Elucidation of Enteric Virulence of Campylobacter Concisus

Elucidation of enteric virulence of Campylobacter concisus

Hoyul Lee

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

Supervisor: Dr. Li Zhang

School of Biotechnology and Biomolecular Science Faculty of Science University of New South Wales

September 2015

THE UNIVERSITY OF NEW SOUTH WALES Thesis/Dissertation Sheet Surname or Family name: Lee First name: Hoyul Other name/s: Leo Abbreviation for degree as given in the University calendar: PhD School: The School of Biotechnology and Biomolecular Science Faculty: Faculty of Science Title: Elucidation of enteric virulence of Campylobacter concisus

Abstract Campylobacter concisus is a Gram-negative mobile oral bacterium that has been shown to be associated with human inflammatory bowel disease. Various experiments were conducted in this PhD project, aiming to elucidate the enteric virulence of C. concisus. The growth of oral C. concisus strains under different atmospheric conditions was examined. It was found that 92 % of strains grew under anaerobic conditions without H2. However, none of the strains grew under microaerobic conditions without H2. The presence H2 greatly increased the growth of C. concisus under anaerobic conditions and enabled C. concisus to grow under microaerobic conditions. The H2 had no effects on the expression of a number of putative virulence factors in C. concisus. The effects of formate and fumarate, on the production of H2S in oral C. concisus were investigated. Supplementation of formate and fumarate significantly increased the positivity of H2S production. In addition, the fumarate significantly increased C. concisus growth. Bioinformatics analysis was conducted to search for potential virulence factors encoded by prophages. Four prophage elements were identified in the genome of C. concisus strain 13826 with putative attachment sites overlapping with tRNA. Each prophage elements contained a novel Xer phage integrase. It was found that CON_phi2 prophage encodes a Zot protein, and CON_phi3 encodes a Zot-like protein. Moreover, the phylogenetic analysis showed the horizontal gene transfer between Campylobacter species. Monoclonal antibody to C. concisus Zot was produced and verified. The impact of bile on the expression of Zot was examined. It was found that a full length and a cleaved fragment of Zot were released from C. concisus in the presence of ox bile. Whether C. concisus strains affect actin in the intestinal epithelial cell line, Caco2 cells, was examined. Some C. concisus strains significantly reduced the levels of β-actin and caused the redistribution of F-actin in Caco2 cells. These findings show that the enteric virulence of C. concisus is determined by bacterial, host and environmental factors.

Declaration relating to disposition of project thesis/dissertation I hereby grant to the University of New South Wales or its agents the right to archive and to make available my thesis or dissertation in whole or in part in the University libraries in all forms of media, now or here after known, subject to the provisions of the Copyright Act 1968. I retain all property rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstracts International (this is applicable to doctoral theses only).

18 March 2016 …………………………… …………………………… …………………………… Signature Witness signature Date The University recognises that there may be exceptional circumstances requiring restrictions on copying or conditions on use. Requests for restriction for a period of up to 2 years must be made in writing. Requests for a longer period of restriction may be considered in exceptional circumstances and require the approval of the Dean of Graduate Research.

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

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

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 18 March 2016

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

Signed

Date 18 March 2016

Authenticity statement

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

Signed

Date 18 March 2016

Acknowledgements

The completion of this PhD project consumed a huge amount of effort on the part of not only myself but also my laboratory colleagues. I would like to acknowledge those who have helped me, and show my gratitude towards them. Studying in another country was probably one of the best decisions that I have ever made. This experience has positively influenced my life as I was able to improve my scientific knowledge, communication skills, my grasp of the English language and culture and perhaps most importantly, I became independent. This experience allowed me to meet a great number of people in Australia including; teachers, friends, families, and academics. The accumulation of all these experiences has made these past 10 years very enjoyable and beneficial.

First of all, I would like to acknowledge each family member. I thank my mother, Sung- Yun Cho, who continuously encouraged my studies overseas with her heartfelt pray. I thank my father, Jong-Ho Lee, who inspired me to follow my dreams. I also thank my beloved sister, Seon-Woo Lee, who always refreshed my mind with funny jokes. Lastly, I thank my family for the financial support.

I would like to express great thanks to my supervisor, Dr. Li Zhang who has offered me a valuable opportunity of doing my PhD research project in her laboratory. I would not have been able to complete my PhD without her supervision, scientific knowledge and experience. Her patience and enthusiasm makes her a fantastic mentor for young scientific researchers and students. I am grateful to her for making this experience possible.

I would also like to thank associate Professor Mark Tanaka and Dr. Wallace Bridge for being members of my panel. Their in-depth comments and constructive criticism in regards to my PhD project. I found their annual progress reviews invaluable.

I am also grateful to Dr. Sophie Octavia from associate Professor Ruiting Lan's laboratory, Mrs. Iveta Slapetova from BMIF, Dr. Ling Zhong from BMSF, Mr. Chris Brownlee from BRIL, as well as the staff members from the Ramaciotti Centre for their advice and technical support for RT-PCR, confocal microscopy, LC/MS/MS analyses, flow cytometry analyses, and gene sequencing respectively.

iii

I would also like to give sincere thanks to my colleagues in my laboratory who have become a team for me; Yazan, Viki, Lisa, Yesing, Omar, Mona, Rena, Connie, Sheryl, Nick, Fang, William, Sandra, Leslie, and Ece. They have all worked with me during my PhD studies. I would especially like to give thanks to my two good friends Yazan Ismail and Vikneswari Mahendran, who have helped me since I first joined this team.

Lastly, I would like to thank my good friends, especially Hwang and Choi, in Australia who have been a continuous source of encouragement for me during my PhD journey. I would like to extend my gratitude to my family, supervisor, academic staff, colleagues and friends for their kind cooperation and encouragement, which have been a significant source of comfort in my completion of my PhD.

Publications and conference proceedings

Publications during PhD

Ma R, Sapwell N, Chung HK, Lee H, Mahendran V, Leong RW, Riordan SM, Grimm MC and Zhang L (2015) Investigation of the effects of pH and bile on the growth of oral Campylobacter concisus strains isolated from patients with inflammatory bowel disease and controls. Journal of medial microbiology. April 64: 438-445. Doi: 10.1099/jmm.0.000013

Lee H, Ma R, Grimm MC, Riordan SM, Lan R, Zhong L, Raftery M, Zhang L (2014) Examination of the anaerobic growth of Campylobacter concisus strains. International Journal of Microbiology. Vol2014. Doi: 10.1155/2014/476047

Zhang L, Lee H, Grimm MC, Riordan SM, Day AS, Lemberg DA (2014) Campylobacter concisus and inflammatory bowel disease. World journal of gastroenterology. 2014 Feb 7;20(5):1259-67. Doi: 10.3748/wjg.v20.i5.1259.

Ismail Y, Lee H, Riordan SM, Grimm MC, Zhang L (2013) The Effects of Oral and Enteric Campylobacter concisus Strains on Expression of TLR4, MD-2, TLR2, TLR5 and COX-2 in HT-29 Cells. PLoS ONE 8(2): e56888. Doi:10.1371/journal.pone.0056888

Manuscripts submitted

Lee H, Liu F, Ma R, Mahendran V, Chung HK, Lee S, Riordan SM, Grimm MC, Leong RW, Zhang L (2015) Campylobacter concisus releases zonula occludens toxin in the presence of bile and the toxin causes prolonged intestinal epithelial barrier defect. Cellular microbiology.

Manuscripts in preparation

Liu F, Lee H, Mahendran V, Ma R, Tanaka M, Zhang L (2015) Identification of prophages that are similar to Campylobacter concisus putative prophages Con_phi2 and Con_phi3 in other Campylobacter species: evidence that C. concisus acquires virulence factors from other Campylobacter species through horizontal gene transfer.

Lee H, Cattanach W, Zhang L (2014) Campylobacter concisus affects actin in Caco2 cells.

Conference proceedings

Lee H, Liu F, Mahendran V, Zhang L (2014) Detection of the zonula occludens toxin expressed in Campylobacter concisus strains. Journal of gastroenterology and hepatology Volume 29, Issue S2 October 2014 Pages 10–26

Lee H, Cattanach W, Zhang L (2014) Campylobacter concisus affects actin in Caco2 cells. CEUR workshop proceedings. ISSN:1613-0073.

Mahendran V, Lee H, Tan YS, Riordan SM, Grimm MC, Day AS, Lemberg DA, Octavia S, Lan R and Zhang L (2013) The association between a tight junction toxin gene in Campylobacter concisus and inflammatory bowel disease. Journal of Gastroenterology and Hepatology. Volume 28, Page 14, Oct

List of contents

Originality statement ················································································· i

Authenticity statement ·············································································· ii

Acknowledgements ·················································································· iii

Publications and conference proceedings ······················································· v

Publications during PhD ··········································································· v

Manuscripts submitted ············································································· v

Manuscripts in preparation ······································································· vi

Conference proceedings ·········································································· vi

List of contents ······················································································ vii

List of tables xv

List of figures ······················································································ xviii

Abstract xx

Chapter 1: General introduction ······························································· 1

1.1. Campylobacter genus ······································································· 1

1.2. Campylobacter concisus ··································································· 9

1.3. Human oral cavity for C. concisus colonisation ·······································10

1.4. Association with disease in human ······················································12

1.4.1. Oral disease ···················································································· 12 1.4.2. Intestinal disease ·············································································· 13 1.4.2.1. Enteritis ················································································ 13 1.4.2.2. Inflammatory bowel disease ························································ 14 1.4.2.2.1. Intestinal microbiota ···························································· 15

vii

1.4.2.2.2. C. concisus as a potential initiator of IBD ··································· 15

1.5. Virulence properties of C. concisus ·····················································19

1.5.1. Motility, adherence, invasion and translocation ··········································· 19 1.5.2. Toxins in C. concisus ········································································· 20 1.5.2.1. Phospholipase A ······································································ 20 1.5.2.2. S-layer repeats-in-toxin ······························································ 21 1.5.2.3. Zonula occludens toxin ······························································ 22 1.5.3. Production of hydrogen sulfide······························································ 22 1.5.4. Impaired tight junctions by C. concisus ···················································· 25

1.6. Hypothesis and aims ·······································································28

Chapter 2: General materials and methodology ···········································29

2.1. Buffers ·······················································································29

2.1.1. Phosphate buffer saline (PBS) ······························································· 29 2.1.2. Tris-ethylenediaminetetraacetic acid (TE) Buffer········································· 29 2.1.3. 50× Tris acetic acid EDTA (TAE) Buffer ················································· 29

2.2. Bacterial culture ············································································30

2.2.1. Media for bacterial culture ··································································· 30 2.2.2. Bacterial cryopreservation ··································································· 30 2.2.3. C. concisus culture ············································································ 30 2.2.4. E. coli culture ·················································································· 31 2.2.5. Quantification of the number of bacteria ··················································· 31 2.2.5.1. Colony forming unit (CFU) counting ·············································· 31 2.2.5.2. Absorbance measurement at 600nm ················································ 32 2.2.6. Preparation of Caco2 cell lysates ··························································· 32 2.2.7. Preparation of bacterial lysates ······························································ 32

2.3. Caco2 cell culture ··········································································33

2.3.1. Media for cell culture ········································································· 33 2.3.2. Maintenance of the Caco2 Cells ···························································· 33

viii

2.3.3. Determination of cell number and viability ················································ 33 2.3.4. Cryopreservation of Caco2 cells ···························································· 34 2.3.5. Revising cells from liquid nitrogen ························································· 34

2.4. Bicinchoninic acid (BCA) assay ·························································35

2.5. Sodium dodecyl sulfate poly-acrylamide gel electrophoresis (SDS-PAGE) ······36

2.5.1. Reagents for SDS-PAGE····································································· 36 2.5.2. Preparation of SDS-polyacrylamide gel ···················································· 36 2.5.3. Protein denaturation ·········································································· 37 2.5.4. Electrophoresis ················································································ 37 2.5.5. Coomassie brilliant blue staining ··························································· 38

2.6. Western blot ·················································································39

2.6.1. Reagents for western blot ···································································· 39 2.6.2. Transferring to the polyvinylidene fluoride (PVDF) membrane ························ 39 2.6.3. Immunoblotting ··············································································· 40 2.6.4. Visualisation ··················································································· 40

Chapter 3: The anaerobic growth of Campylobacter concisus ··························41

3.1. Introduction ·················································································41

3.2. Materials and methods ·····································································42

3.2.1. C. concisus strains used in this study ······················································· 42 3.2.2. Examination of C. concisus growth under various atmospheric conditions ··········· 42 3.2.3. Quantitative comparison of C. concisus growth under AnaeroH2- and AnaeroH2+ conditions ······················································································ 43 3.2.4. Examination of hydrogen dependent growth of C. concisus under anaerobic and microaerobic conditions ······································································ 44 3.2.5. Proteins expressed by C. concisus cultured under AnaeroH2- and AnaeroH2+ conditions 44 3.2.6. Statistical analysis ············································································ 46

3.3. Results ·······················································································47

3.3.1. The growth of C. concisus strains under anaerobic condition with and without H2 ·· 47 3.3.2. Quantitative comparison of C. concisus growth under AnaeroH2- and AnaeroH2+ conditions ······················································································ 48 3.3.3. The growth of C. concisus under anaerobic and microaerobic conditions containing

different concentrations of H2 ······························································· 49 3.3.4. Proteins expressed by C. concisus strain P6CDO-S1 cultured under AnaeroH2- and AnaeroH2+ conditions ········································································· 51

