenicYersinia species, Yersinia pseudotuberculosis and Yersinia enterocolitica was reported by Rebeilet al 373.

Refer to Figure 4.1. The inner and outer cores are made of sugars and sugar derivatives. The inner core contains L-glycero-D-mannohepatopyranose (Hep) and 3-deoxyD-manno- octulosonic acid (KDO)379, while the outer core consists of glycose residues, differing in the specific sugarresidues and the linkages between them380. The O- antigen is a polysaccharide

Chapter 4 made of repeating oligosaccharide units generally composed of 3–6 sugars, O-antigen is the part that extends from the cell surface 379.

The inner core usually tends to be conserved within a genus or even a family. However, the outer core has a more structural diversity although this diversity is limited within given specie, or even a genus.The highest variable part of LPS is the O-antigen, this variability is determined by the structure and the length (repeating unit) of the O-polysaccharide chain 364. The individual antigenicity of each bacteria serotype is characterised by changes in the O- antigen 381. It was found that the diversity in the saccharide structure (core and the O-antigen) of LPS is responsible for the serotypes of C. jejuni382,383. The O-chain was found to protect the organism against complement and other serum components 384,385 and was found to be involved in the resistance to the bactericidal activity of the polymorphonuclear leukocytes 386.

LPS is the source of endotoxin activity of Gram negative bacteria and is recognised as the major antigen 381. LPSs can induce many pathophysiological host responses including fever, hypotension, circulatory abnormalities, multi-organ failure, and death. Most of these responses are mediated by cytokine production by stimulated mononuclearphagocytes 387.

Early studies showed that, injecting purified LPS intravenously in experimental animals caused an increased production of many pro-inflammatory cytokines (detected in the serum) such as tumour necrosis factor TNF 388-390, interleukin (IL)-1 391,392, and IL-6 393. An early study on human volunteers by Michie et al in 1988 showed that administrating LPS extracted from Escherichia coli intravenously into human volunteers have elevated their plasma level of TNF 394.

Studies have shown that LPS activates a pattern recognition receptor called toll-like receptor 4 (TLR4) 395, activating TLR4 will lead to the activating of a transcription factor called nuclear factor kappa B (NFκB)396 that leads to the production of pro-inflammatory cytokines, which affect the immune response and host survival (TLRs will be discussed in chapter 5) 397. In active Inflammatory bowel disease (IBD), the levels of TLR4 were strongly upregulated in the intestinal epithelial cells (biopsy samples from small intestine and colon) taken from both Crohn’sdisease(CD) and ulcerative colitis (UC) patients, indicating that alterations in the innate response system may contribute to the pathogenesis of IBD 92.

LPS plays an important role in the pathogenesis of Campylobacter398,399, Moran et al found that the 50% lethal dose (LD50) of Campylobacter jejuni LPS is almost 9 ng/mouse weighing

Chapter 4

20-30 g, which is similar to the LD50 of the enteric pathogen Salmonella typhimurium(almost 4.2 ng/mouse). They have also found that a reduced dose (0.1 ng/ml) of Campylobacter jejuni LPS have stimulated the production of TNF by mouse macrophages reaching to 300 ng/ml, the control Salmonella abortusequi (non-pathogenic or rarely pathogenic Salmonella) LPS stimulated the production of 400ng/ml using the same conditions398, these findings illustrates the importance of LPS in the pathogenesis of Campylobacter jejuni.

Analysing LPS profiles using electrophoresis and staining by silver stain, reveals bands in specific arrangement. Depending on this arrangement LPS could be classified into three different forms:

I. Rough (R)-form in which the profile shows one fast migrating band consists of LPS lacking the O repeating unit 400. II. Semi-rough (SR)- form consisting of the fast migration band and a second band represent the core plus one repeating unit 401. III. Smooth (S)-form in which the profile shows numerous bands arranged in a ladder like pattern whereas the first is the core lacking the O repeating unit and the second band is the core-plus-one repeating unit and so forth402. The three forms are presented in Figure 4.2.

Figure 4.2 Schematic representation of the detailed structure of LPS. Rough (R)-LPS consists of lipid A and the core. Semi-rough (SR)-form LPS consists of lipid A, the core and one repeating unit. The Smooth (S)-form LPS consist of lipid A, the core and many repeating units 369. OM, Outer membrane; PG, Peptidoglycan layer; PM, Plasma membrane.

LPS forms vary between Campylobacter species. Studies have shown that C. coli403has a smooth (S) form of LPS, while C. pyloridis404and C. fetus384,405,406 and C. jejuni407,408showed

Chapter 4 different LPS profiles ranged from rough to smooth. There are currently no studies on C. concisus LPS forms. The aim of this study was to examine Lipopolysaccharides (LPS) extracted from C. concisus strains; their profile forms and its possible association to patients with IBD, the heterogeneity between strains, comparing oral isolates to enteric isolates of the same patients, and grouping C. concisus according to LPS profile.

4.2 Materials and Methods

4.2.1 Study subjects:

C. concisus was isolated from 32 human study subjects, 15 of which were patients with IBD (9 CD and 6 UC) and 17 were healthy individuals, Table 4.1.

4.2.2 C. concisus strains used in this study

C. concisus isolated previously in Chapter 3 (from the 32 study subjects) were identified by bacterial morphology and confirmed by C. concisus specific-PCR. Out of the C. concisus isolated, 56 different C. concisus strains were identified according to the whole cell protein profile as shown in Chapter 3. List of the C. concisus strains used in this study is shown in Table 4.1.

The 56 C. concisus strains used in this study includes: 52 C. concisus strains isolated from the oral cavity, two C. concisus strains isolated from intestinal biopsies, and two C. concisus strains isolated from luminal-washout fluid. Table 4.1

Oral C. concisus refers to strains cultured from saliva, biopsy C. concise refers to strains cultured from intestinal biopsies, and luminal-washout fluid C. concisus refers to strains cultured from luminal-washout fluid. Luminal-washout fluid was the fluid collected from luminal fluid draining tube prior to the start of the colonoscopy, which contains faecal bacteria and the mucosa associated bacteria flushed out from the intestinal mucus, due to the severe diarrhoea induced during the preparation for colonoscopy.

Chapter 4

The details of C. concisus strains are listed in Table 4.1. The strain ID contains information including diagnosis, source of the strain and the number of strains isolated from a given individual. For example, P1CDO1 means that this strain was isolated from patient No. 1 (P1) who was diagnosed with CD (CD), it is an oral strain (O) and it is the first oral isolate from this patient (1). P1CDO2 is the second oral C. concisus isolate from patient No. 1. P1CDB1 is the first isolate from intestinal biopsies (B) of patient No. 1 with CD. H1O1 means that this strain is the first oral isolate from healthy control No. 1.

Table 4.1 C. concisus strains used in LPS extraction, their LPS form and LPS form group if applicable.

Individual ID Sample Strain ID LPS LPS pattern LPS Group and clinical source form number if applicable condition Patient No.1, Saliva P1CDO2 S 7 CD

Saliva P1CDO3 S 8

Intestinal P1CDB1(UNSWCD) S n/a Biopsy

Patient No.2, Saliva P2CDO3 S 10 CD

Saliva P2CDO4 S 9

Patients No.3, Saliva P3UCO1 S 32 UC

Intestinal P3UCB1 S n/a Biopsy

luminal- P3UCLW1 S n/a washout fluid

luminal- P3UCLW2 S n/a washout fluid

Patient No. 4, Saliva P4CDO1 S 13 CD

Saliva P4CDO2 SR 12

Chapter 4

Saliva P4CDO3 S 1 Group A

Saliva P4CDO4 SR 11

Patient No. 5, Saliva P5CDO1 S 2 Group B CD

Patient No. 6, Saliva P6CDO1 S 14 CD

Patient No. 7, Saliva P7UCO1 S 33 UC

Patient No. Saliva P10CDO1 S 2 Group B 10,CD

Saliva P10CDO2 S 1 Group A

Patient No. Saliva P11CDO1 S 3 Group C 11,CD

Patient No. Saliva P12CDO1 S 2 Group B 12,CD

Saliva P12CDO2 S 4 Group D

Saliva P12CDO3 S 2 Group B

Patient No. Saliva P13CDO1 S 1 Group A 13,CD

Patient No. Saliva P14UCO1 S 1 Group A 14,UC

Saliva P14UCO2 S 2 Group B

Saliva P14UCO3 S 3 Group C

Patient No. Saliva P15UCO1 S 31 15,UC

Saliva P15UCO2 S 30

Patient No. Saliva P16UCO1 S 1 Group A 16,UC

Saliva P16UCO2 S 29

Saliva P16UCO3 S 6 Group F

Chapter 4

Patient No. Saliva P17UCO1 S 1 Group A 17,UC

Saliva P17UCO2 S 6 Group F

Healthy No.1 Saliva H1O1 S 22

Healthy No.2 Saliva H2O1 S 2 Group B

Healthy No.3 Saliva H3O1 S 23

Healthy No.4 Saliva H4O1 S 17

Healthy No.5 Saliva H5O1 S 16

Healthy No.6 Saliva H6O1 S 1 Group A

Healthy No.7 Saliva H7O1 S 28

Healthy No.8 Saliva H8O1 SR 26

Saliva H8O2 S 27

Saliva H8O3 S 2 Group B

Healthy No.9 Saliva H9O1 S 1 Group A

Saliva H9O2 S 25

Saliva H9O3 S 24

Healthy No.10 Saliva H10O1 S 2 Group B

Healthy No.11 Saliva H11O1 S 5 Group E

Saliva H11O2 S 15

Saliva H11O3 S 5 Group E

Healthy No.12 Saliva H12O1 S 19

Healthy No.13 Saliva H13O1 SR 18

Healthy No.14 Saliva H14O1 S 2 - Group B

Healthy No.15 Saliva H15O1 S 21

Healthy No.16 Saliva H16O1 S 20

Healthy No.17 Saliva H17O1 S 4 Group D

n/a: not applicable,

Chapter 4

4.2.3 C. concisus growth condition

C. concisus strains were grown on horse blood agar (HBA) (Oxoid, UK) at 370 C, for 24 or 48 hours as indicated in section 4.2.5.1, under microaerobic conditions generated by Campylobacter gas generating kit (Oxoid)

4.2.4 Extraction of LPS and separation on Polyacrylamide Gel Electrophoresis

4.2.4.1 LPS extraction

The LPS was prepared by the rapid phenol micro method described by Fomsgaard et al in 1993409 with some modification. Briefly, Bacteria were grown on HBA for 24, then harvested and re-inoculated onto fresh HBA for 48 hours in the condition previously described in section 4.2.4. Following incubation bacteria were harvested and washed twice using PBS then re-suspended in 500ul pyrogen free water. Equal volume of 90% (v/v) phenol (Merck Millipore, USA) solution was added and the mixture was then vortexed for 45 seconds and heated for 10 minutes in a 70 0C water bath. The mixture was cooled in an ice bath for 5 minutes, which was then centrifuged at 3000g for 5 minutes at 4 0C. After centrifugation the upper clear phase containing the LPS was collected. 1 ml of pure aceton (Ajax Chemicals, Australia) was added to the pooled upper clear phase to precipitate the LPS. The LPS was separated from the solution by centrifugation at 12,000g for 10 minutes, and the pellet of LPS was dried on a 37 0C heat block for 5 minutes. The LPS pellet was then suspended in pyrogen free water (100 µl) (Ambion, USA) and stored at -20 0C until use. An Escherishia coli strain (strain 98) that was previously shown to have a smooth-form LPS was used as a control 538

4.2.4.2 Separation of LPS by polyacrylamide gel electrophoresis

LPS extracted from C. concisus strains was analysed by polyacrylamide gel electrophoresis (PAGE). Equal volumes of the LPS solution and 2X sample buffer were mixed and heated to 1000 C for 3 minutes. LPS was then separated on 1mm thick 12% PAGE separating gel (stacking gel: 5% PAGE gel). Electrophoresis was carried out in 1X electrophoresis buffer with a constant voltage of 80 Volt for 2 hours (until the bromophenol blue had reached the bottom of the gel).

Chapter 4

4.2.4.3 Visualizing LPS patterns

Following electrophoresis, gels were fixed in a fixing solution (25% propan-2-ol (Ajax Chemicals) and 7% glacial acetic acid(Ajax Chemicals) overnight, then visualised by silver staining (Chapter 2), and scanned using UMAX powerlook 1000 (MIAF, UNSW, Sydney, Australia).

4.2.5 Band intensity measurement:

The intensity of LPS bands was analysed using ImageJ v. 1.45 package (NIH).

4.2.6 Statistical analysis

Fisher’s exact test (two tailed) was used to examine if any of the LPS pattern groups is related to IBD, significance was defined as P < 0.05. Unpaired t test was used to examine statistical difference between the ages of patients with IBD and healthy controls. Significance was defined as P < 0.05. The analyses were made using GraphPad Prism 5 software (San Diego, CA).

4.3 Results

4.3.1 Patient characteristics

C. concisus were isolated from 32 study subjects, 15 of which were patients (11 adults and 4 children) with IBD (9 CD and 6 UC) and 17 were healthy controls (12 adults and 5 children), Table 3.1 (chapter 3).

The 15 patients with IBD (11 males and 4 females) ranged 5-73 years old (mean±SD, 33±5.7), and the 17 healthy controls (8 males and 9 females) were 4-67 years old (26±4.7). The age of patients and controls was not statistically different (unpaired t test, P > 0.05).

Out of the 15 patients; 14 patients had active IBD (13 newly diagnosed IBD and one relapsing) while one patient was in remission. None of the patients had received antibiotic treatment for their IBD except for the relapsed patient who received metronidazole and ciprofloxacin two years before the time of their initial diagnosis. Patients were recruited from the Prince of Wales Hospital, the St George Hospital and Sydney Children’s Hospital at Sydney, Australia. All patients had not received antibiotic treatment during the six months prior to sample collection. Adding to that, all controls also had not received antibiotic treatment during the six months prior to sample collection.

Chapter 4

4.3.2 Oral C. concisus LPS profile

The 52 oral C. concisus strains were subjected to LPS extraction using rapid phenol micro method 409. The extracted LPS were then separated on PAGE and visualised by silver staining. LPS forms were then determined by analysing the LPS profile (patterns). The fastest migrating band was LPS lacking the O repeating units, whereas the second band represents the core-plus-one repeating unit and so forth. According to the LPS bands shown on the silver-stained PAGE gels, LPS patterns could be distinguish to R-form LPS, S-form LPS 402 and SR- form.

R-form LPS was used to describe the LPS profile that lack the O polysaccharide and may have incomplete core oligosaccharides (OS)400, which is characterized by a zone of fast migrating low molecular weight band that lack the O-antigen chain length. S-form LPS was used to describe those profiles with the O-polysaccharide chains with characteristic laddering. SR- form LPS describes strains that have core OS and a single O-chain 401. S-form LPS is demonstrated by the control (E. coli) as show in Figure 4.3A.

Of the 52 oral C. concise strains used, 33 different LPS profile (patterns) were recorded, 29 (88%) LPS profiles showed a S-form LPS, 4 (12%) LPS profiles showed a SR- form LPS, and no S-form LPS (0%) was recorded (Figure 4.3A). Strains that showed the SR-form LPS profile were; P4CDO2, P4CDO4, H8O1, and H13O1.

Of the 32 cases studied (15 IBD and 17 healthy), all 15 patients with IBD (100%) and 16 healthy controls (94%) had at least one C. concisus strain with S-form LPS in their oral cavity. while only 1 case which is a healthy control (H13) had a C. concisus strain in his oral cavity showing SR-form LPS Figure 4.3A and Table 4.1.

Of the 52 oral C. concisus strains used, 25 strains (48%) were assorted in 6 different groups (A-F). Group A and B contained 8 (32% of the grouped strains) and 9 (36% of the grouped strains) strains respectively, which are the largest two groups; 33% of the strains isolated were either in group A (pattern 1) or group B (pattern 2). The other groups (C-F) had 2 strains in each group, Table 4.1.

Figure 4.3 Oral C. concisus LPS profiles, LPS was extracted by the rapid phenol micro method, then analysed by 12% PAGE and resolved by silver stain. A. Out of the 52 oral strains used in this study 33 different LPS profiles were recorded and 25 strains were assigned into 6 different groups (group A-F, LPS patterns 1-6). Patterns 11, 12, 18 and 26 showed a SR- form LPS (12% of the total 33 different LPS profiles), while the rest patterns show a S-form LPS (88% of the total 33 different LPS profiles). No R-form LPS is presented. Refer to table 4.1.for the ID of each strains that are in each group; B. E. coli (strain 98) 538 shows a S-form LPS (used as a control). M: Molecular weight marker (kDa).

Chapter 4

4.3.3 LPS profile associated with IBD disease

To determine if any of the LPS pattern groups were associated with IBD rather than healthy control, this study compared the number of patients with IBD to the number of healthy controls in each pattern group (A-F) using two-tailed Fishers exact test.

As shown in Table 4.2, there was no significant difference between the number of patients with IBD and the number of healthy individuals in any pattern group. The results are as follow:

Group A LPS pattern was observed in 6 patients (3 CD and 3 UC) out of 15 patients with a prevalence of 40% of patients with IBD, and was observed in 2 controls out of 17 controls with a prevalence of 11.8% of healthy controls no significance was observed between patients and controls (P = 0.1).

Group B LPS pattern was observed in 4 patients (3 CD and 1 UC) out of 15 patients showing a prevalence of 27% of patients with IBD (note, P12CD had 2 strains with the same LPS profile from group B), and was observed in 4 controls out of 17 controls with a prevalence of 24% of healthy controls. No significance was observed between patients and controls (P > 0.05).

Group C and F LPS pattern was observed in 2 patients (C: 1CD and 1UC, F: 2 UC) showing a prevalence of 13% of patients with IBD, although the pattern did not show in any of the healthy individual the p-value was >0.05.

Group D LPS pattern showed on one patient and one control with a prevalence of 6.7% of patients with IBD and 5.9% of healthy controls respectively (P > 0.05).

Group E was observed only in one healthy individuals showing a prevalence of 6% of healthy controls (note, the healthy control H11 had 2 oral C. concisus strains with the same LPS profile from group E).

Chapter 4

Table 4.2 Comparison of LPS pattern groups presented in patients with IBD and in healthy controls.

LPS Pattern Group Group Group Group Group Group Groups A B C D E F

CD 3 3* 1 1 0 0 (n=9) (33%) (33%) (11%) (11%) (0%) (0%)

Patients with UC 3 1 1 0 0 2 IBD (n=15) (n=6) (50%) (17%) (17%) (0%) (0%) (33%)

Total 6 4 2 1 0 2 IBD (40%) (27%) (13%) (7%) (0%) (13%) (n=15)

Healthy individuals (n=17) 2 4 0 1 1 0

(12%) (24%) (0%) (6%) (6%)** (0%)

Total study subjects (n=32) 8 8 2 2 1 2

(25%) (25%) (6%) (6%) (3%) (6%)

P-value IBD to Healthy 0.1 1 0.2 1 1 0.2

(Two tailed Fisher’s exact test) *Group B pattern was observed in 3 patients with CD, but in 4 different strains, 2 of which is in patient P12CD. ** Group E pattern was observed only in the control H11, but in 2 different strains isolated from his (H11) own oral cavity. 4.3.4 Comparing LPS profiles between Oral and enteric C. concisus Isolates

A patient with CD (P1CD) and another patient with UC (P3UC) were the case groups for comparing LPS profiles between Oral and enteric (biopsy and luminal-washout fluid) C. concisus isolates. In the case of the patient with CD (P1CD), LPS was extracted from two oral strains P1CDO3 and P1CDO4 and from the biopsy strain P1CDB1(UNSWCD) (Table 4.1). The LPS profile

Chapter 4 presented in Figure 4.4A shows that all 3 strains has a S-form LPS, and that P1CDO4 LPS pattern is more similar to P1CDB1(UNSWCD). In the case of the UC patient (P3UC), LPS was extracted from one oral strain (P3UCO1), one biopsy strain (P3UCB1) and two luminal-washout fluid strains (P3UCLW2 and P3UCLW1). The LPS pattern were similar between the isolates P3UCO1, P3UCB1 and P3UCF2 but with a slight difference in the lower band (presented in Figure 4.4 B). The lower band of the oral strain has a lower intensity in comparison to the lower band of the enteric isolates (P3UCB1 and P3UCF2) Figure 4.4 B. By using Image J, it is found that the lower LPS band intensity normalized by the upper middle LPS band intensity (represented in Figure 4.5 A) of each isolate was 0.65 (±0.21) for P3UCO1, 2.33 (±0.57) for P3UCB1 and 2.16 (±0.49) for P3UCLW2 (Figure 4.5 B). Using unpaired t test, it was found that the lower band intensity normalized by the upper middle band intensity ratio was significantly higher in P3UCB1 and P3UCLW2 compared to the oral strain P3UCO1 (Figure 4.5 B). Which indicates that the lower band in the oral isolate P3UCO1 has a lower density in comparison to the enteric isolates (P3UCB1 and P3UCLW2). P3UCF1 has a different LPS pattern compared to the other strains (P3UCO1, P3UCB1 and P3UCF2), represented in Figure 4.4 B. All strains isolated from the UC patient (P3UC) had a S-form LPS, as shown in Figure 4.4 B.

Chapter 4

A. B. Figure 4.4 LPS profile of the two IBD cases P1CD and P3UC (a) The LPS profile of the oral isolates P1CDO2 and P1CDO3 and the biopsy isolate P1CDB1(UNSWCD), isolated from the patient P1CD. (b)The LPS profile of the oral isolates P3UCO1, the biopsy isolate P3B5, and the two luminal-washout fluid isolates P3UCLW1 and P3UCLW2, isolated from the patient P3UC (details in results and discussion). M: Molecular weight marker (kDa).

A. B.

Figure 4.5 Comparing the intensity of the lower LPS band between P3UCO1, P3UCB1 and P3UCLW2. A. A representative image showing the lower LPS band (arrow 2) and the upper middle LPS band (arrow 1); B. The lower band intensity normalized by the upper middle band intensity of P3UCO1, P3UCB1, and P3UCLW2. The P3UCB1 and P3UCLW2 have stronger lower band intensity compared to P3UCO1. The values represent the average of three independent experiments ±SE. Intensity of bands was measured using ImageJ software. *: Significant(P < 0.05, unpaired t test). M: Molecular weight marker (kDa).

Chapter 4

4.4 Discussion

The 56 C. concisus strains discussed in this chapter were isolated from 15 patients with IBD and 17 Healthy controls (The age of patients and controls was not statistically different (unpaired t test, P > 0.05). C. concisus was identified by bacterial morphology and confirmed by C. concisus specific-PCR, and strains were identified using whole cell protein profiling as shown in Chapter 3.

In Chapter 3 we subgrouped C. concisus according to the whole protein profile and we found that, whole protein profile was not suitable for discriminating oral C. concisus strains isolated from patients with IBD from C. concisus strains isolated healthy controls. More over C. concisus strains isolated from each individual revealed a unique whole protein profile reflecting the high heterogeneity between strains. In this chapter we show that, using LPS profiling is more successful in subgrouping C. concisus strains, many strains isolated from different individual had identical LPS profile. Results show that 48% of the orals trains were successfully assigned to 6 groups (A-F), 68% of the grouped strains were from group A and B (32% and 36% respectively). Although we were able to subgroup the oral C. concisus strains according to their LPS profiles, it was found that none of the LPS patterns was associated with IBD, as results show (Table 4.2) that no significant difference between number of patients with IBD and the number of healthy controls in each group (A-F) was found (P > 0.05).

C. concisus heterogeneity was shown in different studies using genetic345-347, phenotypic studies319,329, and by whole protein profile as shown in Chapter 3. In this chapter, LPS profiling also showed heterogeneity between C. concisus strains. 33 different LPS patterns were obtained from 54 strains. 29 (88%) LPS pattern showed a S-form LPS, 4 (12%) LPS profiles showed a SR- form LPS, and No S-form LPS was recorded (Figure 4.3 A). This shows LPS heterogeneity, furthermore all of these LPS profiles have been found to have at least one repeating O-antigen unit. Supported with the finding that the diversity in the saccharide structure (core and the O-antigen) of LPS gives out the serotypes of C. jejuni 382,383 we expect that C. concisus will have many serotypes.

Campylobacter with R-form LPS were found to be serum-sensitive while Campylobacter with smooth LPS form were serum-resistant384. Both forms of LPS R or S could activate the innate immune system through TLR4/MD-2 pathway which leads to cytokine production and

Chapter 4 inflammation. Interestingly it was found that the R form of LPS could activate TLR4/MD-2 complex directly while S form of LPS needs CD14 to mediate the activation of TLR4/MD-2 complex, so R form of LPS could stimulate inflammation faster and stronger than what the S form of LPS could do410-413. But the delayed activation of the innate immune system by the S form LPS plays in favour of bacterial survival and proliferation412.

Interestingly, this study showed that 100% of patients and 94% of the healthy controls had at least one oral C. concisus showing a S-form LPS, adding to that all the strains isolated from the enteric parts showed S-form LPS (Figure 3A and B). Combining our finding with the fact that some C. concisus strains specially the strains isolated from the CD patients have the abilities to attach and invade the epithelial cells (Chapter 3), we assume that C. concisus has the potential ability to cause chronic inflammation.

In order to compare oral C. concisus to enteric C. concisus, strains from the oral and the enteric side of two patients with IBD (P1CD and P3UC) were studied.

In the case of the patient P1CD (patient with CD), the two oral strains P1CDO2 and P1CDO3, and the biopsy strain P1CDB1(UNSWCD) all showed S-form LPS. Using multilocus sequence typing (MLST) it was found that P1CDB1(UNSWCD) is a recombinant strain of the oral strains P1CDO2 and P1CDO3 but more closely related to P1CDO2 351. Comparing the two LPS profiles of the oral strains (P1CDO2 and P1CDO3) to the LPS profile of the biopsy strain (P1CDB1(UNSWCD)) we found that, P1CDO2 LPS pattern is more similar to the LPS pattern of the biopsy strain P1CDB1(UNSWCD), as demonstrated in Figure 4.4A, which supports the MLST findings344.

The other case P3UC (patient with UC), a published paper by our group found that P3UCO1, P3UCB1, and P3UCLW2 have identical whole protein profile and were grouped as the same strain using MLST analysis (six housekeeping genes) 351. It was also found that P3UCO1, P3UCB1, and P3UCLW2 are different strains from P3UCLW1 which was also shown by whole protein profile and MLST analysis 351. The results of the LPS profiling shown in this chapter, supports the whole protein profile and MLST analysis findings, as illustrated by Figure 4.3A. It was found that P3UCLW1 has a different LPS profile compared to the other 3 strains (P3UCO1, P3UCB1, and P3UCLW2), and that the strains that showed identical whole protein profile and were group as the same strain using MLST (P3UCO1, P3UCB1 and P3UCLW2) had identical LPS profiles; except that that the enteric strains P3UCB1 and

Chapter 4

P3UCLW2 had more intense lower bands when compared to the oral strain P3UCO1 which is shown in Figure 4.5. Knowing that the lower band in the LPS profile represents the lipid A and core oligosaccharide.402,414, it seems that when oral C. concisus enters the intestinal side it adapts to the new environment by promoting the production or/and change the composition of lipid A/core.

Studies have shown that bacteria could induce lipid A structure modifications in response to altered environment giving the bacteria advantages within host tissue. 370-373. It was shown that Salmonella typhimurium could response to environmental changes by inducing lipid A structural modifications, this modification is regulated by the regulatory system PhoP-PhoQ 370. The regulatory system PhoP-PhoQ sense host microenvironments by sensing low level Mg2+ 374 and response by regulating virulence gene transcription375,376. It was also shown that the enteropathogenic Yersinia species (Yersinia pseudotuberculosis and Yersinia enterocolitica) have modified their lipid A structure when grown at two different temperatures, 270 C and 370 C. The modification of lipid A when grown at 370 C by these enteropathogenic Yersinia species have reduce the activation of human macrophages 373.

Adding our finding which showed the increase production of lipid A/core in the enteric strain compared to the oral strain, to the fact that LPS bestows its endotoxic properties to lipid A 398,415, one could suggest that some oral C. concisus strains produces low concentration of lipid A, but when introduced into the intestinal environment production of lipid A increases which results in increasing its toxicity (This suggestion needs further investigations).

In conclusion, this is the first study on LPS extracted from C. concisus strains. It was found that all (100%) patients with IBD and 94% of healthy individuals have at least one oral strain of C. concisus with S-form LPS, and that none (0%) of the C. concisus strains had a R-form LPS. Grouping C. concisus strains depending on the LPS profile was more successful than using whole protein profile (48% of the orals trains were successfully assigned to 6 groups (A to F)) and the LPS patterns 1 and 2 demonstrated by group A and B (Figure 4.3A) is the most two abundant LPS profiles among C. concisus strains; 33% of the strains isolated were either in group A (pattern 1) or group B (pattern 2). It was found that none of the LPS profiles was associated with IBD, and finally we suggested that oral C. concisus could increase the production or changes the composition of core/lipid A when introduced to the enteric environment.

Chapter 5

Chapter 5: The effects of C. concisus on intestinal epithelial expression of MD-2, TLR2, 4 and 5

5.1 Introduction

Campylobacter concisus is a Gram-negative curved rod; motile by means of a single polar flagellum 306.

Recently, C. concisus has gained attention as an emerging human enteric pathogen 336,347,349,416,417. A number of research groups have reported that C. concisus and the established human enteric pathogen Campylobacter jejuni are the most frequently isolated Campylobacter species from diarrheal stool samples 336,347,349,417. In addition to diarrheal disease, C. concisus has been associated with inflammatory bowel disease (IBD) 321,338-340.

IBD is a chronic inflammatory disorder of the gastrointestinal tract (GIT) 418, which is associated with unrestrained immune cell activation and proinflammatory cytokine production 50,113,421.

Inappropriate activation of innate immune system contributes to the pathogenesis of IBD 422,423. The innate immune system recognizes microorganisms by pattern-recognition receptors (PRRs). PRRs activate the innate immune system by recognizing conserved microbial motifs called pathogen-associated molecular patterns (PAMPs) 424. PRRs are classified into four main families: Toll-like receptors (TLRs), nucleotide-binding oligomerisation domain (NOD) receptors, retinoic acid-inducible gene I (RIG-I)-like receptors and the C-type lectin receptors88. Many studies have shown that TLRs expression levels on intestinal epithelial cells (IECs) are different between patients with IBD and healthy controls 92,425-427, implying that TLRs have a role in the pathogenesis of IBD.

A study by Cario et al in 2000 found that IECs from patients with IBD (CD and UC) strongly upregulated TLR4 while TLR5 expression remained unchanged when compared to the IECs from healthy controls, implying that alterations in the innate response system may contribute to the pathogenesis of IBD. Cario et al used biopsy specimens from the small intestine and colon of patients with IBD and healthy controls, and the TLR expression was assessed by immunohistochemical analysis 92. A recent study by Vamadevan et al in 2010 found that IECs in patients with IBD exhibited increased expression of TLR4 and its co-receptor myeloid differentiation-2 (MD-2) mRNA425. Adding to that, Szebeni et al found that mRNA and protein levels of TLR4 in the inflamed colonic mucosa of children with freshly diagnosed 95

Chapter 5

IBD and with relapsed IBD are higher than healthy controls, but the non-inflamed colonic mucosa of the children with IBD were similar to healthy controls. The TLR3 level was not changed in any of the groups 426. Szebeni et al concluded that innate immunity has an important role in the pathogenesis IBD426.

TLR2 expression in IBD showed conflicting results between studies. The study by Cario et al using immunohistochemical analysis, reported that the expression of TLR2 remained unchanged between the biopsy specimens from small intestine and colon of patients with IBD and healthy controls 92. On the other hand, Szebeni et al found higher TLR2 mRNA and protein levels in the inflamed colonic mucosa of children with IBD compared to healthy controls 426. The Immunohistochemical analysis by Frolova et al showed that the expression of TLR2 was only significantly higher in the terminal ileum of patients with inactive and active UC compared to controls, but the expression of TLR2 in the terminal ileum, cecum and rectum of patients with inactive and active CD and the cecum and rectum of patients with inactive and active UC were not significantly higher than the healthy controls 413.

Two types of TLRs are known: first are TLRs which are found on the cell membrane which includes TLR1, 2, 4, 5 and 9, and the second are TLRs which are found on intracellular organelles including TLR3, 7 and 8 428,429. TLRs are expressed throughout the GIT on IECs, enteroendocrine cells, myofibroblasts, and on immune cells within the lamina propria, such as dendritic cells and T cells 429. Following is a brief overview of TLR2, TLR4, TLR5, and MD-2, and their PAMP ligands.

Lipopolysaccharide (LPS) is the major immune stimulator of the Gram negative bacteria cell wall 387,430. The predominant sensor for LPS is TLR4 in association with its co-receptor MD- 2. MD-2 is an essential co-receptor that binds TLR4 at its extracellular domain431,432. LPS recognition begins when LPS is transferred to CD14 via the opsonin LPS-binding protein 433, LPS/CD14 will stimulate TLR4 via MD-2 on the cell surface which will then result in NFκB (nuclear factor kappa-light-chain-enhancer of activated B cells) activation causing the secretion of pro-inflammatory cytokines 434. No study has yet shown if C. concisus affects the expression level of TLR4 on human IEC.

Gewirtz et al found that flagellin (monomeric subunit of flagella) is the ligand for TLR5 which causes NFκB activation, and that TLR5 is expressed exclusively on the basolateral surface of IECs 435. Gewirtz et al found that only ﬂagellin that contacts the basolateral

Chapter 5 epithelial surface activates epithelial proinflammatory gene expression; ﬂagellin on the apical epithelial surface had no effect 435. Gewirtz et al concluded that only bacteria that could invade and translocate through the IECs such as Salmonella typhimurium, will induce IECs to orchestrate an inﬂammatory response 435. Interestingly, a study by Andersen-Nissen et al found that CHO K1 cells (cell line initiated from a biopsy of an ovary of an adult Chinese hamster, American Type Culture Collection) transfected with human TLR5 cDNA did not stimulate human TLR5 when adding purified flagellin from C. jejuni at wide concentration range, whereas the flagellins from Salmonella typhimurium, Escherichia coli, Pseudomonas aeruginosa, Listeria monocytogenes and Serratia marcescens did 436. To date, no studies have examined the effect of C. concisus on the expression level of TLR5 on human IEC

Bacterial peptidoglycan, lipopeptide, and lipotechoic acid are the ligands that activate TLR2. Activating TLR2 will induce the activation of the transcription factor NFκB in host cells 424,437-439. No study has shown if C. concisus affects the expression level of TLR5 on human IEC.

A breakdown in tolerance of the gut immune system to commensal intestinal bacteria in patients with IBD has been detected 117,220. However, the factor(s) that triggers the gut immune system to attack the intestinal commensal bacteria, a cohort of organisms that the gut immune system has co-evolved and lived peacefully with, perhaps for decades, is unknown.

One of the mechanisms that ensures the tolerance of the mucosal immune system towards the luminal commensal bacteria is decreasing the surface expression of toll like receptors and co- receptors 440. Cario et al reported in 2000 by using immunohistochemical analysis that TLR2 and TLR4 are barely detectable in small intestinal and colonic biopsy specimens taken from healthy individuals while TLR3 and TLR5 are constitutively expressed 92. MD-2 was found to be expressed at a low level in normal colonic epithelial cells 441. It was also found that the surface protein expression and the mRNA levels of TLR4 and MD-2 in the human IEC lines HT-29 and Caco-2 are expressed at low level 442,443. The low level expression of TLR4 and MD-2 will avoid the excessive inflammatory response by minimising the recognition of luminal bacterial in the healthy intestine. Furthermore, it was found that decreased expression of TLR4 and MD-2 correlated with the downregulation of downstream immune responses 444,445. It was found by Cario et al that LPS activates NFκB in the human IEC HT-29 in a time- and dose- dependent manner. IEC express TLRs that mediate LPS stimulation of specific intracellular signal transduction pathways in IEC. Thus, IEC plays a frontline role in

Chapter 5 monitoring lumenal bacteria 445. Interestingly it was found by Abreu et al that cotransfection of TLR4 and MD-2 in IEC cell lines (HT-29, Caco-2 and T84) will lead to synergistic activation of NFκB (T84, 82-fold; HT-29, 13-fold; Caco-2, 168-fold) and IL-8 (Caco-2, 12- fold) in response to LPS (compared to empty-vector control) 446.

The intestinal microbial flora plays a key role in the development of IBD. Studies from both human and animal models of IBD have demonstrated that colitis does not occur in the absence of intestinal microbiota 211,212,447,448. Despite these advances in understanding the role of intestinal microbiota in the development of IBD, the exact causative agent(s) of human IBD still remains unknown. Accumulated evidence suggests that some intestinal commensal bacterial species are involved in the pathogenesis of human IBD 419,449. Changes in the composition of the intestinal bacterial community have also been reported 271,450-452.

In 2009, Zhang and collaborators detected a significantly higher prevalence of C. concisus in intestinal biopsies and fecal samples of children with CD as compared with the controls 321,338. Recently, studies from the same group and others showed that such an association between C. concisus and IBD also exists in the adult population 339,340.

Despite the high intestinal prevalence of C. concisus in patients with IBD, whether C. concisus has contributed to the pathogenesis of IBD is not clear. We hypothesize that C. concisus may act as a trigger to initiate the development of human IBD. To test our hypothesis, in this study, we examined the effects of C. concisus strains isolated from patients with IBD and controls on human intestinal epithelial expression of TLR2, TLR4, and its co- receptor MD-2, and TLR5 using an in vitro human IEC model (HT-29).

5.2 Material and methods

5.2.1 C. concisus strains and cultivation conditions

Eleven oral (isolated from saliva) and enteric (isolated from intestinal biopsies and faeces) C. concisus strains we previously isolated were included in the studies described in this chapter. Of the 11 C. concisus strains studied, eight strains were from three patients with IBD (some patients were colonized with multiple strains) and three strains were from healthy controls. Details of the C. concisus strains used in this study are listed in Table 5.1.

The isolation and identification of the oral C. concisus strains used in this chapter were presented in chapter 3, except for C. concisus strains P1CDB1(UNSWCD) 321, P3UCB1 and

Chapter 5

P3UCLW1 344 which were previously isolated by our group. P1CDO13 was defined as a strain by using multilocus sequence typing (MLST) 344. To maintain the consistency with the previous publications and the naming system, we used P1CDB1 (UNSWCD) to label this strain in this study. All strains used in this chapter are presented in Table 5.1.

All C. concisus strains were grown on horse blood agar (HBA) (for preparation refer to chapter 2) at 37 °C under microaerobic condition. The microaerobic condition was generated using the BR0056A gas generating system (Oxoid, Hants, United Kingdom).

Table 5.1 C. concisus strains used in this study.

Strain ID Sample source Clinical condition P1CDO2 Saliva Crohn’s disease P1CDO3 Saliva P1CDO13 Saliva P1CDB1(UNSWCD) Intestinal Biopsy P2CDO1 Saliva Crohn’s disease P3UCO1 Saliva Ulcerative colitis P3UCB1 Intestinal Biopsy P3UCLW1 Laminal wash out H1O1 Saliva Healthy H4O1 Saliva Healthy H5O1 Saliva Healthy P1CDO2, P1CDO3, P1CDO13 and P1CDB1(UNSWCD) were isolated from the same patient (P1CD). P3UCO1, P3UCB1 and P3UCLW1 were isolated from the same patient (P3UC). The remaining strains were isolated from individual patients with IBD and healthy controls; for explaining the code given for each strain refer to chapter 3 section (3.2.3)\

5.2.2 Cultivation of HT-29 cells

Human intestinal epithelial cell line HT-29 cells (ATCC No. HTB-38), were maintained in McCoy’s 5A medium (Invitrogen, California, USA) supplemented with 10% heat inactivated foetal bovine serum (FBS) (Bovogen Biologicals, Melbourne, Australia), 100U/ml penicillin and 100 µg/ml streptomycin (Invitrogen,). The cells were grown at 37°C in a humidified incubator containing 5% CO2.

Chapter 5

5.2.3 Western blot

Western blot (WB) was used to examine the expression of TLR2, MD-2, TLR4 and TLR5 in response to C. concisus infection in HT-29 cells. WB assays were performed in triplicate and repeated at least twice.

5.2.3.1 Antibodies used for Western blotting

All antibodies used for Western blotting were purchased from Santa Cruz Biotechnology Inc (Santa Cruz biotechnology Inc, California, USA). Primary antibodies used were monoclonal anti-TLR2 (sc-166900), and polyclonal anti-TLR4 (sc-10741), anti-MD-2 (sc-20668), anti- TLR5 (sc-16243), anti-α tubulin (sc-31779). Secondary antibodies conjugated with horseradish peroxidase (HRP) were bovine anti-goat IgG (sc-2352), goat anti-mouse IgG (sc- 2031), and goat anti-rabbit IgG (sc-2054).

5.2.3.2 Infection of HT-29 cells with C. concisus

HT-29 epithelial cells were seeded at an initial concentration of 5x105 cell/ml in 6-well cell culture plates (Nunc, Roskilde, Denmark). The cells were grown for 48 hours to form a monolayer.

HT-29 monolayer cells were washed 5 times using Dulbecco's Phosphate-Buffered Saline (DPBS) (Invitrogen) and then added with McCoy’s 5A medium supplemented with 10% FBS without antibiotics. HT-29 monolayer cells were infected with C. concisus at a multiplicity of infection (MOI) of 25 and further incubated for 24 hours. HT-29 cells without C. concisus were used as the negative control.

5.2.3.3 Preparation of HT-29 cells whole cell proteins

HT-29 cells were harvested from culture plates and washed with pre-cooled DPBS. The cells were lysed using RIPA Buffer (50 mM Tris, 150 mM NaCl, 1% Triton X-100, 0.1% SDS) containing a mixture of protease inhibitors (Sigma-Aldrich, Castle Hill Australia). Whole cell lysates were centrifuged twice at 14000g for 20 minutes at 4 0C. Supernatant was collected and stored at -80 0C till use. Protein concentrations were determined using Pierce® BCA Protein Assay Kit (Thermo Fisher Scientific, Scoresby, Australia).

100

Chapter 5

5.2.3.4 Western blot

Whole cell proteins were separated on 12% sodium dodecyl sulphate (SDS)-polyacrylamide gel under reducing conditions, and transferred onto polyvinylidine difluoride (PVDF) membranes (Bio-Rad, California, USA). PVDF membranes were blocked with the blocking solution (5% skim milk in PBS) for 1 hour and 30 minutes at room temperature then probed with primary antibodies (dilution 1:300) overnight at 4 0C, followed by secondary antibody conjugated with horseradish peroxidase (HRP) (dilution 1:2500) for 1 hour and 40 minutes at room temperature. The HRP-labelled antibody was detected using Immun-Star™ WesterC™ Chemiluminescence Kits (Bio-Rad laboratory, Gladesville, Australia) and a LAS-3000 imaging system (Fujifilm, Tokyo, Japan). The intensity of protein bands was analysed using ImageJ, v. 1.45 package (National Institutes of Health, USA, http://rsb.info.nih.gov/ij/index.html).

5.2.4 Immunofluorescence staining and confocal microscopy

Expression of TLR2, MD-2, TLR4 and TLR5 in responses to C. concisus infection in HT-29 was visualized using immunofluorescence staining and confocal microscopy. Immunofluorescence experiments were repeated at least three times.

5.2.4.1 Antibodies used for immunofluorescence staining

Primary antibodies used were anti-TLR2 (sc-166900), anti-TLR4 (sc-10741), anti-MD-2 (sc- 20668), and anti-TLR5 (sc-16243) (Santa Cruz). Secondary antibodies used were Alexa Fluor® 488 donkey anti-goat IgG (A11055), Alexa Fluor® 488 goat anti-mouse IgG, and Alexa Fluor® 594 goat anti-rabbit IgG (A11037) (Invitrogen).

5.2.4.2 Infection of HT-29 cells with C. concisus

HT-29 cells (1x105/ml) were seeded onto sterile cover-slips placed in 6-well cell culture plates and allowed to grow for 48 hours. HT-29 cells were then infected with C. concisus at an MOI 25 and incubated for a further 24 hours. HT-29 cells without C. concisus were used as the negative control.

5.2.4.3 Immunostaining and visualization of TLR4, MD-2 and TLR5 by confocal microscopy HT-29 cells grown on cover-slips were fixed in 3.7% paraformaldehyde, permeabilized with 0.1% Triton X-100 in PBS. The cover-slips carrying HT-29 cells were blocked with blocking

101

Chapter 5 buffer (1% bovine serum albumin (Invitrogen) in PBS) for 1 hour at room temperature. HT- 29 cells were then sequentially incubated at room temperature with a primary antibody (dilution 1:40) and an Alexa Fluor conjugated secondary antibody (dilution 1:400) for 1 hour, followed by staining with Hoechst 33342 (1µg/ml) (Invitrogen) for 10 minutes at room tempreture. The cover slips were washed three times with PBS between incubations for 10 minutes each time. The cover slips were mounted using AF1 antifadent (Citifluor Ltd,

London, UK) and HT-29 cells were observed using an Olympus FluoView FV1000 confocal laser scanning microscope.

5.2.5 Flow cytometry

Flow cytometry (FC) was used to examine the surface and the whole cell expression of TLR2, MD-2, TLR4 and TLR5 in response to C. concisus infection in HT-29 cells. FC experiments were performed in triplicate and repeated at least twice.

5.2.5.1 Antibodies used for immunofluorescence staining

Primary antibodies used were anti-TLR2 (sc-166900), anti-TLR4 (sc-10741), anti-MD-2 (sc- 20668), and anti-TLR5 (sc-16243) (Santa Cruz). Secondary antibodies used were Alexa Fluor® 488 donkey anti-goat IgG (A11055), Alexa Fluor® 488 goat anti-mouse IgG (A11029), and Alexa Fluor® 594 goat anti-rabbit IgG (A11037) (Invitrogen).

5.2.5.2 Infection of HT-29 cells with C. concisus

HT-29 cells (5x105/ml) were seeded onto T-25 tissue culture flask (Nunc). The cells were grown for 48 hours to form a monolayer.

HT-29 monolayer cells were washed 5 times using Dulbecco's Phosphate-Buffered Saline (DPBS)(Invitrogen) and then added with McCoy’s 5A medium supplemented with 10% FBS without antibiotics, HT-29 monolayer cells were infected with C. concisus at an MOI of 25 and further incubated for 24 hours. HT-29 cells without C. concisus were used as the negative control.

5.2.5.3 Flow cytometry

HT-29 cells were detached from culture flasks by incubating with 0.25% trypsin (Invitrogen) for 5 minutes then deactivated with McCoy’s 5A medium supplemented with 10% FBS without antibiotics, and washed 2 times with pre-cooled DPBS.

102

Chapter 5

HT-29 cells were fixed for 12 minutes in 3.7% paraformaldehyde in PBS, then permeabilized for 10 minutes with 0.1% Triton X-100 in PBS when analysing the total expression of the target protein in cells. Cells that were not permeabilized were used to analyse the surface expression of the target protein. Cells were then washed twice using blocking solution (1% Bovine Serum albumin in PBS), and blocked with the blocking solution for 25 minutes. HT- 29 cells were then sequentially incubated at room temperature with a primary antibody (dilution 1:40) and an Alexa Fluor conjugated secondary antibody (dilution 1:400) for 1 hour and 45 minutes respectively. Cells were washed three times with the blocking solution between incubations for 10 minutes each time. HT-29 cells that were not exposed to antibodies were used to assess the background signal. Data were acquired by BD LSRFortessa™ SORP cell analyser (BD Biosciences, San Jose, USA) and analysed in Flow Jo software (http://www.flowjo.com/).

5.2.6 Statistical analysis Data were analysed by means of unpaired t test using GraphPad Prism version 5.1 (San Diego, CA). P-values < 0.05 (two tailed, 95% confidence interval) were considered significant.

5.3 Results The results shown are the mean of triplicates; the experiments were repeated at least twice.

5.3.1 Effects of C. concisus on TLR4 expression in the intestinal epithelial cells HT-29

Expression of TLR4 induced by C. concisus strains in HT-29 cells was assessed by WB and FC.

WB revealed two protein bands, the glycosylated TLR4 (Gly-TLR4) and non-glycosylated TLR4 (Non-GlyTLR4). Gly-TLR4 and Non-GlyTLR4 were analysed separately. The intensity of Gly-TLR4 and Non-GlyTLR4 bands were normalized to the intensity of α- Tubulin (internal control, 55 kDa) of the same sample. The levels of Gly-TLR4 and Non- GlyTLR4 in each sample were expressed as the fold change of the normalized band intensity relative to the normalized band intensity of the non-infected HT-29 cells (HT-29 cells without C. concisus). The representative WB of Gly-TLR4 (120 kDa) and Non-GlyTLR4 (90 kDa) is shown in Figure 5.1A. The levels of Gly-TLR4 and Non-GlyTLR4 (mean ± SE) in HT-29 cells induced by C. concisus strains are shown in Figure 5.1 D.

103

Chapter 5

The levels of Gly-TLR4 in HT-29 cells induced by the 11 C. concisus strains examined (measured by WB) were all significantly higher than the non-infected HT-29 cells (t test, P < 0.05); P1CDO2 (5.71±1.34), P1CDO3 (8.50±0.80), P1CDO13 (6.55±0.61), P1CDB1(UNSWCD) (7.31±0.93), P2CDO1 (1.63±0.15), P3UCO1 (5.45±0.22), P3UCB1 (4.40±0.50), P3UCLW1 (4.34±0.64), H1O1 (2.68±0.43), H4O1 (1.60±0.14), and H5O1 (2.45±0.36) (Figure 5.1 D). The average level of Gly-TLR4 induced by the eight C. concisus strains isolated from patients with IBD was significantly higher than that induced by the three C. concisus strains isolated from healthy controls (5.49±0.74 vs 2.24±0.32, P < 0.05) (Figure 5.2 A). The details of C. concisus strains used in this study are shown in Table 5.1.

The levels of Non-GlyTLR4 induced by the 11 C. concisus strains examined (measured by WB) were all significantly higher than the non-infected HT-29 cells (t test, P < 0.05); P1CDO2 (3.24±0.29), P1CDO3 (3.10±0.47), P1CDO13 (2.61±0.42), P1CDB1(UNSWCD) (2.21±0.37), P2CDO1 (1.27±0.08), P3UCO1 (3.12±0.87), P3UCB1 (1.97±0.38), P3UCLW1 (1.81±0.31), H1O1 (2.97±0.66), H4O1(2.25±0.03), and H5O1 (6.25±0.48) (Figure 5.1 D).The average level of Non-GlyTLR4 induced by the eight C. concisus strains isolated from patients with IBD was not statistically different from that induced by the C. concisus strains isolated from healthy controls (2.41±0.25 vs 3.82±1.23, P > 0.05) (Figure 5.2 B).

Using FC, the levels of surface TLR4 (non-permeabilized cells) and total TRL4 (permeabilized cells) were expressed as the fold change of the mean channel fluorescence intensity (MFI) derived from fluorescence histogram of a sample relative to the MFI of the non-infected HT-29 cells (HT-29 cells without C. concisus). The representative FC histogram of surface TLR4 and total TLR4 is shown in Figure 5.1C. The levels of surface TLR4 and total TLR4 (mean ± SE) in HT-29 cells induced by C. concisus strains are shown in Figure 5.1 D.

The levels of surface TLR4 on HT-29 cells induced by the 11 C. concisus strains examined (measured by FC) were all significantly higher than the non-infected HT-29 cells (t test, P < 0.05); P1CDO2 (4.18±0.56), P1CDO3 (5.03±0.67), P1CDO13 (5.40±0.71), P1CDB1(UNSWCD) (3.5±0.26), P2CDO1 (2.30±0.29), P3UCO1 (2.23±0.22), P3UCB1 (4.61±0.61), P3UCLW1 (2.30±0.23), H1O1 (1.84±0.20), H4O1 (2.02±0.23), and H5O1 (1.92±0.21) (Figure 5.1 D). The average level of surface TLR4 induced by the eight C. concisus strains isolated from patients with IBD was significantly higher than that induced by

104

Chapter 5 the three C. concisus strains isolated from healthy controls (3.70±0.46 vs 1.93±0.05, P < 0.05) (Figure 5.2 C)

The levels of total TLR4 induced by the 11 C. concisus strains examined (measured by FC) were all significantly higher than the non-infected HT-29 cells (t test, P < 0.05); P1CDO2 (2.10±0.19), P1CDO3 (1.92±0.18), P1CDO13 (1.74±0.15), P1CDB1(UNSWCD) (2.12±0.09), P2CDO1 (1.65±0.09), P3UCO1 (1.62±0.11), P3UCB1 (1.51±0.11), P3UCLW1 (1.82±0.07), H1O1 (1.64±0.08), H4O1(1.71±0.09), and H5O1 (1.54±0.05) (Figure 5.1 D).The average level of total TLR4 induced by C. concisus strains by the eight C. concisus strains isolated from patients with IBD was not statistically different from that induced by the C. concisus strains from the healthy controls (1.81±0.08 vs 1.63±0.05, P > 0.05) (Figure 5.2 D).

In addition to detection by WB and FC, expression of TLR4 in HT-29 cells was visualized using immunofluorescence staining and confocal microscopy. Confocal microscopy image showed an increased expression of TLR4 in HT-29 cells after co-incubation with a representative C. concisus strain (P1CDB1(UNSWCD)) for 24 hours (Figure 5.1B).

105

Chapter 5 D

Figure 5.1 Detection of TLR4 in HT-29 cells infected with C. concisus strains for 24 hours by: A. Western blot (WB): a representative Western blot image revealing two bands, the Glycosylated TLR4 (Gly-TLR4, 120 kDa) and non-glycosylated TLR4 (Non-GlyTLR4, 95 kDa). The intensity of Gly-TLR4 and Non-GlyTLR4 bands of each sample was normalized to the intensity of the internal control α-Tubulin (55 kDa) of the same sample. The levels of Gly-TLR4 and Non-GlyTLR4 of each sample were expressed as the fold change of the normalized band intensity relative to the normalized band intensity of the HT-29 cells without C. concisus (N) and shown in part D; B. Immunofluorescence microscopy: a representative confocal microscope image showing the expression of TLR4 in HT-29 cells with and without C. concisus (P1CDB1(UNSWCD) strain) infection. Detection was by anti-TLR4 antibody followed by Alexa Fluor 594 labelled secondary antibody. Scale Bar=10µm; C. Flow Cytometry (FC): a representative FC histogram showing the expression of surface TLR4 and total TLR4 in HT-29 cells with and without C. concisus infection (grey, N). Detection was by anti-TLR4 antibody followed by Alexa Fluor 594 labelled secondary antibody. Background (purple) was from HT-29 cells that were not exposed to antibodies. The levels of TLR4 were expressed as the fold change of the mean channel fluorescence intensity (MFI) derived from fluorescence histogram of a sample relative to the MFI of the HT-29 cells without C. concisus (N), the fold change is shown in the part D; D. The fold change of TLR4 expression in HT- 29 cells induced by C. concisus strains N: HT-29 cells without C. concisus; P1CDB1 refers to P1CDB1(UNSWCD)

106

Chapter 5

Figure 5.2 Comparison of mean expression (fold change) of TLR4 in HT-29 cells infected with C. concisus strains isolated from patient with IBD and healthy controls. A. Gly-TLR4 expression (Mean ± SEM) in infected HT-29 by C. concisus strains isolated from patients with IBD and healthy controls (using WB); B. Non-GlyTLR4 expression (Mean ± SEM) in HT-29 infected by C. concisus strains isolated from patients with IBD and healthy controls (using WB); C. Surface TLR4 expression (Mean ± SEM) on infected HT-29 by C. concisus strains isolated from patients with IBD and healthy controls (using FC); D. Total TLR4 expression (Mean ± SEM) in infected HT-29 by C. concisus strains isolated from patients with IBD and healthy controls (using FC). n.s, not significant; * = significant P <0.05 (unpaired t test).

5.3.2 Effects of C. concisus on MD-2 expression in HT-29 cells

Expression of MD-2 induced by C. concisus strains in HT-29 cells was assessed by WB and FC.

WB revealed two bands, the glycosylated MD-2 and the non-glycosylated MD-2. The narrow separation of glycosylated MD-2 and the non-glycosylated MD-2 bands on WB made it difficult to analyze the two bands separately; the level of total MD-2 (including both the

107

Chapter 5

glycosylated and non-glycosylated MD-2) of each sample was therefore analyzed. The level of MD-2 (23-25 kDa) in each sample was calculated as described for TLR4. The representative WB of MD-2 is shown in Figure 5.3 A. The levels of MD-2 in HT-29 cells induced by C. concisus strains are shown in Figure 5.3 D.

The levels of MD-2 in HT-29 cells (measured by WB) induced by the 11 C. concisus strains examined were all significantly higher than the non-infected HT-29 cells (t test, P < 0.05); P1CDO2 (1.68±0.18), P1CDO3 (1.63±0.25), P1CDO13 (2.32±0.05), P1CDB1(UNSWCD) (2.35±0.12), P2CDO1 (1.15±0.02), P3UCO1 (2.06±0.1), P3UCB1 (1.90±0.19), P3UCLW1 (2.27±0.03), H1O1 (1.24±0.05), H4O1 (1.56±0.23), and H5O1 (1.97±0.25) (Figure 5.3 D). The average level of MD-2 induced by the eight C. concisus strains isolated from patients with IBD was not statistically different from that induced by the C. concisus strains isolated from healthy controls (1.92±0.15 vs 1.59±0.21, P > 0.05) (Figure 5.4 A).

Using FC, the level of surface MD-2 (non-permeabilized cells) and total MD-2 (permeabilized cells) was calculated as described for TLR4. The representative FC histogram of surface MD-2 and total MD-2 is shown in Figure 5.3C. The levels of surface MD-2and total MD-2 (mean ± SE) in HT-29 cells induced by C. concisus strains are shown in Figure 5.3 D.

The levels of surface MD-2 on HT-29 cells induced by the 11 C. concisus strains examined (measured by FC) were all significantly higher than the non-infected HT-29 cells (t test, P < 0.05); P1CDO2 (2.46±0.32), P1CDO3 (2.96±0.39), P1CDO13 (2.66±0.30), P1CDB1(UNSWCD) (1.90±0.18), P2CDO1 (1.54±0.15), P3UCO1 (1.51±0.04), P3UCB1 (2.29±0.29), P3UCLW1 (1.46±0.03), H1O1 (1.36±0.06), H4O1 (1.32±0.05), and H5O1 (1.40±0.07) (Figure 5.3 D). The average level of surface MD-2 induced by the eight C. concisus strains isolated from patients with IBD was higher than that induced by the three C. concisus strains isolated from healthy controls (2.10±0.20 vs 1.36±0.02), but the comparison of the two groups did not quite reach significance (P = 0.06), Figure 5.4 B.

The levels of total MD-2 on HT-29 cells induced by the 11 C. concisus strains examined (measured by FC) were all significantly higher than the non-infected HT-29 cells (t test, P < 0.05); P1CDO2 (1.87±0.07), P1CDO3 (1.86±0.07), P1CDO13 (1.49±0.09), P1CDB1(UNSWCD) (1.69±0.08), P2CDO1 (1.91±0.10), P3UCO1 (2.00±0.17), P3UCB1 (1.77±0.04), P3UCLW1 (1.61±0.06), H1O1 (1.68±0.10), H4O1(1.40±0.07), and H5O1

108

Chapter 5

(1.76±0.10) (Figure 5.3 D).The average level of total MD-2 induced by the eight C. concisus strains isolated from patients with IBD was not statistically different from that induced by the C. concisus strains isolated from healthy controls (1.78±0.06 vs 1.61±0.11, P > 0.05) (Figure 5.4 C).

In addition to detection by WB and FC, expression of MD-2 in HT-29 cells was visualized using immunofluorescence staining and confocal microscopy. Confocal microscopy image showed an increased expression of MD-2 in HT-29 cells after co-incubation with a representative C. concisus strain (P1CDB1(UNSWCD)) for 24 hours (Figure 5.3B).

109

Figure 5.3 Detection of MD-2 in HT-29 cells infected with C. concisus strains for 24 hours by: A. Western blot (WB): a representative Western blot image revealing two bands, the glycosylated MD-2 and the non-glycosylated MD-2. The narrow distance of glycosylated MD-2 and the non-glycosylated MD-2 made it difficult to analyse the two bands separately, these two protein bands were analysed together. The intensity of MD-2 (23-25 kDa) bands of each sample was normalized to the intensity of the internal control α-Tubulin (55 kDa) of the same sample. The levels of MD-2 of each sample were expressed as the fold change of the normalized band intensity relative to the normalized band intensity of the HT-29 cells without C. concisus (N) and shown in part D; B. Immunofluorescence microscopy: a representative confocal microscope image showing the expression of MD-2 in HT-29 cells with and without C. concisus (P1CDB1(UNSWCD) strain) infection. Detection was by anti-MD-2 antibody followed by Alexa Fluor 594 labelled secondary antibody. Scale Bar=10µm; C. Flow Cytometry (FC): a representative FC histogram showing the expression of surface MD-2 and total MD-2 in HT-29 cells with and without C. concisus infection (grey, N). Detection was by anti-MD-2 antibody followed by Alexa Fluor 594 labelled secondary antibody. Background (purple) was from HT-29 cells that were not exposed to antibodies. The levels of MD-2 were expressed as the fold change of the mean channel fluorescence intensity (MFI) derived from fluorescence histogram of a sample relative to the MFI of the HT-29 cells without C. concisus (N), the fold change is shown in the part D; D. The fold change of MD-2 expression in HT-29 cells induced by C. concisus strains. N: HT-29 cells without C. concisus; P1CDB1 refers to P1CDB1(UNSWCD)

110

Chapter 5

Figure 5.4 Comparison of mean expression (fold change) of MD-2 in HT-29 cells infected with C. concisus strains isolated from patient with IBD and healthy controls A. MD-2 expression (Mean ± SEM) in infected HT-29 by C. concisus strains isolated from patients with IBD and healthy controls (using WB); B. Surface MD-2 expression (Mean ± SEM) on HT-29 infected by C. concisus strains isolated from patients with IBD and healthy controls (using FC); C. Total MD-2 expression (Mean ± SEM) in HT-29 infected by C. concisus strains isolated from patients with IBD and healthy controls (using FC). n.s, not significant; P, value (t test).

5.3.3 Effects of C. concisus on TLR5 expression in HT-29 cells

Expression of TLR5 induced by C. concisus strains in HT-29 cells was assessed by WB and FC.

WB revealed one band. The level of TLR5 in each sample was calculated as described for TLR4. The representative WB with TLR5 bands (111 kDa) is shown in Figure 5.5 A. The levels of TLR5 in HT-29 cells induced by C. concisus strains are shown in Figure 5.5 D.

111

Chapter 5

The levels of TLR5 on HT-29 cells induced by the 11 C. concisus strains examined (measured by WB) were not statistically different from non-infected HT-29 cells (t test, P > 0.05); P1CDO2 (1.10±0.05), P1CDO3 (1.31±0.16), P1CDO13 (1.41±0.21), P1CDB1(UNSWCD) (0.90±0.05), P2CDO1 (0.95±0.05), P3UCO1 (1.21±0.06), P3UCB1 (0.87±0.05), P3UCLW1 (0.77±0.11), H1O1 (1.05±0.09), H4O1(1.07±0.11), and H5O1 (1.23±0.11) (Figure 5.5 D). The average level of TLR5 induced by the eight C. concisus strains isolated from patients with IBD was not statistically different from that induced by the C. concisus strains isolated from the healthy controls (1.07±0.08 vs 1.12±0.06, P > 0.05), Figure 5.6A.