3.4. Discussion ···················································································54

Chapter 4: The effects of formate and fumarate on Campylobacter concisus hydrogen sulfide production and growth ·····································58

4.1. Introduction ·················································································58

4.2. Materials and methods ·····································································60

4.2.1. Examination of H2S production in C. concisus strains using modified HBA plates and the effects of supplementation of both formate and fumarate in culture media as well

as H2 gas under atmospheric conditions on the production of H2S by C. concisus strains ··························································································· 60

4.2.2. Examination of individual effect of formate and fumarate on the production of H2S in C. concisus strains ············································································ 61 4.2.3. Quantification of C. concisus growth with and without supplementation of formate and fumarate ··················································································· 61 4.2.4. Statistical analysis ············································································ 62

4.3. Results ·······················································································63

Fe/S H2+ 4.3.1. H2S positivity in oral C. concisus strains cultured on HBA plates under Anaero and AnaeroH2- conditions ····································································· 63

4.3.2. H2S positivity in C. concisus strains isolated from patients with IBD and controls ·· 63 4.3.3. The effects of supplementation of both fumarate and formate into culture media on the

production of H2S in C. concisus strains ··················································· 63

4.3.4. H2S positivity in C. concisus strains of different subpopulations ······················· 65

4.3.5. Individual effect of formate or fumarate on H2S production in oral C. concisus ······ 67 4.3.6. Quantitative examination of formate and fumarate supplementation on C. concisus growth ·························································································· 69

4.4. Discussion ···················································································71

Chapter 5: In silico identification analysis of Campylobacter concisus prophages ·75

5.1. Introduction ·················································································75

5.2. Materials and methods ·····································································77

5.2.1. Identification of prophage in C. concisus 13826 reference strain ······················· 77 5.2.2. Re-annotation of some proteins in the prophage region ·································· 77 5.2.3. Examination of secretion proteins ·························································· 78 5.2.4. Examination of operon in prophage ························································ 78 5.2.5. Examination of paralogue proteins in prophage ··········································· 78 5.2.6. Consensus sequence of XerC or XerD binding sites of prophages ····················· 79 5.2.7. Protein identity of phage integrases in CON_phi with XerC, XerD, XerH and XerT 80 5.2.8. Characteristics of phage integrase in prophage in C. concisus strain 13826 ··········· 80 5.2.9. Identification of CON_phi-like prophages in Campylobacter genus ··················· 80 5.2.10. Phylogenetic tree construction of the zot gene and phage maturation protein of Campylobacter species ······································································· 81

5.3. Results ·······················································································82

5.3.1. Identification of prophage in C. concisus strain 13826 ··································· 82 5.3.1.1. Re-annotation of proteins in the putative prophage identified by PHAST····· 84 5.3.1.2. Multiple prophages within the prophage··········································· 91 5.3.1.3. Identification of CON_phi4 in C. concisus strain 13826 and re-annotation of proteins ····················································································· 94 5.3.1.4. Paralogue search and MLST ························································ 97 5.3.1.5. Structure of integrase in C. concisus prophage ·································· 102 5.3.1.6. Consensus sequence of XerC or XerD binding sites in various organisms ··· 104 5.3.1.7. Protein identity (%) of phage integrase in CON_phi comparing with XerC, XerD, XerH and XerT ··································································· 105 5.3.1.8. Putative attachment sites ···························································· 106

5.3.1.9. Putative operons ····································································· 108 5.3.1.10. Putative secreted proteins encoded by genes in the prophage region ········ 109 5.3.2. Identification of CON_phi2-like prophage and CON_phi3-like prophage in Campylobacter genus other than C. concisus ············································· 111 5.3.3. Phylogenetic relationship of Zot and phage maturation protein identified in Campylobacter spp. ·········································································· 118

5.4. Discussion ················································································· 122

Chapter 6: The release of Zot protein from Campylobacter concisus in the presence of bile ··············································································· 127

6.1. Introduction ··············································································· 127

6.2. Materials and methods ··································································· 129

6.2.1. Development of a monoclonal antibody targeting C. concisus Zot ···················· 129 6.2.2. Cloning of zot gene into pETBlue2 E. coli expression system ························· 129 6.2.3. Induction of Zot protein in E. coli ························································· 131 6.2.4. Detection of secreted Zot protein ·························································· 132 6.2.5. The measurement of mRNA level of zot in C. concisus strain 13826 ················· 133 6.2.5.1. List of primers ······································································· 133 6.2.5.2. Efficiency test of each primer pairs ··············································· 134 6.2.5.3. Total RNA preparation for RT-PCR ·············································· 135 6.2.5.4. Removal of bacterial DNA ························································· 136 6.2.5.5. RNA re-purification ································································· 136 6.2.5.6. Synthesis of cDNA ·································································· 136 6.2.5.7. Quantification of RT-PCR ·························································· 136 6.2.5.8. Analysis of RT-PCR ································································ 137 6.2.6. Effect of the C. concisus culture supernatant on intestinal epithelial barrier ········· 137 6.2.7. Molecular weight prediction of Zot protein in C. concisus ····························· 139

6.3. Results ····················································································· 140

6.3.1. Expression of Zot using E. coli expression system ······································ 140 6.3.2. Secretion of C. concisus Zot protein ······················································· 142

xii

6.3.3. Selection of housekeeping genes that can be used as internal control for examination of the effects of bile on Zot expression in C. concisus by RT-PCR ··················· 143 6.3.4. The mRNA level of zot in C. concisus strain 13826 cultured in media with and without bile ··················································································· 146 6.3.5. The effects of C. concisus Zot released in the presence of bile on intestinal epithelial permeability ·················································································· 147

6.4. Discussion ················································································· 150

Chapter 7: Investigation of the possible effects of C. concisus on actin in Caco2 cells ·················································································· 153

7.1. Introduction ··············································································· 153

7.2. Materials and methods ··································································· 155

7.2.1. Detection of β-actin and α-tubulin by flow cytometry analysis ························ 155 7.2.2. Detection of β-actin in Caco2 cells by western blot ····································· 156 7.2.3. Examination of effects of C. concisus on F-actin arrangement in Caco2 cells using confocal microscopy ········································································· 157 7.2.4. Examination of F-actin in the Caco2 cells treated with different concentrations of C. concisus using confocal microscopy ······················································· 157 7.2.5. Statistical analysis ··········································································· 158

7.3. Results ····················································································· 160

7.3.1. Decreased protein level of β-actin in Caco2 cells in the presence of C. concisus detected using flow cytometry······························································ 160 7.3.2. Decreased protein level of β-actin in Caco2 cells in the presence of C. concisus detected using western blot ································································· 163 7.3.3. F-actin alteration induced by C. concisus ················································· 165 7.3.4. Dose-dependent effects of C. concisus infection on F-actin in Caco2 cells ·········· 167

7.4. Discussion ················································································· 169

General discussion and future directions ····················································· 172

xiii

General discussion ·············································································· 172

Future directions ················································································· 173

Appendix 1. The alignment of nucleotide sequences of five phage paralogue proteins used in MLST analysis ·························································· 176

References 178

xiv

List of tables

Table 1-1. Human hosted Campylobacter species and their disease associations ...... 2

Table 1-2. Animal hosted Campylobacter species and their disease associations ...... 4

Table 1-3. Significantly higher prevalence of C. concisus in patients with IBD as compared to that in healthy individuals ...... 18

Table 3-1. Positive growth rates of C. concisus strains under AnaeroH2- and AnaeroH2+ conditions ...... 47

Table 3-2. CFU of C. concisus strains cultured under AnaeroH2+ and AnaeroH2- conditions ...... 48

Table 3-3. Virulence proteins expressed by C. concisus strain P6CDO-S1 cultured under H2+ H2- * Anaero and Anaero conditions ...... 52

Table 3-4. A list of four proteins in which a significant change in protein expression was found via mass spectrometry analysis ...... 53

H2+ H2- Table 4-1. H2S positivity of oral C. concisus strains under Anaero and Anaero # conditions with or without supplementation ...... 64

^ Table 4-2. H2S positivity in C. concisus strains from five subpopulation groups ...... 66

Table 4-3. H2S positivity of six C. concisus strains with supplementation of formate or @ fumarate ...... 68

Table 4-4. List of the redox potential at pH 7.0 ...... 73

Table 5-1. DNA binding sites to the XerC or XerD site-specific recombinase ...... 79

Table 5-2. Phage-like proteins in the genome of C. concisus strain 13826 ...... 85

Table 5-3. Proteins encoded by putative prophage island in C. concisus strain 13826 ...... 89

Table 5-4. The identity (%) of proteins in CON_phi3 to that in the CON_phi4 ...... 95

Table 5-5. A list of phage-like proteins in the CON_phi4 ...... 96

Table 5-6. Paralogous proteins in four prophage elements* ...... 100

Table 5-7. Protein identity (%) of Xer type recombinases in C. concisus strain 13826 with

XerC, XerD, XerH and XerT@ ...... 105

Table 5-8. The predicted attachment sites of CON_phi prophage ...... 107

Table 5-9. Predicted operons in the PHAST identified prophage region of C. concisus strain 13826 genome ...... 108

Table 5-10. Predicted secreted proteins encoded by prophage genes in the genome of C. concisus strain 13826 ...... 110

Table 5-11. List of tRNA types at upstream of prophage in Campylobacter species ...... 112

Table 5-12. Protein identity percentage between CON_phi2 proteins and corresponding phage-like proteins in C. ureolyticus ACS-301-V-Sch3b, C. concisus UNSWCS, C. concisus UNSW3, and C. corcagiensis CIT045 ...... 114

Table 5-13. Protein identity percentage between CON_phi3 proteins and corresponding phage-like proteins C. jejuni subsp. doylei 269.97, C. gracilis RM3268, Campylobacter sp. FOBRC14, C. jejuni subsp. jejuni 60004, C. jejuni subsp. jejuni 86605, C. corcagiensis CIT0 and C. hyointestinalis subsp. hyointestinalis DSM 19053 ...... 116

Table 6-1. List of primer pairs used in this study ...... 134

Table 6-2. Efficiency of four genes used in this study ...... 144

xvi

Table 6-3. The expression level difference of three candidate housekeeping genes in C. # concisus strain 13826 cultured in media with bile versus without bile ...... 145

Table 6-4. The fold increase of zot mRNA induced by bile in C. concisus strain 13826 . 146

xvii

List of figures

Figure 1-1. Schematic view of role of tight junction in intestinal epithelial cells against luminal antigens ...... 26

Figure 3-1. Growth of C. concisus strain P3UCO-S1 at various H2 concentrations ...... 50

Figure 4-1. Quantification of the growth of C. concisus strain UNSWCD with and without the supplementation of formate and fumarate ...... 70

Figure 5-1. Prophage identified by PHAST search in the genome map of C. concisus strain 13826 ...... 83

Figure 5-2. The location of prophages in C. concisus strain 13826 on the genome...... 92

Figure 5-3. Genetic structures of the prophages identified in C. concisus strain 13826 ..... 93

Figure 5-4. Phylogenetic relationship between genes encoding paralogous protein in four prophages...... 101

Figure 5-5. The 3D structure of three integrase found in CON_phi prophage...... 103

Figure 5-6. Frequency of nucleotide of XerC or XerD binding sites ...... 104

Figure 5-7. Schematic view of the potential CON_phi2-like prophage genome structure among the Campylobacter genus...... 115

Figure 5-8. Schematic view of the potential CON_phi3-like prophage genome structure among the Campylobacter genus...... 117

Figure 5-9. Phylogenetic tree of 16S rRNA in Campylobacter genus...... 120

xviii

Figure 5-10. The phylogenetic trees based on the gene sequence encoding the phage maturation protein and Zot protein in the Campylobacter genus...... 121

Figure 6-1. Expressed C. concisus Zot in the transformed E. coli was detected by anti penta-His antibody and 1-4C381 mAb...... 141

Figure 6-2. The Zot protein in C. concisus strain 13826 released into supernatant in the presence of bile...... 142

Figure 6-3. The standard curve of Ct value vs genomic DNA for four genes, 16S rRNA, flgB, rpoA and zot...... 143

Figure 6-4. The TEER in Caco2 cells treated with the C. concisus culture supernatant. .. 149

Figure 7-1. The flow cytometry analysis of β-actin in Caco2 cells in response to oral C. concisus strains ...... 161

Figure 7-2. The flow cytometry analysis of α-tubulin in Caco2 cells in response to oral C. concisus strains ...... 162

Figure 7-3. Western blot analysis of β-Actin level in Caco2 cells in response to oral C. concisus strains ...... 164

Figure 7-4. Effects of C. concisus infection on F-actin in Caco2 cells ...... 166

Figure 7-5. Differential F-actin morphology in Caco2 cells infected by C. concisus strain P2CDO4 with various MOI...... 168

xix

Abstract

Inflammatory bowel disease (IBD) is a group of chronic inflammatory diseases of the gastrointestinal tract. The two major forms of IBD are Crohn’s disease (CD) and ulcerative colitis (UC). The aetiology of IBD is thought to include a combination of gene susceptibility, environmental factors, immune response and microbiota factors. It has also been suggested that the microbiota plays a key role in the development of IBD.