Using FC, the level of surface TLR5 (non-permeabilized cells) and total TLR5 (permeabilized cells) was calculated as described for TLR4. The representative FC histogram of surface TLR5 and total TLR5 is shown in Figure 5.5C. The levels of surface TLR5 and total TLR5 (mean ± SE) in HT-29 cells induced by C. concisus strains are shown in Figure 5.5 D.

The levels of surface TLR5 on HT-29 cells induced by the 11 C. concisus strains examined (measured by FC) were not statistically different from the non-infected HT-29 cells (t test, P > 0.05); P1CDO2 (1.11±0.06), P1CDO3 (0.99±0.03), P1CDO13 (1.01±0.06), P1CDB1(UNSWCD) (0.91±0.12), P2CDO1 (1.14±0.10), P3UCO1 (1.13±0.03), P3UCB1 (0.73±0.09), P3UCLW1 (1.10±0.13), H1O1 (1.02±0.06), H4O1 (1.02±0.05), and H5O1 (1.00±0.03) (Figure 5.5 D). The average level of surface TLR5 induced by the eight C. concisus strains isolated from patients with IBD was not statistically different from that induced by the C. concisus strains isolated from the healthy controls (1.02±0.05 vs 1.01±0.01, P > 0.05), Figure 5.6B.

The levels of total TLR5 on HT-29 cells induced by the 11 C. concisus strains examined (measured by FC) were not statistically different from the non-infected HT-29 cells (t test, P > 0.05); P1CDO2 (1.16±0.11), P1CDO3 (1.07±0.05), P1CDO13 (1.12±0.09), P1CDB1(UNSWCD) (1.08±0.05), P2CDO1 (0.97±0.05), P3UCO1 (1.06±0.05), P3UCB1 (0.85±0.19), P3UCLW1 (1.08±0.0.05), H1O1 (0.97±0.04), H4O1(0.96±0.04), and H5O1 (1.24±0.10) (Figure 5.5 D).The average level of total TLR5 induced by the eight C. concisus strains isolated from patients with IBD was not statistically different from that induced by the C. concisus strains isolated from the healthy controls (1.05±0.03 vs 1.06±0.09, P > 0.05), Figure 5.6 C.

112

Chapter 5

Confocal microscope image showed no apparent change of TLR5 in HT-29 cells after incubation with a representative C. concisus strain (P1CDB1(UNSWCD)) for 24 hours (Figure 5.5.B).

113

Figure 5.5 Detection of TLR5 in HT-29 cells infected with C. concisus strains for 24 hours by: A. Western blot (WB): a representative Western blot image revealing one TLR5 (110 kDa) band. The intensity of TLR5 of each sample was normalized to the intensity of the internal control α-Tubulin (55 kDa) of the same sample. The levels TLR5 of each sample were expressed as the fold change of the normalized band intensity relative to the normalized band intensity of the HT-29 cells without C. concisus (N) and shown in part D; B. Immunofluorescence microscopy: a representative confocal microscope image showing the expression of TLR5 in HT-29 cells with and without C. concisus (P1CDB1(UNSWCD) strain) infection. Detection was by anti-TLR5 antibody followed by Alexa Fluor 488 labelled secondary antibody. Scale Bar=10µm; C. Flow Cytometry (FC): a representative FC histogram showing the expression of surface TLR5 and total TLR5 in HT-29 cells with and without C. concisus infection (grey, N). Detection was by anti-TLR5 antibody followed by Alexa Fluor 488 labelled secondary antibody. Background (purple) was from HT-29 cells that were not exposed to antibodies. The levels of TLR5 were expressed as the fold change of the mean channel fluorescence intensity (MFI) derived from fluorescence histogram of a sample relative to the MFI of the HT-29 cells without C. concisus (N), the fold change is shown in the part D. D. The fold change of TLR5 expression in HT-29 cells induced by C. concisus strains. N: HT-29 cells without C. concisus; P1CDB1 refers to P1CDB1(UNSWCD) D

114

Chapter 5

Figure 5.6 Comparison of mean expression (fold change) of TLR5 in HT-29 cells infected with C. concisus strains isolated from patient with IBD and healthy controls A. TLR5 expression (Mean ± SEM) in infected HT-29 by C. concisus strains from patients with IBD and healthy controls (using WB); B. Surface TLR5 expression (Mean ± SEM) on HT-29 infected by C. concisus strains isolated from patients with IBD and healthy controls (using FC); C. Total TLR5 expression (Mean ± SEM) in HT-29 infected by C. concisus strains isolated from patients with IBD and healthy controls (using FC). n.s, not significant.

5.3.4 Effects of C. concisus on TLR2 expression in HT-29 cells

Expression of TLR2 induced by C. concisus strains in HT-29 cells was assessed by WB and FC.

WB revealed two protein bands, the glycosylated TLR2 (Gly-TLR2) and non-glycosylated TLR2 (Non-GlyTLR2). Gly-TLR2 and Non-GlyTLR2 were analysed separately. The intensity of Gly-TLR2 (90 kDa) and Non-GlyTLR2 (66 kDa) bands were normalized to the

115

Chapter 5 intensity of α-Tubulin (internal control, 55 kDa) of the same sample. The level of Gly-TLR2 and Non-GlyTLR2 in each sample was calculated as described for TLR4. The representative WB of Gly-TLR2 (90 kD) and Non-GlyTLR2 (66 kD) is shown in Figure 5.7 A. The levels of Gly-TLR2 and Non-GlyTLR2 (mean ± SE) in HT-29 cells induced by C. concisus strains are shown in Figure 5.7 D.

The levels of Gly-TLR2 in HT-29 cells induced by the 11 C. concisus strains examined (measured by WB) were not statistically different from the non-infected HT-29 cells (t test, P > 0.05); P1CDO2 (0.87±0.06), P1CDO3 (0.83±0.09), P1CDO13 (1.11±0.09), P1CDB1(UNSWCD) (1.15±0.11), P2CDO1 (1.06±0.23), P3UCO1 (1.14±0.8), P3UCB1 (0.87±0.09), P3UCLW1 (0.74±0.13), H1O1 (1.02±0.17), H4O1 (1.06±0.16), and H5O1 (0.88±0.10) (Figure 5.7 D). The average level of Gly-TLR2 induced by the eight C. concisus strains isolated from patients with IBD was not statistically different from that induced by the C. concisus strains isolated from the healthy controls (0.97±0.06 vs 0.99±0.05, P > 0.05), Figure 5.8 A.

The levels of Non-GlyTLR2 in HT-29 cells induced by the 11 C. concisus strains examined (measured by WB) were not statistically different from the non-infected HT-29 cells (t test, P > 0.05); P1CDO2 (1.11±0.08), P1CDO3 (0.95±0.03), P1CDO13 (1.25±0.13), P1CDB1(UNSWCD) (1.28±0.15), P2CDO1 (0.95±0.06), P3UCO1 (1.30±0.15), P3UCB1 (0.92±0.04), P3UCLW1 (0.91±0.09), H1O1 (0.97±0.02), H4O1(0.85±0.18), and H5O1 (0.93±0.07) (Figure 5.7 D). The average level of Non-GlyTLR2 induced by the eight C. concisus strains isolated from patients with IBD was not statistically different from that induced by the C. concisus strains isolated from the healthy controls (1.09±0.06 vs 0.92±0.04, P > 0.05), Figure 5.8 B.

Using FC, the level of surface TLR2 (non-permeabilized cells) and total TRL2 (permeabilized cells) was calculated as described for TLR4. The representative FC histogram of surface TLR2 and total TLR2 is shown in Figure 5.7 C. The levels of surface TLR2 and total TLR2 (mean ± SE) in HT-29 cells induced by C. concisus strains are shown in Figure 5.7 D.

The levels of surface TRL5 on HT-29 cells induced by the 11 C. concisus strains examined (measured by FC) were not statistically different from the non-infected HT-29 cells (t test, P > 0.05); P1CDO2 (1.11±0.16), P1CDO3 (0.99±0.07), P1CDO13 (1.12±0.16),

116

Chapter 5

P1CDB1(UNSWCD) (1.13±0.14), P2CDO1 (1.19±0.14), P3UCO1 (1.32±0.33), P3UCB1 (1.21±0.11), P3UCLW1 (1.24±0.12), H1O1 (1.20±0.19), H4O1 (1.14±0.11), and H5O1 (1.36±0.13) (Figure 5.7 D). The average level of surface TLR2 induced by the eight C. concisus strains isolated from patients with IBD was not statistically different from that induced by the C. concisus strains isolated from the healthy controls (1.16±0.04 vs 1.23±0.07, P > 0.05), Figure 5.8C.

The levels of total TLR2 on HT-29 cells induced by the 11 C. concisus strains examined (measured by FC) were not statistically different from the non-infected HT-29 cells (t test, P > 0.05); P1CDO2 (0.99±0.02), P1CDO3 (0.87±0.05), P1CDO13 (0.97±0.02), P1CDB1(UNSWCD) (0.94±0.04), P2CDO1 (1.14±0.04), P3UCO1 (1.28±0.24), P3UCB1 (1.01±0.11), P3UCLW1 (1.10±0.05), H1O1 (0.95±0.04), H4O1(1.17±0.04), and H5O1 (1.05±0.06) (Figure 5.7 D).The average level of total TLR2 induced by the eight C. concisus strains isolated from patients with IBD was not statistically different from that induced by the C. concisus strains isolated from the healthy controls (1.04±0.05 vs 1.06±0.06, P > 0.05), Figure 5.8 D.

Confocal microscope image showed no apparent change of TLR2 in HT-29 cells after incubation with a representative C. concisus strain (P1CDB1(UNSWCD)) for 24 hours (Figure 5.7.B).

117

Figure 5.7 Detection of TLR2 in HT-29 cells infected with C. concisus strains for 24 hours by: A. Western blot (WB): a representative Western blot image revealing two bands, the Glycosylated TLR2 (Gly-TLR2, 90 kDa) and non-glycosylated TLR2 (Non- GlyTLR2, 66 kDa). The intensity of Gly-TLR2 and Non-GlyTLR2 bands of each sample was normalized to the intensity of the internal control α-Tubulin (55 kDa) of the same sample. The levels of Gly-TLR2 and Non-GlyTLR2 of each sample were expressed as the fold change of the normalized band intensity relative to the normalized band intensity of the HT-29 cells without C. concisus (N) and shown in part D; B. Immunofluorescence microscopy: a representative confocal microscope image showing the expression of TLR2 in HT-29 cells with and without C. concisus (P1CDB1(UNSWCD)strain) infection. Detection was by anti-TLR2 antibody followed by Alexa Fluor 488 labelled secondary antibody. Scale Bar=10µm; C. Flow Cytometry (FC): a representative FC histogram showing the expression of surface TLR2 and total TLR2 in HT-29 cells with and without C. concisus infection (grey, N). Detection was by anti-TLR2 antibody followed by Alexa Fluor 488 labelled secondary antibody. Background (purple) was from HT-29 cells that were not exposed to antibodies. The levels of TLR2 were expressed as the fold change of the mean channel fluorescence intensity (MFI) derived from fluorescence histogram of a sample relative to the MFI of the HT-29 cells without C. concisus (N), the fold change is shown in the part D; D. The fold change of TLR2 expression in HT-29 induced by C. concisus strains. N: HT-29 cells without C. concisus; P1CDB1 refers to P1CDB1(UNSWCD)

118

Chapter 5

Figure 5.8 Comparison of mean expression (fold change) of TLR2 in HT-29 cells infected with C. concisus strains isolated from patient with IBD and healthy controls A. Gly-TLR2 expression (Mean ± SEM) in infected HT-29 by C. concisus strains isolated from patients with IBD and healthy controls (using WB); B. Non-GlyTLR4 expression (Mean ± SEM) in infected HT-29 by C. concisus strains isolated from patients with IBD and healthy controls (using WB); C. Surface TLR4 expression (Mean ± SEM) on HT-29 infected by C. concisus strains isolated from patients with IBD and healthy controls (using FC); D. Total TLR4 expression (Mean ± SEM) in HT-29 infected by C. concisus strains isolated from patients with IBD and healthy controls. n.s, not significant.

5.3.5 The effects of oral and enteric C. concisus strains isolated from individual patients with IBD on expression of TLR4 and MD-2 in HT-29 cells

In order to assess whether oral and enteric C. concisus strains obtained from individual patients with IBD exert the same biological effects on intestinal epithelial cells, the levels of TLR4 and MD-2, induced by enteric C. concisus strains obtained from two patients with IBD

119

Chapter 5

(P1CD and P3UC) were compared to the levels induced by the oral C. concisus strains isolated from these two patients. The levels of TLR4 and MD-2 induced by enteric C. concisus strains of these two patients (P1CDB1(UNSWCD) vs P1CDO2, P1CDO3, PACDO13 and P3UCB1, P3UCLW1 vs P3UCO1) were not significantly different from that induced by the patient’s own oral C. concisus strains (Figures 5.1D and 5.3D).

5.4 Discussion This study aimed to examine if C. concisus has the potential to initiate the development of human IBD by examining the immune modulatory effects of C. consisus using a human intestinal cell line (HT-29 cells).

Responses of the gut immune system to enteric bacterial infections begin with the recognition of bacterial components by innate immune receptors including TLRs expressed on the intestinal epithelial cells 453. Recognition of bacterial components by TLRs leads to the development of acute inflammation, a process required to eradicate enteric pathogens. In addition to defending the host against enteric pathogens, an additional challenge that the gut immune system has to face is the trillions commensal bacteria lining the surface of the intestinal epithelial cells. Given that the bacterial components recognized by TLRs are present not only on pathogenic bacteria, but also on commensal bacterial species, the gut immune system has evolved strategies to maintain its tolerance or low response to the intestinal commensal bacterial species. Examples of such strategies include low level expression of TLR2 and TLR4 in intestinal epithelial cells and expression of TLR5 at the basolateral rather than the apical surface of the enterocytes under normal physiological conditions 92,446,454.

TLR4 recognizes lipopolysacchride (LPS) found in Gram-negative bacteria. In patients with IBD, increased intestinal epithelial expression of TLR4 has been detected 92. It was found that Salmonella enterica serovar Typhimurium (S. Typhimurium) and different Lactobacillus strains did not stimulate TLR4 production at the mRNA level in HT-29 455. In our study, we found that all C. concisus strains significantly (P < 0.05) upregulated the expression of TLR4 in HT-29 cells using two methods the WB and FC. Using WB, the levels of TLR4 induced by different C. concisus strains varied greatly. The levels of Gly-TLR4 and Non-GlyTLR4 induced by different C. concisus strains arranged from 1.60 to 8.50 and 1.27 to 6.25 respectively, reflecting strain differences (Figure 5.1A and D). Using FC, The levels of TLR4 induced by different C. concisus strains also varied greatly. The levels of Surface TLR4 and

120

Chapter 5

Total TLR4 induced by different C. concisus strains ranged from 1.84 to 5.40 and 1.51 to 2.12 respectively, reflecting strain differences (Figure 5.1C and D). Interestingly, we found that the average level of Gly-TLR4 (measured by WB) and surface TLR4 (measured by FC) induced by the eight C. concisus strains isolated from patients with IBD was significantly higher than that induced by the three C. concisus strains isolated from healthy controls (Figure5.2 A and C).

Human TLR4 contains 9 N-linked glycosylation sites and glycosylations are important for the transportation of TLR4 to cell surface and for TLR4 to function as the LPS receptor [34]. A similar phenomenon was previously observed in Helicobacer pylori. In a study examining the effects of H. pylori LC11 and LC20 strains on expression of TLR4 in MKN45 gastric epithelial cells, Su et al found that the H. pylori LC11 strain upregulated the expression of Gly-TLR4 at a much greater level than that induced by the LC20 strain 456. The difference between the LC11 and LC20 strains is that LC11 contains the pathogenicity island 457. It is possible that a common virulence factor(s) shared by pathogenic H. pylori and C. concisus strains may have contributed to the upregulation of Gly-TLR4. Our finding that C. concisus upregulated intestinal epithelial expression of TLR4, particularly the Gly-TLR4, suggests that some C. concisus strains (which were mostly isolated from patients with IBD) have the potential to enhance the responses of the gut immune system to Gram-negative bacteria including both pathogenic and commensal bacterial species.

MD-2 is a co-receptor that is associated with the extracellular domain of TLR4 432. Previous studies have shown that MD-2 plays an essential role in TLR4 glycosylation and cell surface distribution, which enables TLR4 to function as the LPS receptor 458,459. Intestinal epithelial cells express low level of MD-2 in normal physiological conditions 441. Increased intestinal epithelial expression and serum MD-2 activity in patients with IBD have been reported 425,460. In our study, we found that all C. concisus strains significantly (P < 0.05) upregulated the expression of MD-2 (measured by WB), surface and total MD-2 (measured by FC) in HT-29 cells (Figure 5.3).The average level of MD-2 (measured by WB), total MD-2 and surface MD-2 (measured by FC) induced by the eight C. concisus strains isolated from patients with IBD was higher than that induced by the three C. concisus strains isolated from healthy controls. Although the difference was not significant (P > 0.05), it was found that the difference in the case of the expression of surface MD-2 between the two groups did not quite reach significance (P = 0.06, 2.10±0.20 vs 1.36±0.02), Figure 5.4 B. Our finding that C.

121

Chapter 5 concisus upregulated MD-2 expression in HT-29 cells further supports the possible role of C. concisus in augmenting inflammatory responses.

TLR-5 recognizes a conserved site on bacterial flagellin 461. It has been found that the flagellin of Salmonella typhimurim is the major epithelial proinflammatory determinant, activating intestinal epithelial expression of TLR5 435,462. Furthermore, it was found by using FC that enteropathogenic Escherichia coli, a bacterium that causes diarrhoeal disease, upregulated the expression of surface TLR5 and total TRL5 in HT-29 cells 463. However, the flagellin of members of ε Proteobacteria, which include genera of Helicobacter, Campylobacter and Wollinella, is able to evade the recognition by TLR-5, owing to changed amino acids at the TLR5 recognition site 436. Indeed, in our study C. concisus strains showed no significant (P > 0.05) effects on TLR5 expression in HT-29 cells (Figure 5.5). Previously Man et al showed that C. concisus attached to Caco2 cells using thier flagellum 315. The evasion of TLR5 would allow C. concisus to attach to intestinal epithelial cells using the flagellum without being noticed by the innate immune system which makes it possible for this bacterium to modulate the innate gut immune system as discussed above. In patients with CD, bacterial flagellin has been identified as the dominant antigen 119. However, the intestinal epithelial expression of TLR5 in these CD patients did not change as compared with the controls 92. These findings suggest that flagellin specific antibodies detected in patients with Crohn’s disease were most likely induced by bacterial species whose flagellin has evaded the detection of TLR5; thus members of ε Proteobacteria representing good candidates. Indeed, a high prevalence of some Helicobacter species and C. concisus has been detected in patients with IBD 464-467.

TLR2 responds to lipoproteins and components of Gram-positive bacteria, such as peptidoglycan. In IBD, there have been conflicting reports of the expression of TLR2 on IECs. Although it was reported that TLR2 expression on IEC of patients with IBD remained unchanged when compared healthy controls 92, another study reported a higher TLR2 expression on IECs of patients with IBD 426. Using HT-29 cells, it was found that S. Typhimurium did not stimulate TLR2 production at the mRNA level 455. In our study C. concisus strains showed no significant (P > 0.05) effects on TLR2 expression in HT-29 cells (Figure 5.7).

122

Chapter 5

In summary, this is the first study to examine the effect of C. concisus on the expression of TLR2, TLR4, TLR5 and MD-2 on the human IECs (in vitro). We have provided the first evidence that C. concisus has the potential to modulate the gut innate immune system. We speculate that the impact of C. concisus on human health is determined by the host’s genetic makeup, C. concisus strain specificity, local defence mechanisms and the surrounding microbes. The upregulation of epithelial expression of TLR4 (especially the surface and Gly- TLR4) and its co-receptor MD-2 (especially the surface MD-2) by C. concisus in human intestine, may enhance the responses of the gut immune system to Gram-negative commensal bacterial species and lead to the breakdown in tolerance of the gut immune system to intestinal commensal bacteria, which in genetically predisposed individuals may result in chronic intestinal inflammations such as IBD. These speculations require further investigation.

123

Chapter 6

Chapter 6: Study of C. concisus proinflammatory properties using human intestinal epithelial cell line HT-29 and Caco-2

6.1 Introduction

In healthy individuals, the gut immune system preserves the balance between the proinflammatory and the anti-inflammatory mediators, this harmony between the mediators prevents the gut immune system from attacking the intestinal commensal bacteria while promoting a strong and effective defence against enteric pathogens. In inflammatory bowel disease (IBD) an imbalance between the proinflammatory and the anti-inflammatory mediators causes the gut immune system to attack the intestinal commensal bacteria which leads to an uncontrolled inflammation of the intestines that cannot be down-regulated1,468

In chapter 5 we presented for the first time the effect (in vitro) of C. concisus strains in modulating the gut innate immune system. It was found that TLR4 and its co-receptor MD-2 were significantly upregulated in the human intestinal epithelial cells (IECs) by C. concisus strains while there was no effect on the expression of TLR5 and TLR2.

TLR4 together with its co-receptor MD-2 (binds to the extracellular domain of TLR-4) recognizes LPS of Gram-negative bacteria. TLR4 is activated when LPS is transferred to CD14, LPS/CD14 will stimulate TLR4/MD-2 on the cell surface432,469. TLR4 activation will induce the production of pro-inflammatory cytokines such as interleukin (IL)-8 and the proinflammatory mediator cyclooxygenase-2 (COX-2)470,471. TLR4 has two downstream pathways the MyD88-dependent pathway 472 and the TRIF-related adaptor molecule (TRAM) pathway 473. Following is an overview of these two pathways which are also presented in Figure 6.1:

I. MyD88-dependent pathway: MyD88 is an immediate downstream adaptor molecule that interacts directly with the Toll/IL-1R resistance (TIR) domain of TLR4, MyD88 recruits interleukin-1 receptor-associated kinase (IRAK) and phosphatidylinositol 3- kinase (PI3K)474: a. Activated IRAK recruits tumor necrosis factor receptor-associated factor-6 (TRAF6), activated TRAF6 will then requite NIK and mitogen activated protein kinases (MAPKs) pathway 474,475.

124

Chapter 6

NIK activation leads subsequently to the degradation of inhibitor κB (IκB) which causes the activation of NFκB, activated NFκB translocate to the nucleus where it transcribe genes responsible of proinflammatory cytokines468 such IL-8 476,477 and the proinflammatory mediator COX-2 478,479. MAPKs pathway leads to the activation of p38MAPK which regulates COX-2 at transcription and mRNA stability levels 480,481and cytokine production such IL-8 482,483. p38MAPK is involved in regulating several transcription factors including ATF2, STAT1 and could activate NFκB 483, it was found that p38 MAPK pathway leads to the production of IL-8 possibly through NFκB activation 484.

b. Activated PI3K leads to the recruitment of AKT485, AKT will then activate its downstream signalling that eventually leads to the p65 phosphorylation, resulting in enhanced NFκB transactivation 486. NFκB activation leads to the expression of proinflammatory cytokines and COX-2.482,483

II. TRAM pathway: TRAM is recruited to the TIR domain of TLR4 487. Activating TRAM will activate TRIF which leads to the TANK-binding kinase-1 (TBK1) activation 487,488. TBK1 activates the transcription factor interferon-regulated factor 3 (IRF3), activated IRF3 form a complex with NFκB subunit p65 489. p65/IRF3 complex could function in two ways depending on the nature of signalling, either IRF3 functions as an essential co-activator of p65 for the transcription of NFκB- dependent genes or p65 functions as a cofactor of IRF3 for the transcription of IRF- dependent genes (from a review by 490, references used in the review 489,491,492) . IRF3 dependent genes are responsible for promoting antibacterial and antiviral innate immunity, genes such as IFNB, IFNA4, IFIT1, CXCL9, CXCL10 and CCL5 (reviewed by 490). It was found that p65/IRF3 complex cooperate with the glucocorticoid receptor (GR) to synergistically transrepress distinct subsets of TLR- responsive genes 489.

125

Chapter 6

Figure 6.1 TLR4 downstream signalling pathway induced by LPS leads to the expression of target genes, including IL-8 and COX2.

NFκB has a principal role as a transcription factor in chronic inflammatory diseases 493494. It regulates the transcription of an exceptionally large number of genes, particularly those involved in immune and inflammatory responses 495 such as interleukin (IL)-8, IL-1β, IL-6 and cyclooxygenase-2 (COX-2) genes 470,471. NFκB is a heterogeneous collection of dimers, with p65:p50 being the most abundant dimer in most cell types 495. In resting cells, NFκB dimers are found in the inactive form associating with regulatory proteins called inhibitors of nuclear factor κB (IκB) such as IκBα. When epithelial cells are activated by a potent activator, a cascade of events leading to the activation of inhibitor κB kinases (IKKs) is induced. Activation of IKKs phosphorylates IκB and leads to its degradation by the 26S proteasome (within minutes), thereby releasing NFκB dimers (which also will be phosphorylated by IKKs) from the cytoplasmic NFκB–IκB complex. The released NFκB dimers translocate to the nucleus where they activate the transcription of target genes (Figure 6.2) 471,495,496.

126

Chapter 6

Figure 6.2 Recognising bacterial PAMPs by PRRs (such TLRs) causes activation of the transcription factor NFκB through phosphorylation and then degradation of the NFκB inhibitor IκB (such as IκBα). Activated NFκB translocates into the nucleus to regulate transcription of an exceptionally large number of genes, especially those responsible for immune and inflammatory responses (see text).

Schreiber et al found an increased activation of NFκB in lamina propria biopsy specimens taken from patients with CD and UC compared to healthy controls 497. Further, Rogler et al found that not only was NFκB found to be over expressed in mucosal macrophages and epithelial cells in patients with IBD, but it was also in a state of activation 498. Rogler et al also found that NFκB activation correlated with the degree of mucosal inflammation 498. As illustrated previously NFκB activation leads to the expression of proinflammatory cytokines such as IL-8476,477 and the inflammatory mediator COX-2.482,483

COX-2 is the key enzyme responsible for the synthesis of prostaglandins (PGs) associated with the mediation of inflammation 499. COX-2 is expressed at very low levels in most tissues

127

Chapter 6 such as the gastrointestinal tissues 500, and is frequently detected in tissues involved in the inflammatory process 501. In patients with IBD, Singer et al found that COX-2 was induced in intestinal apical epithelial cells of inflamed areas 502. COX-2 was also suggested to play an important role in colorectal cancer, as it was found that 80–90% of colorectal adenocarcinomas and 40–50% of premalignant adenomas had an overexpression of COX- 2503. On the other hand, the other Cyclooxygenase enzyme COX-1 is constitutively expressed in most tissue and is important for its homeostatic function 504. COX-2 is induced by proinflammatory cytokines such as interleukin (IL)-1β505, phorbol esters 506, environmental stress such as hyperosmolarity 507 invasive bacteria such as Salmonella 508and Helicobacter pylori 509, and LPS 480,481. The COX-2 signalling pathway may incorporate toll-like receptors and their adapter proteins 510, p38 480 and extracellular-regulated kinase (ERK) 511 which are members of the mitogen- activated protein kinases (MAPKs) family 512, and the transcription factor NFκB 478,479. IL-8 was one of the earliest chemokine to be identified, sequenced and characterized 513-515. IL-8 is a novel mediator of inflammation 516. It was found that nanomolar concentration of IL-8 is chemotactic and activates neutrophils, monocytes and memory T-cells and inhibits hematopoietic stem cell proliferation 515,517,518. IL-8 can stimulate neutrophils to undergo a respiratory burst and release superoxide anions 519. IL-8 protein 520 and mRNA521 levels were found to be elevated in the intestinal mucosal biopsies taken from the inflamed mucosa of patients with CD and UC. Mazzucchelli et al found that cells expressing IL-8 in patients with UC were diffusely distributed over the entire affected mucosa while it was shown a focal distribution pattern in the case of patients with CD 522. Mazzucchelli et al consistently failed to detect IL-8 messenger RNA in the mucosa of uninvolved bowel segments and in normal-appearing control mucosa of patients with colon cancer. Mazzucchelli et al also found that epithelial cells, macrophages and neutrophils are the sources of this cytokine in active IBD. Mazzucchelli et al concluded that IL-8 plays an important but nonspecific role in the pathogenesis of IBD 522. Mitsuyama et al found that colonic IL-8 levels significantly correlate with the macroscopic grade of local inflammation, especially in the case of patients with UC, and that colonic IL-8 levels also correlated well with the neutrophil numbers in mucosal tissue. Mitsuyama et al concluded that IL-8 mediates neutrophil infiltration of the gut wall in IBD 520.

128

Chapter 6

Chemokines (such as IL-8) are involved in activating and regulating the process of selective adhesion, cell activation, and migration of leukocytes 513,523. In the inflamed intestines, intestinal epithelial cells will produce chemokines that begins the chemoattractant process in neighbouring capillaries by causing shape changes to granulocytes and macrophages called diapedesis 515. Chemokines also cause the capillary endothelial cells to upregulate the expression of selectin and integrin 524. Selectin and integrin causes the circulating granulocytes and macrophages to attach, roll, adhere, and consequently migrate between the endothelial cells and into the intestinal mucosa 523, beginning the inflammatory response. In patients with IBD, it was found that the human colonic chemokine expression is non- selectively up-regulated in IBD but the up-regulation of chemokine expression (including IL- 8) was shown to correlate with the increasing activity of IBD 525. In this chapter, we investigate (in vitro) the proinflammatory properties of C. concisus strains on human IECs and the signalling pathway involved. To achieve this purpose, we investigated the activation of NFκB, the expression of the proinflammatory mediator COX-2 and the secretion of the chemokine IL-8 by the human IEC line HT-29 after exposure to different strains of C. concisus bacteria.

6.2 Materials and methods

6.2.1 C. concisus strains and cultivation conditions

The 11 oral (isolated from saliva) and enteric (isolated from intestinal biopsies and faeces) C. concisus strains which were used in chapter 5 are used in this chapter. All C. concisus strains were grown as shown in chapter 5.

6.2.2 Cultivation of HT-29 cells Human intestinal epithelial cell line HT-29 cells (ATCC No. HTB-38), were maintained in McCoy’s 5A medium (Invitrogen, California, USA) supplemented with 10% heat inactivated foetal bovine serum (FBS) (Bovogen Biologicals, Melbourne, Australia), 100U/ml penicillin and 100 µg/ml streptomycin (Invitrogen,). The cells were grown at 37°C in a humidified incubator containing 5% CO2.

6.2.3 Cultivation of Caco-2 cells Caco-2 cells were cultivated as shown in chapter 5.

129

Chapter 6

6.2.4 Western blot Western blot (WB) assays were performed in triplicate and repeated at least twice

WB was used to examine the:

a. Effect of different C. concisus strains on the expression of COX-2 by HT-29 cells. b. Effect of C. concisus infection on IκBα expression by HT-29. c. Relationship between the dose of C. concisus and the expression COX-2, TLR4 and IκBα by the HT-29 cells.

6.2.4.1 Antibodies used for Western blotting All antibodies used for WB were purchased from Santa Cruz Biotechnology Inc (Santa Cruz biotechnology Inc, California, USA). Primary antibodies used were polyclonal anti-TLR4 (sc-10741), anti-COX-2 (sc-1746), anti-IκB-α (sc-371) and anti-α tubulin (sc-31779). Secondary antibodies conjugated with horseradish peroxidase (HRP) were bovine anti-goat IgG (sc-2352), and goat anti-rabbit IgG (sc-2054).