Campylobacter concisus is a Gram negative bacterium, which naturally occurs in the human oral cavity as a commensal bacterium. In addition, this bacterium is currently recognised as a potential initiator of a subgroup of IBD. It is unclear why C. concisus behaves as a commensal oral bacterium in general but may initiate enteric diseases such as IBD in some individuals. It was hypothesised that the enteric virulence of C. concisus is determined by multiple factors including the physiological properties of individual C. concisus strains, the virulence factors carried by different C. concisus strains and the intestinal environmental factors of different individuals. The aims of this thesis were 1) To examine the impact of anaerobic conditions and hydrogen gas (H2) on the growth of C. concisus and protein expression, 2) To examine the effects of formate and fumarate supplementation on the H2S production and growth of different C. concisus strains, 3) To identify prophages as a vehicle of virulence gene acquisition via horizontal gene transfer, 4) To examine the secretion of zonular occludens toxin (Zot) in C. concisus, and 5) To examine the impact of C. concisus strains on β-actin and F-actin in Caco2 cells.

In chapter 3, the impacts of anaerobic conditions with H2 were examined. In this study, 60 oral strains were used. It was found that 92% of the strains grew under anaerobic conditions without H2, while 100% of the strains grew under anaerobic conditions with H2.

The incubation with H2 greatly increased the colony forming units (CFU). The incubation with H2 had no differential effects on the expression of virulence factors in C. concisus strain P6CDO-S1 grown under anaerobic condition. However, the expression level of some metabolic enzymes in the strain grown with H2 differed from that in the same strain grown without H2.

In chapter 4, the effects of two food additives, formate and fumarate, on the production of

H2S in C. concisus strains were investigated. A total of 57 oral C. concisus strains were used. The results showed that the formate and fumarate supplementation significantly increased the positivity of H2S production in the C. concisus strains (40% vs 72%). Moreover, the supplementation with fumarate alone significantly increased the growth of C. concisus.

Chapter 5 used bioinformatics tools to analyse the prophages in C. concisus strain 13826 genome. Four CON_phi prophage elements were identified with putative attachment sites overlapping with tRNA. These included CON_phi1, CON_phi2, CON_phi3 and CON_phi4. Each prophage element contained a phage integrase. The protein alignment suggests that these were novel Xer recombinases. Moreover, these prophages were divided into two categories, CON_phi2 type with Zot and CON_phi3 type with a Zot-like protein based on the protein alignments and paralogue search. Furthermore, the phylogenetic analysis showed the horizontal gene transfer of the operon with open reading frames for Zot and phage maturation protein between Campylobacter species.

In chapter 6, the secretion of Zot protein from C. concisus was examined. A monoclonal antibody 1-4C381 was commercially produced and its specificity was validated using a Escherichia coli expression system and Western blot. The effects of bile on the release of C. concisus Zot protein were examined. It was found that both the full length and a cleaved fragment of Zot were released from C. concisus strain 13826 in the presence of bile. The mRNA levels of C. concisus Zot was detected by real time PCR and it was found that bile did not increase the synthesis of Zot in C. concisus.

Chapter 7 examined whether the infection of oral C. concisus strains altered β-actin level in the intestinal epithelial cell line using Caco2 cell culture model. In this study, five C. concisus strains were used. The result of flow cytometry analyses showed that the β-actin levels in Caco2 cells decreased following 24 hours incubation with C. concisus strains. However, the α-tubulin levels were not affected. Western blot analysis showed similar results. Furthermore, confocal microscopy analysis showed that C. concisus affects F-actin arrangement in Caco2 cells.

xxi

In conclusion, the findings in this thesis show that C. concisus is an opportunistic enteric pathogen and its enteric virulence is determined by multiple factors such as the properties of individual strains, prophages and host intestinal factors.

xxii

Chapter 1: General introduction

1.1. Campylobacter genus

Campylobacter species are Gram-negative bacteria. In general, members of Campylobacter genus have a curved or spiral shape. Some species are mobile. Their movement usually is facilitated by corkscrew-like motion using a unipolar or bipolar unsheathed flagellum (1). Most of the Campylobacter species are oxidase positive, catalase positive, urease negative (1). While most of the Campylobacter species are able to grow under microaerobic conditions, some prefer anaerobic conditions for their growth (2). Some Campylobacter species, including Campylobacter concisus, Campylobacter showae, Campylobacter curvus, Campylobacter rectus, Campylobacter gracilis, Campylobacter sputorum and Campylobacter hominis require gaseous hydrogen as an electron donor to support their growth. Those species have been shown to be more closely related according to the 16S rRNA phylogenetic tree (3, 4)

Up to date, 33 bacterial species were found to belong to Campylobacter genus. Some Campylobacter species are human hosted. Most of the human hosted Campylobacter species colonise the human oral cavity, while C. hominis was found to be a commensal bacterium of the human intestinal tract (5) (Table 1-1). The majority of known species in the Campylobacter genus has animals as their natural host, such as domestic, wild and marine animals (Table 1-2).

Most of the animal hosted Campylobacter species live primarily as commensal bacteria in the intestinal tract of their hosts without causing any harm (1, 6, 7) (Table 1-2). However, a number of Campylobacter species, including C. fetus subsp. fetus, C. fetus subsp. venerealis C. hyointestinalis subsp. hyointestinalis, C. jejuni subsp. jejuni and C. upsaliensis were reported to be associated with animal diseases such as spontaneous abortion and gastroenteritis (Table1-2).

Table 1-1. Human hosted Campylobacter species and their disease associations

Species Colonisation site Human disease association Animal disease association References C. concisus Oral cavity Inflammatory bowel disease, periodontal Enteritis in dog (3, 8-14) disease, enteritis, septicaemia, Barrett’s oesophagus, and foot ulcers

C. curvus Oral cavity Hepatic abscesses, alveolar abscesses, No case reported (3, 9, 10, 12, 13, gastroenteritis, and Guillain-Barre 15-17) syndrome

C. gracilis Oral cavity Septicaemia No case reported (3, 9, 10, 13, 18)

C. hominis Intestine No case reported No case reported (3, 9, 10)

C. rectus Oral cavity Ulcerative colitis, periodontitis No case reported (3, 9, 10, 12, 13)

C. showae Oral cavity Axillary abscesses, septicaemia, No case reported (3, 9, 10, 13, 19) periodontitis and gingivitis

C. sputorum biovar Oral cavity Palate abscesses, lung abscesses, scrotal No case reported (3, 8, 9, 12) sputorum abscesses, and axillary abscesses

C. ureolyticus Oral cavity Periradicular abscesses, wound infections, No case reported (3, 9, 10) enteritis, and urethritis

Table 1-2. Animal hosted Campylobacter species and their disease associations

Species Animal host Human disease association Animal disease association References C. avium Chickens and turkeys No case reported No case reported (3) C. canadensis Whooping cranes No case reported No case reported (3) C. coli Chickens, dogs, pigs, sheep Enteritis and septicaemia No case reported (3, 7, 9, 12, 20) and ostriches C. corcagiensis Macaques No case reported No case reported (21) C. cuniculorum Rabbits No case reported No case reported (3, 22) C. fetus Cattle, dogs, sheep and Septicaemia, meningitis, Spontaneous abortion in cattle (3, 7, 9, 12, 23-25) subsp. fetus turtles vascular infection, and abortion and sheep C. fetus Skinks, snakes and turtles Septicaemia, enteritis No case reported (3, 7, 26) subsp. testudinum and pulmonary oedema C. fetus Cattle and sheep Septicaemia Infectious infertility in cattle (3, 9, 25) subsp. venerealis C. helveticus Dogs, cats and pig No case reported No case reported (3, 9, 20) C. hyointestinalis Pigs, cattle, hamsters, Enteritis and septicaemia Enteritis in reindeers (3, 7, 9, 12, 27) subsp. monkeys, elephants and reindeers hyointestinalis

Table 1-2. (Continue) Species Animal host Human disease association Animal disease association References C. hyointestinalis Pigs, chickens, birds Enteritis No case reported (3, 7, 9, 20, subsp. lawsonii and cattle 28) C. insulaenigrae Seals and porpoises Enteritis and septicaemia No case reported (3, 9, 29) C. jejuni Unknown Enteritis and septicaemia No case reported (3, 9, 10, subsp. doylei 12, 20, 30) C. jejuni Cattle, sheep, Enteritis, septicaemia, abortion, Spontaneous abortion in cattle and (3, 7, 9, 10, subsp. jejuni chickens, turkeys, appendicitis, colitis, myocarditis, sheep, gastroenteritis in dogs and cats 12, 30-32) dogs, cats, mink, reactive arthritis, Reiter’s syndrome and ferrets, crows and non- Guillain-Barre´ syndrome human primates C. lanienae Pigs and cattle No case reported No case reported (3, 9, 33) C. lari Shellfish and seagulls No case reported No case reported (3, 9, 12, subsp. concheus 34) C. lari Cats, dogs, chickens, Enteritis and septicaemia No case reported (3, 9, 12, subsp. lari seals, mussels, oysters 20, 34) and horse C. mucosalis Pigs and dogs No case reported No case reported (3, 9, 20) C. peloridis Shellfish No case reported No case reported (3, 34) C. sputorum Cattle and sheep No case reported No case reported (3, 9) biovar faecalis

Table 1-2. (Continue) Species Animal host Human disease association Animal disease association References C. subantarcticus Penguins and No case reported No case reported (3) albatrosses C. troglodytis Chimpanzees Enteritis No case reported (3) C. upsaliensis Cats, dogs, ducks and Enteritis, septicaemia, abortion, Breast Gastroenteritis in dogs and cats (3, 7, 9, 12, monkeys abscesses and Guillain-Barre syndrome 14, 17, 35) C. volucris Black-headed gulls and No case reported No case reported (3) crows

Animal hosted Campylobacter specie may cause human diseases on infection. Consumption of contaminated food or water is the major infection route (36). Among Campylobacter species, the outstanding human pathogen is C. jejuni, causing bacterial gastroenteritis. The clinical data have shown that the isolation positive rate of C. jejuni in a faecal sample taken from patients with diarrhoea was significantly higher than that from healthy controls in a worldwide scale (37). In addition to gastroenteritis, the infection with C. jejuni can be linked to the Guillain-Barre syndrome (GBS) or Miller Fisher syndrome. In 1995, Rees et al. reported that 26 % (n = 103) of patients with GBS were positive for C. jejuni infection, as compared to 2 % (n = 94) of household control (P < 0.001) (38). In 2004, Kuroki et al. recovered C. jejuni in stool samples collected from 30 % (n = 46) of patients with GBS, which was significantly higher than the healthy control (1.2 %, n = 503) (39). Moreover, C. jejuni has been shown to be associated with axonal degeneration, slow recovery, and severe residual disability (38, 40).

Most of the reported diseases related to the human hosted Campylobacter species are human diseases. They are mainly inflammatory diseases of the gastrointestinal tract (GIT) such as gastroenteritis, inflammatory bowel disease, periodontal diseases and Barrett’s oesophagus. Some species were isolated from blood samples of patients with septicaemia (Table 1-1). Of the human hosted Campylobacter species, C. concisus has gained increasingly attention in the past a number of years. Accumulated evidence suggests that some strains of this bacterium may play a role in a number of chronic inflammatory conditions of the human GIT, in particular, inflammatory bowel disease.

1.2. Campylobacter concisus

In 1981, Tanner et al. characterised sixteen isolates of motile bacteria from inflamed oral tissues in patients with gingivitis and periodontitis. Among this sixteen isolates, six isolates were C. concisus (8). The author observed C. concisus as Gram-negative bacteria having rapid motility with a polar flagellum, and used a microaerobic atmosphere containing 7.5 % hydrogen gas to culture. Formate oxidation and fumarate reduction were observed (8). In addition, oxidase and hydrogen sulfide were produced, but catalase, hydrogen peroxide, urease, lecithinase, lipase, indole, ammonia, and acetylmethylcarbinol were not produced (41).

1.3. Human oral cavity for C. concisus colonisation

Following the isolation of C. concisus by Tanner et al. in 1981, a number of other researchers have also isolated C. concisus from human dental samples. The reported isolation rates from healthy adults and juveniles were approximately 40 % (13) and from 15 to 40 % (42), respectively. Based on these earlier studies, only some individuals in the human population were colonised by C. concisus. However, this conclusion was reconsidered due to the studies reported by Zhang et al. and other researchers (43). Zhang et al. in 2010 compared the presence of C. concisus between saliva samples taken from different age groups of healthy individuals (43). They found that 75 % (44/59) and 97 % (57/59) of saliva samples were positive for C. concisus by culture method and 16S rRNA PCR method, respectively. Petersen et al. and Nielsen et al. also reported similar findings, although their sample's sizes were small. Petersen et al. in 2007 conducted a study to examine the presence of C. concisus presence in saliva samples taken from eleven dentally healthy individuals using a semi-nested PCR method and showed 100 % detection rate of C. concisus (44). Nielsen et al. showed that 100 % (3/3) isolation rate of C. concisus in saliva from healthy individuals (45).

In contrast to the high isolation rate of C. concisus from saliva samples, the detection rates of C. concisus in faeces or intestinal biopsies from healthy individuals were low. In 2000, Engberg et al. collected faecal samples from 107 healthy individuals and examined the presence of C. concisus by C. concisus species-specific PCR (46). They found C. concisus DNA in faecal samples collected from three healthy individuals (2.8 %) (46). A study by Nielsen et al. showed that no C. concisus were detected in faecal samples taken from three healthy individuals using the filtration-culture method (47).

Up to date, C. concisus has not been isolated healthy animals. In 2009, Chaban et al. detected C. concisus in stool samples taken from domestic dogs using species-specific quantitative PCR assays targeting the 60 kDa chaperonin gene (48). Using this method, C. concisus was detected from six stool samples collected from 65 diarrheic dogs, but not in faecal samples collected from 70 healthy dogs (49). There is one report that C. concisus detection was positive in a saliva sample of one two-year-old cat with oral disease using semi-nested PCR amplification targeting 16S rRNA (44).

C. concisus was isolated from chicken and beef meat. In 2011, Lynch et al. reported the isolation of C. concisus was from 10 % chick meat (n = 185) and 3 % beef meat (n = 186) (50). The author pointed out that this may result from the contamination during manufacturing. However, it is not clear whether the contamination source was from animals or humans.

Taken together, these findings suggest that human oral cavity is a natural reservoir of C. concisus.

1.4. Association with disease in human

1.4.1. Oral disease

Tanner et al. for the first time isolated C. concisus from inflamed dental samples collected from patients with periodontitis and gingivitis and pointed out the possible association of C. concisus with oral disease (8). Subsequently, there were studies reporting the isolation of C. concisus in from dental samples of patients with periodontitis or gingivitis. In 1982, Moore et al. isolated C. concisus in samples taken from young adult patients (age 22-32) with gingivitis by cotton rolls and tooth pick, while no C. concisus was isolated in samples from uninflamed sites taken from the same individuals (51). Moreover, C. concisus was also detected in a dorsum of a tongue in a patient with moderate odour in the mouth (52). However, from these earlier reports, any statically significant conclusion has not been completed due to lack of comparisons with healthy subjects.

In 1987, Moore et al. showed that the average percentage of C. concisus in oral microbiota population from ten juvenile patients with naturally-occurring gingivitis was 1.3 ± 1.36 %, while healthy juvenile subjects had the negative result. In addition, patients with juvenile periodontitis have a higher percentage of C. concisus in oral microbiota than juvenile patient with moderate periodontitis (0.4 ± 0.24 % and 0.2 ± 0.10 %, respectively) (53). In 2000, Macuch and Tanner collected bacterial samples from shallow and deeper pockets from 12 patients with initial periodontitis and from 18 healthy individuals. They found that C. concisus was frequently isolated from patients with periodontitis at early stage than the healthy individual group (70 % versus 40 %, respectively). In that paper, however, the prevalence of C. concisus over total Campylobacters microbiota was significantly higher in healthy individuals than that in patients with initial periodontitis (18.9 % and 9.0 % respectively with P value of 0.04) (13).

In a further study, Kamma et al. studied the composition of the sub-gingival microbiota in dental plaques collected from permanent teeth and deciduous teeth of 40 healthy juveniles (42). The detection rate of C. concisus in plaque samples from permanent teeth (40 %, 16/40) was significantly higher than that from deciduous teeth (15%, 6/40). Moreover, the relative percentage of C. concisus among 160 bacterial species colonising in permanent

teeth was also significantly higher than that in deciduous teeth (5.5 % and 2.7 %, respectively) (42).

Despite the accumulated clinical statistics and scientific research over the last three decades, the role of C. concisus periodontitis and gingivitis is still not conclusive. This was largely due to the fact that most of the previous studies did not include samples collected from healthy controls.

1.4.2. Intestinal disease

1.4.2.1. Enteritis

Evidence suggests that C. concisus may be involved in enteritis. In a study from Lastovica et al. published in 1993, 6,111 diarrheic stool samples were subjected to enteric pathogen isolation by a filtration method and culture method. Of these 6111 samples, 1,519 samples were positive for Campylobacter, Helicobacter or Arcobacter species (54). Among these bacterial species, C. concisus had 3rd highest prevalence (187/1,519, 12.3 %) (54). The established enteric pathogen Campylobacter jejuni had a prevalence of 42.5 % (646/1519).

Among patients with diarrhoea, C. concisus was more often found in young juveniles. In 1991, Lauwers et al. collected stool samples from 3,165 juvenile patients and 1,265 adult patients with diarrhoea and vomiting. The isolated C. concisus was from 2.4 % of stool samples collected from juvenile patients and 1.5 % of stool samples collected from adult patients. Of the juvenile patients who were positive for C. concisus isolation, 54 % of them were younger than one-year-old (55). However, most of these studies investigating the prevalence of C. concisus in diarrheal stool samples did not include control faecal samples from healthy individuals, which makes it difficult to conclude whether C. concisus is associated with diarrhoeal disease based on these data.

Some other studies have included control groups. Van Etterrijck et al. examined the presence of C. concisus by culture method in stool samples collected from 174 juvenile patients with enteritis and 686 healthy juveniles. They found that the prevalence of C. concisus in patients and controls was not statistically different (56). Engberg et al. in 2000 also showed that the isolation rate of C. concisus in faecal samples between diarrheic

patient group and healthy controls is insignificantly different (3/107 versus 5/107 respectively) (46).

In contrast to previous two studies, a study from Nielsen et al. in 2013 isolated Campylobacter species from stool samples collected from 1,532 patients with diarrhoea and from 108 healthy volunteers. In this study, a significantly higher isolation of C. concisus in patients with diarrhoea (26 %, 400/1,532) as compared with the healthy volunteers (0/108) was reported (47). In addition to isolation of C. concisus from stool samples, the level of C. concisus specific antibody was also measured previously. In 1996, Zhi et al. measured the antibody specific to C. concisus in sear samples collected from 88 patients with diarrhoea (24 juveniles and 42 adults) by the enzyme-linked immunosorbent assay (ELISA). They found that the C. concisus specific antibody level in juvenile patients with enteritis was significantly higher than that in healthy juvenile group (P < 0.001). Similarly, adult patients group was reported to have a significantly higher C. concisus specific antibody level in sera than healthy adult group (P < 0.001) (57). These data support a role of C. concisus in diarrheal disease.

1.4.2.2. Inflammatory bowel disease

Inflammatory bowel disease (IBD) is a group of chronic gastrointestinal disorders, which are characterised by uncontrolled mucosal immune responses to uncertain causes (58). Crohn’s disease (CD) and Ulcerative colitis (UC) are two main types of IBD. The incidence rate of CD and UC was reported to increase globally and reviewed in many scientific journals (59, 60). The main symptoms of IBD include abdominal pain, bleeding, fever, arthritis, growth failure, renal disease and bone abnormality (61, 62). Moreover, patients with IBD may also develop other chronic inflammatory diseases such as psoriasis, ankylosing spondylitis (63) and primary sclerosing cholangitis (64) as well as colorectal cancer (65) and small intestine adenocarcinoma (66).

The exact aetiology of IBD still remains unclear. It is generally believed that IBD is multi factorial disease affected by a number of risk factors, including environmental factors, immunological dysregulation, genetic factors and enteric microbiota.

1.4.2.2.1. Intestinal microbiota

Animal model studies showed that the intestinal microbiota contributed to the development of IBD. In the mid-1990s, a number of IBD mice models were established. These include mice lacking T-cell receptor (TCR) chains such as α, β, or δ chains (67), transforming growth factor (TGF)-β (68), interleukin (IL)-10 (69) and IL-2 (70). It was reported that these mice spontaneously developed colitis that had some of the features of IBD in human when grown under conventional conditions or specific pathogen free conditions. Specific pathogen free conditions are defined as conditions where known viral and bacterial murine pathogens are not detected. However, when these mutants were kept in germ-free conditions, they did not develop intestinal inflammation (67, 70-73).

In humans, faecal stream diversion, which prevents passage of intestinal contents through the intestinal inflamed tissue by means of surgical procedure in patients with CD ceased their intestinal inflammatory lesions. After receiving re-anastomosis, these patients had recurrence of disease both endoscopically and histologically (74). Furthermore, clinical amelioration of CD by probiotics and antibiotics supports a role of intestinal bacteria in IBD (75-77). However, the exact bacterial specie among intestinal micro-flora contributing to the onset of IBD is unclear.

1.4.2.2.2. C. concisus as a potential initiator of IBD

It was reported that a significantly higher prevalence of C. concisus in intestinal biopsy samples and faecal samples taken from paediatric patients with IBD. In 2009, Zhang et al. demonstrated a significantly higher prevalence of C. concisus DNA in the biopsy specimens collected from paediatric patients with CD (82 %) as compared with healthy individuals (23 %), but not other Campylobacter species. In addition, the level of C. concisus specific immunoglobulin G (IgG) in sera collected from patients with CD was compared with that from healthy controls, and the difference was significant (P value < 0.001) (78). In 2011, the same laboratory group found the presence of C. concisus DNA in faecal samples collected from paediatric patients with CD (54 subjects), non-IBD controls (27 subjects) and from healthy controls (33 individuals). This study showed the significantly higher prevalence of C. concisus in paediatric patients with CD than the other

groups (P = 0.008 with healthy controls and P = 0.03 with non-IBD controls (79). In contrast, Hansen et al. observed insignificantly difference of C. concisus prevalence in colonic mucosal biopsies collected from paediatric patients with CD as compared with endoscopically healthy controls (80).

The significantly higher prevalence of C. concisus DNA was also reported in adult patients with IBD. Mahendran et al. compared the C. concisus detection rates in ileum, caecum, colon and rectum biopsy samples in adult patients with CD and control group, and the significantly higher prevalence of C. concisus DNA in colonic biopsy collected from patients with CD was shown as compared with the control group (P < 0.05). However, the prevalence of C. concisus in biopsy samples collected from other sites was not significant between patients with CD and controls (81), suggesting that the colonic environment may be more suitable for C. concisus to grow.

In addition to patients with CD, the prevalence of C. concisus in patients with UC was studied. In 2011, Mukhopadhya et al. compared the presence of C. concisus DNA from the intestinal biopsy samples between 69 adult patients with UC and 65 healthy individuals, and showed a significantly higher prevalence of 16S rRNA gene in patients with UC as compared with healthy individuals by PCR detection and nested PCR (P = 0.0001 and P = 0.0019, respectively) (82). Mahendran et al. confirmed the significantly higher prevalence of C. concisus in biopsy samples from adult patients with UC in comparison with the healthy group (P < 0.05) (81).

In contrast to intestinal biopsy and faecal samples, the detection rates of C. concisus DNA in a saliva sample between the patient groups with IBD and healthy controls were not significantly different (97 % and 100 % respectively) (43). The isolation rate of C. concisus in a saliva sample from patients with IBD was not significant different from that in healthy controls (88.8 % vs 75 % respectively) (43). This is because human oral cavity is the natural colonisation site for C. concisus as mentioned in Section 1.3.

In summary, the significantly higher detection rates of C. concisus were found in intestinal biopsies and faecal samples of patients with IBD, including CD and UC, suggesting that this bacterium may be associated with enteric IBD (Table 1-3).

Moreover, this disease association is more likely to be an initiator. The development of IBD can be aggravated by an bacterial initiator, triggering the early stage of IBD, and exacerbator, sustaining the inflammation (10). It has been suggested that the pathogenic role of C. concisus in IBD is more likely to be an initiator regarding a clinical manifestation of IBD (10), which is that (1) CD is sporadically distributed throughout the GIT, suggesting that the initiators are likely to exist at the upper GIT, and that (2) the relapse in patients with IBD is frequent, indicating that the patients with IBD are frequently exposed to those initiators (10).

Table 1-3. Significantly higher prevalence of C. concisus in patients with IBD as compared to that in healthy individuals

Year Author Method Sample Age Disease P value Reference 2009 Zhang PCR Inflamed tissue biopsy Paediatric CD < 0.0001 (83) IgG level Blood Paediatric CD < 0.001

2010 Zhang PCR Saliva All range IBD NS# (43)

2011 Man PCR Faecal sample Paediatric CD 0.008 (79)

2011 Mahendran PCR Colonic biopsy^ Adult CD < 0.05 (81) PCR Biopsy Adult CD NS# PCR Biopsy Adult UC 0.05 PCR Colonic biopsy^ Adult IBD < 0.05

2011 Mukhopadhya PCR Colonic biopsy^ Adult UC 0.0001 (82) and 0.0019* ^ Colonic biopsy sample is a biopsy specimen specifically taken from colon in subject * P values are 0.0001 and 0.0019 by the means of PCR detection and nested PCR detection, respectively # NS indicates insignificant differences or P value > 0.05

1.5. Virulence properties of C. concisus

Various potential virulence properties of C. concisus, including motility, chemotaxis, adherence, invasion, translocation, production of toxins, production of hydrogen sulfide and host immune response to C. concisus infection and breakdown of tight junctions have been suggested. While some of these potential virulence factors have been experimental characterised, the others were isolated through bioinformatics analysis.

1.5.1. Motility, adherence, invasion and translocation

C. concisus has a screw-like motility with a unipolar flagellum, which provides the ability to penetrate an innate immune barrier in an intestine such as mucus layer or epithelial fence. In 2012, Lavrencic et al. performed the motility assay on C. concisus strain UNSWCD strain in the liquid media with various viscosities, and found that C. concisus was able to maintain the velocity in the liquid media with from 20 cp to 80 cp of viscosity (84). Bacteria such as Vibrio cholerae, Pseudomonas aeruginosa, and Salmonella typhimurium have the highest moving velocity at 1.5 – 3.0 cp and can gradually increase their velocity to 10 to 100 cp (85). Although these bacteria are flagellated, they are not spiral-shaped like C. concisus. This morphology of C. concisus may assist in the penetration of the bacterium into the thick mucus layer of intestinal epithelial cells (84). Another virulence mechanism of C. concisus is its ability to adhere, invade and translocate to host cells. Using scanning electron microscopy, Man et al. observed adherence and invasion of C. concisus strain UNSWCD in Caco2 cells (human epithelial colorectal cell line) (86). In addition, the preferential adherence of C. concisus strain to intercellular junction space between HT-29 cells, human colorectal epithelial cell line was also observed (86).