6.2.4.2 Infection of HT-29 cells with C. concisus

HT-29 cells were infected as shown in chapter 5, except for the following differences:

a. For detection of COX-2, HT-29 cells were infected with C. concisus at a multiplicity of infection (MOI) of 25 and further incubated for 24 hours. b. For detection of IκBα, HT-29 cells were infected with C. concisus at an MOI of 100 and further incubated for 10, 15, 20, 30 and 60 minutes (min) respectively. c. To examine the relationship between the dose of C. concisus and the expression of TLR4 COX-2, and IκBα in HT-29 cells, HT-29 cells were infected with C. concisus at five different MOIs including MOI 5, 12.5, 25, 50, and 100 for 24 hours.

HT-29 cells without infection of C. concisus were used as the negative control.

6.2.4.3 Preparation of HT-29 cells whole cell proteins

As shown in chapter 5.

130

Chapter 6

6.2.4.4 Western blot

As shown in chapter 5. Except that the anti-COX-2 was diluted at 1:250 and incubated overnight at 4 0C, and the anti- IκBα was diluted at 1:1000 and incubated for 90 minutes at room temperature.

6.2.5 Immunofluorescence staining and confocal microscopy

Immunofluorescence staining and confocal microscopy was used to visualise the:

a. Different expressions of COX-2 by HT-29 cells before and after C. concisus infection. b. Effect of C. concisus infection on IκBα expression by HT-29 at different time intervals.

Immunofluorescence experiments were repeated at least three times.

6.2.5.1 Antibodies used for immunofluorescence staining

Primary antibodies used were polyclonal anti-COX-2 (sc-1746), and anti-IκB-α (sc-371). Secondary antibodies used were Alexa Fluor® 488 donkey anti-goat IgG (A11055), and Alexa Fluor® 594 goat anti-rabbit IgG (A11037) (Invitrogen).

6.2.5.2 Infection of HT-29 cells with C. concisus

HT-29 cells were infected as shown in chapter 5, except for the following differences:

a. For detection of COX-2, HT-29 cells were infected with C. concisus at an MOI of 25 and further incubated for 24 hours. b. For detection of IκBα, HT-29 cells were infected with C. concisus at an MOI of 100 and further incubated for 10, 15, 20, 30 and 60 minutes (min) respectively.

HT-29 cells without infection of C. concisus were used as the negative control.

6.2.5.3 Immunostaining and visualization of COX-2 and IκBα by confocal microscopy

As shown in chapter 5.

131

Chapter 6

6.2.6 Flow cytometry

Flow cytometry (FC) was used to examine the expression of COX-2 in response to different C. concisus strains infection in HT-29 cells. FC experiments were in triplicate and repeated at least twice.

6.2.6.1 Antibodies used for immunofluorescence staining

Primary antibody used was polyclonal anti-COX-2 (Santa Cruz). Secondary antibody used was Alexa Fluor® 488 donkey anti-goat IgG (A11055) (Invitrogen).

6.2.6.2 Infection of HT-29 cells with C. concisus

HT-29 cells were infected as shown in chapter 5,

6.2.6.3 Flow cytometry As shown in chapter 5. Except that all cells were permeabilized for 10 minutes with 0.1% Triton X-100 in PBS.

6.2.7 Measurement of IL-8 in HT-29 cell culture supernatant by ELISA

HT-29 epithelial cells were seeded at an initial concentration of 5x105 cell/ml in 6-well cell culture plates. The cells were grown for 48 hours to form a monolayer.

To examine the production of IL-8 by HT-29 cells induced by different C. concisus strains, HT-29 monolayer were infected with C. concisus strains at an MOI of 100 and further incubated for 24 hours.

To examine the relationship between the dose of C. concisus and the production of IL-8 in HT-29 cells, HT-29 cells were infected with three C. concisus strains (two oral and one biopsy C. concisus strains isolated from a patient with IBD) at four different MOIs including MOI 5, 25, 50, 100 and 200 for 24 hours.

Supernatants were collected and centrifuged twice for 2 minutes at 10,000g. IL-8 secreted in the supernatants was measured using Human IL-8 CytoSet™ (Invitrogen) according to manufacturer’s instructions.

Supernatant collected from HT-29 cells incubated with S. typhimurium and supernatant from HT-29 cells without bacteria were used as the positive and negative control respectively.

132

Chapter 6

Experiments were in triplicate and repeated at least twice.

6.2.8 The effect of TNF and IFN-γ on adhesion and invasion ability of C. concisus strains to Human intestinal epithelial cells

Three different C. concisus strains were used in this section: P1CDO2, P1CDO3, P1CDB1(UNSWCD).

5 × 105 Caco-2 cells (human intestinal epithelial) were seeded onto 24-well plates (Nunc). Plates were pre-coated with 0.05 mg/ml of rat tail collagen type 1 (BD, New Jersey, USA) for o 45 minutes at room temperature. Cells were then incubated at 37 C and 5% CO2 for 4 days to form a monolayer.

The monolayer was washed 4 times using phosphate buffer saline (PBS) and Caco-2 culture media containing no antibiotics but treated with the designated concentration (20,40 or 80ng/ml) of TNF (Sigma, Missouri, USA) or IFN-γ (Sigma), which was added for 1 hour. Cells without treatment with TNF or IFN-γ were used as the untreated controls. C. concisus strains were then added to the Caco-2 monolayer at an MOI of 200 (with the presence of the cytokines). Cells were then centrifuged at 320 RCF for 5 minutes to promote the adherence of the bacteria to the surface of the epithelial cells. Infected monolayers were 0 then incubated at 37 C and 5% CO2 for 2 hours.

Eight wells of Caco-2 cells were infected with each C. concisus strain. Two wells were the untreated controls and the other six were the monolayer cells treated with either TNF or IFN- γ at the three different concentrations, two wells for each cytokine concentration. Following the 2 hour incubation, the wells were washed 5 times with PBS. The Caco-2 monolayer of four wells was lysed with 1% Triton X-100 (Gibco, California, USA) for 5 minutes. Serial dilutions of cell lysates were inoculated onto HBA plates and incubated at 37 0C for 48 hours in a microaerobic condition. Colony-forming units (CFU) of C. concisus were recorded, which were regarded as the numbers of C. concisus that were associated with the Caco-2 cells (the sum of adhering bacteria and invading bacteria).

The remaining four wells of Caco-2 cells infected with C. concisus were incubated with 1 ml of MEM containing 200 µg/ml of gentamicin to kill the extracellular bacteria. Following washing of the Caco-2 monolayer for 5 times using PBS, the cells were lysed with 1% Triton X-100 for 5 minutes. Serial dilutions of cell lysates were inoculated onto HBA plates and

133

Chapter 6 incubated for two days. CFU of C. concisus were recorded, which were regarded as the numbers of C. concisus that had invaded (internalized) Caco-2 cells.

Prior to cell lysis, the number of viable extracellular bacteria was determined by culturing the supernatant on HBA. This was to insure that gentamicin had killed all extracellular bacteria which had not invaded the Caco-2 cells.

C. concisus adhesion index was calculated as follow:

The invasion index was calculated using the formula described by Larson et al in 2008 350, which is as follow:

Results were reported as the fold change of the adhesion index or the invasion index of the bacteria in the presence of TNF or IFN-γ relative to the adhesion index or the invasion index of the bacteria without the presence of TNF or IFN-γ (untreated control). Experiments were repeated three times.

6.2.9 Statistical analysis Data were analysed by means of unpaired t test using GraphPad Prism version 5.1 (San Diego, CA). P-values < 0.05 (two tailed, 95% confidence interval) were considered significant.

6.3 Results 6.3.1 Effects of C. concisus on COX-2 expression in HT-29 cells

Expression of COX-2 induced by different C. concisus strains in HT-29 cells was assessed by WB and FC.

The representative WB of COX-2 (70 kDa) is shown in Figure 6.3A. The intensity of COX-2 bands was normalized to the intensity of α-Tubulin (internal control, 55 kDa) of the same sample. The levels of COX-2 in each sample were expressed as the fold change of the normalized band intensity relative to the normalized band intensity of the non-infected HT-29

134

Chapter 6 cells (HT-29 cells without C. concisus). The levels of COX-2 (mean ± SE) in HT-29 cells induced by C. concisus strains are shown in Figure 6.3 D.

The levels of COX-2 induced by the 11 C. concisus strains examined (measured by WB) were all significantly higher than the non-infected HT-29 cells (P < 0.05, t test); P1CDO2 (3.36±0.68), P1CDO3 (2.71±0.02), P1CDO13 (4.56±0.60), P1CDB1(UNSWCD) (5.30±1.02), P2CDO1 (1.49±0.10), P3UCO1 (4.24±1.00), P3UCB1 (2.49±0.45), P3UCLW1 (2.28±0.42), H1O1 (1.21±0.07), H4O1 (1.52±0.05), H5O1 (1.85±0.04). The average level of COX-2 induced by the eight IBD strains was significantly higher than that induced by the five C. concisus strains isolated from the healthy controls (3.30±0.46 vs 1.53±0.18, P < 0.05) (Figure 6.4A).

Using FC, the level of total COX-2 (permeabilized cells) was expressed as the fold change of the mean channel fluorescence intensity (MFI) derived from fluorescence histogram of a sample relative to the MFI of the non-infected HT-29 cells (HT-29 cells without C. concisus). The representative FC histogram of total COX-2 is shown in Figure 6.3C. The levels of total COX-2 (mean ± SE) in HT-29 cells induced by C. concisus strains are shown in Figure 6.3 D.

The levels of total COX-2 induced by the 11 C. concisus strains examined (measured by FC) were all significantly higher than the non-infected HT-29 cells (P < 0.05, t test); P1CDO2 (1.43±0.08), P1CDO3 (1.38±0.08), P1CDO13 (1.38±0.09), P1CDB1(UNSWCD) (1.53±0.04), P2CDO1 (1.33±0.02), P3UCO1 (1.58±0.13), P3UCB1 (1.33±0.08), P3UCLW1 (1.44±0.06), H1O1 (1.45±0.03), H4O1 (1.35±0.07), and H5O1 (1.27±0.05) (Figure 6.3 D). The average level of total COX-2 induced by the eight C. concisus strains isolated from patients with IBD was higher but not statistically different from that induced by the C. concisus strains isolated from healthy controls (1.43±0.03 vs 1.36±0.05, P > 0.05) (Figure 6.4 B)

In addition to detection by WB and FC, expression of COX-2 in HT-29 cells was visualized using immunofluorescence staining and confocal microscopy. Confocal microscopy image showed an increased expression of COX-2 in HT-29 cells after co-incubation with a representative C. concisus strain (P1CDB1(UNSWCD)) for 24 hours (Figure 6.3B).

135

Figure 6.3 Detection of COX-2 in HT-29 cells infected with C. concisus strains for 24 hours by: A. Western blot (WB): a representative WB image revealing one COX-2 (70 kDa) band. The intensity of COX-2 of each sample was normalized to the intensity of the internal control α-Tubulin (55 kDa) of the same sample. The levels COX-2 of each sample were expressed as the fold change of the normalized band intensity relative to the normalized band intensity of the HT-29 cells without C. concisus (N) and shown in part D; B. Immunofluorescence microscopy: a representative confocal microscope image showing the expression of COX-2 in HT- 29 cells with and without C. concisus (P1CDB1(UNSWCD) strain) infection. Detection was by polyclonal anti-COX-2 antibody followed by Alexa Fluor 488 labelled secondary antibody. Scale Bar=10µm; C. Flow Cytometry (FC): a representative FC histogram showing the expression of total COX-2 in HT-29 cells (permeabilized cells) with and without C. concisus infection (grey, N). Detection was by polyclonal anti-COX-2 antibody followed by Alexa Fluor 488 labelled secondary antibody. Background (purple) was from HT-29 cells that were not exposed to antibodies. The levels of COX-2 were expressed as the fold change of the mean channel fluorescence intensity (MFI) derived from fluorescence histogram of a sample relative to the MFI of the HT-29 cells without C. concisus (N), the fold change is shown in the part D. D. The fold change of COX-2 expression in HT-29 cells induced by C. concisus strains. N: HT-29 cells without C. concisus; P1CDB1 refers to P1CDB1(UNSWCD).

136

Chapter 6

A. B.

Figure 6.4 Comparison of mean expression of COX-2 in infected HT-29 cells between C. concisus strains isolated from patient with IBD and C. concisus strains isolated from healthy controls. A. COX-2 levels (Mean ± SEM) in infected HT-29 by C. concisus strains isolated from patients with IBD and healthy controls (using Western blot); B. COX-2 expression (Mean ± SEM) in infected HT-29 by C. concisus strains isolated from patients with IBD and healthy controls (using FC). n.s, not significant; * = Significant results (P <0.05 (unpaired t test)).

The relationship between the dose of C. concisus and the level of COX-2 in HT-29 cells was investigated by WB. HT-29 cells were infected with a representative C. concisus strains (P1CDB1(UNSWCD)) at five different MOIs (5, 12.5, 25, 50, and 100). The intensity of COX-2 bands was normalized to the intensity of α-Tubulin of the same sample. The levels of COX-2 in each sample were expressed as the fold change of the normalized band intensity relative to the normalized band intensity of the non-infected HT-29 cells (HT-29 cells without C. concisus). The representative WB of COX-2 is shown in Figure 6.5A.

The levels of COX-2 induced by the five different MOIs examined were all significantly higher than the non-infected HT-29 cells (P < 0.05, t test); MOI 5 (5.18±0.57), MOI 12.5 (8.32±0.71), MOI 25 (8.91±0.50), MOI 50 (6.44±0.33) and MOI 100 (3.04±0.10) (Figure 6.5B). The data were the average of triplicate experiments ± standard error. Experiments were repeated at least twice. The highest level of COX-2 was expressed at MOI 25 while the lowest expression was at MOI 100.

137

Chapter 6

Figure 6.5 Levels of COX-2 in HT-29 induced by the C. concisus strain P1CDB1(UNSWCD) at five different MOIs (5, 12.5, 25, 50, 100). A. Representative Western blot of COX-2 (70kD), The intensity of COX-2 of each sample was normalized to the intensity of the internal control α-Tubulin (55 kDa) of the same sample. The levels of COX-2 of each sample were expressed as the fold change of the normalized band intensity relative to the normalized band intensity of the HT-29 cells without C. concisus (N).; B. Levels of COX-2 induced by the C. concisus strain P1CDB1(UNSWCD): data were the average of triplicate experiments ± standard error. The highest induction level was at MOI 25, all the induction levels by the five different MOI were significantly higher than the non-infected HT-29 cells (P < 0.05, t test). N: HT-29 cells without C. concisus infection.

6.3.2 Effects of C. concisus on IL-8 production in HT-29 cells

The concentrations of IL-8 in HT-29 cell culture supernatants were determined by enzyme linked immunosorbent assay (ELISA). The value of natural production of IL-8 by HT-29 cells (HT-29 cells without C. concisus infection) was subtracted from the concentration of IL- 8 in each sample (HT-29 cells incubated with C. concisus strains) and the results are shown in Figure 6.6 A (Data were the average of triplicate experiments ± standard error. Experiments were repeated at least twice).

The concentrations of IL-8 (pg/ml) induced by the 11 C. concisus strains at an MOI of 100 were all significantly higher than the non-infected HT-29 cells (P < 0.05, t test); P1CDO2(133±31), P1CDO3(276±20), P1CDO13(274±9), P1CDB1(UNSWCD)(171±19), P2CDO1(144±16), P3UCO1 (230±11), P3UCB1 (234±26), P3UCLW1(344±14), H1O1(238±23), H4O1(229±43), H5O1(288±30). The concentration of IL-8 induced by the positive control Salmonella typhimurium was 903±130. The average concentration of IL-8

138

Chapter 6 induced by C. concisus strains from patients with IBD was not significantly different from that induced by C. concisus strains from healthy controls (226±26 vs 251±18, P > 0.05).

The relationship between the dose of C. concisus and the production of IL-8 in HT-29 cells was assessed by measurement of IL-8 concentrations in cell culture supernatants of HT-29 cells infected with three representative C. concisus strains (P1CDO2, P1CDO3 and P1CDB1(UNSWCD)) of five different MOIs (5, 12.5, 25, 50, 100 and 200). Concentrations of IL-8 induced by P1CDO2 C. concisus strain at the five MOIs were 187±26, 217±20, 227±12, 276±5 and 482±45 respectively. Concentrations of IL-8 induced by P1CDO3 C. concisus strain at the five MOIs were 201±29, 222±16, 220±5, 274±8 and 454±16 respectively. Concentrations of IL-8 induced by P1CDB1(UNSWCD) C. concisus strain at the five MOIs were 143±13, 158±10, 172±13, 171±19 and 293±25 respectively. The results are shown in Figure 6.6 B. No statistical differences were found between the levels of IL-8 at the MOIs 5, 25, 50 and 100 induced by P1CDO2, P1CDO3 or P1CDB1(UNSWCD) (P > 0.05, t test). The induction of IL-8 at MOI200 was significantly increased compared to the other MOIs (5, 25, 50 and 100) induced by P1CDO2, P1CDO3 and P1CDB1(UNSWCD) (P < 0.05, t test)

139

Chapter 6

Figure 6.6 Production of IL-8 by HT-29 cells induced by C. concisus strains. A. HT-29 cells were incubated with C. concisus strains at an MOI of 100 for 24 hours. Concentrations of IL-8 in the cell culture supernatants were measured by ELISA. The basal production of IL-8 (HT-29 cells without C. concisus) has been subtracted from values shown in A. Data were the average of triplicates ± standard error. The level of IL-8 induced by C. concisus strains from patients with IBD was not statistically different from that induced by C. concisus strains from the healthy controls (P > 0.05); B. HT-29 cells were incubated with three representative C. concisus strains (P1CDO2- blue, P1CDO3-red and P1CDB-green) at four different MOIs (5, 25, 50, and 100) for 24 hours. Concentrations of IL-8 in the cell culture supernatants were measured by ELISA. The basal production of IL-8 (HT-29 cells without C. concisus) has been subtracted from values shown in B. Data were the average of triplicates ± standard error. Note: P1CDB1 in the Figure refers to P1CDB1(UNSWCD).

140

Chapter 6

6.3.3 Effects of C. concisus on activation of NFκB in HT-29 cells

Given that all C. concisus isolates induced the production of IL-8, the effect of C. concisus on activation of NFκB was assessed by examination of the degradation of IκBα in HT-29 cells following incubation with a representative C. concisus strain (P1CDB1(UNSWCD)).

Expression of IκBα in HT-29 cells was detected by WB and the level of IκBα in each sample was calculated as described for COX-2. The representative WB of IκBα (35 kD) is shown in Figure 6.7A. The levels of IκBα in HT-29 cells following incubation with C. concisus strain P1CDB1(UNSWCD) for different time points relative to the 0 min (HT-29 cells without C. concisus collected at 0 time) were significantly lower in 10 min (0.31±0.09), 15 min (0.59±0.09), and 20 min (0.56±0.05) (P < 0.05, t test). While there was no significant difference between 30 min (1.26±0.19), and 60 min (1.20±0.08) compared to the negative control (P > 0.05, t test). Results are shown in Figure 6.7 B. The data shown are the average of triplicate experiments ± standard error.

In addition to detection by WB, the levels of IκBα in HT-29 cells was visualized using immunofluorescence staining and confocal microscopy. Confocal microscopy image showed a decreased expression of IκBα at 10 min and 20 min compared to the 0 min (HT-29 cells without C. concisus collected at 0 min). While at 30 min and 60 min a similar expression to the non-infected HT-29 cells (0 min) was observed.

141

Chapter 6

Figure 6.7 Effects of C. concisus on degradation of IκBα in HT-29 cells. HT-29 cells were incubated with C. concisus strain P1CDB1(UNSWCD) at an MOI of 100 for different time points. HT-29 cells were then lysed and expression of IκBα was detected by Western blot (WB). The intensity of IκBα band of each sample was normalized to the intensity of the internal control α-Tubulin of the same sample. The level of IκBα was expressed as the fold change of the normalized band intensity of a sample relative to the normalized band intensity of the negative control (HT-29 cells without C. concisus). A decreased level of IκBα indicates activation of NFκB. A: Representative WB of IκBα (35 kDa) and α-Tubulin (55 kDa); B: Levels of IκBα in HT-29 cells following incubation with a representative C. concisus strain (P1CDB1(UNSWCD)) for different time points. Data shown are the average of triplicate experiments ± standard error; C. Immunofluorescence microscopy: a representative confocal microscope image showing the IκBα level in HT-29 cells following incubation with a representative C. concisus strain (P1CDB1(UNSWCD)) for different time points. Detection was by polyclonal anti- IκBα antibody followed by Alexa Fluor 594 labelled secondary antibody. Scale Bar=10µm.

The relationship between the dose of C. concisus and the activation of NFκB was assessed by examination of the degradation of IκBα in HT-29 cells following incubation with a

142

Chapter 6 representative C. concisus strain (P1CDB1(UNSWCD)) at five different MOIs (5, 12.5, 25, 50, and 100). IκBα in HT-29 cells was assessed by WB, and the level of IκBα in each sample was calculated as described for COX-2. The representative WB of IκBα is shown in Figure 6.8 A.

The levels of IκBα induced by the five different MOIs examined were all significantly lower than the non-infected HT-29 cells (P < 0.05, t test); MOI 5 (0.71±0.05), MOI 12.5 (0.64±0.06), MOI 25 (0.66±0.05), MOI 50 (0.77±0.06) and MOI 100 (0.68±0.06) (Figure 6.8 B). The data were the average of triplicate experiments ± standard error, experiments were repeated at least twice. The levels of IκBα induced by C. concisus (P1CDB1(UNSWCD)) between the different MOI concentrations were not statistically different (P > 0.05, t test).

143

Chapter 6

Figure 6.8 Levels of IκBα in HT-29 induced by C. concisus strain P1CDB1(UNSWCD) at five different MOIs (5, 12.5, 25, 50, 100) after 15 min of infection. A. Representative Western blot (WB) of IκBα (35 kDa). HT-29 cells were infected with C. concisus strain (P1CDB1(UNSWCD)) at various concentrations (5, 12.5, 25, 50 and 100 MOI) for 15 min, the level of IκBα in HT-29 cells were measured by WB. The intensity of IκBα (35 kDa) of each sample was normalized to the intensity of the internal control α-Tubulin (55 kDa) of the same sample. The levels of IκBα of each sample were expressed as the fold change of the normalized band intensity relative to the normalized band intensity of the HT-29 cells without C. concisus (N) ; B. Levels of IκBα induced by the C. concisus strain P1CDB1(UNSWCD): data were the average of triplicate experiments ± standard error. All the levels of IκBα by the five different MOI were significantly lower than the non-infected HT-29 cells (N) (P < 0.05, t test), No significant differences were found in the level of IκBα between the different MOI (P >0.05, t test).

6.3.4 Effects of C. concisus concentration on the expression of TLR4 in HT-29 cells

HT-29 cells were infected with a representative C. concisus strain (P1CDB1(UNSWCD)) at five different MOIs (5, 12.5, 25, 50, and 100), and the relationship between the dose of C. concisus and the expression TLR4 in HT-29 cells was detected by WB.

144

Chapter 6

WB revealed two protein bands, the glycosylated TLR4 (Gly-TLR4) and non-glycosylated TLR4 (Non-GlyTLR4). Gly-TLR4 and Non-GlyTLR4 were analysed separately. The expression of Gly-TLR4 and Non-GlyTLR4 in HT-29 cells was detected by WB, and the level of Gly-TLR4 and Non-GlyTLR4in each sample was calculated as described for COX-2. The representative WB of Gly-TLR4 and Non-GlyTLR4 is shown in Figure 6.9 A

The levels of Gly-TLR4 induced by the five different MOIs examined were all significantly higher than the non-infected HT-29 cells (P < 0.05, t test); MOI 5 (6.89±1.08), MOI 12.5 (7.96±0.74), MOI 25 (5.40±0.19), MOI 50 (4.77±0.30) and MOI 100 (2.67±0.28) (Figure 6.9 B). The data shown are the average of triplicate experiments ± standard error. Experiments were repeated at least twice.

The levels of Non-GlyTLR4 induced by the five different MOIs examined were all significantly higher than the non-infected HT-29 cells (P < 0.05, t test) except for the MOI 100 which was not significantly different from the non-infected HT-29 cells (P > 0.05, t test); MOI 5 (2.49±0.13), MOI 12.5 (2.87±0.30), MOI 25 (2.51±0.21), MOI 50 (1.57±0.15) and MOI 100 (0.91±0.03) (Figure 6.9 B). The data shown are the average of triplicate experiments ± standard error, experiments were repeated at least twice. The highest level of Gly-TLR4 and Non-GlyTLR4 was expressed at MOI 12.5 while the lowest expression was at MOI 100.

145

Chapter 6

Figure 6.9 Levels of Gly-TLR4 and Non-GlyTLR4 in HT-29 induced by the C. concisus strain P1CDB1(UNSWCD) at five different MOIs (5, 12.5, 25, 50, 100). A. Representative Western blot of Gly-TLR4 (120 kDa) and Non-GlyTLR4 (95 kDa), the intensity of Gly-TLR4 and Non-GlyTLR4 of each sample was normalized to the intensity of the internal control α-Tubulin (55 kDa) of the same sample. The levels Gly- TLR4 and Non-GlyTLR4 of each sample were expressed as the fold change of the normalized band intensity relative to the normalized band intensity of the HT-29 cells without C. concisus (N).; B. Levels of Gly-TLR4 and Non-GlyTLR4 induced by the C. concisus strain P1CDB1(UNSWCD): data were the average of triplicate experiments ± standard error. The highest induction level was at MOI 12.5. All the induction levels of the Gly-TLR4 and Non-GlyTLR4 by the five different MOI were significantly higher than the non-infected HT-29 cells (P < 0.05, t test), except for the Non-GlyTLR4 at MOI 100 which was not significantly different from the non-infected HT-29 cells (P > 0.05, t test).

6.3.5 The effects of oral and enteric C. concisus strains isolated from individual patients with IBD on expression of COX-2 and IL-8 in HT-29 cells

In order to assess whether oral and enteric C. concisus strains obtained from individual patients with IBD exert the same biological effects on intestinal epithelial cells, the levels of COX-2 and IL-8, induced by enteric C. concisus strains obtained from two patients with IBD

146

Chapter 6

(P1CD and P3UC) were compared to the levels induced by the oral C. concisus strains isolated from these two patients. The levels of COX-2 and IL-8 induced by enteric C. concisus strains of these two patients (P1CDB1(UNSWCD) vs P1CDO2, P1CDO3, PACDO13 and P3UCB1, P3UCLW1 vs P3UCO1) were not significantly different from that induced by the patient’s own oral C. concisus strains (Figures 6.3D and Figure 6.6A).

6.3.6 The effect of pre-existing inflammation on Campylobacter concisus adhering and invading HT-29 cells To investigate the effect of gastrointestinal inflammation (presented by the two cytokines TNF and IFN-γ) on the adhesion and invasion abilities of C. concisus, Caco-2 human intestinal epithelial cell monolayers were treated with TNF and IFN-γ at three different doses, 20ng/ml, 40ng/ml and 80ng/ml respectively, 1 hour prior to infection by the C. concisus strain. TNF and IFN-γ were maintained through the 2 hours of the adhesion and invasion assay.

The ability of C. concisus strains (P1CDO2, P1CDO3, and P1CDB1(UNSWCD)) to adhere and invade the human intestinal epithelial cells Caco-2 treated with TNF and IFN-γ at the three different concentrations were all increased. Results are shown in Table 6.1 and 6.2.

147

Chapter 6

Table 6.1 The effect of TNF at three different concentrations (20, 40, and 80 ng/ml) on the adhesion and invasion ability of C. concisus strains (P1CDO2, P1CDO3, and P1CDB1(UNSWCD)) to the human intestinal epithelial cells Caco-2.

20 ng/ml 40 ng/ml 80 ng/ml

Adhesion Invasion Adhesion Invasion Adhesion Invasion

P1CDO2 1.21±0.09 1.41±0.14 1.24±0.06 1.25±0.04 1.50±0.03 1.27±0.06

P1CDO3 1.50±0.11 1.21±0.10 1.73±0.13 1.29±0.16 1.68±0.23 1.23±0.29

P1CDB1(UNSWCD) 1.73±0.19 1.67±0.22 1.21±0.05 1.23±0.07 1.20±0.23 1.38±0.20

Results are reported as the fold change of the adhesion index or the invasion index of the bacteria in the presence of TNF relative to the adhesion index or the invasion index of the bacteria without the presence of TNF. Experiments were repeated three times. Results are the mean of three independent experiments ± SE.

All the adhesion and invasion abilities of the bacteria to the Caco-2 cells treated with TNF showed significant increase (P < 0.05, t test) when compared to the adhesion and invasion abilities of the bacteria to the Caco-2 cells without the presence of TNF (untreated control).

148

Chapter 6

Table 6.2 The effect of IFN-γ at three different concentrations (20, 40, and 80 ng/ml) on the adhesion and invasion ability of C. concisus strains (P1CDO2, P1CDO3, and P1CDB1(UNSWCD) to the human intestinal epithelial cells Caco-2.

20 ng/ml 40 ng/ml 80 ng/ml

Adhesion Invasion Adhesion Invasion Adhesion Invasion

P1CDO2 1.50±0.16 1.40±0.26 1.80±0.28 2.21±0.25 1.95 2.63±0.23

P1CDO3 1.05±0.03* 1.33±0.08 1.08±0.01* 2.17±0.22 1.54±0.11 1.50±0.14

P1CDB1(UNSWCD) 1.45±0.36 1.29±0.20 1.93±0.31 1.41±0.14 2.50±0.24 1.65±0.18

Results are reported as the fold change of the adhesion index or the invasion index of the bacteria in the presence of IFN-γ relative to the adhesion index or the invasion index of the bacteria without the presence of IFN-γ. Experiments were repeated three times. Results are the mean of three independent experiments ± SE.

All the adhesion and invasion abilities of the bacteria to the Caco-2 cells treated with IFN-γ showed significant increase (P < 0.05, t test) when compared to the adhesion and invasion abilities of the bacteria to the Caco-2 cells without the presence of IFN-γ (untreated control). Except for the two cases shown by *.

149

Chapter 6

6.4 Discussion

In chapter 5 it was found that C. concisus has the ability to modulate the gut immune system, it was found that C. concisus strains could upregulate the expression of TLR4 and MD-2. In this chapter we investigate the proinflammatory properties of C. concisus strains on human IECs and the signalling pathway involved.

COX-2 is an inducible enzyme responsible for producing prostaglandins and other important inflammatory mediators under the regulation of the TLR4 pathway 526,527. Singer et al showed that COX-2 was not detected in normal colonic epithelial cells but was induced in patients with IBD 502. COX-2 has also been shown to be associated with enteric pathogen induced intestinal epithelial high fluid secretion 528,529. In addition to its involvement in inflammation, COX-2 has been linked to several malignancies including colorectal cancer 530. In this study, we found by using the two methods WB and FC that all C. concisus strains used in this study have significantly (P < 0.05) induced a higher level of COX-2 in the human intestinal cell line HT-29 as compared to the non-infected HT-29 cells (Figure 6.3). Interestingly, It was found that C. concisus strains isolated from patients with IBD induced a higher level of COX-2 in HT-29 cells as compared with C. concisus strains isolated from healthy controls (Figure 6.4), which was significant in the case of WB (P < 0.05), suggesting a possible role for C. concisus in the pathogenesis of IBD. Given that IBD is a risk factor for colorectal cancer, the possible contribution of C. concisus to the increased incidence of colorectal cancer in patients with IBD should be further investigated 531.

In our study, all C. concisus strains significantly induced the production of IL-8, consistent with previous observations [48,61]. The level of IL-8 induced by C. concisus strains from patients with IBD was not statistically different from that induced by C. concisus strains from healthy controls (Figure 6.6). When comparing this finding to the COX-2 finding which showed significant increase induction of COX-2 by C. concisus strains from patients with IBD compared to controls suggest that; IL-8 and COX-2 may have one or more downstream signalling pathway that differs from each other and/or C. concisus strains may have different mechanism by which they induce IL-8 and COX-2 production from intestinal epithelial cells.