Following this, the same team conducted adherence and invasion assay using strains including C. concisus strain UNSWCD (isolated from an intestinal biopsy sample in patients with CD), C. concisus strains ATCC-51561 and ATCC-51562 (isolated from a faecal sample in patients with acute gastroenteritis). This investigation showed C. concisus strain UNSWCD exhibiting higher invasion and adherence rate as compared with the other

strains (87). This suggests that some C. concisus has a higher ability to invade and adhere to intestinal epithelial cells and could be associated with chronic inflammation (87).

Previously, Kalischuk and Inglis isolated 23 C. concisus strains from patients with diarrhoea and healthy individuals. The authors measured their adherence, invasion, and translocation in T84 cell line (human colon epithelial cell line) using these bacterial strains (88). As the result, there were no significant differences in the adhesion, invasion and translocation rates between two subject groups (88). This suggests that the ability of invasion, adherence and translocation of C. concisus has an unconvincing disease correlation with diarrhoea.

Oral C. concisus strains isolated from patients with IBD and healthy controls were also examined for their ability to invade Caco2 cells (89). Fifty percent of oral C. concisus strains isolated from patients with IBD were invasive (enteric invasive C. concisus (EICC)) to Caco2 cells, while none of the oral C. concisus strains isolated from healthy controls were invasive (89). Moreover, the EICC strains were genetically closely related and were grouped in the same cluster (Cluster 1) (89). 87.5 % of individuals having the strains belonged to Cluster 1 had inflammatory enteric diseases, and this was significantly higher as compared with the remaining individuals (28.6 %) (P < 0.05) (89).

1.5.2. Toxins in C. concisus

Over the years, a number of potential toxins have been implicated in the pathogenesis of C. concisus.

1.5.2.1. Phospholipase A

Phospholipids represent the major chemical constituents of the host cell membrane. In 1997, Songer et al. found that phospholipases constituted a very diverse subgroup of lipolytic enzymes that have the ability to hydrolyse ester linkages in phospholipids. Consequently, a wide variety of bacteria has evolved to produce enzymes, which are capable of hydrolysing phospholipids (90). In 2004, Istivan et al. detected haemolytic

PLA2 activity in C. concisus strains isolated from paediatric patients with gastroenteritis

(91). In 2008, the same author conducted the cloning of the pldA gene encoded for PLA2

enzyme of C. concisus strain and confirmed the haemolytic activity of pldA in E. coli (92). This data was further supported by Kalischuk and Inglis, who measured the haemolytic activity of C. concisus strains in sheep erythrocytes (88).

1.5.2.2. S-layer repeats-in-toxin

Surface-layer (S-layer) is composed of a single secreted protein, and provides a 2D array surrounding the entire cell surface. This layer is found in Gram positive, Gram negative and Archaea and may provide an important functional property involved in pathogenicity (93). S-layer toxins were shown to provide an evasive mechanism against complement C3b in C. fetus strain 23D, while a spontaneous mutant C. fetus strain 23B lacking S-layer gene was more vulnerable to complement C3b (94).

Ausiello et al. showed that S-layer protein A in C. difficile induced the secretion of proinflammatory cytokines IL-1β and IL-6 from monocytes, maturation of monocyte- derived dendritic cells and enhanced proliferation of allogeneic T cells (95).

In 1993, Gillespie et al. isolated a secreted cytotoxin from C. rectus (96). This toxin was later characterised to be S-layer having repeats-in-toxin (RTX) at the C-terminus (97). RTX is secreted by several Gram-negative bacteria in the host cells. It contains repetitions of glycine- and aspartate-rich sequences, binds to Ca2+ ions, and can be secreted out by ATP-binding cassette transporter-based secretion via type I secretion system (98). It acts as a pore-forming exotoxin and haemolysin, and it also causes elevation of Ca2+ ion concentration in target cells (98). In addition, the RTX is also present in several pathogenic organisms such as HlyA in uropathogenic E. coli, LktC in Pasteurella haemolytica, CyaA in Bordetella pertussis, RtxA in Vibrio cholerae as well as the protein encoded by CCC13826_1838 in C. concisus (98).

Bioinformatics’ analysis by Kaakoush et al. predicted S-layer RTX to be secreted by C. concisus strain UNSWCD (99). Kalischuk and Inglis also detected the presence of a gene which encoded for S-layer RTX by PCR in two C. concisus isolates (CHRB3287 and CHRB2004) (88).

1.5.2.3. Zonula occludens toxin

The gene encoded for Zonula occludens toxin (Zot) was first discovered in human enteric pathogen V. cholerae (100). The zot gene in V. cholerae is a component of CTXφ prophage integrated into the V. cholerae genome (101). Interestingly, only V. cholerae strains carrying CTXφ caused epidemic and pandemic cholera in the past (102).

In 2000, Wang et al. showed that V. cholerae Zot protein was a homologous protein of human Zonulin protein (103). Studies have shown that V. cholerae Zot modulates paracellular flows by regulating tight junctions in a reversible manner following activation of protein kinase C (PKC) -α (104).

Recently, it was reported that C. concisus also carries the zot gene. In 2011, Kalischuk and Inglis detected the zot gene in 80 % and 20 % C. concisus strains isolated from faecal samples of healthy and diarrheic human individuals respectively. However, the sample size in this study was very small In 2013, Mahendran et al. detected zot gene in C. concisus strains isolated from saliva samples from 40 % of healthy individuals and 54.5 % of patients with active IBD, the prevalence of zot positive oral C. concisus strain was not significantly different in patients and controls (P = 0.22) (105). However, this study found that a specific polymorphism of Zot protein in C. concisus was associated with active IBD. Thus, Zot with the substitution of valine at position 270 was significantly higher prevalent in C. concisus collected from patients with active IBD (36.4 %) compared with that from the healthy controls (0/20) (P = 0.011) (105).

However, currently there is no available experimental data regarding whether C. concisus has Zot pathogenic effects on humans.

1.5.3. Production of hydrogen sulfide

2- Sulfate-reducing bacteria (SRB) utilise inorganic sulfate (SO4 ) as a terminal electron acceptor in the fermentation process of various substrates, resulting in the production of hydrogen sulfide (H2S). C. concisus is able to undergo sulfur respiration to form H2S (106), which can classify them as SRB specie. SRB plays a physiological role in anaerobic degradation of undigested and unabsorbed carbohydrates, fermentation of human derived

substrates such as mucin, reducing sulfate and removal of hydrogen (107). The typical H2S concentrations ranged from 0.18 to 0.29 µmol/gram of wet weight of colonic contents in mouse intestine (108), from 0.29 to 3.4 µmol/mL (10 to 115 ppm) in human faeces (109) and from 0 to 1.9 µmol/gram of wet weight of colonic contents in human (110).

The possible association between SRB and IBD has been described. Gibson et al. showed that there was a more than a two-fold increase in sulfate reduction in faecal samples collected from patients with UC in the comparison with healthy control, suggesting an increased prevalence of SRB species in faecal collection from UC group (111). This study was confirmed by Levine et al. (112). Mills et al. detected elevated SRB population in the microbiota from patients with UC as compared with non-UC volunteers (113). The Desulfovibrio genus, well known SRB species in a human gut, was reported to be involved in the onset of UC (111, 114, 115). However, other studies did not find the association between SRB and UC (116, 117).

Exposure to exogenous H2S may affect the function of various organs and systems in both humans and animals (118). It was shown that the exposure to H2S with more than 500 ppm can cause respiratory paralysis, neural paralysis, cardiac arrhythmias and even death in humans (119). In addition, it was reported that the exposure of sodium hydrosulfide, NaHS, (10 to 1,000 μM = 0.56 to 56.06 ppm) to intestine affected the intestinal motility through reduction of acetylcholine-mediated contraction using rabbit, guinea pig, and mouse intestines (120).

H2S exhibits cytotoxicity and genotoxicity to human cells. Butyrate is a main energy source to colonic epithelial cells. Roediger et al. in 1993 examined the oxidation of butyrate in presence of various concentrations of H2S in rats. This study found that the exposure to H2S inhibited butyrate ketogenesis (or β-oxidation) in a dose-dependent manner in colonic epithelial cells, suggesting that exposure to H2S may induce an energy- deficiency state in colonic epithelial cells (121). Christl et al. measured the proliferation rates in the mucosal biopsy sample from the sigmoid colon of patients with CD in the presence of either sodium chloride, butyrate or sodium hydrogen sulfide by bromodeoxyuridine labelling, which can be combined with the newly synthesised DNA of replicating cells during the S-phase of the cell cycle. It was found that sodium hydrogen

sulfide significantly enhanced the hyper-proliferation rate at crypts by 19 % (P<0.05) as compared with sodium chloride and butyrate, suggesting that H2S can affect the cellular metabolism in a colonic epithelium (122).

H2S also exhibits genotoxicity in human cell lines. Baskar et al. observed that H2S induced DNA damage; exposure of human lung fibroblast cells resulted in the formation of micronuclei, apoptosis, time dependent and dose dependent induction of p53, p21, ku70 and ku80 and release of cytochrome c into cytosol. In 2006, Attene-Ramos et al. showed that H2S induced chronic cytotoxicity in CHO cells and HT29-Cl.16E cells, and even the low concentration of H2S (250µM), which was similar to what found in the colon, can significantly cause genotoxicity in CHO cells (123). The same author treated the isolated free nuclei from CHO cells with H2S and observed that H2S directly caused the genotoxicity without the involvement of cellular metabolism (124). Further studies showed that the PTGS2 and WNT2 genes were also up-regulated in CHO cells after four-hour exposure to hydrogen sulfide (125). PTG2 gene is known to be participated in carcinogenesis, immune response suppression, inhibition of apoptosis, angiogenesis and tumour cell invasion and metastasis (126). WNT2 gene is also a well characterised gene that has a role in carcinogenesis (127). Furthermore, gingival epithelial cells exposed to hydrogen sulfide had suppression of DNA synthesis, cell cycle arrest at G1 phase, Rb phosphorylation decrease and p21 elevation (128).

In addition to cytotoxicity and genotoxicity, the exposure of H2S to epithelial cells promotes the production of cytokines. Using human gingival epithelial cell line OBA9 and cheek epithelial cell line OKF6, the significantly higher secretion of IL-8 in epithelial cells exposed to exogenous H2S, which was produced by Porphyromonas gingivalis (107cells/mL), was reported (129).

In contrast, the endogenous H2S, converted from L-cystein by cystathionine β-synthase, showed some beneficial effects. In 2006, Distrutti et al. measured the response to pain stimuli (colorectal distension; insertion of the balloon to rectal capacity) in rats given intraperitoneal injection of either sodium hydrogen sulfide or L-cystein using a semi- quantitative abdominal withdrawal reflex (AWR) scoring system. They found that administration of sodium hydrogen sulfide and L-cystein decreased the AWR as compared

with the non-treated group, suggesting antinociceptive effects (reduction of painful stimuli) in rodent colon tissue (130). In 2006, Zanardo et al. showed that endogenous H2S produced by endothelium suppressed the adherence of leukocytes and infiltration via ATP- sensitive potassium channel (131), suggesting endogenous H2S production is involved in the natural healing process associated with inflammation. Furthermore, the endogenous

H2S was considered as a neuromodulator, Abe et al. detected abundant expression of H2S- producing enzyme in the hippocampus and enhanced current by sodium hydrogen sulfide in hippocampal CA1 pyramidal cells (132).

Previously, Tanner et al. reported that C. concisus was able to produce H2S under microaerobic conditions enriched with hydrogen (8). However, the production of H2S by C. concisus strains under different atmospheric conditions has not been investigated.

1.5.4. Impaired tight junctions by C. concisus

The intestinal epithelial tight junctions maintain the cellular polarity and integrity of the intestinal epithelial barrier. The tight junction consists of multiple sub-structural molecules such as ZO-1, ZO-2, ZO-3, occludens, claudins and junctional adhesion molecules (JAM) (Figure 1-1), and these molecules are connected to filamentous actin (F-actin). The disruption of this structure leads to exposure of intestinal tissues to increased load of antigens from the lumen through paracellular spaces, which may result in inappropriate immune responses (Figure 1-1). Damaged intestinal epithelial barrier function is associated with patients with IBD (133-135).

Previously, it was shown that C. concisus was able to induce the impairment of tight junction molecules using in vitro intestinal epithelial cell models. In 2010, Man et al. examined the impact of C. concisus infection on ZO-1 and occludens expression level in Caco2 cells by western blot. The authors found that the protein level of membrane- associated ZO-1 and occluden in Caco2 cells significantly decreased as compared with untreated Caco2 cells, while the protein levels of total ZO-1 and occludens in Caco2 were not affected by C. concisus infection (86), suggesting that C. concisus may induce the internalisation of membrane-associated tight junction molecules to cytosol.

Figure 1-1. Schematic view of role of tight junction in intestinal epithelial cells against luminal antigens Tight junctions are composed of several transmembrane and intracellular molecules. The most extensively studied cell surface molecules of the tight junctions are occludin, claudins, and JAM. These proteins interacted with many intracellular molecules such as ZO-1, ZO-2, and ZO-3, which were in turn link to the F-actin cytoskeleton (left side). Defeats in intestinal epithelial barrier function lead to the influx of luminal antigens, which may result in inappropriate immune responses (right side with red arrow).

Nielsen et al. measured the expression of variety of claudin molecules in HT-29/B6 cells in response to C. concisus, and showed that the infected HT-29/B6 cells expressed a significantly reduced level of claudin-5 after 48-hour infection, as compared with untreated cells (P < 0.05), while the expression level of claudin-1, -2, -3, -4 and -8 were not changed significantly (45). In this paper, the apoptosis, necrosis and activation of caspase-3 in HT-29/B6 cells in response to C. concisus were also observed as compared with the negative control (45). The same authors also examined intestinal permeability in the presence and absence of C. concisus on HT-29/B6 cells by measuring Trans endothelial electrical resistance (TEER) and fluorescein translocation through the monolayer. They found that C. concisus decreased TEER value in a dose-dependent manner, suggesting an enhanced intestinal permeability. The permeability for fluorescein was also significantly increased after C. concisus infection (P < 0.01) in comparison with the negative control. However, there was no significant difference between C. concisus isolates from oral and faecal samples (45). The increased intestinal permeability by measurement of TEER was also observed in Caco2 cells infected by C. concisus strain UNSWCD (86).

Some human pathogens can directly affect the distribution of actin in the host cell such as Salmonella spp. and Yersinia pseudotuberculosis, (136) or utilise actin filament to facilitate its movement such as Listeria, Rickettsia, Burkholderia, Shigella and Mycobacterium (137). However, there is no experimental data showing whether C. concisus impacts on actin expression in the host cell.

1.6. Hypothesis and aims

Accumulated evidence suggests that C. concisus is involved in the initiation of inflammatory bowel disease (IBD). The mechanisms by which C. concisus, an oral bacterium, initiates enteric diseases are not known. The hypothesis of this project is that C. concisus is an opportunistic pathogen, and its enteric virulence is determined by multiple factors. This project has five specific aims, through which to elucidate the enteric virulence of C. concisus.

The specific aims of this thesis are;

 To examine the impact of hydrogen gas on the growth of C. concisus under anaerobic conditions

 To examine effects of hydrogen gas, formate and fumarate on the production of hydrogen sulfide in C. concisus strains

 To examine prophages in the genome of C. concisus

 To investigate the expression and secretion of zonula occludens toxin in C. concisus and its effects on Caco2 cell monolayer permeability

 To investigate the effects of C. concisus on the level of β-actin in Caco2 cells

Chapter 2: General materials and methodology

2.1. Buffers

2.1.1. Phosphate buffer saline (PBS)

One litre of PBS was made by dissolving following mixture in 1.0 litre of distilled water; 8.0 g sodium chloride (Ajax Finechem, NSW, AUS), 0.2 g potassium chloride (Ajax Finechem), 1.44 g disodium phosphate (Ajax Finechem), and 0.24 g monopotassium phosphate. The pH was adjusted to 7.4. The solution was then autoclaved for 15 minutes at 120 °C.

2.1.2. Tris-ethylenediaminetetraacetic acid (TE) Buffer

TE buffer contained 10mM Trizma-base (Sigma-Aldrich, MO, US) and 1 mM ethylenediaminetetraacetic acid (EDTA) 161-0729 (Bio-Rad, CA, US) in a final volume of 1 litre. The autoclave was followed to sterilise the buffer.

2.1.3. 50× Tris acetic acid EDTA (TAE) Buffer

50× TAE buffer stock solution was prepared by dissolving 242 g Trizma base (Sigma- Aldrich), 57.1 mL of glacial acetic acid (Ajex Finechem) and 100 mL of 500 mM EDTA pH 8.0 solution in distilled water. The final volume was 1.0 litre. This stock solution was diluted to 1× working solution (50 time dilution).

2.2. Bacterial culture

2.2.1. Media for bacterial culture

Horse blood agar (HBA) was used to culture Campylobacter concisus strains. HBA was prepared by dissolving 20 g of Blood Agar Base No.2 (Oxoid, SA, AUS) in 500 mL of distilled water. The mixture was autoclaved for 15 minutes at 120 °C. Defibrinated horse blood (30 mL) (Oxoid) was added after cooling down to 50 °C.

Horse blood agar with vancomycin (HBAV) is a HBA supplemented with 10 µg/mL of vancomycin (Sigma-Aldrich, MO, US), which was used to prevent any contamination from Gram positive bacteria in these experiments.

Nutrient agar (NA) was used to cultivate Escherichia coli. Fourteen grams of nutrient agar powder CM0003 (Oxoid) was dissolved in 500 mL of distilled water, and sterilizing by autoclaving for 15 minutes at 120 °C.

Bacteria freezing media contained 10 mL of heart infusion broth (Oxoid) and 3.03 g of glycerol (Ajax Finechem). The media was autoclaved for 15 minutes at 120 °C before use. If necessary, antibiotics were supplemented with the appropriate amount.

2.2.2. Bacterial cryopreservation

Bacteria were cultured on appropriate agar plate at 37 °C. The morphology was confirmed by wet prep under the microscope. One or two loops of bacteria from the agar plate were inoculated into 2.0 mL cryotube containing 500 µL of bacterial freezing media. In order to provide the cooling rate of -1 °C/minute (138), the cryotube was incubated in -20 °C freezer for two hours and then transferred to -80 °C freezer.

2.2.3. C. concisus culture

In our previous studies, all the C. concisus strains used in this thesis were isolated by filtering methods and characterised by 16S rRNA sequencing and protein profiling (89, 105).

C. concisus was cultured on either HBA or HBAV plates under either microaerobic or anaerobic gas condition supplemented with 5 % (v/v) hydrogen at 37 °C for two days according to the purpose. A 2.5-litre anaerobic jar (Oxoid) was used. Hydrogen gas was generated using commercially available gas packs from Oxoid. These gas packs were used according to the manufacturer’s instruction. Moreover, this study was carried out in accordance with the Health & Safety procedures of the UNSW. Microaerobic gas condition with hydrogen gas was generated by gas generation kit Campylobacter system BR56A (Oxoid) and anaerobic conditions with hydrogen gas were generated by AnaeroGenTM system AN25A (Oxoid). Hydrogen gas was generated by dissolving 0.42 g of sodium borohydride (Fisher Biotech, WA, AUS) in 10 mL of water. The chemical reaction equation is as following,

NaBH4 (aq) + 4 H2O = 4 H2 (g) + NaB(OH)4 (aq)

The purity of C. concisus culture was confirmed by a wet mount microscopy prior to each experiment.

2.2.4. E. coli culture

E. coli were cultured on NA plate for overnight at 37 °C in aerobic condition.

2.2.5. Quantification of the number of bacteria

The number of bacteria was estimated by two methods; colony-forming unit (CFU) counting or optician density (OD) measurement at 600 nm.

2.2.5.1. Colony forming unit (CFU) counting

Bacterial sample was diluted by 10 fold serial dilution from 100 to 10-8. Five microliter from each dilution was blotted on the appropriate agar plate in quadruplicates. After appropriate incubation time (for instance, two days for C. concisus and overnight for E. coli), CFU on each blot was counted. CFU per mL was calculated as followings,

CFU per mL = [average CFU number from quadruplicates] × 200 × [dilution factor from the serial dilution

2.2.5.2. Absorbance measurement at 600nm

The bacterial suspension was collected, washed with PBS twice and resuspended with 1 mL of PBS. Of this bacterial suspension, 200 μL was subjected to the spectrophotometre analysis. The OD was measured at 600 nm using the SPECTRAmax 340 microplate reader was used (Molecular Devices, CA, US). The CFU number was estimated using the standard curve of CFU number of C. concisus in 200 μL suspension over optical absorbance reading at 600 nm, generated by the previous PhD student.

2.2.6. Preparation of Caco2 cell lysates

Monolayer of either HT-29 cells or Caco2 cells was washed five times using DPBS (Invitrogen, CA, US) to remove non-viable cells and antibiotics. Adherent cells were detached by scrubbing with a sterile P1000 pipette tip (Axygen, CA, US) in 1.0 mL of DPBS. The cell suspension was transferred to 10 mL or 25 mL falcon tubes (Sigma- Aldrich or Fisher Biotech), and then was centrifuge at 300xg for five minutes. Cell pellet was then resuspended with 400 µL of radioimmunoprecipitation assay (RIPA) buffer (25 mM Tris-HCl pH 7.6, 150 mM NaCl, 5 mM EDTA, 1 % (w/v) Triton X-100, 1 % (w/v) sodium deoxycholate, 0.1 % (w/v) sodium dodecyl sulfate (SDS)) or RIPA lysis and extraction buffer (Cat. No. 89901) purchased from (Thermo Fisher Scientific. MA. US). Protease cocktail (Invitrogen) was added. Prepared samples were stored at -80 °C.

2.2.7. Preparation of bacterial lysates

Bacteria were collected from culture plates with 1.0 mL of PBS. Bacterial cell was further washed with 1.0 mL of PBS twice by centrifugation at 9,000 xg for 2 minutes and resuspension with 1.0 mL of PBS. Following this, bacterial pellet was resuspended in 300 µL to 600 µL of PBS. On this cell suspension, freezing/thaw cycle was performed three times. For the freezing step, the bacterial suspension was kept in liquid nitrogen for 1 minute, and for the thawing step, the frozen bacterial suspension was kept in water bath at 20°C till complete defrosting. After thawing, sonication method was used (3 second on/off interval, 40 % amplitude, 3 minutes) (139).

2.3. Caco2 cell culture

2.3.1. Media for cell culture

For Caco2 cell culture, minimum essential medium 11095-098 (MEM) (Invitrogen), supplemented with 10 % (v/v) heat-inactivated FBS (Invitrogen), 1 mM sodium pyruvate (Invitrogen), 20mM L-glutamine 25030-081 (Invitrogen), 0.1 mM non-essential amino acids (Invitrogen), 2.25mg/L sodium bicarbonate (Invitrogen), 100 unit/mL penicillin and 100 µg/mL streptomycin (Invitrogen) was used.

2.3.2. Maintenance of the Caco2 Cells

The Caco2 cells were maintained in the T25 cm2 tissue culture flask (Corning, NY, US) with 10 mL of the MEM media under the aerobic condition with 5 % (v/v) CO2 at 37 ºC.

The CO2 incubator model C 150 (Binder, Tuttlingen, Germany) was used.

When the confluence of cells reach 85~95 %, passage of Caco2 cell was performed. The monolayer of cells was washed with 10 mL of Dulbecco's PBS (DPBS) (Invitrogen) once. Trypsin-EDTA 0.25 % (Invitrogen) with the volume of 3.0 mL was added and incubated for 7 minutes at 37 °C. Trypsin was diluted by adding 10 mL of MEM media. Cells were pelleted at 300 xg for 5 minutes and resuspended with 10 mL of MEM media. After the determination of the cell number and viability, the cells were seeded into the new T25 cm2 tissue culture flask at the concentration of 1x105 cell/mL in 10 mL of MEM media. In this manner, 85~95 % confluence was achieved every 7 days. Culture media was replenished with the fresh media every three days.

All the cell culture was performed aseptically under the biosafety cabinet hood.

2.3.3. Determination of cell number and viability

Cell numbers and viability were determined using hemocytometer (Biobase, Wolfenbüttel, Germany). Prior to counting, cell suspension (100µL) was stained with equal volume (100µL) of trypan blue solution 15250-061 (Invitrogen) in order to distinguish viable cells.

 The number of cells per mL = [average number of cells in each large square]×[dilution factor]×104  Viable cells % = 100×[1-(number of blue cells)÷(number of total cells)]

2.3.4. Cryopreservation of Caco2 cells

For the cryopreservation, only cells with the viability more than 90 % were used. The cells were collected, washed with DPBS, and pelleted by centrifugation at 300 xg for five minutes. The cell number was determined by the hemocytometer. The cells were resuspended with MEM media containing 10 % (v/v) dimethyl sulfoxide (DMSO) (Sigma- Aldrich) to make 1x 106 viable cells/mL. Aliquots at 1.0 mL were prepared in a 2 mL cryogenic storage vials (Greiner Bio One, Frickenhausen, Germany). Each vial was placed in a freezing container Mr. Frosty (Nalgene) in -80 °C freezer, which provides 1 °C/min cooling rate required for successful cryopreservation.

2.3.5. Revising cells from liquid nitrogen

Frozen cells from liquid nitrogen was placed in a 37 °C water bath until cells were about to thaw. Cells were mixed 10 mL of MEM media containing 20 % (v/v) FBS aseptically under the biosafety cabinet hood and pelleted at 300 xg for 5 minutes. Cells were re- resuspended with 10 mL of the same 20 % (v/v) FBS media and seeded in the T25 cm2 tissue culture flask. Cells were then maintained at 37 °C.

2.4. Bicinchoninic acid (BCA) assay

Pierce™ BCA Protein Assay Kit #23227 (Thermo Fisher Scientific) was used to measure the total protein concentration in a solution. This kit contains BCA reagent A and BCA reagent B. Copper (II) ion will be reduced to copper (I) ion and in turn BCA molecule with the copper (I) ion will form BCA-Cu(I) complex, which gives purple colour. First, the standard curve was generated using 2.0 mg/mL bovine serum albumin (BSA) stock standard by 2 fold serial dilutions. Second, BCA working solution was prepared by adding 1 potion of BCA reagent B to 50 potion of BCA reagent A. For each microwell plate, 200 μL of working solution is required. For example, 10 samples with triplicates will require 200 μL × 10 × 3 = 6 mL of working solution. For the standard, 200 μL× 8 × 3 = 4.8 mL of working solution will be required. Thus, 15 mL of working solution may be required roughly and will be prepared by adding 0.3 mL of BCA reagent B to 15 mL of BCA reagent A. Third, the unknown protein sample was diluted by 2, 5 or 10 time dilution in triplicates to get the working range from 20 to 2,000 μg/mL. Forth, 25 μL of the BSA standard and unknown sample will be loaded onto the 96 well micro plates with blank (the diluent, PBS in this thesis, was used) and subsequently 200 µl of BCA working solution was added. After following incubation for 30 minutes at 37 °C, the absorbance was measured at 562nm using SpectraMax 340 microplate reader and analysed by Softmax software (Molecular devices).