In addition, we found that the levels of IL-8 induced by C. concisus strains were significantly (P < 0.05) lower than that induced by the enteric pathogen S. typhimurium of the same MOI; S. typhimurium induced almost 3 times more IL-8 compared to C. concisus strains (Figure

150

Chapter 6

6.6). Stimulating a low level of IL-8 may be related to the infectious ability of C. concisus, IL-8 is an important chemokine that could activate the immune system by its having a chemotactic and cellular activation activities for neutrophils, monocytes and memory T-cells 515,518. Leukocyte migration to the sight of an infection causes acute inflammation which would affect the C. concisus colonization and reproduction in the gastrointestinal tract. So the reduction in IL-8 production plays in favour of C. concisus survival by enabling their infection and colonizing without being detected by the host defence system. The flagellin evasion of TLR5 (as shown in chapter 5) and the LPS composition and structure may have at least in part contributed to the low production of IL-8 induced by C. concisus strains.

NFκB plays a key role in inflammation 471. Activation of NFκB leads to transcription of proinflammatory cytokines including IL-8 532 and the proinflammatory mediator COX-2 478,479. IκBα is a molecule regulating the activation of NFκB and a decreased level (degradation) of IκBα indicates an activation of NFκB 471. In our study, a decreased level of IκBα in HT-29 cells following incubation with a representative C. concisus strain (P1CDB1(UNSWCD)) was observed (Figure 6.7). The increased level of IκBα after the initial reduction may be due to the auto feedback regulation of IκBα on NFκB activity 533 (Figure 6.7). Kaakoush et al found by using protein functional analysis that C. concisus infection regulates processes related to NFκB activation348. In this study we confirm, using WB and Immunofluorescence staining, that C. concisus proinflammatory property involves the NFκB signaling pathway.

Further examination of the relationship between C. concisus dose and the production of IL-8 and NFκB activation (presented by the degradation of IκBα at 15 min) in HT-29 cells showed that at lower MOIs (MOI 5-100), an increase in bacterial dose did not affect the production of IL-8 (no significant difference (P > 0.05, t test) by the HT-29 between the MOI 5, 12.5, 25, 50, and 100 was found), presented in Figure 6.6, neither the activation of NFκB; MOI 5-100 showed similar fold reduction (no significant difference in the degradation of IκBα between the MOI 5, 12.5, 25, 50, and 100 was found), presented in Figure 6.8 . Interestingly, upregulation of Gly-TLR4 and Non-glyTLR4 in HT-29 cells by C. concisus strains does not require a high dose of C. concisus; as shown the strain P1CDB1(UNSWCD) upregulated Non-glyTLR4 and more importantly the Gly-TLR4 at MOI of 5. These results suggest that when C. concisus begins colonizing the IECs (low bacterial count), the bacteria adapted its

151

Chapter 6 self not to be detected by the host immune defence system, but at the same time is able to modulate the gut immune system.

At high bacterial concentration represented by the MOI 200, C. concisus dose dependent production of IL-8 in HT-29 cells was significantly increased (P < 0.05) compared to the lower MOIs (MOI5-100) (Figure 6.6), which suggests that C. concisus would need to multiply to high numbers to directly induce more severe intestinal inflammation. The induction of the inflammatory response at the high bacterial concentration will play in favour of C. concisus colonizing at that stage (the high bacterial count stage), as we show in this study that there is significantly increase (P < 0.05) in the adhesion and invasion abilities of C. concisus strains to treated human intestinal epithelial cells Caco-2 with the cytokines TNF and IFN-γ when compared to the adhesion and invasion abilities of C. concisus strains to the untreated human intestinal epithelial cells Caco-2 with the cytokines TNF and IFN-γ (untreated control cells).

The mechanisms by which the cytokines TNF and IFN-γ increases susceptibility to C. concisus adherence and invasion to human intestinal epithelial cells remain to be elucidated Man et al in 2010 studied the effect of TNF and IFN-γ at a concentration of 40 ng/ml on the invasion ability of C. concisus strain (UNSWCD) on Caco-2 cells and found that both cytokines significantly increased the invasion ability of C. concisus strain (UNSWCD), which further supports our findings315.

As illustrated previously, an increase in C. concisus dose from MOI 5 to MOI 100 did not affect the production of IL-8 which does not correlate with the COX-2 expressions in response to C. concisus dose infection from MOI 5 to MOI 100 which showed different fold expression rate having the highest expression at MOI 12.5 while decreasing expression at the MOI100. This finding suggests that COX-2 could have one more different downstream signalling pathway that differs from IL-8 production.

Oral and intestinal C. concisus strains isolated from the same patients with IBD showed a similar pattern in regulating the expression of TLR4 (chapter 5), MD-2 (chapter 5), COX-2 and IL-8 in HT-29 cells (P1CDO2, P1CDO3 and P1CDO13 vs P1CDB1(UNSWCD); P3UCO2 vs P3UCB2 and P3UCLW1). As compared to the oral cavity, human intestine is less favorable for C. concisus growth; as indicated by the 75% positive isolation rate from saliva and only 2.8% from fecal samples of healthy controls 328,336. C. concisus is fastidious and requires an hydrogen enriched microaerobic atmosphere for growth 349. In the human

152

Chapter 6 intestine, hydrogen is produced through fermentation of unabsorbed carbohydrates by anaerobes, with 99% of the hydrogen being produced in colon 534. The amount of hydrogen produced in the intestine is affected by the type of food ingested and the intestinal bacterial composition 534-536. Our recent study examining the genetic relationship of oral and intestinal C. concisus strains showed that the C. concisus strains colonizing intestinal tissues of patients with IBD are closely related to the patients’ own oral C. concisus strains, suggesting that C. concisus strains found in a particular patient’s intestine originate in the oral cavity 351. C. concisus is a bacterium with a great diversity 328,329,345,346,537. It is possible that some specific oral C. concisus strains have a better ability to survive the intestinal environment and are more likely to be involved in enteric diseases. Very recently, Nielsen et al showed that oral C. concisus increased intestinal epithelial leakage 361.

This study demonstrates a direct proinflammatory effect of C. concisus strains from patients with IBD and healthy controls on human intestinal epithelial cells (HT-29), by showing the bacterial activation of nuclear factor kappa B (NFκB), the expression of the proinflammatory mediator cyclooxygenase-2 (COX-2) and the secretion of the chemokine interleukin (IL)-8. These finding supported finding of chapter 5 in which we proposed that C. concisus could modulate the gut innate immune system, concludes that C. concisus could act as atrigger of chronic intestinal inflammation such as IBD.

153

Chapter 7

Chapter 7: General discussion and future directions

7.1 General discussion

Inflammatory Bowel Disease (IBD) is a chronic inflammatory disease of the gastrointestinal tract (GIT) 1, the main symptoms of abdominal pain, rectal bleeding and severe diarrhoea 2. Patients with IBD are at higher risk of developing colorectal cancer (CRC) which accounts for 15% of total deaths in IBD patients 4. IBD covers a group of disorders, with two major types: Crohn’s disease (CD) and ulcerative colitis (UC) 91.

To date, the aetiology of IBD is unclear but it is widely accepted that IBD is an outcome of the interaction of environmental factors, dysregulated immunological response, genetic susceptibility and bacterial factors.

Many studies have shown that IBD arises as a result of a host response to bacterial factors 211,213-215,217-219. In patients with IBD, T-cell 220 and serological 117 responses to enteric bacteria were displayed.

However, what triggered the gut immune system to attack the intestinal commensal bacteria (a cohort of organisms that the gut immune system has co-evolved and lived peacefully with) remains an enigma.

C. concisus is gaining interest as an emerging enteric pathogen that may be involved in human IBD, following recent studies that reported the association between C. concisus and IBD 321,322,328,329,338-340. C. concisus is a Gram-negative rods, with small curved cells that could be helical or curved in shape, and motile by means of a single polar flagellum 316,306. C. concisus colonizes the human oral cavity 306. To date, the role of C. concisus in the pathogenesis of IBD is still unknown. Therefore, the main focus of this PhD study was to investigate the potential role that C. concisus plays in the pathogenesis of IBD.

Studies carried out in this PhD project were described in chapters 3 to 6.

Studies in Chapter 3 included: isolation of C. concisus from the oral cavity of patients with IBD and from healthy control individuals; identification and subgrouping of C. concisus strains by whole protein profile; and examination of the adhesion and invasion properties of the C. concisus strains isolated. It was found for the first time that patients with IBD have

154

Chapter 7

significantly higher prevalence of multiple (≥2) oral C. concisus strains colonizing their oral cavity as compared to controls, and that oral C. concisus strains have the ability to adhere to and to invade the intestinal epithelial cells (IECs) in vitro. Interestingly, it was found that the mean invasion index of oral C. concisus strains isolated from patients with CD was significantly higher than that of the controls and that the mean adhesion index of oral C. concisus strains isolated from patients with CD was also higher than that of the controls. These findings suggest that patients with CD are colonized with specific oral C. concisus strains that have the potential to cause enteric disease when colonizing the intestinal tract. Furthermore, it was found that obtaining the whole protein profile is a rapid, easy and cost effective method of identifying C. concisus strains. It was also found that C. concisus strains have heterogeneous whole protein profiles, and that using this method was not suitable for discriminating oral C. concisus strains isolated from patients with IBD from oral C. concisus strains isolated from healthy individuals.

Studies in Chapter 4 examined the lipopolysaccharide (LPS) extracted from C. concisus strains. It was found for the first time that all of the C. concisus strains have a smooth-form LPS profile. It was also found that subgrouping C. concisus strains according to their LPS profile was more successful than using whole protein profiles. Out of the 52 oral strains used in this study, 33 different LPS profiles were recorded and 25 strains were assigned into 6 different groups (group A-F, LPS patterns 1-6). The largest two groups were Group A and B which contained 8 (32% of the grouped strains) and 9 (36% of the grouped strains) strains respectively; 33% of the strains isolated were either in group A (pattern 1) or group B (pattern 2). Unfortunately, the LPS profile was unable to discriminate between oral C. concisus strains isolated from patients with IBD from that isolated from healthy individuals. Furthermore, it was suggested that oral C. concisus could increase the production or changes the composition of core/lipid A of the LPS when introduced to the enteric environment.

Chapter 5 investigated if C. concisus strains have the potential to modulate the gut innate immune system by examining the effect of C. concisus on HT-29 cells (human IEC line) expression of TLR2, TLR4, TLR5 and MD-2. It was found for the first time that C. concisus strains significantly increased the expression of TLR4 and MD-2 but not TLR5 or TLR2. Interestingly, it was found that the average level of Gly-TLR4 [detected by Western blot(WB)], surface TLR4 [detected by flow cytometry (FC)] and surface MD-2 (detected by FC) induced by C. concisus strains isolated from patients with IBD was higher than that induced by C. concisus strains isolated from healthy controls. The increased level was

155

Chapter 7

significant in the case of Gly-TLR4 and surface TLR4, while it showed a borderline result (P = 0.06) in the case of surface MD-2.

Chapter 6 investigated the proinflammatory properties of C. concisus strains on human IECs HT-29. It was found that C. concisus strains stimulated the HT-29 cells to secrete the chemokine IL-8, express the inflammatory modulator COX-2, and activate the nuclear factor kappa B (NFκB) even at low bacterial concentration shown by multiple of infection (MOI) of 5. Interestingly, TLR4 (especially Gly-TLR4) was also significantly (P < 0.05) upregulated by the HT-29 at a low C. concisus dose (MOI 5). Furthermore, the research showed that there was no significant difference in the production of IL-8 nor the activation of NFκB by the HT- 29 induced by C. concisus between concentrations from MOI 5 up to MOI 100. However, at high C. concisus concentration (MOI 200) IL-8 production was significantly increased compared to the lower bacterial concentrations (MOI5-100). Finally this research showed that the cytokines TNF and IFN-γ increased the adhesion and invasion ability of C. concisus strains to Caco-2 cells (human IEC line).

Based on these findings we proposed a model as to how C. concisus may trigger the onset of IBD by the following steps:

1. Oral C. concisus undergoes gene recombination in the oral cavity; the recombination lead to the creation of a more virulent C. concisus strain that could adhere and invade the IECs. This hypothesis is supported by the following findings:

a. Significantly higher number of patients with IBD were colonized with multiple (≥2) oral C. concisus strains as compared to the healthy controls: the presence of multiple oral strains increases the chance of lateral (horizontal) gene transfer between these oral C. concisus strains. A Master thesis from our lab by Mahendran V., showed that by using MLST that the biopsy strain P1CDB1(UNSWCD) is a recombinant of the oral strains P1CDO2 and P1CDO3, and that P1CDO2 is more closely related to the biopsy strain P1CDB1(UNSWCD) 344. Adding to that, the significantly higher presence of multiple strains in patients with IBD suggests that C. concisus strains colonizing the oral cavity of patients with IBD are more capable of natural recombination as compared to the C. concisus strains colonizing healthy individuals.

156

Chapter 7

b. It was found by Kaakoush et al in 2011 that the enteric C. concisus strains that are highly invasive carry a certain plasmid 348. This finding further supports the hypothesis that lateral (horizontal) gene transfer between multiple (≥2) oral C. concisus strains could be via through transferral of the plasmid between oral C. concisus.

c. Results of the adhesion and invasion assay showed that the C. concisus biopsy strain P1CDB1(UNSWCD) has both high adhesion and invasion abilities, while P1CDO2 strain has a high adhesion capability but a very low invasion capability. On the other hand, P1CDO3 has a low adhesion capability but a very high invasion capability. Knowing that P1CDB1(UNSWCD) is a recombinant of the oral strains P1CDO2 and P1CDO3 and that P1CDO2 is more closely related to the biopsy strain P1CDB1(UNSWCD), it is possible that the gene responsible for invasion could be transferred through lateral (horizontal) gene transfer from the donor P1CDO3 to the recipient P1CDO2. The plasmid could play a role in the lateral gene transfer.

2. The new recombinant oral C. concisus strains colonize intestinal tract and penetrate the mucus layer that separate the luminal bacteria from the IECs. This is supported by the spiral shape and small size of C. concisus; the fact that GIT is a suitable environment in which C. concisus can grow (C. concisus is a microaerobic bacterium); and the isolation of C. concisus from intestinal biopsy samples taken from the inflamed area of patients with IBD 321,322.

3. C. concisus strains adhere, and some may invade, the IECs: it has been shown in this study that oral C. concisus strains have the ability to adhere and invade the human IECs. Interestingly, the mean invasion index of oral C. concisus strains isolated from patients with CD was significantly higher than that of the healthy controls; and the number of patients with CD who have an oral C. concisus strain with higher invasion index than 0.39 (the highest invasion index for C. concisus strain isolated from healthy controls) is significantly higher than the healthy control (this finding supports the hypothesis that patients with IBD are colonized with specific oral C. concisus strains which may have the potential to cause intestinal inflammation if colonizing the intestinal tract).

157

Chapter 7

It was found that adhesion 354-356 and invasion 357-360 plays an important role in the Campylobacter jejuni pathogenesis in the GIT.

4. At the first stages of intestinal colonization, C. concisus begins modulating the gut innate immune system by upregulating the expression of TLR4 and its co-receptor MD-2 without being detected by the host immune defence system. This favours the colonization process of the recombinant C. concisus. This hypothesis is supported by the following findings:

a. It was found that C. concisus strains were able to significantly upregulate the expression of TLR4 [especially the Gly-TLR4 (WB) and surface TLR4 (FC)] and its co-receptor MD-2 [especially the surface MD-2 (FC)] but not the expression of TLR2 and TLR5 at the HT-29 cells. Interestingly, the average level of Gly-TLR4 (measured by WB), surface TLR4 (measured by FC) and surface TLR4 (measured by FC) induced by C. concisus strains isolated from patients with IBD were significantly higher than that induced by the C. concisus strains isolated from healthy controls. Adding to that, the average level surface MD-2 (measured by FC) induced by the C. concisus strains isolated from patients with IBD was higher than that induced by the C. concisus strains isolated from healthy controls (statistical analysis showed a borderline result P=0.06). This finding suggest that C. concisus strains have the potential to enhance the responses of the gut immune system to Gram- negative bacteria, including both pathogenic and commensal bacterial species. This finding further supports the previous hypothesis that patients with IBD are colonized with specific oral C. concisus strains that have the potential to cause intestinal inflammation if colonizing the intestinal tract.

b. It was found that C. concisus could induce the production of IL-8 by HT-29 cells, but the production was significantly lower as compared to the production in response to the pathogenic enteric bacteria Salmonella typhimurium. S. typhimurium induced the production of IL-8 almost three-fold more than C. concisus at the same MOI. It was also found that C. concisus activated the nuclear factor-κB (NFκB).

c. It was found that the production of IL-8 and NFκB activation in the HT-29 cells at low bacterial concentrations presented by the MOI 5 up to MOI 100

158

Chapter 7

did not affect the production of IL-8 nor the activation of NFκB. Interestingly, we have also shown that the upregulation of Gly-TLR4 and Non-glyTLR4 in HT-29 cells by C. concisus strains does not require a high dose of C. concisus; MOI5 was able to be significantly upregulated by Non-glyTLR4 and more importantly by the Gly-TLR4.

5. Triggering the development of chronic intestinal inflammation such as IBD by the intestinal colonizing C. concisus, chronic inflammation could be trigger by one of the two following possibilities (individually or combined):

I. When the C. concisus colonizing intestinal tract reaches high bacterial concentration level, the host immune defence system will be activated causing acute inflammation. Acute inflammation will damage the mucosal barrier resulting in increased severity of the inflammation. In genetically predisposed individuals chronic inflammation will then result as an outcome of the high expression of surface TLR4 and its co-receptor MD-2 on the surface of IECs, in which the Gram-negative intestinal commensal bacterial species will be in direct contact after the mucus layer has been destroyed.

It was shown in chapter 6 that the high bacterial concentration represented by the MOI 200, C. concisus dose dependent production of IL-8 in HT-29 cells was significantly increased (P > 0.05) compared to the lower MOIs (MOI5- 100).

II. The other possibility is that at low bacterial concentration levels, when C. concisus begins to colonize the IECs, the expression of the surface and Gly- TLR4 and its co-receptor MD-2 by the IECs will be highly upregulated (promoted by C. concisus, as explained in step 4, above). This upregulation on the surface of IECs, a site where Gram-negative bacteria are the dominant indigenous bacterial species, may enhance the responses of the gut immune system to Gram-negative intestinal commensal bacterial species, causing the breakdown in tolerance of the gut immune system to the intestinal commensal bacteria. This breakdown in the tolerance will lead to chronic intestinal inflammation in genetically predisposed individuals. It was reported in Chapter 6 that C. concisus could upregulate the expression of Gly-TLR4 on

159

Chapter 7

HT-29 cells up to seven-fold at MOI 5 and eight-fold at MOI 12.5 compared to non-infected IECs.

At the chronic intestinal inflammation stage, the increase in the production of cytokines will increase the C. concisus adhesion and invasion abilities to the inflamed IECs. This increase in the adhesion and invasion abilities will further increase the severity of the chronic intestinal inflammation. It was found that TNF and IFN-γ have significantly increased the adhesion and invasion abilities of C. concisus strains to human IECs (chapter 6).

In patients with IBD, it seems that the high cytokine levels produced in the intestinal inflamed area are causing C. concisus to increased adhesion and invasion the IECs of the inflamed area. This observation could explain the higher detection of C. concisus in the biopsy samples taken from patients with IBD compared to healthy controls 321,339, and the isolation of C. concisus from the intestinal biopsies taken from the inflamed area of patients with IBD 321,322 .

7.2 Future Directions

More work is needed in the following areas:

1. Further investigation of the horizontal (lateral) gene transfer between the C. concisus strains in the oral cavity, and the role of the plasmid reported by Kaakoush et al 348 in the horizontal (lateral) gene transfer.

2. Investigation of the mechanisms by which C. concisus penetrates the mucus layer that separates the intestinal commensal bacteria from the IECs.

3. Further study of the chemical composition of C. concisus LPS, and investigating the hypothesis that oral C. concisus could increase the production of or change in the composition of core/lipid A when introduced to the enteric environment. This could explain the different TLR-4 and MD-2 expressions induced by the different C. concisus strains.

4. Further investigation of the downstream singling pathways by the IECs in response to C. concisus. We have reported that the IECs have different patterns in the production of IL-8 and COX-2 in response to C. concisus dose. It was suggested

160

Chapter 7

that the reason for this difference is that there could be some differences in the downstream singling pathways.

5. Further study of the mechanisms by which some C. concisus strains selectively upregulate the glycosylation of TLR4 and MD-2. This will further explain the finding that C. concisus isolated from patients with IBD selectively upregulate the glycosylated TLR4 and the surface MD-2 more efficiently than C. concisus strains isolated from healthy controls.

6. Study of the antibody serum levels to C. concisus strains in patients with IBD and healthy controls. One suggestion is to use LPS extracted from C. concisus strains isolated from intestinal biopsies as the target antigen in the experiments. Preliminary studies using protein as an antigen showed high cross reactions (data not shown). This will provide evidence that C. concisus is involved in the chronic inflammation of IBD or just plays a role in triggering the host immune system without activating the humeral immune response.

7. A larger sample size is needed to further investigate the possibility of grouping C. concisus by using LPS profiles. Grouping oral C. concisus strains will help in differentiating C. concisus strains isolated from patients with IBD from oral C. concisus strains isolated from healthy individuals. This is proposed after this study has shown that C. concisus strains isolated from patients with IBD have different properties such as invading IECs, activating TLR4 and MD-2 and other properties as compared to C. concisus strains isolated from healthy controls, which indicates that there are different groups.

8. Give that the potential role of C. concisus in modulating the gut innate immune system and inducing inflammation was conducted under in vitro conditions, in vivo experiments is needed to confirm these results.

In summary, this study provides evidence that explain the potential role of C. concisus in the pathogenesis of IBD. Further evidence is needed to confirm these novel findings. The future directions proposed will further provide evidence regarding the pathological mechanisms of C. concisus in inducing chronic inflammation such as IBD.

161

References

1. Hanauer SB. Inflammatory bowel disease: epidemiology, pathogenesis, and therapeutic opportunities. Inflamm Bowel Dis. 2006;12: S3-S9.

2. Strober W, Fuss I, Mannon P. The fundamental basis of inflammatory bowel disease. J Clin Invest. 2007;117: 514.

3. Cho JH. The genetics and immunopathogenesis of inflammatory bowel disease. Nat Rev Immunol. 2008;8: 458-466.

4. Munkholm P. Review article: the incidence and prevalence of colorectal cancer in inflammatory bowel disease. Aliment Pharmacol Ther. 2003;18: 1-5.

5. Geier MS, Butler RN, Howarth GS. Inflammatory bowel disease: current insights into pathogenesis and new therapeutic options; probiotics, prebiotics and synbiotics. Int J Food Microbiol. 2007;115: 1-11.

6. Monteleone G, Biancone L, Marasco R, et al. Interleukin 12 is expressed and actively released by Crohn's disease intestinal lamina propria mononuclear cells. Gastroenterology. 1997;112: 1169-1178.

7. MacDonald T, Hutchings P, Choy MY, et al. Tumour necrosis factor‐alpha and interferon‐gamma production measured at the single cell level in normal and inflamed human intestine. Clin Exp Immunol. 1990;81: 301-305.

8. Peluso I, Pallone F, Monteleone G. Interleukin-12 and Th1 immune response in Crohn's disease: Pathogenetic relevance and therapeutic inplication. World J Gastroenterol. 2006;12: 5606.

9. Pallone F, Monteleone G. Interleukin 12 and Th1 responses in inflammatory bowel disease. Gut. 1998;43: 735-736.

10. Giulia R, Margherita M, Alessandro S, et al. Cytokine Networks in Ulcerative Colitis. Ulcers. 2011;2011:1-5.

11. Kitto L. Inflammatory Bowel Disease. J Sch Nurs. 2010;26: 102.

12. Bouma G, Strober W. The immunological and genetic basis of inflammatory bowel disease. Nat Rev Immunol. 2003;3: 521-533.

162

References

13. Cosnes J, Gower-Rousseau C, Seksik P, et al. Epidemiology and natural history of inflammatory bowel diseases. Gastroenterology. 2011;140: 1785-1794. e4.

14. Molodecky NA, Soon IS, Rabi DM, et al. Increasing incidence and prevalence of the inflammatory bowel diseases with time, based on systematic review. Gastroenterology. 2012; 142: 46-54.

15. Loftus EV. Clinical epidemiology of inflammatory bowel disease: incidence, prevalence, and environmental influences. Gastroenterology. 2004;126: 1504-1517.

16. Loftus Jr EV, Sandborn WJ. Epidemiology of inflammatory bowel disease. Gastroenterol Clin North Am. 2002;31: 1.

17. Loftus CG, Loftus Jr EV, Harmsen WS, et al. Update on the incidence and prevalence of Crohn's disease and ulcerative colitis in Olmsted County, Minnesota, 1940–2000. Inflamm Bowel Dis. 2007;13: 254-261.

18. Bernstein CN, Wajda A, Svenson LW, et al. The epidemiology of inflammatory bowel disease in Canada: a population-based study. Am J Gastroenterol. 2006;101: 1559-1568.

19. Pinchbeck B, Kirdeikis J, Thomson A. Inflammatory bowel disease in northern Alberta. An epidemiologic study. J Clin Gastroenterol. 1988;10: 505.

20. Devlin HB, Datta D, Dellipiani A. The incidence and prevalence of inflammatory bowel disease in North Tees Health District. World J Surg. 1980;4: 183-192.

21. Rubin G, Hungin A, Kelly P, et al. Inflammatory bowel disease: epidemiology and management in an English general practice population. Aliment Pharmacol Ther. 2000;14: 1553-1559.

22. Lapidus A, Bernell O, Hellers G, et al. Incidence of Crohn’s disease in Stockholm County 1955–1989. Gut. 1997;41: 480-486.

23. Lapidus A. Crohn's disease in Stockholm County during 1990-2001: an epidemiological update. World J Gastroenterol. 2006;12: 75.

163

References

24. Rönnblom A, Samuelsson SM, Ekbom A. Ulcerative colitis in the county of Uppsala 1945–2007: Incidence and clinical characteristics. J Crohns Colitis. 2010;4: 532-536.

25. Ouyang Q, Tandon R, Goh KL, et al. The emergence of inflammatory bowel disease in the Asian Pacific region. Curr Opin Gastroenterol. 2005;21: 408.

26. Yoshida Y, Murata Y. Inflammatory bowel disease in Japan: studies of epidemiology and etiopathogenesis. Med Clin North Am. 1990;74: 67.

27. Morita N, Toki S, Hirohashi T, et al. Incidence and prevalence of inflammatory bowel disease in Japan: nationwide epidemiological survey during the year 1991. J Gastroenterol. 1995;30 Suppl 8: 1-4.

28. Yang SK, Hong WS, Min YI, et al. Incidence and prevalence of ulcerative colitis in the Songpa‐Kangdong District, Seoul, Korea, 1986–1997. J Gastroenterol Hepatol. 2000;15: 1037-1042.

29. Yang SK, Yun S, Kim JH, et al. Epidemiology of inflammatory bowel disease in the Songpa‐Kangdong district, Seoul, Korea, 1986–2005: A KASID study. Inflamm Bowel Dis. 2008;14: 542-549.

30. Sood A, Midha V, Sood N, et al. Incidence and prevalence of ulcerative colitis in Punjab, North India. Gut. 2003;52: 1587.

31. Linares CJA, Canton C, Hermida C, et al. Estimated incidence of inflammatory bowel disease in Argentina and Panama (1987-1993). Revista espanola de enfermedades digestivas: organo oficial de la Sociedad Espanola de Patologia Digestiva. 1999;91: 277.

32. Appleyard CB, Hernández G, Ríos‐Bedoya CF. Basic epidemiology of inflammatory bowel disease in Puerto Rico. Inflamm Bowel Dis. 2004;10: 106-111.

33. Myren J, Gjone E, Hertzberg J, et al. Epidemiology of ulcerative colitis and regional enterocolitis (Crohn's disease) in Norway. Scand J Gastroenterol. 1971;6: 511- 514.

164

References

34. Moum B, Vain M, Ekbom A, et al. Incidence of inflammatory bowel disease in southeastern Norway: evaluation of methods after 1 year of registration. Digestion. 1995;56: 377-381.

35. Shivananda S, Pena A, Nap M, et al. Epidemiology of Crohn's disease in Regio Leiden, the Netherlands: a population study from 1979 to 1983. Gastroenterology. 1987;93: 966-974.

36. Shivananda S, Pena A, Mayberry J, et al. Epidemiology of proctocolitis in the region of Leiden, the Netherlands: a population study from 1979 to 1983. Scand J Gastroenterol. 1987;22: 993-1002.

37. Romberg-Camps MJL, Hesselink-van de Kruijs MAM, Schouten LJ, et al. Inflammatory bowel disease in South Limburg (the Netherlands) 1991–2002: incidence, diagnostic delay, and seasonal variations in onset of symptoms. J Crohns Colitis. 2009;3: 115-124.

38. Binder V, Both H, Hansen P, et al. Incidence and prevalence of ulcerative colitis and Crohn's disease in the County of Copenhagen, 1962 to 1978. Gastroenterology. 1982;83: 563.

39. Munkholm P, Langholz E, Nielsen OH, et al. Incidence and prevalence of Crohn's disease in the county of Copenhagen, 1962-87: a sixfold increase in incidence. Scand J Gastroenterol. 1992;27: 609-614.

40. 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. Am J Gastroenterol. 2006;101: 1274-1282.

41. Madunil N, De Silva Arjuna DA, Madurangi A, et al. Prevalence of inflammatory bowel disease in two districts of Sri Lanka: a hospital based survey. BMC Gastroenterol 2010;10: 32.

42. Al-Shamali MA, Kalaoui M, Patty I, et al. Ulcerative colitis in Kuwait: a review of 90 cases. Digestion. 2003;67: 218-224.

165

References

43. Al-Nakib B, Radhakrishnan S, Jacob G, et al. Inflammatory bowel disease in Kuwait. Am J Gastroenterol. 1984;79: 191.

44. Tozun N, Atug O, Imeryuz N, et al. Clinical characteristics of inflammatory bowel disease in Turkey: A multicenter epidemiologic survey. J Clin Gastroenterol. 2009;43: 51.

45. Radhakrishnan S, Zubaidi G, Daniel M, et al. Ulcerative Colitis in Oman. Digestion. 1997;58: 266-270.

46. Anseline PF. Crohn's disease in the Hunter Valley region of Australia. Aust N Z J Surg. 1995;65: 564-569.

47. 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: 1550-1556.

48. Eason R, Lee S, Tasman‐Jones C. Inflammatory bowel disease in Auckland, New Zealand. Aust N Z J Med. 1982;12: 125-131.

49. Gearry RB, Richardson A, Frampton C, et al. High incidence of Crohn's disease in Canterbury, New Zealand: results of an epidemiologic study. Inflamm Bowel Dis. 2006;12: 936-943.

50. Sartor RB. Mechanisms of disease: pathogenesis of Crohn's disease and ulcerative colitis. Nature Clinical Practice Gastroenterology and Hepatology. 2006;3: 390.

51. Mahid SS, Minor KS, Soto RE, et al. Smoking and inflammatory bowel disease: a meta-analysis. 2006;81: 1462.

52. Sher ME, Bank S, Greenberg R, et al. The influence of cigarette smoking on cytokine levels in patients with inflammatory bowel disease. Inflamm Bowel Dis. 1999;5: 73-78.

53. Zijlstra F, Srivastava E, Rhodes M, et al. Effect of nicotine on rectal mucus and mucosal eicosanoids. Gut. 1994;35: 247-251.

54. Srivastava E, Barton J, O'Mahony S, et al. Smoking, humoral immunity, and ulcerative colitis. Gut. 1991;32: 1016-1019.

166

References

55. Mahmud N, Weir DG. The urban diet and Crohn's disease: is there a relationship? Eur J Gastroenterol Hepatol. 2001;13: 93.

56. Amre DK, D'Souza S, Morgan K, et al. Imbalances in dietary consumption of fatty acids, vegetables, and fruits are associated with risk for Crohn's disease in children. Am J Gastroenterol. 2007;102: 2016-2025.

57. Sakamoto N, Kono S, Wakai K, et al. Dietary risk factors for inflammatory bowel disease A Multicenter Case‐Control Study in Japan. Inflamm Bowel Dis. 2005;11: 154-163.

58. Lakatos PL. Recent trends in the epidemiology of inflammatory bowel diseases: up or down?. World J Gastroenterol. 2006;12: 6102.

59. Shoda R, Matsueda K, Yamato S, et al. Epidemiologic analysis of Crohn disease in Japan: increased dietary intake of n-6 polyunsaturated fatty acids and animal protein relates to the increased incidence of Crohn disease in Japan. Am J Clin Nutr. 1996;63: 741-745.

60. Sherry S, William J A, R James B. Analyzing serum-stimulated prostate cancer cell lines after low-fat, high-fiber diet and exercise intervention. J Evid Based Complementary Altern Med. 2011;2011: Article ID 529053.

61. Bartsch H, Nair J, Owen RW. Dietary polyunsaturated fatty acids and cancers of the breast and colorectum: emerging evidence for their role as risk modifiers. Carcinogenesis. 1999;20: 2209-2218.

62. Wilson LM, Baldwin AL. Environmental stress causes mast cell degranulation, endothelial and epithelial changes, and edema in the rat intestinal mucosa. Microcirculation. 1999;6: 189-198.

63. Santos J, Saperas E, Nogueiras C, et al. Release of mast cell mediators into the jejunum by cold pain stress in humans. Gastroenterology. 1998;114: 640-648.

64. Cenac N, Chin AC, Garcia‐Villar R, et al. PAR2 activation alters colonic paracellular permeability in mice via IFN‐γ‐dependent and‐independent pathways. J Physiol (Lond). 2004;558: 913-925.

167

References

65. Farhadi A, Fields JZ, Keshavarzian A. Mucosal mast cells are pivotal elements in inflammatory bowel disease that connect the dots: stress, intestinal hyperpermeability and inflammation. World J Gastroenterol. 2007;13: 3027.

66. Mawdsley JE, Macey MG, Feakins RM, et al. The effect of acute psychologic stress on systemic and rectal mucosal measures of inflammation in ulcerative colitis. Gastroenterology. 2006;131: 410-419.

67. Shanahan F. Probiotics in inflammatory bowel disease--therapeutic rationale and role. Adv Drug Deliv Rev. 2004;56: 809-818.

68. Gent A, Hellier M, Grace R, et al. Inflammatory bowel disease and domestic hygiene in infancy. The Lancet. 1994;343: 766-767.

69. Duggan A, Usmani I, Neal K, et al. Appendicectomy, childhood hygiene, Helicobacter pylori status, and risk of inflammatory bowel disease: a case control study. Gut. 1998;43: 494-498.

70. Card T, Logan R, Rodrigues L, et al. Antibiotic use and the development of Crohn’s disease. Gut. 2004;53: 246-250.

71. Shaw SY, Blanchard JF, Bernstein CN. Association between the use of antibiotics in the first year of life and pediatric inflammatory bowel disease. Am J Gastroenterol. 2010;105: 2687-2692.

72. Trinchieri G. Interleukin-12 and the regulation of innate resistance and adaptive immunity. Nat Rev Immunol. 2003;3: 133-146.

73. Cooke A. Th17 cells in inflammatory conditions. The Review of Diabetic Studies. 2006;3: 72.

74. Brynskov J, Tvede N, Andersen C, et al. Increased concentrations of interleukin 1 beta, interleukin-2, and soluble interleukin-2 receptors in endoscopical mucosal biopsy specimens with active inflammatory bowel disease. Gut. 1992;33: 55-58.

75. Reinecker HC, Steffen M, Witthoeft T, et al. Enhand secretion of tumour necrosis factor‐alpha, IL‐6, and IL‐1β by isolated lamina ropria monouclear cells from patients with ulcretive cilitis and Crohn's disease. Clin Exp Immunol. 1993;94: 174-181.

168

References

76. Daig R, Andus T, Aschenbrenner E, et al. Increased interleukin 8 expression in the colon mucosa of patients with inflammatory bowel disease. Gut. 1996;38: 216-222.

77. Monteleone G, Trapasso F, Parrello T, et al. Bioactive IL-18 expression is upregulated in Crohn’s disease. J Immunol. 1999;163: 143-147.

78. Schmidt C, Giese T, Ludwig B, et al. Expression of interleukin‐12‐related cytokine transcripts in inflammatory bowel disease: Elevated interleukin‐23p19 and interleukin‐27p28 in Crohn's disease but not in ulcerative colitis. Inflamm Bowel Dis. 2005;11: 16-23.

79. Artis D. Epithelial-cell recognition of commensal bacteria and maintenance of immune homeostasis in the gut. Nat Rev Immunol. 2008;8: 411-420.

80. Hyun JG, Mayer L. Mechanisms underlying inflammatory bowel disease. Drug Discovery Today. Disease Mechanisms. 2007;3: 457-462.

81. Jung HC, Eckmann L, Yang SK, et al. A distinct array of proinflammatory cytokines is expressed in human colon epithelial cells in response to bacterial invasion. J Clin Invest. 1995;95: 55.

82. Mueller T, Podolsky DK. Nucleotide-binding-oligomerization domain proteins and toll-like receptors: sensors of the inflammatory bowel diseases' microbial environment. Curr Opin Gastroenterol. 2005;21: 419.

83. Bevins CL. The Paneth cell and the innate immune response. Curr Opin Gastroenterol. 2004;20: 572.

84. Kontoyiannis D, Pasparakis M, Pizarro TT, et al. Impaired on/off regulation of TNF biosynthesis in mice lacking TNF AU-rich elements: implications for joint and gut-associated immunopathologies. Immunity. 1999;10: 387-398.

85. 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 pathology. Proc Natl Acad Sci U S A. 2011;108: 5396.

86. Stadnyk AW. Intestinal epithelial cells as a source of inflammatory cytokines and chemokines. Can J Gastroenterol. 2002;16: 241-246.

169

References

87. Shih DQ, Targan SR. Immunopathogenesis of inflammatory bowel disease. World J Gastroenterol. 2008;14: 390.

88. Jopeace A, Howard C, Murton B, et al. Pattern Recognition Receptors and Infectious Diseases. Kumar P, ed. Insight and Control of Infectious Disease in Global Scenario. Rijeka, Croatia: InTech; 2012:383.

89. Sun SM, Wang XL, Wu XP, et al. Toll-like receptor activation by helminths or helminth products to alleviate inflammatory bowel disease. Parasit Vectors. 2011;4: 1- 8.

90. Shibolet O, Podolsky DK. TLRs in the Gut. IV. Negative regulation of Toll-like receptors and intestinal homeostasis: addition by subtraction. Am J Physiol Gastrointest Liver Physiol. 2007;292: G1469-G1473.

91. Goel GA, Kandiel A, Achkar JP, et al. Molecular Pathways Underlying IBD- Associated Colorectal Neoplasia: Therapeutic Implications. Am J Gastroenterol. 2011;106: 719-730.

92. Cario E, Podolsky DK. Differential alteration in intestinal epithelial cell expression of toll-like receptor 3 (TLR3) and TLR4 in inflammatory bowel disease. Infect Immun. 2000;68: 7010.

93. Strober W, Murray PJ, Kitani A, et al. Signalling pathways and molecular interactions of NOD1 and NOD2. Nat Rev Immunol. 2005;6: 9-20.

94. McGovern DPB, Hysi P, Ahmad T, et al. Association between a complex insertion/deletion polymorphism in NOD1 (CARD4) and susceptibility to inflammatory bowel disease. Hum Mol Genet. 2005;14: 1245-1250.

95. Eckmann L, Karin M. NOD2 and Crohn’s disease: loss or gain of function? Immunity. 2005;22: 661-667.

96. Inohara N, Nuñez G. NODs: intracellular proteins involved in inflammation and apoptosis. Nat Rev Immunol. 2003;3: 371-382.

97. Xavier R, Podolsky D. Unravelling the pathogenesis of inflammatory bowel disease. Nature. 2007;448: 427-434.

170

References

98. Kamada N, Hisamatsu T, Okamoto S, et al. Abnormally differentiated subsets of intestinal macrophage play a key role in Th1-dominant chronic colitis through excess production of IL-12 and IL-23 in response to bacteria. J Immunol. 2005;175: 6900- 6908.

99. Duerr RH, Taylor KD, Brant SR, et al. A genome-wide association study identifies IL23R as an inflammatory bowel disease gene. Science's STKE. 2006;314: 1461.

100. Targan SR, Hanauer SB, van Deventer SJH, et al. A short-term study of chimeric monoclonal antibody cA2 to tumor necrosis factor α for Crohn's disease. N Engl J Med. 1997;337: 1029-1036.

101. Seow CH, Newman A, Irwin SP, et al. Trough serum infliximab: a predictive factor of clinical outcome for infliximab treatment in acute ulcerative colitis. Gut. 2010;59: 49.

102. Himmel ME, Hardenberg G, Piccirillo CA, et al. The role of T‐regulatory cells and Toll‐like receptors in the pathogenesis of human inflammatory bowel disease. Immunology. 2008;125: 145-153.

103. Murugananthan AU, Bernardo D, Tozer P, et al. TH1/TH17 profiles in crohn's disease: a cross sectional single centre study in postoperative crohn's disease. Gut. 2011;60: A212.

104. 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: 6276-6282.

105. Parrello T, Monteleone G, Cucchiara S, et al. Up-Regulation of the IL-12 Receptor ß2 Chain in Crohn’s Disease. J Immunol. 2000;165: 7234.

106. Neurath M, Weigmann B, Finotto S, et al. The transcription factor T-bet regulates mucosal T cell activation in experimental colitis and Crohn's disease. J Exp Med. 2002;195: 1129-1143.

107. Kolls JK, Lindén A. Interleukin-17 family members and inflammation. Immunity. 2004;21: 467-476.

108. Ouyang W, Kolls JK, Zheng Y. The biological functions of T helper 17 cell effector cytokines in inflammation. Immunity. 2008;28: 454-467.

171

References

109. Weaver CT, Hatton RD, Mangan PR, et al. IL-17 family cytokines and the expanding diversity of effector T cell lineages. Annu Rev Immunol. 2007;25: 821-852.

110. Fujino S, Andoh A, Bamba S, et al. Increased expression of interleukin 17 in inflammatory bowel disease. Gut. 2003;52: 65-70.

111. Yen D, Cheung J, Scheerens H, et al. IL-23 is essential for T cell-mediated colitis and promotes inflammation via IL-17 and IL-6. J Clin Invest. 2006;116: 1310.

112. Neurath MF, Finotto S, Glimcher LH. The role of Th1/Th2 polarization in mucosal immunity. Nat Med. 2002;8: 567-573.

113. 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.

114. 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: 1490-1497.

115. Monteleone G, Monteleone I, Fina D, et al. Interleukin-21 enhances T-helper cell type I signaling and interferon-γ production in Crohn’s disease. Gastroenterology. 2005;128: 687-694.

116. Rosekrans P, Meijer C, Van der Wal A, et al. Immunoglobulin containing cells in inflammatory bowel disease of the colon: a morphometric and immunohistochemical study. Gut. 1980;21: 941-947.

117. Macpherson A, Khoo U, Forgacs I, et al. Mucosal antibodies in inflammatory bowel disease are directed against intestinal bacteria. Gut. 1996;38: 365-375.

118. Brandwein SL, McCabe RP, Cong Y, et al. Spontaneously colitic C3H/HeJBir mice demonstrate selective antibody reactivity to antigens of the enteric bacterial flora. J Immunol. 1997;159: 44-52.

119. Lodes MJ, Cong Y, Elson CO, et al. Bacterial flagellin is a dominant antigen in Crohn disease. J Clin Invest. 2004;113: 1296-1306.

172

References

120. Howells B, Matthews N, Mayberry J, et al. Agglutinins to anaerobic bacteria in Crohn's disease and in Indian patients with diarrhoea. J Med Microbiol. 1984;17: 207- 209.

121. Auer I, Röder A, Wensinck F, et al. Selected bacterial antibodies in Crohn's disease and ulcerative colitis. Scand J Gastroenterol. 1983;18: 217-223.

122. Matthews N, Mayberry J, Rhodes J, et al. Agglutinins to bacteria in Crohn's disease. Gut. 1980;21: 376-380.

123. Tabaqchali S, O'Donoghue D, Bettelheim K. Escherichia coli antibodies in patients with inflammatory bowel disease. Gut. 1978;19: 108-113.

124. Saxon A, Shanahan F, Landers C, et al. A distinct subset of antineutrophil cytoplasmic antibodies is associated with inflammatory bowel disease. J Allergy Clin Immunol. 1990;86: 202-210.

125. Gigase P, Clerck LSD, Cotthem KAV, et al. Anti-neutrophil cytoplasmic antibodies in inflammatory bowel disease with special attention for IgA-class antibodies. Dig Dis Sci. 1997;42: 2171-2174.

126. Stöcker W, Otte M, Ulrich S, et al. [Autoantibodies against the exocrine pancreas and against intestinal goblet cells in the diagnosis of Crohn's disease and ulcerative colitis]. Deutsche medizinische Wochenschrift. 1984;109: 1963.

127. Mayet W, Press A, Hermann E, et al. Antibodies to cytoskeletal proteins in patients with Crohn's disease. Eur J Clin Invest. 1990;20: 516-524.

128. Khor B, Gardet A, Xavier RJ. Genetics and pathogenesis of inflammatory bowel disease. Nature. 2011;474: 307-317.

129. Mathew CG, Lewis CM. Genetics of inflammatory bowel disease: progress and prospects. Hum Mol Genet. 2004;13: R161.

130. Orholm M, Binder V, Sørensen T, et al. Concordance of inflammatory bowel disease among Danish twins: results of a nationwide study. Scand J Gastroenterol. 2000;35: 1075-1081.

173

References

131. Tysk C, Lindberg E, Järnerot G, et al. Ulcerative colitis and Crohn's disease in an unselected population of monozygotic and dizygotic twins. A study of heritability and the influence of smoking. Gut. 1988;29: 990-996.

132. Farrell R, Peppercorn M. Ulcerative colitis. Lancet. 2002;359: 331-340.

133. Roussomoustakaki M, Satsangi J, Welsh K, et al. Genetic markers may predict disease behavior in patients with ulcerative colitis. Gastroenterology. 1997;112: 1845- 1853.

134. Girardin SE, Boneca IG, Viala J, et al. Nod2 is a general sensor of peptidoglycan through muramyl dipeptide (MDP) detection. J Biol Chem. 2003;278: 8869-8872.

135. 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: 821-823.

136. Lipinski S, Till A, Sina C, et al. DUOX2-derived reactive oxygen species are effectors of NOD2-mediated antibacterial responses. J Cell Sci. 2009;122: 3522-3530.

137. Neurath MF, Pettersson S, Zum Büschenfelde KHM, et al. Local administration of antisense phosphorothiate olignucleotides to the p65 subunit of NF–κB abrogates established experimental colitis in mice. Nat Med. 1996;2: 998-1004.

138. Cho JH. The genetics and immunopathogenesis of inflammatory bowel disease. Nat Rev Immunol. 2008;8: 458-466.

139. Hisamatsu T, Suzuki M, Reinecker HC, et al. CARD15/NOD2 functions as an antibacterial factor in human intestinal epithelial cells. Gastroenterology. 2003;124: 993-1000.

140. Kim YG, Park JH, Shaw MH, et al. The cytosolic sensors Nod1 and Nod2 are critical for bacterial recognition and host defense after exposure to Toll-like receptor ligands. Immunity. 2008;28: 246-257.

141. Curran ME, Lau KF, Hampe J, et al. Genetic analysis of inflammatory bowel disease in a large European cohort supports linkage to chromosomes 12 and 16. Gastroenterology. 1998;115: 1066-1071.

174

References

142. Cavanaugh J, Callen D, Wilson S, et al. Analysis of Australian Crohn's disease pedigrees refines the localization for susceptibility to inflammatory bowel disease on chromosome 16. Ann Hum Genet. 1998;62: 291-298.

143. 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.

144. Ogura Y, Bonen DK, Inohara N, et al. A frameshift mutation in NOD2 associated with susceptibility to Crohn's disease. Nature. 2001;411: 603-606.

145. Cuthbert AP, Fisher SA, Mirza MM, et al. The contribution of NOD2 gene mutations to the risk and site of disease in inflammatory bowel disease**. Gastroenterology. 2002;122: 867-874.

146. Newman B, Gu X, Wintle R, et al. A risk haplotype in the Solute Carrier Family 22A4/22A5 gene cluster influences phenotypic expression of Crohn's disease. Gastroenterology. 2005;128: 260-269.

147. Yazdanyar S, Weischer M, Nordestgaard BG. Genotyping for NOD2 genetic variants and crohn disease: a metaanalysis. Clin Chem. 2009;55: 1950-1957.

148. Hampe J, Cuthbert A, Croucher PJP, et al. Association between insertion mutation in NOD2 gene and Crohn's disease in German and British populations. The Lancet. 2001;357: 1925-1928.

149. Kugathasan S, Loizides A, Babusukumar U, et al. Comparative phenotypic and CARD15 mutational analysis among African American, Hispanic, and White children with Crohn's disease. Inflamm Bowel Dis. 2005;11: 631-638.

150. Özen SC, Daǧlı Ü, Kiliç MY, et al. NOD2/CARD15, NOD1/CARD4, and ICAM- 1 gene polymorphisms in Turkish patients with inflammatory bowel disease. J Gastroenterol. 2006;41: 304-310.

151. Yamazaki K, Takazoe M, Tanaka T, et al. Absence of mutation in the NOD2/CARD15 gene among 483 Japanese patients with Crohn's disease. J Hum Genet. 2002;47: 469-472.

175

References

152. Leong R, Armuzzi A, Ahmad T, et al. NOD2/CARD15 gene polymorphisms and Crohn's disease in the Chinese population. Aliment Pharmacol Ther. 2003;17: 1465- 1470.

153. Wang ZW, Ji F, Teng WJ, et al. Risk factors and gene polymorphisms of inflammatory bowel disease in population of Zhejiang, China. World J Gastroenterol. 2011;17: 118.

154. Inohara N, Ogura Y, Chen FF, et al. Human Nod1 confers responsiveness to bacterial lipopolysaccharides. J Biol Chem. 2001;276: 2551-2554.

155. Abraham C, Cho JH. Functional consequences of NOD2 (CARD15) mutations. Inflamm Bowel Dis. 2006;12: 641-650.

156. Gazouli M. NOD2/CARD15 mediates the inflammatory responses in inflammatory bowel diseases (IBDs). Ann Gastroentol. 2007;18 (1): 16-18.

157. Wehkamp J, Harder J, Weichenthal M, et al. NOD2 (CARD15) mutations in Crohn’s disease are associated with diminished mucosal α-defensin expression. Gut. 2004;53: 1658-1664.

158. Kobayashi KS, Chamaillard M, Ogura Y, et al. Nod2-dependent regulation of innate and adaptive immunity in the intestinal tract. Science. 2005;307: 731.

159. Newman B, Siminovitch KA. Recent advances in the genetics of inflammatory bowel disease. Curr Opin Gastroenterol. 2005;21: 401.

160. Peltekova VD, Wintle RF, Rubin LA, et al. Functional variants of OCTN cation transporter genes are associated with Crohn disease. Nat Genet. 2004;36: 471-475.

161. Xuan C, Zhang BB, Yang T, et al. Association between OCTN1/2 gene polymorphisms (1672C-T, 207G-C) and susceptibility of Crohn's disease: a meta- analysis. Int J Colorectal Dis. 2012: 1-9.

162. Vermeire S, Pierik M, Hlavaty T, et al. Association of organic cation transporter risk haplotype with perianal penetrating Crohn’s disease but not with susceptibility to IBD. Gastroenterology. 2005;129: 1845-1853.

176

References

163. Cucchiara S, Latiano A, Palmieri O, et al. Role of CARD15, DLG5 and OCTN genes polymorphisms in children with inflammatory bowel diseases. World J Gastroenterol. 2007;13.

164. Palmieri O, Latiano A, Valvano R, et al. Variants of OCTN1–2 cation transporter genes are associated with both Crohn's disease and ulcerative colitis. Aliment Pharmacol Ther. 2006;23: 497-506.

165. Waller S, Tremelling M, Bredin F, et al. Evidence for association of OCTN genes and IBD5 with ulcerative colitis. Gut. 2006;55: 809-814.

166. Török HP, Glas J, Tonenchi L, et al. Polymorphisms in the DLG5 and OCTN cation transporter genes in Crohn’s disease. Gut. 2005;54: 1421-1427.

167. Leung E, Hong J, Fraser AG, et al. Polymorphisms in the organic cation transporter genes SLC22A4 and SLC22A5 and Crohn's disease in a New Zealand Caucasian cohort. Immunol Cell Biol. 2006;84: 233-236.

168. Martínez A, del Carmen Martín M, Mendoza JL, et al. Association of the organic cation transporter OCTN genes with Crohn's disease in the Spanish population. Eur J Hum Genet. 2005;14: 222-226.

169. Törkvist L, Noble CL, Lördal M, et al. Contribution of the IBD5 locus to Crohn's disease in the Swedish population. Scand J Gastroenterol. 2007;42: 200-206.

170. de Ridder L, Weersma RK, Dijkstra G, et al. Genetic susceptibility has a more important role in pediatric‐onset Crohn's disease than in adult‐onset Crohn's disease. Inflamm Bowel Dis. 2007;13: 1083-1092.

171. Gaj P, Habior A, Mikula M, et al. Lack of evidence for association of primary sclerosing cholangitis and primary biliary cirrhosis with risk alleles for Crohn's disease in Polish patients. BMC medical genetics. 2008;9: 81.

172. Yamazaki K, Takazoe M, Tanaka T, et al. Association analysis of SLC22A4, SLC22A5 and DLG5 in Japanese patients with Crohn disease. J Hum Genet. 2004;49: 664-668.

177

References

173. Tosa M, Negoro K, Kinouchi Y, et al. Lack of association between IBD5 and Crohn's disease in Japanese patients demonstrates population-specific differences in inflammatory bowel disease. Scand J Gastroenterol. 2006;41: 48-53.

174. Li M, Gao X, Guo CC, et al. OCTN and CARD15 gene polymorphism in Chinese patients with inflammatory bowel disease. World J Gastroenterol. 2008;14: 4923.

175. Chua KH, Lian LH, Kee BP, et al. Identification of DLG5 and SLC22A5 gene polymorphisms in Malaysian patients with Crohn's disease. J Dig Dis. 2011;12: 459- 466.

176. Repnik K, Potočnik U. Haplotype in the IBD5 region is associated with refractory Crohn’s disease in Slovenian patients and modulates expression of the SLC22A5 gene. J Gastroenterol. 2011: 1-11.

177. Peltekova VD, Wintle RF, Rubin LA, et al. Functional variants of OCTN cation transporter genes are associated with Crohn disease. Nat Genet. 2004;36: 471-475.

178. Feng Y, Zheng P, Zhaoh, et al. SLC22A4 and SLC22A5 gene polymorphisms and Crohn's disease in the Chinese Han population. J Dig Dis. 2009;10: 181-187.

179. Newman B, Gu X, Wintle R, et al. A risk haplotype in the Solute Carrier Family 22A4/22A5 gene cluster influences phenotypic expression of Crohn's disease. Gastroenterology. 2005;128: 260-269.

180. Stoll M, Corneliussen B, Costello CM, et al. Genetic variation in DLG5 is associated with inflammatory bowel disease. Nat Genet. 2004;36: 476-480.

181. Lin Z, Poritz L, Franke A, et al. Genetic association of DLG5 R30Q with familial and sporadic inflammatory bowel disease in men. Dis Markers. 2009;27: 193-201.

182. Daly MJ, Pearce AV, Farwell L, et al. Association of DLG5 R30Q variant with inflammatory bowel disease. Eur J Hum Genet. 2005;13: 835-839.

183. Newman WG, Gu X, Wintle RF, et al. DLG5 variants contribute to Crohn disease risk in a Canadian population. Hum Mutat. 2006;27: 353-358.

178

References

184. Lin Z, Hegarty JP, Berg A, et al. DLG5 P1371Q is associated with inflammatory bowel disease and complementary to R30Q in disease susceptibility. Swiss medical weekly. 2011;141: w13290.

185. Friedrichs F, Brescianini S, Annese V, et al. Evidence of transmission ratio distortion of DLG5 R30Q variant in general and implication of an association with Crohn disease in men. Hum Genet. 2006;119: 305-311.

186. Humbert P, Russell S, Richardson H. Dlg, Scribble and Lgl in cell polarity, cell proliferation and cancer. Bioessays. 2003;25: 542-553.

187. Wakabayashi M, Ito T, Mitsushima M, et al. Interaction of lp-dlg/KIAA0583, a membrane-associated guanylate kinase family protein, with vinexin and β-catenin at sites of cell-cell contact. J Biol Chem. 2003;278: 21709.

188. Stoll M, Corneliussen B, Costello CM, et al. Genetic variation in DLG5 is associated with inflammatory bowel disease. Nat Genet. 2004;36: 476-480.

189. Walczak R, Tontonoz P. PPARadigms and PPARadoxes: expanding roles for PPARγ in the control of lipid metabolism. J Lipid Res. 2002;43: 177.

190. Rousseaux C, Lefebvre B, Dubuquoy L, et al. Intestinal antiinflammatory effect of 5-aminosalicylic acid is dependent on peroxisome proliferator–activated receptor-γ. J Exp Med. 2005;201: 1205.

191. Dubuquoy L, Dharancy S, Nutten S, et al. Role of peroxisome proliferator- activated receptor [gamma] and retinoid X receptor heterodimer in hepatogastroenterological diseases. The Lancet. 2002;360: 1410-1418.

192. Sugawara K, Olson TS, Moskaluk CA, et al. Linkage to peroxisome proliferator- activated receptor-[gamma] in SAMP1/YitFc mice and in human Crohn's disease. Gastroenterology. 2005;128: 351-360.

193. Glas J, Seiderer J, Markus C, et al. Role of PPARG gene variants in inflammatory bowel disease. Inflamm Bowel Dis. 2011;17: 1057-1058.

194. Atug O, Tahan V, Eren F, et al. Pro12Ala polymorphism in the peroxisome proliferator-activated receptor-gamma (PPARgamma) gene in inflammatory bowel disease. J Gastrointestin Liver Dis. 2008;17: 433-437.

179

References

195. Lytle C, Tod TJ, Vo KT, et al. The peroxisome proliferator‐activated receptor γ ligand rosiglitazone delays the onset of inflammatory bowel disease in mice with interleukin 10 deficiency. Inflamm Bowel Dis. 2005;11: 231-243.

196. Poto&ccaron U. Polymorphisms in multidrug resistance 1 (MDR1) gene are associated with refractory Crohn disease and ulcerative colitis. Genes Immun. 2004;5: 530-539.

197. Fojo A, Ueda K, Slamon D, et al. Expression of a multidrug-resistance gene in human tumors and tissues. Proc Natl Acad Sci U S A. 1987;84: 265.

198. Huebner C, Browning BL, Petermann I, et al. Genetic analysis of MDR1 and inflammatory bowel disease reveals protective effect of heterozygous variants for ulcerative colitis. Inflamm Bowel Dis. 2009;15: 1784-1793.

199. Ho G, Moodie F, Satsangi J. Multidrug resistance 1 gene (P-glycoprotein 170): an important determinant in gastrointestinal disease?. Gut. 2003;52: 759.

200. Panwala CM, Jones JC, Viney JL. A novel model of inflammatory bowel disease: mice deficient for the multiple drug resistance gene, mdr1a, spontaneously develop colitis. J Immunol. 1998;161: 5733-5744.

201. Satsangi J, Parkes M, Louis E, et al. Two stage genome–wide search in inflammatory bowel disease provides evidence for susceptibility loci on chromosomes 3, 7 and 12. Nat Genet. 1996;14: 199-202.

202. Hoffmeyer S, Burk O, Von Richter O, et al. Functional polymorphisms of the human multidrug-resistance gene: multiple sequence variations and correlation of one allele with P-glycoprotein expression and activity in vivo. Proc Natl Acad Sci U S A. 2000;97: 3473.

203. Brant SR, Panhuysen CIM, Nicolae D, et al. MDR1 Ala893 Polymorphism Is Associated with Inflammatory Bowel Disease. Am J Hum Genet. 2003;73: 1282-1292.

204. Onnie CM, Fisher SA, Pattni R, et al. Associations of allelic variants of the multidrug resistance gene (ABCB1 or MDR1) and Inflammatory Bowel Disease and their effects on disease behavior: A case‐control and meta‐analysis study. Inflamm Bowel Dis. 2006;12: 263-271.

180

References

205. Ho GT, Nimmo ER, Tenesa A, et al. Allelic variations of the multidrug resistance gene determine susceptibility and disease behavior in ulcerative colitis. Gastroenterology. 2005;128: 288-296.

206. Palmieri O, Latiano A, Valvano R, et al. Multidrug resistance 1 gene polymorphisms are not associated with inflammatory bowel disease and response to therapy in Italian patients. Aliment Pharmacol Ther. 2005;22: 1129-1138.

207. Urcelay E, Mendoza JL, Martín MC, et al. MDR1 gene: susceptibility in Spanish Crohn's disease and ulcerative colitis patients. Inflamm Bowel Dis. 2006;12: 33-37.

208. Fischer S, Lakatos PL, Lakatos L, et al. ATP-binding cassette transporter ABCG2 (BCRP) and ABCB1 (MDR1) variants are not associated with disease susceptibility, disease phenotype response to medical therapy or need for surgeryin Hungarian patients with inflammatory bowel diseases. Scand J Gastroenterol. 2007;42: 726-733.

209. Schwab M, Schaeffeler E, Marx C, et al. Association between the C3435T MDR1 gene polymorphism and susceptibility for ulcerative colitis. Gastroenterology. 2003;124: 26-33.

210. Huebner C, Browning BL, Petermann I, et al. Genetic analysis of MDR1 and inflammatory bowel disease reveals protective effect of heterozygous variants for ulcerative colitis. Inflamm Bowel Dis. 2009;15: 1784-1793.

211. 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: 262-267.

212. Rutgeerts P, Peeters M, Hiele M, et al. Effect of faecal stream diversion on recurrence of Crohn's disease in the neoterminal ileum. The Lancet. 1991;338: 771- 774.

213. Wellmann W, Fink P, Benner F, et al. Endotoxaemia in active Crohn's disease. Treatment with whole gut irrigation and 5-aminosalicylic acid. Gut. 1986;27: 814-820.

214. Selby W, Pavli P, Crotty B, et al. Two-year combination antibiotic therapy with clarithromycin, rifabutin, and clofazimine for Crohn's disease. Gastroenterology. 2007;132: 2313-2319.

181

References

215. Gionchetti P, Rizzello F, Lammers KM, et al. Antibiotics and probiotics in treatment of inflammatory bowel disease. World J Gastroenterol. 2006;12: 3306.

216. Gionchetti P, Rizzello F, Helwig U, et al. Prophylaxis of pouchitis onset with probiotic therapy: a double-blind, placebo-controlled trial. Gastroenterology. 2003;124: 1202-1209.

217. Gupta P, Andrew H, Kirschner BS, et al. Is lactobacillus GG helpful in children with Crohn's disease? Results of a preliminary, open-label study. J Pediatr Gastroenterol Nutr. 2000;31: 453.

218. 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.