2.5. Sodium dodecyl sulfate poly-acrylamide gel electrophoresis (SDS-PAGE)

The SDS PAGE with 12 % (w/v) polyacrylamide separating gel and 5 % (w/v) polyacrylamide stacking gel was used to separate the protein according to the molecular weight.

2.5.1. Reagents for SDS-PAGE

12 % SDS-polyacrylamide separating gel contained 2.45 mL of milli-Q (Millipore) water, 2.25 mL of 40 % Acrylamide/Bis solution #161-0148 (Bio-Rad), 1.9 mL of 1.5M Tris pH 8.8 (Sigma-Aldrich) and 75 µL of 10 % SDS (Sigma-Aldrich). After addition of 50 µL of 10 % (w/v) ammonium persulfate (APS) (Sigma-Aldrich), and 5µL of N,N,N,N- tetramethylenediamine (TEMED) (Sigma-Aldrich) into the solution, the solution was immediately loaded to the glass plates.

5 % SDS-Polyacrylamide stacking gel contained 1.7 mL of milli Q water (Millipore), 310 µL of 40 % Acrylamide/Bis solution #161-0148 (Bio-Rad), 473 µL of 0.5M Tris pH 6.8 (Sigma-Aldrich) and 25 µL of 10 % SDS (Sigma-Aldrich). After addition of 20 µL of 10 % (w/v) APS (Sigma-Aldrich) and 2.5 µL of TEMED (Sigma-Aldrich) into the solution, the solution was immediately loaded to the glass plates.

2X sample buffer contained 25 mL of 0.5M Tris buffer (Sigma-Aldrich), 1 mL of 10 % SDS (Sigma-Aldrich), 20 mL of glacial glycerol (Ajax Finechem), 4 g of SDS (Bio-Rad), 2 mL of 2-mercaptoethanol (Sigma-Aldrich), 1mg of bromophenol blue (Sigma-Aldrich) and filled with milli-Q water (Millipore) in a final volume of 100 mL.

5X SDS running buffer contained 15.1 g of Tris (Sigma-Aldrich), 72.0 g of glycine (Ajax Finechem) and 5.0 g of SDS (Bio-Rad) in a final volume of 1 litre of milli-Q water (Millipore).

2.5.2. Preparation of SDS-polyacrylamide gel

Two glass plates, spacer plates with 1.0 mm integrated spacer #165-3311 (Bio-Rad) and short plates #165-3308 (Bio-Rad), were used to cast SDS-polyacrylamide gel. Depending on the volume of samples, the 1.5 mm integrated spacer #165-3312 (Bio-Rad) was used

instead of the spacer plates with 1.0 mm spacer. Two glass plates were held together with Mini-PROTEAN® Casting Frame #165-3304 (Bio-Rad). Before casting gel, the leakage between two plates was checked by adding propanol. After propanol removal, the 12 % SDS-polyacrylamide separating gel solution was gently mixed and loaded between glass plates to about 3/4 inch below the short plate. A small amount (500~1,000 µL) of propanol layer was applied over the top of the 12 % separating gel solution. The gel was allowed to stand for about 40 minutes. Once the gel was polymerized, the propanol was removed by the filter paper. The top layer of the gel was rinsed with milli-Q water. After removal of remaining milli-Q water, the 5 % SDS-polyacrylamide stacking gel solution was gently mixed and fully loaded between glass plates. The mini-PROTEAN® Comb 10-well, 1.0 mm (Bio-Rad) or mini-PROTEAN® Comb 10-well, 1.5mm (Bio-Rad) were inserted. The gel was allowed to stand for 40 minutes at room temperature. After complete polymerisation, the comb was removed, and the gel is ready for electrophoresis.

2.5.3. Protein denaturation

Protein samples were mixed with 2X sample buffer in the 1.5 mL eppendorf tube with 1:1 ratio and then boiled at 95 °C for five minutes. The tube was allowed to stand at room temperature for 1~2 minutes to cool down, and then was briefly spin-down.

2.5.4. Electrophoresis

The prepared SDS polyacrylamide gel cassette was inserted into the mini PROTEAN® tetra cell #165-8004 (Bio-Rad). The gel cassette and tank were filled with 1X SDS running buffer (diluted from 5X SDS running buffer stock solution). The wells of the gel were washed by gentle pipetting with 1X SDS running buffer.

The denatured protein samples were loaded onto the well. The electrophoresis was run at 30 voltages for 40 minutes, and then the voltage was changed to 80 volts. The electrophoresis was stopped when the reference dye (bromophenol blue in this study) reached the bottom of the gel.

Prestained SDS-PAGE standards #161-0318 (Bio-Rad) or Precision Plus Protein™ WesternC™ Standards #161-0376 (Bio-Rad) was included.

2.5.5. Coomassie brilliant blue staining

The separated protein by 12 % SDS-PAGE was stained in Coomassie brilliant blue R-250 solution. Simply, the gel was incubated in that solution for overnight on the orbital shaker with 30 rpm. The Coomassie brilliant blue staining solution was containing 0.25 g of Coomassie brilliant blue R-250 #161-0400 (Bio-Rad), 25 mL of methanol (Ajax Finechem), 10 mL of glacial acetic acid (Ajax Finechem) and 65 mL of milli-Q water.

The gels were destained in the distaining solution (300 mL of methanol, 100 mL of glacial acetic acid and 600 mL of milli-Q water) on the orbital shaker with 30 rpm till the protein bands of interest were clearly visible. The destaining solution was changed every 30 minutes.

2.6. Western blot

2.6.1. Reagents for western blot

Transfer buffer contained 3.03 g of Tris base (Sigma-Aldrich) and 14.4 g of glycine (Ajex Finechem) was dissolved in 800 mL of milli-Q water. The addition of 200 mL of methanol was followed to make a final volume of 1 litre.

Wash buffer contained 500 µL of Tween-20 (Sigma-Aldrich) was added into 1 litre of PBS buffer. This solution was not undergone through autoclaving process.

Blocking buffer was prepared by mixing 50 mL of the wash buffer with 2.5 g of skim-milk powder (genuine brand) to make 5 % (w/v) skim-milk solution. This buffer was not processed by autoclaving.

2.6.2. Transferring to the polyvinylidene fluoride (PVDF) membrane

The immune-blot PVDF membrane #162-0177 (Bio-Rad) was cut to fit the cassette. The PVDF membrane was soaked in 100 % methanol for 10 seconds to increase hydrophobicity. The membrane was quickly transferred to the transfer buffer and incubated for 15 minutes to equilibrate. Two 3M thick blot paper #170-3932 (Bio-Rad) and two foam pads #170-3933 (Bio-Rad) as well as the non-stained SDS-PAGE gel having the separated protein were soaked in the transfer buffer for 15 minutes to equilibrate.

The transfer-sandwich cassette was prepared by stacking soaked materials in the order; (anode side) the clear side of mini-gel holder cassette, foam pad, thick blot paper, SDS- PAGE gel, PVDF membrane, thick blot paper, foam pad and the black side of mini-gel holder cassette (cathode side). Between layers, bubble was removed by a roller. The transfer cassette was inserted into the mini trans-blot Central Core #170-3812 (Bio-Rad) in which the black side of the cassette faced toward the anode and the clear side faced toward the cathode. The Mini Trans-Blot Central Core was settled in the mini-PROTEAN tetra cell (Bio-Rad) and was filled with the transfer buffer. The magnetic stirring bar and ice pack were used. The transfer was run at 100 voltages. After 1.5 hours, the PVDF membrane was ready for immunoblotting.

2.6.3. Immunoblotting

The PVDF membrane was incubated with the mild agitation (rpm 40) in the blocking buffer overnight at 4 ºC. The PVDF membrane was washed once with the wash buffer, and treated with the primary antibody with the optimal dilution (e.g. 1:200, 1:500, 1:1000 and 1:2000) in the blocking buffer for 1 hour and 15 minutes. The PVDF membrane was washed three times with the wash buffer and treated with secondary antibody conjugated to horseradish peroxidase (HRP) with the dilution (e.g. 1:500, 1:1000 and 1:2000) in the blocking buffer for 1 hour and 15 minutes. After three-time washes, the PVDF membrane was kept in the wash buffer or PBS.

2.6.4. Visualisation

The probed PVDF membrane was incubated with Clarity™ ECL western blotting substrate (Bio-Rad) for 5 minutes in the dark. The image was achieved by LAS-3000 imaging system (Fujifilm, NSW, AUS).

Chapter 3: The anaerobic growth of Campylobacter concisus

3.1. Introduction

Campylobacter concisus is a Gram-negative bacterium that is commonly present in the human oral cavity (8, 43). In some individuals, C. concisus may colonise the intestinal tract and was found to be associated with inflammatory bowel disease (IBD) due to its significantly higher prevalence in the intestinal tract of patients with IBD as compared with controls (78, 79, 81, 140). IBD is a chronic inflammatory disease of the gastrointestinal tract (GIT) with unknown aetiology. Crohn’s disease (CD) and ulcerative colitis (UC) are the two major clinical forms of IBD (141). In addition to IBD, C. concisus was also often isolated from diarrheal stool samples, suggesting its possible involvement in diarrheal disease (46, 47, 142).

In the literatures, it was described that C. concisus requires hydrogen (H2)-enriched microaerobic conditions for growth (143, 144). In the laboratory cultivation of C. concisus, microaerobic conditions enriched with 5-10 % (v/v) of H2 have been used (8, 46, 91, 145). The primary colonisation site of C. concisus is the human oral cavity (8, 43). The level of

H2 in the human oral cavity is extremely low (146). Given this, it is unlikely that C. concisus grow microaerobically in the human oral cavity. In previous studies, C. concisus was isolated from gingival plaque and saliva, places where a large number of anaerobes were found, suggesting that C. concisus is more likely to grow anaerobically in the human oral cavity (8, 42, 43).

Up to date, there are no studies systemically examining the growth of C. concisus under anaerobic conditions. Furthermore, there is no information available regarding the impact of H2 on C. concisus growth under anaerobic conditions. In this study, these issues were investigated. Moreover, the expression of putative virulence proteins of an oral C. concisus strain grown under anaerobic conditions was examined.

3.2. Materials and methods

3.2.1. C. concisus strains used in this study

A total of 60 C. concisus strains were used in this study, including 20 strains from patients with CD (19 oral strains and one enteric strain), 16 strains from patients with UC (14 oral strains and two enteric strains) and 24 oral strains from healthy individuals. These C. concisus strains were isolated in our previous studies (43, 78, 81).

3.2.2. Examination of C. concisus growth under various atmospheric conditions

The growth of the 60 C. concisus strains under the following atmospheric conditions was examined.

Original isolation condition: the C. concisus strains used in this study were isolated in our previous studies (43, 78, 81). The atmospheric condition used in isolation of these C. concisus strains from clinical samples in our previous studies was to place the BR56A gas- generation-system into a 2.5 L jar with a catalyst, which has modified the manufacturer’s instruction. In this study, this atmospheric condition was named as the original isolation condition (Oricon).

H2- H2- Microaerobic condition without H2 (Micro ): A Micro condition was generated using two different gas-generation-systems, following the manufacturer’s instruction (Oxoid). The first gas-generation-system was BR56A (MicroH2-a), which was placed into a 3.5 L jar in the presence a catalyst. The second gas-generation-system was CN25A (MicroH2-b), which was placed into a 2.5 L jar.

H2+ H2+ Anaerobic condition containing 9 % (v/v) of H2 (Anaero ): Anaero condition was generated using BR38B gas-generation-system, which was placed into a 3.5 L jar in the presence of a catalyst, following the manufacturer’s instruction (Oxoid).

H2- H2- Anaerobic condition without H2 (Anaero ): Anaero condition was generated using AN25A gas-generation-system as instructed by the manufacturer (Oxoid, Hampshire, UK).

Each of the 60 C. concisus strains was streaked on three horse blood agar plates supplemented with vancomycin (HBAV). Vancomycin was added to prevent the Gram positive bacterial contamination. Following incubation at 37°C under AnaeroH2-, AnaeroH2+, MicroH2-a, MicroH2-b and Oricon conditions respectively for 48 hours, plates were examined for the appearance of colonies under a stereo microscope. The morphology of C. concisus strains grown under each condition was examined using a phase-contrast microscope.

3.2.3. Quantitative comparison of C. concisus growth under AnaeroH2- and AnaeroH2+ conditions

To further quantitatively compare the growth of C. concisus under AnaeroH2- and AnaeroH2+ conditions, the colony forming unit (CFU) of 12 C. concisus strains grown under these two conditions were determined. These strains were chosen randomly from the 60 strains above.

C. concisus strains were first cultured on HBAV plates under Oricon condition at 37 °C for 48 hours. The bacteria were collected and washed once with PBS. The bacterial pellet of each strain was resuspended in PBS and the optical density at 600 nm (OD600) was adjusted to 0.05, which was used as the initial inoculum for further assessment of the growth of C. concisus under AnaeroH2- and AnaeroH2+ conditions.