219. Burich A, Hershberg R, Waggie K, et al. Helicobacter-induced inflammatory bowel disease in IL-10-and T cell-deficient mice. Am J Physiol Gastrointest Liver Physiol. 2001;281: G764-G778.

220. Duchmann R, Kaiser I, Hermann E, et al. Tolerance exists towards resident intestinal flora but is broken in active inflammatory bowel disease (IBD). Clin Exp Immunol. 1995;102: 448-455.

221. Bannantine J, Zhang Q, Li LL, et al. Genomic homogeneity between Mycobacterium avium subsp. avium and Mycobacterium avium subsp. paratuberculosis belies their divergent growth rates. BMC microbiology. 2003;3: 10.

222. Thorel MF, Krichevsky M, Lévy-Frébault VV. Numerical Taxonomy of Mycobactin-Dependent Mycobacteria, Emended Description of Mycobacterium avium, and Description of Mycobacterium avium subsp. avium subsp. nov., Mycobacterium avium subsp. paratuberculosis subsp. nov., and Mycobacterium avium subsp. silvaticum subsp. nov. Int J Syst Bacteriol. 1990;40: 254-260.

223. Hermon-Taylor J. Mycobacterium avium subspecies paratuberculosis is a cause of Crohn's disease. Gut. 2001;49: 755.

224. Hermon-Taylor J, Bull TJ, Sheridan JM, et al. Causation of Crohn's disease by Mycobacterium avium subspecies paratuberculosis. Can J Gastroenterol. 2000;14: 521.

182

References

225. Dalziel T. Chronic interstitial enteritis. The British Medical Journal. 1913;2: 1068-1070.

226. Chiodini R. Crohn's disease and the mycobacterioses: a review and comparison of two disease entities. Clin Microbiol Rev. 1989;2: 90-117.

227. Chiodini RJ, Van Kruiningen HJ, Merkal RS, et al. Characteristics of an unclassified Mycobacterium species isolated from patients with Crohn's disease. J Clin Microbiol. 1984;20: 966-971.

228. Chiodini RJ, Van Kruiningen HJ, Thayer WR, et al. Possible role of mycobacteria in inflammatory bowel disease. Dig Dis Sci. 1984;29: 1073-1079.

229. Chiodini RJ, Van Kruiningen HJ, Thayer WR, et al. Spheroplastic phase of mycobacteria isolated from patients with Crohn's disease. J Clin Microbiol. 1986;24: 357.

230. Mendoza JL, Lana R, Díaz-Rubio M. Mycobacterium avium subspecies paratuberculosis and its relationship with Crohn’s disease. World J Gastroenterol. 2009;15: 417.

231. Naser SA, Ghobrial G, Romero C, et al. Culture of Mycobacterium avium subspecies paratuberculosis from the blood of patients with Crohn's disease. The Lancet. 2004;364: 1039-1044.

232. Mendoza JL, San-Pedro A, Culebras E, et al. High prevalence of viable Mycobacterium avium subspecies paratuberculosis in Crohn’s disease. World J Gastroenterol. 2010;16: 4558.

233. Green E, Tizard M, Moss M, et al. Sequence and characteristics or IS900, an insertion element identified in a human Crohn's disease isolate or Mycobacterium paratuberculosis. Nucleic Acids Res. 1989;17: 9063-9073.

234. Sanderson J, Moss M, Tizard M, et al. Mycobacterium paratuberculosis DNA in Crohn's disease tissue. Gut. 1992;33: 890-896.

235. Lisby G, Andersen J, Engbaek K, et al. Mycobacterium paratuberculosis in intestinal tissue from patients with Crohn's disease demonstrated by a nested primer polymerase chain reaction. Scand J Gastroenterol. 1994;29: 923-929.

183

References

236. Bull TJ, McMinn EJ, Sidi-Boumedine K, et al. Detection and verification of Mycobacterium avium subsp. paratuberculosis in fresh ileocolonic mucosal biopsy specimens from individuals with and without Crohn's disease. J Clin Microbiol. 2003;41: 2915-2923.

237. Ryan P, Bennett M, Aarons S, et al. PCR detection of Mycobacterium paratuberculosis in Crohn’s disease granulomas isolated by laser capture microdissection. Gut. 2002;51: 665.

238. Scanu AM, Bull TJ, Cannas S, et al. Mycobacterium avium subspecies paratuberculosis infection in cases of irritable bowel syndrome and comparison with Crohn's disease and Johne's disease: common neural and immune pathogenicities. J Clin Microbiol. 2007;45: 3883-3890.

239. Sechi LA, Scanu AM, Molicotti P, et al. Detection and isolation of Mycobacterium avium subspecies paratuberculosis from intestinal mucosal biopsies of patients with and without Crohn's disease in Sardinia. Am J Gastroenterol. 2005;100: 1529-1536.

240. Elsaghier A, Prantera C, Moreno C, et al. Antibodies to Mycobacterium paratuberculosis‐specific protein antigens in Crohn's disease. Clin Exp Immunol. 1992;90: 503-508.

241. Stainsby K, Lowes J, Allan R, et al. Antibodies to Mycobacterium paratuberculosis and nine species of environmental mycobacteria in Crohn's disease and control subjects. Gut. 1993;34: 371-374.

242. Stevens T, Winrow V, Blake D, et al. Circulating antibodies to heat‐shock protein 60 in Crohn's disease and ulcerative colitis. Clin Exp Immunol. 1992;90: 271-274.

243. Wayne L, Hollander D, Anderson B, et al. Immunoglobulin A (IgA) and IgG serum antibodies to mycobacterial antigens in Crohn's disease patients and their relatives. J Clin Microbiol. 1992;30: 2013-2018.

244. Vannuffel P, Dieterich C, Naerhuyzen B, et al. Occurrence, in Crohn's disease, of antibodies directed against a species-specific recombinant polypeptide of Mycobacterium paratuberculosis. Clin Diagn Lab Immunol. 1994;1: 241-243.

184

References

245. Ellingson JLE, Cheville JC, Brees D, et al. Absence of Mycobacterium avium subspecies paratuberculosis components from Crohn’s disease intestinal biopsy tissues. Clin med res. 2003;1: 217-226.

246. Baksh FK, Finkelstein SD, Ariyanayagam-Baksh SM, et al. Absence of Mycobacterium avium subsp. paratuberculosis in the microdissected granulomas of Crohn's disease. Modern pathology. 2004;17: 1289-1294.

247. Bernstein CN, Nayar G, Hamel A, et al. Study of animal-borne infections in the mucosas of patients with inflammatory bowel disease and population-based controls. J Clin Microbiol. 2003;41: 4986-4990.

248. Suenaga K, Yokoyama Y, Okazaki K, et al. Mycobacteria in the intestine of Japanese patients with inflammatory bowel disease. Am J Gastroenterol. 1995;90: 76.

249. Walmsley R, Ibbotson J, Chahal H, et al. Antibodies against Mycobacterium paratuberculosis in Crohn's disease. QJM. 1996;89: 217-222.

250. Ishii K, Ohkusa T, Takashimizu I. Serum antibodies to Mycobacterium paratuberculosis in Japanese patients with Crohn’s disease: an immunoblot assay. Gastroenterology. 1999;116: G3220.

251. Jones P, Farver T, Beaman B, et al. Crohn's disease in people exposed to clinical cases of bovine paratuberculosis. Epidemiol Infect. 2006;134: 49-56.

252. Järnerot G, Rolny P, Wickbom P, et al. Antimycobacterial therapy ineffective in Crohn's disease after a year. Lancet. 1989;1: 164-165.

253. Hampson S, Parker M, Saverymuttu S, et al. Results of quadruple antimycobacterial chemotherapy in 17 Crohn's disease patients completing six months treatment. Gastroenterology. 1988;94: 170.

254. Picciotto A, Gesu G, Schito G, et al. Antimycobacterial chemotherapy in two cases of inflammatory bowel disease. Lancet. 1988;1: 536-537.

255. Prantera C, Kohn A, Mangiarotti R, et al. Antimycobacterial therapy in Crohn's disease: results of a controlled, double-blind trial with a multiple antibiotic regimen. Am J Gastroenterol. 1994;89: 513-518.

185

References

256. Prantera C, Zannoni F, Scribano M, et al. An antibiotic regimen for the treatment of active Crohn's disease: a randomized, controlled clinical trial of metronidazole plus ciprofloxacin. Am J Gastroenterol. 1996;91: 328.

257. Swift G, Srivastava E, Stone R, et al. Controlled trial of anti-tuberculous chemotherapy for two years in Crohn's disease. Gut. 1994;35: 363-368.

258. Thomas G, Swift G, Green J, et al. Controlled trial of antituberculous chemotherapy in Crohn’s disease: a five year follow up study. Gut. 1998;42: 497.

259. Gui G, Thomas P, Tizard M, et al. Two-year-outcomes analysis of Crohn's disease treated with rifabutin and macrolide antibiotics. J Antimicrob Chemother. 1997;39: 393-400.

260. Douglass A, Cann P, Bramble M. An open pilot study of antimicrobial therapy in patients with unresponsive Crohn’s disease. Gut. 2000;46: A11.

261. Shafran I, Kugler L, Kl-Zaatari F, et al. Open clinical trial of rifabutin and clarithromycin therapy in Crohn's disease. Digestive and Liver Disease. 2002;34: 22- 28.

262. Borody T, Leis S, Warren E, et al. Treatment of severe Crohn's disease using antimycobacterial triple therapy--approaching a cure. Digestive and Liver Disease. 2002;34: 29-38.

263. Darfeuille-Michaud A, Colombel JF. Pathogenic Escherichia coli in inflammatory bowel diseases:: Proceedings of the 1st International Meeting on E. coli and IBD, June 2007, Lille, France. J Crohns Colitis. 2008;2: 255-262.

264. Verma R, Verma AK, Ahuja V, et al. Real-time analysis of mucosal flora in patients with inflammatory bowel disease in India. J Clin Microbiol. 2010;48: 4279- 4282.

265. Li Q, Wang C, Tang C, et al. Molecular-Phylogenetic Characterization of the Microbiota in Ulcerated and Non-Ulcerated Regions in the Patients with Crohn's Disease. PloS one. 2012;7: e34939.

266. Martinez‐Medina M, Aldeguer X, Gonzalez‐Huix F, et al. Abnormal microbiota composition in the ileocolonic mucosa of Crohn's disease patients as revealed by

186

References

polymerase chain reaction‐denaturing gradient gel electrophoresis. Inflamm Bowel Dis. 2006;12: 1136-1145.

267. 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. The ISME journal. 2007;1: 403-418.

268. 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.

269. Darfeuille-Michaud A, Neut C, Barnich N, et al. Presence of adherent Escherichia coli strains in ileal mucosa of patients with Crohn's disease. Gastroenterology. 1998;115: 1405-1413.

270. Boudeau J, Glasser AL, Masseret E, et al. Invasive Ability of an Escherichia coli Strain Isolated from the Ileal Mucosa of a Patient with Crohn’s Disease. Infect Immun. 1999;67: 4499-4509.

271. 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: 412-421.

272. Martin HM, Campbell BJ, Hart CA, et al. Enhanced Escherichia coli adherence and invasion in Crohn's disease and colon cancer 1. Gastroenterology. 2004;127: 80- 93.

273. Mylonaki M, Rayment NB, Rampton DS, et al. Molecular characterization of rectal mucosa‐associated bacterial flora in inflammatory bowel disease. Inflamm Bowel Dis. 2005;11: 481-487.

274. Alpern J, Sasaki M, Sitaraman S, et al. Invasive E coli strains are increased in Crohn’s disease. Gastroenterology. 2006;130: A362.

275. Kotlowski R, Bernstein CN, Sepehri S, et al. High prevalence of Escherichia coli belonging to the B2 D phylogenetic group in inflammatory bowel disease. Gut. 2007;56: 669-675.

187

References

276. Sasaki M, Sitaraman SV, Babbin BA, et al. Invasive Escherichia coli are a feature of Crohn's disease. Laboratory Investigation. 2007;87: 1042-1054.

277. La Ferla K, Seegert D, Schreiber S. Activation of NF-κ B in intestinal epithelial cells by E. coli strains isolated from the colonic mucosa of IBD patients. Int J Colorectal Dis. 2004;19: 334-342.

278. Glasser AL, Boudeau J, Barnich N, et al. Adherent invasive Escherichia coli strains from patients with Crohn's disease survive and replicate within macrophages without inducing host cell death. Infect Immun. 2001;69: 5529.

279. Finegold SM, George WL. Anaerobic Infections in Humans. : Academic Press, Inc.; 1989.

280. Brondz I, Olsen I, Haapasalo M, et al. Multivariate analyses of fatty acid data from whole-cell methanolysates of Prevotella, Bacteroides and Porphyromonas spp. J Gen Microbiol. 1991;137: 1445-1452.

281. Wexler HM. Bacteroides: the good, the bad, and the nitty-gritty. Clin Microbiol Rev. 2007;20: 593-621.

282. Brook I. Indigenous microbial flora of humans. In: Surgical Infectious Diseases. 1995;3rd ed.

283. Bamba T, Matsuda H, Endo M, et al. The pathogenic role of Bacteroides vulgatus in patients with ulcerative colitis. J Gastroenterol. 1995;30 Suppl 8: 45-47.

284. Matsuda H, Fujiyama Y, Andoh A, et al. Characterization of antibody responses against rectal mucosa‐associated bacterial flora in patients with ulcerative colitis. J Gastroenterol Hepatol. 2000;15: 61-68.

285. Fujita H, Eishi Y, Ishige I, et al. Quantitative analysis of bacterial DNA from Mycobacteria spp., Bacteroides vulgatus, and Escherichia coli in tissue samples from patients with inflammatory bowel diseases. J Gastroenterol. 2002;37: 509-516.

286. Kim S, Tonkonogy S, Albright C, et al. Different host genetic backgrounds determine disease phenotypes induced by selective bacterial colonization. Gastroenterology. 2005;128: A512.

188

References

287. Sellon RK, Tonkonogy S, Schultz M, et al. Resident enteric bacteria are necessary for development of spontaneous colitis and immune system activation in interleukin- 10-deficient mice. Infect Immun. 1998;66: 5224-5231.

288. Waidmann M, Bechtold O, Frick J, et al. Bacteroides vulgatus protects against Escherichia coli-induced colitis in gnotobiotic interleukin-2-deficient mice. Gastroenterology. 2003;125: 162-177.

289. Bohn E, Bechtold O, Zahir N, et al. Host gene expression in the colon of gnotobiotic interleukin‐2‐deficient mice colonized with commensal colitogenic or noncolitogenic bacterial strains: Common patterns and bacteria strain specific signatures. Inflamm Bowel Dis. 2006;12: 853-862.

290. Gironella M, Iovanna JL, Sans M, et al. Anti-inflammatory effects of pancreatitis associated protein in inflammatory bowel disease. Gut. 2005;54: 1244-1253.

291. Mazmanian SK, Round JL, Kasper DL. A microbial symbiosis factor prevents intestinal inflammatory disease. Nature. 2008;453: 620-625.

292. Skirrow MB. John McFadyean and the centenary of the first isolation of Campylobacter species. Clin Infect Dis. 2006;43: 1213.

293. Jones F, Orcutt M, Little RB. Vibrios (Vibrio jejuni, n. sp.) associated with intestinal disorders of cows and calves. J Exp Med. 1931;53: 853.

294. Levy A. A gastro-enteritis outbreak probably due to a bovine strain of vibrio. Yale J Biol Med. 1946;18: 243.

295. Boivin J, Piemont Y, Christmann D, et al. Septicémie à Campylobacter fetus subsp. intestinalis avec localisation méningée. Médecine et Maladies Infectieuses. 1982;12: 86-88.

296. Butzler JP. Campylobacter, from obscurity to celebrity. Clin Microbiol Infect. 2004;10: 868-876.

297. King EO. Human infections with Vibrio fetus and a closely related vibrio. J Infect Dis. 1957;101: 119-128.

189

References

298. Sebald M, Veron M. Teneur en bases de l'ADN et classification des vibrions. Ann Inst Pasteur. 1963;105: 897-910.

299. Sebald M, Veron M. Base DNA Content and Classification of Vibrios. Ann Inst Pasteur. 1963;105: 897-910.

300. Veron M, Chatelain R. Taxonomic study of the genus Campylobacter Sebald and Véron and designation of the neotype strain for the type species, Campylobacter fetus (Smith and Taylor) Sebald and Véron. Int J Syst Evol Microbiol. 1973;23: 122.

301. Dekeyser P, Gossuin-Detrain M, Butzler J, et al. Acute enteritis due to related vibrio: first positive stool cultures. J Infect Dis. 1972;125: 390.

302. Skirrow M. Campylobacter enteritis: a" new" disease. Br Med J. 1977;2: 9.

303. Tanner ACR, Badger S, Lai CH, et al. Wolinella gen. nov., Wolinella succinogenes (Vibrio succinogenes Wolin et al.) comb. nov., and Description of Bacteroides gracilis sp. nov., Wolinella recta sp. nov., Campylobacter concisus sp. nov., and Eikenella corrodens from Humans with Periodontal Disease. Int J Syst Bacteriol. 1981;31: 432.

304. Zhang L, Man SM, Day AS, et al. Detection and isolation of Campylobacter species other than C. jejuni from children with Crohn's disease. J Clin Microbiol. 2009;47: 453.

305. Garrity G, Bell J, Lilburn T. Epsilonproteobacteria. Brenner D, Krieg N, Staley J, et al, eds. Bergey's Manual of Systematic Bacteriology. second edition, Volume two, Part C ed. New York, USA: Springer; 2005:1145.

306. Vandamme P, Dewhirst F, Paster B, et al. Campylobacteraceae. Brenner D, Krieg N, Staley J, et al, eds. Bergey's Manual of Systematic Bacteriology second Edition, Volume two, Part C ed. New York, USA: Springer; 2005:1145-1160.

307. Willey MJ, Sherwood ML, Woolverton JC,. Prescott’s Microbiology. Journal of Microbiology & Biology Education. 2011;8th edition.

308. Debruyne L, Gevers D, Vandamme P. Taxonomy of the Family Campylobacteraceae. Nachamkin I, Szymanski C, Blaser MJ, eds. Campylobacter. 3rd ed. Washington DC, USA: ASM Press; 2008:3-25.

190

References

309. Vandamme P. Taxonomy of the family Campylobacteraceae. Campylobacter. 2000;2: 3-26.

310. Vandamme P, Debruyne L, De Brandt E, et al. Reclassification of Bacteroides ureolyticus as Campylobacter ureolyticus comb. nov., and emended description of the genus Campylobacter. Int J Syst Evol Microbiol. 2010;60: 2016-2022.

311. Smibert RM. The genus campylobacter. Annual Reviews in Microbiology. 1978;32: 673-709.

312. Nachamkin I, Szymanski CM, Blaser MJ. Campylobacter. Amer Society for Microbiology; 2008.

313. Penner J. The genus Campylobacter: a decade of progress. Clin Microbiol Rev. 1988;1: 157.

314. Wassenaar TM, Blaser MJ. Pathophysiology of Campylobacter jejuni infections of humans. Microb Infect. 1999;1: 1023-1033.

315. Man SM, Kaakoush NO, Leach ST, et al. Host attachment, invasion, and stimulation of proinflammatory cytokines by Campylobacter concisus and other non- Campylobacter jejuni Campylobacter species. J Infect Dis. 2010;202: 1855.

316. On S. Taxonomy, Phylogeny, and Methods for the Identificationof Campylobacter Species. Ketley JM, Konkel ME, eds. Campylobacter: Molecular and Cellular Biology. : Taylor & Francis; 2005:13-42.

317. Ketley JM, Konkel ME. Campylobacter: Molecular and Cellular Biology. : Taylor & Francis; 2005.

318. Van Etterijck R, Breynaert J, Revets H, et al. Isolation of Campylobacter concisus from feces of children with and without diarrhea. J Clin Microbiol. 1996;34: 2304.

319. Aabenhus R, Permin H, Andersen LP. Characterization and subgrouping of Campylobacter concisus strains using protein profiles, conventional biochemical testing and antibiotic susceptibility. Eur J Gastroenterol Hepatol. 2005;17: 1019.

191

References

320. Engberg J, Bang D, Aabenhus R, et al. Campylobacter concisus: an evaluation of certain phenotypic and genotypic characteristics. Clin Microbiol Infect. 2005;11: 288- 295.

321. Zhang L, Man SM, Day AS, et al. Detection and isolation of Campylobacter species other than C. jejuni from children with Crohn's disease. J Clin Microbiol. 2009;47: 453.

322. Lastovica AJ. Clinical relevance of Campylobacter concisus isolated from pediatric patients. J Clin Microbiol. 2009;47: 2360-2360.

323. Paster BJ, Gibbons RJ. Chemotactic response to formate by Campylobacter concisus and its potential role in gingival colonization. Infect Immun. 1986;52: 378.

324. Kamma JJ, Diamanti‐Kipioti A, Nakou M, et al. Profile of subgingival microbiota in children with primary dentition. J Periodont Res. 2000;35: 33-41.

325. Kamma J, Diamanti‐Kipioti A, Nakou M, et al. Profile of subgingival microbiota in children with mixed dentition. Oral Microbiol Immunol. 2000;15: 103-111.

326. Taubman M, Haffajee A, Socransky S, et al. Longitudinal monitoring of humoral antibody in subjects with destructive periodontal diseases. J Periodont Res. 1992;27: 511-521.

327. Kamma JJ, Nakou M, Persson RG. Association of early onset periodontitis microbiota with aspartate aminotransferase activity in gingival crevicular fluid. J Clin Periodontol. 2001;28: 1096-1105.

328. Zhang L, Budiman V, Day AS, et al. Isolation and detection of Campylobacter concisus from saliva of healthy individuals and patients with inflammatory bowel disease. J Clin Microbiol. 2010;48: 2965.

329. Aabenhus R, Permin H, On SLW, et al. Prevalence of Campylobacter concisus in diarrhoea of immunocompromised patients. Scand J Infect Dis. 2002;34: 248-252.

330. Lindblom GB, Sjögren E, Hansson-Westerberg J, et al. Campylobacter upsaliensis, C. sputorum and C. concisus as common causes of diarrhoea in Swedish children. Scand J Infect Dis. 1995;27: 187-188.

192

References

331. Musmanno R, Russi M, Figura N, et al. Unusual species of campylobacters isolated in the Siena Tuscany area, Italy. New Microbiol. 1998;21: 15.

332. Russell J. Campylobacter like organisms: investigation of clinical and phenotypical aspects. Appl Microbiol Biotechnol. 1995: 15-41.

333. Russell J, Ward P. Adhesion and invasion of HEp2 cells by Campylobacter concisus from children with diarrhoea. Campylobacter, Helicobacter and Related Organisms. 1998: 327-330.

334. Istivan T. Molecular Characterisation of Campylobacter Concisus: A Potential Etiological Agent of Gastroenteritis in Children: PhD Thesis. RMI uni. 2005.

335. Lastovica AJ. Efficient isolation of campylobacteria from stools. J Clin Microbiol. 2000;38: 2798.

336. Engberg J, On SLW, Harrington CS, et al. Prevalence of Campylobacter, Arcobacter, Helicobacter, and Sutterella spp. in human fecal samples as estimated by a reevaluation of isolation methods for campylobacters. J Clin Microbiol. 2000;38: 286.

337. Nielsen HL, Ejlertsen T, Engberg J, et al. High incidence of Campylobacter concisus in gastroenteritis in North Jutland, Denmark: a population‐based study. Clin Microbiol Infect. 2012.

338. Man SM, Zhang L, Day AS, et al. Campylobacter concisus and other Campylobacter species in children with newly diagnosed Crohn's disease. Inflamm Bowel Dis. 2010;16: 1008-1016.

339. Mahendran V, Riordan SM, Grimm MC, et al. Prevalence of Campylobacter Species in Adult Crohn's Disease and the Preferential Colonization Sites of Campylobacter Species in the Human Intestine. PloS one. 2011;6: e25417.

340. Mukhopadhya I, Thomson JM, Hansen R, et al. Detection of Campylobacter concisus and other Campylobacter species in colonic biopsies from adults with ulcerative colitis. PLoS one. 2011;6: e21490.

341. Kamma JJ, Nakou M, Manti FA. Microbiota of rapidly progressive periodontitis lesions in association with clinical parameters. J Periodontol. 1994;65: 1073-1078.

193

References

342. Moore W, Holdeman L, Cato E, et al. Comparative bacteriology of juvenile periodontitis. Infect Immun. 1985;48: 507.

343. Macuch P, Tanner A. Campylobacter species in health, gingivitis, and periodontitis. J Dent Res. 2000;79: 785.

344. Mahendran V. Investigation of the possible role of Campylobacter concisus in inflamatory bowel disease: MSc thesis. UNSW. 2012: 34.

345. Vandamme P, Falsen E, Pot B, et al. Identification of EF group 22 campylobacters from gastroenteritis cases as Campylobacter concisus. J Clin Microbiol. 1989;27: 1775-1781.

346. Bastyns K, Chapelle S, Vandamme P, et al. Specific detection of Campylobacter concisus by PCR amplification of 23S rDNA areas. Mol Cell Probes. 1995;9: 247-250.

347. Aabenhus R, On SLW, Siemer BL, et al. Delineation of Campylobacter concisus genomospecies by amplified fragment length polymorphism analysis and correlation of results with clinical data. J Clin Microbiol. 2005;43: 5091-5096.

348. Kaakoush NO, Deshpande NP, Wilkins MR, et al. The pathogenic potential of Campylobacter concisus strains associated with chronic intestinal diseases. PloS one. 2011;6: e29045.

349. Lastovica AJ. Emerging Campylobacter spp.: The tip of the iceberg. Clin Microbiol Newsl. 2006;28: 49-56.

350. Larson CL, Christensen JE, Pacheco SA, et al. Campylobacter jejuni secretes proteins via the flagellar type III secretion system that contribute to host cell invasion and gastroenteritis. Campylobacter, 3rd ed.ASM Press, Washington, DC. 2008: 315- 332.

351. Ismail Y, Mahendran V, Octavia S, et al. Investigation of the Enteric Pathogenic Potential of Oral Campylobacter concisus Strains Isolated from Patients with Inflammatory Bowel Disease. PloS one. 2012;7: e38217.

352. Habashneh R, Khader Y, Alhumouz M, et al. The association between inflammatory bowel disease and periodontitis among Jordanians: a case–control study. J Periodont Res. 2011.

194

References

353. Ochman H, Lawrence JG, Groisman EA. Lateral gene transfer and the nature of bacterial innovation. Nature. 2000;405: 299-304.

354. Jin S, Joe A, Lynett J, et al. JlpA, a novel surface‐exposed lipoprotein specific to Campylobacter jejuni, mediates adherence to host epithelial cells. Mol Microbiol. 2001;39: 1225-1236.

355. Konkel ME, Garvis SG, Tipton SL, et al. Identification and molecular cloning of a gene encoding a fibronectin‐binding protein (CadF) from Campylobacter jejuni. Mol Microbiol. 1997;24: 953-963.

356. Pei Z, Burucoa C, Grignon B, et al. Mutation in the peb1A Locus ofCampylobacter jejuni Reduces Interactions with Epithelial Cells and Intestinal Colonization of Mice. Infect Immun. 1998;66: 938-943.

357. Everest P, Goossens H, Butzler JP, et al. Differentiated Caco-2 cells as a model for enteric invasion by Campylobacter jejuni and C. coli. J Med Microbiol. 1992;37: 319-325.

358. Konkel ME, Cieplak Jr W. Altered synthetic response of Campylobacter jejuni to cocultivation with human epithelial cells is associated with enhanced internalization. Infect Immun. 1992;60: 4945-4949.

359. Konkel ME, Mead DJ, Cieplak W. Kinetic and antigenic characterization of altered protein synthesis by Campylobacter jejuni during cultivation with human epithelial cells. J Infect Dis. 1993;168: 948.

360. Rivera-Amill V, Kim BJ, Seshu J, et al. Secretion of the virulence-associated Campylobacter invasion antigens from Campylobacter jejuni requires a stimulatory signal. J Infect Dis. 2001;183: 1607-1616.

361. Nielsen HL, Nielsen H, Ejlertsen T, et al. Oral and fecal Campylobacter concisus strains perturb barrier function by apoptosis induction in HT-29/B6 intestinal epithelial cells. PloS One. 2011;6: e23858.

362. Mayer H, Bhat UR, Masoud H, et al. Bacterial lipopolysaccharides. Pure Appl Chem. 1989;61: 1271-1282.

195

References

363. Rietschel ET, Kirikae T, Schade FU, et al. Bacterial endotoxin: molecular relationships of structure to activity and function. The FASEB Journal. 1994;8: 217- 225.

364. Raetz CRH, Whitfield C. Lipopolysaccharide endotoxins. Annu Rev Biochem. 2002;71: 635.

365. Loppnow H, Brade H, Durrbaum I, et al. IL-1 induction-capacity of defined lipopolysaccharide partial structures. J Immunol. 1989;142: 3229.

366. Kotani S, Takada H, Tsujimoto M, et al. Synthetic lipid A with endotoxic and related biological activities comparable to those of a natural lipid A from an Escherichia coli re-mutant. Infect Immun. 1985;49: 225-237.

367. Rietschel ET, Westphal O. Endotoxin: historical perspectives. Endotoxin in Health and Disease. 1999: 1-30.

368. Galanos C. Physical state and biological activity of lipopolysaccharides. Toxicity and immunogenicity of the lipid A component. Z Immunitatsforsch Exp Klin Immunol. 1975;149: 214.

369. Dixon D, Darveau R. Lipopolysaccharide heterogeneity: innate host responses to bacterial modification of lipid a structure. J Dent Res. 2005;84: 584-595.

370. Guo L, Lim KB, Gunn JS, et al. Regulation of lipid A modifications by Salmonella typhimurium virulence genes phoP-phoQ. Science. 1997;276: 250.

371. Trent MS, Pabich W, Raetz CRH, et al. A PhoP/PhoQ-induced Lipase (PagL) That Catalyzes 3-O-Deacylation of Lipid A Precursors in Membranes ofSalmonella typhimurium. J Biol Chem. 2001;276: 9083-9092.

372. Kawahara K, Tsukano H, Watanabe H, et al. Modification of the structure and activity of lipid A in Yersinia pestis lipopolysaccharide by growth temperature. Infect Immun. 2002;70: 4092-4098.

373. Rebeil R, Ernst RK, Gowen BB, et al. Variation in lipid A structure in the pathogenic yersiniae. Mol Microbiol. 2004;52: 1363-1373.

196

References

374. Soncini FC, Vescovi EG, Solomon F, et al. Molecular basis of the magnesium deprivation response in Salmonella typhimurium: identification of PhoP-regulated genes. J Bacteriol. 1996;178: 5092-5099.

375. Miller SI, Kukral AM, Mekalanos JJ. A two-component regulatory system (phoP phoQ) controls Salmonella typhimurium virulence. Proc Natl Acad Sci U S A. 1989;86: 5054.

376. Gunn JS, Miller SI. PhoP-PhoQ activates transcription of pmrAB, encoding a two-component regulatory system involved in Salmonella typhimurium antimicrobial peptide resistance. J Bacteriol. 1996;178: 6857-6864.

377. Roland KL, Martin LE, Esther CR, et al. Spontaneous pmrA mutants of Salmonella typhimurium LT2 define a new two-component regulatory system with a possible role in virulence. J Bacteriol. 1993;175: 4154-4164.

378. Perry RD, Fetherston JD. Yersinia pestis--etiologic agent of plague. Clin Microbiol Rev. 1997;10: 35-66.

379. Reeves PR, Hobbs M, Valvano MA, et al. Bacterial polysaccharide synthesis and gene nomenclature. Trends Microbiol. 1996;4: 495-503.

380. Frirdich E, Whitfield C. Review: Lipopolysaccharide inner core oligosaccharide structure and outer membrane stability in human pathogens belonging to the Enterobacteriaceae. J Endotoxin Res. 2005;11: 133-144.

381. Lüderitz O, Freudenberg MA, Galanos C, et al. Lipopolysaccharides of gram- negative bacteria. Curr Top Membr Transp. 1982;17: 79-151.

382. Moran AP, Kosunen TU. Serological analysis of the heat‐stable antigens involved in serotyping Campylobacter jejuni and Campylobacter coli. APMIS. 1989;97: 253- 260.

383. Penner J, Hennessy J, Congi R. Serotyping ofCampylobacter jejuni andCampylobacter coli on the basis of thermostable antigens. Eur J Clin Microbiol Infect Dis. 1983;2: 378-383.

384. Perez GIP, Blaser M. Lipopolysaccharide characteristics of pathogenic campylobacters. Infect Immun. 1985;47: 353-359.

197

References

385. Hackett J, Wyk P, Reeves P, et al. Mediation of serum resistance in Salmonella typhimurium by an 11-kilodalton polypeptide encoded by the cryptic plasmid. J Infect Dis. 1987;155: 540-549.

386. Stinavage P, Martin LE, Spitznagel JK. O antigen and lipid A phosphoryl groups in resistance of Salmonella typhimurium LT-2 to nonoxidative killing in human polymorphonuclear neutrophils. Infect Immun. 1989;57: 3894-3900.

387. Movat H, Cybulsky M, Colditz I, et al. Acute inflammation in gram-negative infection: endotoxin, interleukin 1, tumor necrosis factor, and neutrophils. Euro PMC. 1987;46: 97-104.

388. Ruff MR, Gifford GE. Purification and physico-chemical characterization of rabbit tumor necrosis factor. J Immunol. 1980;125: 1671-1677.

389. Mathison JC, Wolfson E, Ulevitch RJ. Participation of tumor necrosis factor in the mediation of gram negative bacterial lipopolysaccharide-induced injury in rabbits. J Clin Invest. 1988;81: 1925.

390. Beutler B, Milsark I, Cerami A. Passive immunization against cachectin/tumor necrosis factor protects mice from lethal effect of endotoxin. Science. 1985;229: 869- 871.

391. Tabor DR, Burchett SK, Jacobs RF. Enhanced production of monokines by canine alveolar macrophages in response to endotoxin-induced shock. Exp Biol Med. 1988;187: 408-415.

392. Dinarello CA, Cannon JG, Wolff SM, et al. Tumor necrosis factor (cachectin) is an endogenous pyrogen and induces production of interleukin 1. J Exp Med. 1986;163: 1433-1450.

393. Loppnow H, Libby P, Freudenberg M, et al. Cytokine induction by lipopolysaccharide (LPS) corresponds to lethal toxicity and is inhibited by nontoxic Rhodobacter capsulatus LPS. Infect Immun. 1990;58: 3743-3750.

394. Michie HR, Manogue KR, Spriggs DR, et al. Detection of circulating tumor necrosis factor after endotoxin administration. N Engl J Med. 1988;318: 1481-1486.

198

References

395. Tapping RI, Akashi S, Miyake K, et al. Toll-like receptor 4, but not toll-like receptor 2, is a signaling receptor for Escherichia and Salmonella lipopolysaccharides. J Immunol. 2000;165: 5780.

396. Akashi S, Shimazu R, Ogata H, et al. Cutting edge: cell surface expression and lipopolysaccharide signaling via the toll-like receptor 4-MD-2 complex on mouse peritoneal macrophages. J Immunol. 2000;164: 3471.

397. Lehner MD, Ittner J, Bundschuh DS, et al. Improved innate immunity of endotoxin-tolerant mice increases resistance to Salmonella enterica serovar typhimurium infection despite attenuated cytokine response. Infect Immun. 2001;69: 463-471.

398. Moran AP. Biological and serological characterization of Campylobacter jejuni lipopolysaccharides with deviating core and lipid A structures. FEMS Immunol Med Microbiol. 1995;11: 121-130.

399. Moran AP, O'Malley DT, Vuopio‐Varkila J, et al. Biological characterization of Campylobacter fetus lipopolysaccharides. FEMS Immunol Med Microbiol. 1996;15: 43-50.

400. Hancock I, Poxton I. Bacterial cell surface techniques. John Wiley & Sons Inc.; 1988.

401. Monteiro M. Helicobacter pylori: a wolf in sheep's clothing: the glycotype families of Helicobacter pylori lipopolysaccharides expressing histo-blood groups: structure, biosynthesis, and role in pathogenesis. Horton D, ed. Advances in Carbohydrate Chemistry and Biochemistry. volume 57 ed. California, USA: Academic Press; 2001:99-154.

402. Fomsgaard A, Freudenberg M, Galanos C. Modification of the silver staining technique to detect lipopolysaccharide in polyacrylamide gels. J Clin Microbiol. 1990;28: 2627-2631.

403. Mandatori R, Penner JL. Structural and antigenic properties of Campylobacter coli lipopolysaccharides. Infect Immun. 1989;57: 3506-3511.

199

References

404. Perez-Perez GI, Blaser MJ. Conservation and diversity of Campylobacter pyloridis major antigens. Infect Immun. 1987;55: 1256-1263.

405. Perez G, Hopkins J, Blaser M. Antigenic heterogeneity of lipopolysaccharides from Campylobacter jejuni and Campylobacter fetus. Infect Immun. 1985;48: 528-533.

406. Perez-Perez GI, Hopkins JA, Blaser MJ. Lipopolysaccharide structures in Enterobacteriaceae, Pseudomonas aeruginosa, and Vibrio cholerae are immunologically related to Campylobacter spp. Infect Immun. 1986;51: 204-208.

407. Blake Jr D, Russell R. Demonstration of lipopolysaccharide with O- polysaccharide chains among different heat-stable serotypes of Campylobacter jejuni by silver staining of polyacrylamide gels. Infect Immun. 1993;61: 5384-5387.

408. Preston MA, Penner J. Structural and antigenic properties of lipopolysaccharides from serotype reference strains of Campylobacter jejuni. Infect Immun. 1987;55: 1806- 1812.

409. Fomsgaard A, Shand GH, Freudenberg MA, et al. Antibodies from chronically infected cystic fibrosis patients react with lipopolysaccharides extracted by new micromethods from all serotypes of Pseudomonas aeruginosa. APMIS. 1993;101: 101- 112.

410. Huber M, Kalis C, Keck S, et al. R‐form LPS, the master key to the activation ofTLR4/MD‐2‐positive cells. Eur J Immunol

. 2006;36: 701-711.

411. Jerala R. Structural biology of the LPS recognition. Int J Med Microbiol. 2007;297: 353-363.

412. Duerr CU, Zenk SF, Chassin C, et al. O-antigen delays lipopolysaccharide recognition and impairs antibacterial host defense in murine intestinal epithelial cells. PloS pathogens. 2009;5: e1000567.

413. Frolova L, Drastich P, Rossmann P, et al. Expression of Toll-like receptor 2 (TLR2), TLR4, and CD14 in biopsy samples of patients with inflammatory bowel diseases: upregulated expression of TLR2 in terminal ileum of patients with ulcerative colitis. J Histochem Cytochem. 2008;56: 267.

200

References

414. Giwercman B, Fomsgaard A, Mansa B, et al. Polycrylamide gel electrophoresis analysis of lipopolysaccharide from Pseudomonas aeruginosa growing planktonically and as biofilm. FEMS Microbiol Lett. 1992;89: 225-229.

415. Rietschel ET, Brade L, Lindner B, et al. Biochemistry of lipopolysaccharides. Bacterial endotoxic lipopolysaccharides. 1992;1: 3-41.

416. Newell DG. Campylobacter concisus: an emerging pathogen?. Eur J Gastroenterol Hepatol. 2005;17: 1013.

417. Snijders F, Kuijper E, De Wever B, et al. Prevalence of Campylobacter-associated diarrhea among patients infected with human immunodeficiency virus. Clin Infect Dis. 1997;24: 1107.

418. Podolsky DK. Inflammatory Bowel Disease. N Engl J Med. 2002;347: 417-429.

419. Sartor RB. Microbial influences in inflammatory bowel diseases. Gastroenterology. 2008;134: 577-594.

420. Fiocchi C. Inflammatory bowel disease: etiology and pathogenesis. Gastroenterology. 1998;115: 182-205.

421. Sturm A, de Souza HSP, Fiocchi C. Mucosal T cell proliferation and apoptosis in inflammatory bowel disease. Curr Drug Targets. 2008;9: 381-387.

422. Rakoff-Nahoum S, Paglino J, Eslami-Varzaneh F, et al. Recognition of commensal microflora by toll-like receptors is required for intestinal homeostasis. Cell. 2004;118: 229-241.

423. Lee J, Mo JH, Katakura K, et al. Maintenance of colonic homeostasis by distinctive apical TLR9 signalling in intestinal epithelial cells. Nat Cell Biol. 2006;8: 1327-1336.

424. Akira S, Uematsu S, Takeuchi O. Pathogen recognition and innate immunity. Cell. 2006;124: 783-801.

425. Vamadevan AS, Fukata M, Arnold ET, et al. Regulation of Toll-like receptor 4- associated MD-2 in intestinal epithelial cells: a comprehensive analysis. Innate immunity. 2010;16: 93.

201

References

426. Szebeni B, Veres G, Dezsofi A, et al. Increased expression of Toll‐like receptor (TLR) 2 and TLR4 in the colonic mucosa of children with inflammatory bowel disease. Clin Exp Immunol. 2008;151: 34-41.

427. Stanislawowski M, Wierzbicki P, Golab A, et al. Decreased Toll-like receptor-5 (TLR-5) expression in the mucosa of ulcerative colitis patients. J Physiol Pharmacol. 2009;60: 71-75.

428. Takeda K, Akira S. Toll-like receptors in innate immunity. Int Immunol. 2005;17: 1-14.

429. Levin A, Shibolet O. Toll-like receptors in inflammatory bowel disease-stepping into uncharted territory. World J Gastroenterol. 2008;14: 5149.

430. Aderem A, Ulevitch RJ. Toll-like receptors in the induction of the innate immune response. Nature. 2000;406: 782-787.

431. Teghanemt A, Widstrom RL, Gioannini TL, et al. Isolation of Monomeric and Dimeric Secreted MD-2. J Biol Chem. 2008;283: 21881-21889.

432. Shimazu R, Akashi S, Ogata H, et al. MD-2, a molecule that confers lipopolysaccharide responsiveness on Toll-like receptor 4. J Exp Med. 1999;189: 1777.

433. Ulevitch R, Tobias P. Receptor-dependent mechanisms of cell stimulation by bacterial endotoxin. Annu Rev Immunol. 1995;13: 437-457.

434. Chow JC, Young DW, Golenbock DT, et al. Toll-like receptor-4 mediates lipopolysaccharide-induced signal transduction. J Biol Chem. 1999;274: 10689.

435. Gewirtz AT, Navas TA, Lyons S, et al. Cutting edge: bacterial flagellin activates basolaterally expressed TLR5 to induce epithelial proinflammatory gene expression. J Immunol. 2001;167: 1882-1885.

436. Andersen-Nissen E, Smith KD, Strobe KL, et al. Evasion of Toll-like receptor 5 by flagellated bacteria. Proc Natl Acad Sci U S A. 2005;102: 9247.

437. Schwandner R, Dziarski R, Wesche H, et al. Peptidoglycan-and lipoteichoic acid- induced cell activation is mediated by toll-like receptor 2. J Biol Chem. 1999;274: 17406-17409.

202

References

438. Takeuchi O, Kaufmann A, Grote K, et al. Cutting edge: preferentially the R- stereoisomer of the mycoplasmal lipopeptide macrophage-activating lipopeptide-2 activates immune cells through a toll-like receptor 2-and MyD88-dependent signaling pathway. J Immunol. 2000;164: 554.

439. Lien E, Sellati TJ, Yoshimura A, et al. Toll-like receptor 2 functions as a pattern recognition receptor for diverse bacterial products. J Biol Chem. 1999;274: 33419- 33425.

440. Cario E, Podolsky DK. Intestinal epithelial Tollerance versus intollerance of commensals. Mol Immunol. 2005;42: 887-893.

441. Abreu MT, Arnold ET, Thomas LS, et al. TLR4 and MD-2 expression is regulated by immune-mediated signals in human intestinal epithelial cells. J Biol Chem. 2002;277: 20431-20437.

442. Hsu RYC, Chan CHF, Spicer JD, et al. LPS-Induced TLR4 Signaling in Human Colorectal Cancer Cells Increases β1 Integrin-Mediated Cell Adhesion and Liver Metastasis. Cancer Res. 2011;71: 1989-1998.

443. Lenoir C, Sapin C, Broquet AH, et al. MD-2 controls bacterial lipopolysaccharide hyporesponsiveness in human intestinal epithelial cells. Life Sci. 2008;82: 519-528.

444. Otte JM, Cario E, Podolsky DK. Mechanisms of cross hyporesponsiveness to Toll-like receptor bacterial ligands in intestinal epithelial cells. Gastroenterology. 2004;126: 1054-1070.

445. Cario E, Rosenberg IM, Brandwein SL, et al. Lipopolysaccharide activates distinct signaling pathways in intestinal epithelial cell lines expressing Toll-like receptors. J Immunol. 2000;164: 966.

446. Abreu MT, Vora P, Faure E, et al. Decreased expression of Toll-like receptor-4 and MD-2 correlates with intestinal epithelial cell protection against dysregulated proinflammatory gene expression in response to bacterial lipopolysaccharide. J Immunol. 2001;167: 1609.

203

References

447. Taurog JD, Richardson JA, Croft J, et al. The germfree state prevents development of gut and joint inflammatory disease in HLA-B27 transgenic rats. J Exp Med. 1994;180: 2359.

448. Veltkamp C, Tonkonogy SL, De Jong YP, et al. Continuous stimulation by normal luminal bacteria is essential for the development and perpetuation of colitis in Tg [epsilon] 26 mice. Gastroenterology. 2001;120: 900-913.

449. MacDonald TT, Monteleone G. Immunity, inflammation, and allergy in the gut. Science. 2005;307: 1920-1925.

450. Conte MP, Schippa S, Zamboni I, et al. Gut-associated bacterial microbiota in paediatric patients with inflammatory bowel disease. Gut. 2006;55: 1760.

451. Ott S, Musfeldt M, Wenderoth D, et al. Reduction in diversity of the colonic mucosa associated bacterial microflora in patients with active inflammatory bowel disease. Gut. 2004;53: 685-693.

452. Swidsinski A, Ladhoff A, Pernthaler A, et al. Mucosal flora in inflammatory bowel disease. Gastroenterology. 2002;122: 44-54.

453. Krutzik SR, Sieling PA, Modlin RL. The role of Toll-like receptors in host defense against microbial infection. Curr Opin Immunol. 2001;13: 104-108.

454. Fusunyan RD, Nanthakumar NN, Baldeon ME, et al. Evidence for an innate immune response in the immature human intestine: toll-like receptors on fetal enterocytes. Pediatr Res. 2001;49: 589-593.

455. Vizoso Pinto MG, Rodriguez Gómez M, Seifert S, et al. Lactobacilli stimulate the innate immune response and modulate the TLR expression of HT29 intestinal epithelial cells in vitro. Int J Food Microbiol. 2009;133: 86-93.

456. Su B, Ceponis PJM, Lebel S, et al. Helicobacter pylori activates Toll-like receptor 4 expression in gastrointestinal epithelial cells. Infect Immun. 2003;71: 3496-3502.

457. Jones NL, Day AS, Jennings HA, et al. Helicobacter pylori induces gastric epithelial cell apoptosis in association with increased Fas receptor expression. Infect Immun. 1999;67: 4237-4242.

204

References

458. Nagai Y, Akashi S, Nagafuku M, et al. Essential role of MD-2 in LPS responsiveness and TLR4 distribution. Nat Immunol. 2002;3: 667-672.

459. Ohnishi T, Muroi M, Tanamoto K. MD-2 is necessary for the toll-like receptor 4 protein to undergo glycosylation essential for its translocation to the cell surface. Clin Diagn Lab Immunol. 2003;10: 405-410.

460. Seksik P, Sokol H, Grondin V, et al. Sera from patients with Crohn’s disease break bacterial lipopolysaccharide tolerance of human intestinal epithelial cells via MD-2 activity. Innate Immunity. 2010;16: 381-390.

461. Smith KD, Andersen-Nissen E, Hayashi F, et al. Toll-like receptor 5 recognizes a conserved site on flagellin required for protofilament formation and bacterial motility. Nat Immunol. 2003;4: 1247-1253.

462. Zeng H, Carlson AQ, Guo Y, et al. Flagellin is the major proinflammatory determinant of enteropathogenic Salmonella. J Immunol. 2003;171: 3668.

463. Salazar‐Gonzalez H, Navarro‐Garcia F. Intimate Adherence by Enteropathogenic Escherichia coli Modulates TLR5 Localization and Proinflammatory Host Response in Intestinal Epithelial Cells. Scand J Immunol. 2011;73: 268-283.

464. Zhang L, Day A, McKenzie G, et al. Nongastric Helicobacter species detected in the intestinal tract of children. J Clin Microbiol. 2006;44: 2276-2279.

465. Man SM, Zhang L, Day AS, et al. Detection of enterohepatic and gastric Helicobacter species in fecal specimens of children with Crohn's disease. Helicobacter. 2008;13: 234-238.

466. Kaakoush NO, Holmes J, Octavia S, et al. Detection of Helicobacteraceae in intestinal biopsies of children with Crohn’s disease. Helicobacter. 2010;15: 549-557.

467. Thomson JM, Hansen R, Berry SH, et al. Enterohepatic helicobacter in ulcerative colitis: potential pathogenic entities PloS one. 2011;6: e17184.

468. Atreya I, Atreya R, Neurath MF. NF-κB in inflammatory bowel disease. J Intern Med. 2008;263: 591-596.

205

References

469. Kobayashi M, Saitoh S, Tanimura N, et al. Regulatory roles for MD-2 and TLR4 in ligand-induced receptor clustering. J Immunol. 2006;176: 6211.

470. Kim H, Lim J, Kim K. Helicobacter pylori-induced Expression of Interleukin-8 and Cyclooxygenase-2 in AGS Gastric Epithelial Cells: Mediation by Nuclear Factor-κ B. Scand J Gastroenterol. 2001;36: 706-716.

471. Tak PP, Firestein GS. NF-kappaB: a key role in inflammatory diseases. J Clin Invest. 2001;107: 7-12.

472. Medzhitov R, Preston-Hurlburt P, Kopp E, et al. MyD88 is an adaptor protein in the hToll/IL-1 receptor family signaling pathways. Mol Cell. 1998;2: 253-258.

473. Reis J, Hassan F, Guan XQ, et al. The Immunoproteasomes Regulate LPS- Induced TRIF/TRAM Signaling Pathway in Murine Macrophages. Cell Biochem Biophys. 2011: 1-8.

474. Muzio M, Natoli G, Saccani S, et al. The human Toll signaling pathway: divergence of nuclear factor κB and JNK/SAPK activation upstream of tumor necrosis factor receptor–associated factor 6 (TRAF6). J Exp Med. 1998;187: 2097-2101.

475. Wesche H, Henzel WJ, Shillinglaw W, et al. MyD88: an adapter that recruits IRAK to the IL-1 receptor complex. Immunity. 1997;7: 837-847.

476. Mastronarde JG, He R, Monick MM, et al. Induction of interleukin (IL)-8 gene expression by respiratory syncytial virus involves activation of nuclear factor (NF)-κB and NF-IL-6. J Infect Dis. 1996;174: 262.

477. Savkovic SD, Koutsouris A, Hecht G. Activation of NF-κB in intestinal epithelial cells by enteropathogenic Escherichia coli. American Journal of Physiology-Cell Physiology. 1997;273: C1160-C1167.

478. Cianciulli A, Calvello R, Cavallo P, et al. Modulation of NF-κB activation by resveratrol in LPS treated human intestinal cells results in downregulation of PGE2 production and COX-2 expression. Toxicology in Vitro. .

479. Pan M, Lai C, Wang Y, et al. Acacetin suppressed LPS-induced up-expression of iNOS and COX-2 in murine macrophages and TPA-induced tumor promotion in mice. Biochem Pharmacol. 2006;72: 1293-1303.

206

References

480. Grishin AV, Wang J, Potoka DA, et al. Lipopolysaccharide induces cyclooxygenase-2 in intestinal epithelium via a noncanonical p38 MAPK pathway. J Immunol. 2006;176: 580-588.

481. Grishin A, Wang J, Hackam D, et al. p38 MAP kinase mediates endotoxin- induced expression of cyclooxygenase-2 in enterocytes. Surgery. 2004;136: 329-335.

482. Kotlyarov A, Neininger A, Schubert C, et al. MAPKAP kinase 2 is essential for LPS-induced TNF-alpha biosynthesis. Nat Cell Biol. 1999;1: 94-97.

483. Schett G, Zwerina J, Firestein G. The p38 mitogen-activated protein kinase (MAPK) pathway in rheumatoid arthritis. Annals of the Rheumatic Diseases. 2008;67: 909-916.

484. Anand AR, Cucchiarini M, Terwilliger EF, et al. The tyrosine kinase Pyk2 mediates lipopolysaccharide-induced IL-8 expression in human endothelial cells. J Immunol. 2008;180: 5636-5644.

485. Monick MM, Carter AB, Robeff PK, et al. Lipopolysaccharide activates Akt in human alveolar macrophages resulting in nuclear accumulation and transcriptional activity of β-catenin. J Immunol. 2001;166: 4713.

486. Sizemore N, Lerner N, Dombrowski N, et al. Distinct roles of the IκB kinase α and β subunits in liberating nuclear factor κB (NF-κB) from IκB and in phosphorylating the p65 subunit of NF-κB. J Biol Chem. 2002;277: 3863-3869.

487. Sato S, Sugiyama M, Yamamoto M, et al. Toll/IL-1 receptor domain-containing adaptor inducing IFN-β (TRIF) associates with TNF receptor-associated factor 6 and TANK-binding kinase 1, and activates two distinct transcription factors, NF-κB and IFN-regulatory factor-3, in the Toll-like receptor signaling. J Immunol. 2003;171: 4304.

488. Yamamoto M, Sato S, Hemmi H, et al. Role of adaptor TRIF in the MyD88- independent toll-like receptor signaling pathway. Science's STKE. 2003;301: 640.

489. Ogawa S, Lozach J, Benner C, et al. Molecular determinants of crosstalk between nuclear receptors and toll-like receptors. Cell. 2005;122: 707-721.

207

References

490. Honda K, Taniguchi T. IRFs: master regulators of signalling by Toll-like receptors and cytosolic pattern-recognition receptors. Nat Rev Immunol. 2006;6: 644- 658.

491. O'NEILL LAJ. Interferon regulatory factor-3-mediated activation of the interferon-sensitive response element by Toll-like receptor (TLR) 4 but not TLR3 requires the p65 subunit of NF-kappa. Bio Chem. 2003;278: 50923-50931.

492. Leung TH, Hoffmann A, Baltimore D. One nucleotide in a κB site can determine cofactor specificity for NF-κB dimers. Cell. 2004;118: 453-464.

493. Epstein FH, Barnes PJ, Karin M. Nuclear factor-κB—a pivotal transcription factor in chronic inflammatory diseases. N Engl J Med. 1997;336: 1066-1071.

494. Grivennikov SI, Karin M. Inflammatory cytokines in cancer: tumour necrosis factor and interleukin 6 take the stage. Ann Rheum Dis. 2011;70: i104.

495. Karin M, Ben-Neriah Y. Phosphorylation meets ubiquitination: the control of NF- κB activity. Annu Rev Immunol. 2000;18: 621-663.

496. Hatada EN, Krappmann D, Scheidereit C. NF-[kappa] B and the innate immune response. Curr Opin Immunol. 2000;12: 52-58.

497. Schreiber S, Nikolaus S, Hampe J. Activation of nuclear factor κB in inflammatory bowel disease. Gut. 1998;42: 477.

498. Rogler G, Brand K, Vogl D, et al. Nuclear factor κB is activated in macrophages and epithelial cells of inflamed intestinal mucosa. Gastroenterology. 1998;115: 357- 369.

499. Mitchell JA, Larkin S, Williams TJ. Cyclooxygenase-2: regulation and relevance in inflammation. Biochem Pharmacol. 1995;50: 1535.

500. Kargman S, Charleson S, Cartwright M, et al. Characterization of prostaglandin G/H synthase 1 and 2 in rat, dog, monkey, and human gastrointestinal tracts. Gastroenterology. 1996;111: 445-454.

501. Agarwal S, Reddy GV, Reddanna P. Eicosanoids in inflammation and cancer: the role of COX-2. Expert Rev Clin Immunol. 2009;5: 145-165.

208

References

502. Singer II, Kawka DW, Schloemann S, et al. Cyclooxygenase 2 is induced in colonic epithelial cells in inflammatory bowel disease. Gastroenterology. 1998;115: 297-306.

503. Williams CS, Smalley W, DuBois RN. Aspirin use and potential mechanisms for colorectal cancer prevention. J Clin Invest. 1997;100: 1325.

504. Rouzer CA, Marnett LJ. Cyclooxygenases: structural and functional insights. J Lipid Res. 2009;50: S29-S34.

505. Maier J, Hla T, Maciag T. Cyclooxygenase is an immediate-early gene induced by interleukin-1 in human endothelial cells. J Biol Chem. 1990;265: 10805-10808.

506. Kujubu DA, Fletcher BS, Varnum BC, et al. TIS10, a phorbol ester tumor promoter-inducible mRNA from Swiss 3T3 cells, encodes a novel prostaglandin synthase/cyclooxygenase homologue. J Biol Chem. 1991;266: 12866-12872.

507. Arbabi S, Rosengart MR, Garcia I, et al. Epithelial Cyclooxygenase-2 Expression: A Model for Pathogenesis of Colon Cancer. J Surg Res. 2001;97: 60-64.

508. Eckmann L, Stenson WF, Savidge TC, et al. Role of intestinal epithelial cells in the host secretory response to infection by invasive bacteria. Bacterial entry induces epithelial prostaglandin h synthase-2 expression and prostaglandin E2 and F2alpha production. J Clin Invest. 1997;100: 296.

509. Romano M, Ricci V, Memoli A, et al. Helicobacter pylori up-regulates cyclooxygenase-2 mRNA expression and prostaglandin E2 synthesis in MKN 28 gastric mucosal cells in vitro. J Biol Chem. 1998;273: 28560-28563.

510. Lee JY, Ye J, Gao Z, et al. Reciprocal modulation of Toll-like receptor-4 signaling pathways involving MyD88 and phosphatidylinositol 3-kinase/AKT by saturated and polyunsaturated fatty acids. J Biol Chem. 2003;278: 37041-37051.

511. Elder DJE, Halton DE, Playle LC, et al. The MEK/ERK pathway mediates COX‐2‐selective NSAID‐induced apoptosis and induced COX‐2 protein expression in colorectal carcinoma cells. Int J Cancer. 2002;99: 323-327.

512. Han J, Lee J, Bibbs L, et al. A MAP kinase targeted by endotoxin and hyperosmolarity in mammalian cells. Science. 1994;265: 808-811.

209

References

513. Yoshimura T, Matsushima K, Oppenheim J, et al. Neutrophil chemotactic factor produced by lipopolysaccharide (LPS)-stimulated human blood mononuclear leukocytes: partial characterization and separation from interleukin 1 (IL 1). J Immunol. 1987;139: 788.

514. Oppenheim JJ, Zachariae COC, Mukaida N, et al. Properties of the novel proinflammatory supergene" intercrine" cytokine family. Annu Rev Immunol. 1991;9: 617-648.

515. Baggiolini M, Loetscher P, Moser B. Interleukin-8 and the chemokine family. Int J Immunopharmacol. 1995;17: 103-108.

516. Baggiolini M, Clark-Lewis I. Interleukin-8, a chemotactic and inflammatory cytokine. FEBS Lett. 1992;307: 97-101.

517. Holmes WE, Lee J, Kuang WJ, et al. Structure and functional expression of a human interleukin-8 receptor. Science. 1991;253: 1278-1280.

518. Gross V, Andreesen R, LESER HG, et al. lnterleukin‐8 and neutrophil activation in acute pancreatitis. Eur J Clin Invest. 1992;22: 200-203.

519. Baggiolini M, Moser B, Clark-Lewis I. Interleukin-8 and related chemotactic cytokines. Chest. 1994;105: 95S-98S.

520. Mitsuyama K, Toyonaga A, Sasaki E, et al. IL‐8 as an important chemoattractant for neutrophils in ulcerative colitis and Crohn's disease. Clin Exp Immunol. 1994;96: 432-436.

521. Izutani R, Loh EY, Reinecker HC, et al. Increased expression of interleukin‐8 mRNA in ulcerative colitis and Crohn's disease mucosa and epithelial cells. Inflamm Bowel Dis. 1995;1: 37-47.

522. Mazzucchelli L, Hauser C, Zgraggen K, et al. Expression of interleukin-8 gene in inflammatory bowel disease is related to the histological grade of active inflammation. AM J Pathol. 1994;144: 997.

523. MacDermoit RP. Alterations of the mucosal immune system in inflammatory bowel disease. J Gastroenterol. 1996;31: 907-916.

210

References

524. Butcher EC. Leukocyte-endothelial cell recognition: three (or more) steps to specificity and diversity. Cell. 1991;67: 1033-1036.

525. Banks C, Bateman A, Payne R, et al. Chemokine expression in IBD. Mucosal chemokine expression is unselectively increased in both ulcerative colitis and Crohn's disease. J Pathol. 2003;199: 28-35.

526. Williams CS, Mann M, DuBois RN. The role of cyclooxygenases in inflammation, cancer, and development. Oncogene. 1999;18: 7908.

527. Fukata M, Chen A, Klepper A, et al. Cox-2 is regulated by Toll-like receptor-4 (TLR4) signaling: role in proliferation and apoptosis in the intestine. Gastroenterology. 2006;131: 862-877.

528. Kim JM, Lee JY, Yoon YM, et al. Bacteroides fragilis enterotoxin induces cyclooxygenase‐2 and fluid secretion in intestinal epithelial cells through NF‐κB activation. Eur J Immunol. 2006;36: 2446-2456.

529. Bertelsen LS, Paesold G, Eckmann L, et al. Salmonella infection induces a hypersecretory phenotype in human intestinal xenografts by inducing cyclooxygenase 2. Infect Immun. 2003;71: 2102-2109.

530. Thun MJ, Henley SJ, Patrono C. Nonsteroidal anti-inflammatory drugs as anticancer agents: mechanistic, pharmacologic, and clinical issues. J Natl Cancer Inst. 2002;94: 252-266.

531. Ekbom A, Helmick C, Zack M, et al. Ulcerative colitis and colorectal cancer. N Engl J Med. 1990;323: 1228-1233.

532. Hayden M, West A, Ghosh S. NF-κB and the immune response. Oncogene. 2006;25: 6758-6780.

533. Nelson D, Ihekwaba A, Elliott M, et al. Oscillations in NF-{kappa} B signaling control the dynamics of gene expression. Science's STKE. 2004;306: 704.

534. Levitt MD. Production and excretion of hydrogen gas in man. N Engl J Med. 1969;281: 122-127.

211

References

535. Cummings JH. Fermentation in the human large intestine: evidence and implications for health. Lancet. 1983;1.

536. Christl S, Murgatroyd PR, Gibson GR, et al. Production, metabolism, and excretion of hydrogen in the large intestine. Gastroenterology. 1992;102: 1269.

537. Matsheka M, Lastovica A, Elisha BG. Molecular Identification of Campylobacter concisus. J Clin Microbiol. 2001;39: 3684-3689.

538. Dissanayake D, Wijewardana T, Gunawardena G, et al. Distribution of lipopolysaccharide core types among avian pathogenic Escherichia coli in relation to the major phylogenetic groups. Vet Microbiol. 2008;132: 355-363.

212