The initial inoculum suspension (50 µL) of each C. concisus strain was inoculated onto six HBAV plates using a sterile L-shaped glass rod. Three plates were incubated under AnaeroH2- condition and the remaining three plates were incubated under AnaeroH2+ condition for 48 hours at 37 °C.

The bacterial cells of each C. concisus strain were collected from the three incubated under AnaeroH2- condition and the three plates were incubated under AnaeroH2+ condition by pooling. For collection of bacterial cells from each plate, 1 mL of PBS was used. From each pooled C. concisus suspension, serial dilutions (1:10 to 1:108) were prepared. Each dilution (5 µL) was inoculated onto HBAV plates in four replicates. The plates were further incubated under Oricon condition for 48 hours at 37 °C to determine the CFU numbers.

3.2.4. Examination of hydrogen dependent growth of C. concisus under anaerobic and microaerobic conditions

In order to test the effect of hydrogen gas on C. concisus growth, the bacterial strain which grew at the highest CFU number under anaerobic gas condition was chosen, which means that the growth of strain P3UCO-S1 were least likely to be affected by oxygen as a limiting factor.

The strain was first cultured on a HBAV plate under Oricon condition for 48 hours at 37 °C.

Following this, bacteria were collected and suspended into PBS and OD600 was adjusted to 0.05. The bacterial suspension (30 µL) was inoculated onto HBAV plates. The plates were incubated under both anaerobic conditions and microaerobic conditions containing various concentrations of H2. Anaerobic and microaerobic conditions were generated using gas- generation-system AN25A and CN25A respectively (Oxoid). H2 gas was supplemented by including 0.021g, 0.042g or 0.083g of sodium borohydride into 10 mL of H2O in a container placed in a 2.5L incubation jar, which generated 2.5 %, 5 % and 10 % (v/v) of

H2 respectively.

After 48 hour incubation at 37 °C, C. concisus bacterial cells were collected from each plate using 1 mL of PBS. Serial dilutions (1:10 to 1:108) were prepared and each dilution (5 µL) was inoculated onto HBA plates in four replicates. The plates were further incubated under Oricon condition for 48 hours at 37 °C to determine the CFU numbers.

3.2.5. Proteins expressed by C. concisus cultured under AnaeroH2- and AnaeroH2+ conditions

Proteins expressed by C. concisus cultured under AnaeroH2- and AnaeroH2+ conditions were analysed using mass spectrometry. C. concisus strain P6CDO-S1 was chosen randomly as a representative strain for this experiment.

Briefly, C. concisus strain P6CDO-S1 was grown on HBA plates for 48 hours under AnaeroH2- or AnaeroH2+ conditions. C. concisus bacteria were collected and washed with PBS. The whole cell lysate was prepared by freeze/thaw cycle and sonication (referring to the Section 2.4.2), and the protein amount was determined by Pierce BCA assay Cat. No.

23225 (Thermo Fisher Scientific). In this study, 19 µg whole cell proteins were separated on 12 % SDS-PAGE as described previously and stained by Coomassie blue dye R-250 (Bio-Rad, CA, US) (referring to the Section 2.6) (89).

The lane in the gel was excised by a sterile razor and divided in to 10 small pieces. Each slice was transferred into a 1.5 mL centrifuge tube and spin down briefly. Each tube was filled with 150 µL of destaining solution (25 mM ammonium bicarbonate, (NH4)HCO3) and 25 % (v/v) acrylonitrile (ACN) in distilled water) and incubated for 1 hour at room temperature. The destaining solution was changed every 15 minutes, with the fresh one. After complete destaining and aspirating, the slice was incubated with 40 µL of reducing solution (10 mM dithiothreitol and 50 mM (NH4)HCO3 in distilled water) for 30 minutes at 37 ºC. After aspiration, alkylation solution (25 mM iodoacetamide and 50 mM

(NH4)HCO3 in distilled water) was added, following 30 minutes incubation at 37 ºC. Alkylation solution was removed. 100 % ACN solution was added until the gel become non-transparent. ACN solution was removed. 40 µL of trypsin digestion solution (2 ng/mL trypsin enzyme and 25 mM (NH4)HCO3 in distilled water) was added and incubated overnight at 37 ºC. The digestion activity was stopped by adding 0.1 % (v/v) formic acid. The digested peptide from the gel was released by adding 100 µL of 100 % ACN solution. The solution was transferred into a new eppendorf tube, followed by drying via SpeedVac for one hour. The dried peptide pellet was resuspended with 20 µL of solution containing 0.1 % (v/v) formic acid and 0.05 % (v/v) heptafluorobutyric acid. The extracted peptides were separated by liquid chromatography and analysed by MS/MS as previously described (89, 147).

LTQ FT Ultra mass spectrometer (Thermo Electron, Bremen, Germany) was used. Positive ions were generated by electrospray, and the LTQ FT Ultra was operated in data- dependent acquisition mode. A survey scan (m/z 350–1750) was acquired in the Fourier transform ion cyclotron resonance cell (resolution = 100 000 at m/z 400, with an accumulation target value of 1 000 000 ions in the linear ion trap). Up to six of the most abundant ions (> 3000 counts) with charge states of > +2 were sequentially isolated and fragmented within the linear ion trap, using collisionally induced dissociation with an activation of q = 0.25 and activation time of 30 ms at a target value of 30 000 ions. m/z ratios selected for MS/MS were dynamically excluded for 30 s. Peak lists were generated

using mascot daemon/extract_msn (Matrix Science, London, UK), using the default parameters, and submitted to the database search program mascot (version 2.2; Matrix Science). Search parameters were as follows: precursor tolerance was 4 ppm, and product ion tolerances were ±0.4 Da; Met(O) was specified as a variable modification, enzyme specificity was trypsin, one missed cleavage was possible, and the NCBInr (July, 2009) or C. concisus strain 13826 databases were searched. A false-positive rate of ± 2 % was applied to searches from the LTQ FT-MS data.

The spectral counts of the same proteins expressed by P6CDO-S1 under AnaeroH2- and AnaeroH2+ conditions were compared using Scaffold-3 software (Proteome software, OR, US) (148). The experiment was carried out in duplicates and repeated twice.

Mass spectrometry was conducted at the Bioanalytical Mass Spectrometry Facility, University of New South Wales, Australia.

3.2.6. Statistical analysis

Unpaired t test was used for comparison of CFU numbers. Fisher’s exact test was used for analysis of the growth rate of C. concisus strains isolated from patients with IBD and controls. Graph Pad Prism 5 software was used for statistical analysis (San Diego, CA). P value < 0.05 was considered statistically significant.

3.3. Results

3.3.1. The growth of C. concisus strains under anaerobic condition with and without H2

Under AnaeroH2- condition, 92 % of C. concisus strains grew. All C. concisus strains grew under AnaeroH2+ conditions. The positive growth rates of C. concisus strains isolated from patients with IBD and controls were not statistically different (Table 3-1). The colonies of C. concisus strains grown under AnaeroH2- conditions appeared much smaller than those grown under AnaeroH2+ conditions. The morphology of C. concisus grown under AnaeroH2- and AnaeroH2+ conditions was not different under phase contrast microscopy.

None of C. concisus strains grew under microaerobic condition without H2; no bacterial colonies were observed on plates cultured under both MicroH2-a and MicroH2-b conditions. All strains grew under Oricon condition.

Table 3-1. Positive growth rates of C. concisus strains under AnaeroH2- and AnaeroH2+ conditions Strains Anaero H2- Anaero H2+ Oral strains from CD (n=19) 84 % 100 % Oral strains from UC (n=14) 86 % 100 % Oral strains from control (n=24) 100 % 100 % Enteric strains from IBD (n=3) 100 % 100 % Total strains (n =60) 92 % 100 % H2- Anaero : Anaerobic conditions without H2. H2+ Anaero : Anaerobic conditions with H2. H2-a None of these strains grew under microaerobic conditions without H2 (Micro and MicroH2-b).

3.3.2. Quantitative comparison of C. concisus growth under AnaeroH2- and AnaeroH2+ conditions

To further compare the growth of C. concisus strains under AnaeroH2+ and AnaeroH2- conditions, CFU of 12 C. concisus strains grown under these two atmospheric conditions were determined. All strains had a greatly increased growth under AnaeroH2+ condition in comparison to AnaeroH2- condition. The CFU numbers of all 12 C. concisus strains grown under AnaeroH2+ condition were significantly higher than those of the respective strains grown under AnaeroH2- condition (P < 0.05) (Table 3-2).

Table 3-2. CFU of C. concisus strains cultured under AnaeroH2+ and AnaeroH2- conditions

Strains AnaeroH2+ CFU AnaeroH2- CFU P1CDO-S1 517 ± 23.6 0.863 ± 0.061 P1CDO-S2 328 ± 25.5 0.813 ± 0.0318 P2CDO4 93.0 ± 8.49 1.36 ± 0.212 P4CDO-S1 123 ± 4.71 0.195 ± 0.0252 P6CDO-S1 126 ± 9.55 1.21 ± 0.125 P8CDO-S1 112 ± 14.4 3.15 ± 0.388 P3UCO-S1 550 ± 70.7 1.43 ± 0.0106 P7UCO-S1 311 ± 105 9.42 ± 0.118 P13UCO-S3 111 ± 2.12 6.50 ± 0.707 P3UCB-S1 146 ± 36.8 1.57 ± 0.330 H3O-S1 73.8 ± 15.9 9.50 ± 1.91 H5O-S1 135 ± 5.30 9.50 ± 0.707 The values (means ± standard deviation) were from triplicates. CFU: (colony forming unit) × 108 / mL. The CFU numbers of all 12 strains grown under AnaeroH2+ conditions were significantly higher than those of the respective strains grown under AnaeroH2- conditions (P < 0.05).

3.3.3. The growth of C. concisus under anaerobic and microaerobic conditions containing different concentrations of H2

C. concisus strain P3UCOS-1 was used as a representative strain to evaluate the growth of C. concisus under anaerobic and microaerobic conditions containing different concentrations of H2.

Under anaerobic condition, the CFU numbers of the strain cultured in the presence of 2.5 9 9 %, 5 % and 10 % H2 were (1.10 ± 0.42) ×10 /ml, (9.15 ± 0.82) ×10 /mL, and (1.90 ± 1.33) 9 ×10 /mL respectively. The CFU numbers of 5 % H2 were significantly higher than the

CFU number of 2.5 % and 10 % H2 (both P values < 0.0001). The CFU numbers of 10 %

H2 and 2.5 % H2 were not significantly different (P value = 0.3) (Figure 3-1).

Under microaerobic conditions, the CFU numbers of P3UCOS-1 strain cultured in the 6 7 presence of 2.5 %, 5 % and 10 % H2 were (1.0 ± 1.15) ×10 /mL, (1.60 ± 0.16) ×10 /mL, 8 and (2.67 ± 0.5) × 10 /ml, respectively. The CFU number of 2.5 % H2 was significantly lower than the CFU numbers of 5 % and 10 % H2 (P value < 0.0001 and P value < 0.005, respectively). The CFU number of 5 % H2 was significantly lower than the CFU number of 10 % H2 (P value < 0.005) (Figure 3-1).

In the presence of 2.5 % and 5 % H2, the CFU numbers of the strain P3UCOS-1 cultured under anaerobic conditions were significantly higher than the CFU numbers of this strain cultured under microaerobic conditions (P < 0.005 and P < 0.0001 respectively). In the presence of 10 % H2, the CFU numbers of P3UCOS-1 strain cultured under anaerobic and microaerobic conditions were not significantly different (P value = 0.09) (Figure 3-1).

Figure 3-1. Growth of C. concisus strain P3UCO-S1 at various H2 concentrations CFU of C. concisus were measured after two day incubation under microaerobic (black bar) or anaerobic (grey bar) condition with different H2 gas concentration. The experiment was repeated at least twice. The error bar indicates the standard deviation. In the absence of H2 gas under microaerobic condition, no C. concisus growth was observed. The number of asterisk indicates the significance of CFU numbers (**, P < 0.01 and ***, P < 0.001). NS indicates no significant difference.

3.3.4. Proteins expressed by C. concisus strain P6CDO-S1 cultured under AnaeroH2- and AnaeroH2+ conditions

Proteins expressed by the strain P6CDO-S1 under AnaeroH2+ and AnaeroH2- conditions were subjected to mass spectrometry analysis. Previously reported virulence proteins such as outer membrane fibronectin-binding protein, major outer membrane protein, protease Htpx, S-layer-RTX protein, putative hemagglutinin/hemolysin-related protein, CjaC and EvpB family type VI secretion protein were identified. However, the spectral counts of these proteins were not statistically different in C. concisus strain P6CDO-S1 grown under AnaeroH2+ and AnaeroH2- conditions (Table 3-3).

Besides virulence proteins, the level of four proteins in C. concisus strain P6CDO-S1 was determined to be significantly different between AnaeroH2+ and AnaeroH2- conditions. The spectral counts of chemotaxis protein CheA in C. concisus grown under AnaeroH2+ condition was significantly increased (P < 0.05) comparing with that under AnaeroH2- condition (Table 3-4). Moreover, proteins involved in metabolism such as flavocytochrome c flavin subunit, succinate dehydrogenase flavoprotein subunit and formate dehydrogenase increased significantly under AnaeroH2+ condition with P values less than 0.05, 0.01 and 0.05 respectively, as comparing with that under AnaeroH2- condition.

Table 3-3. Virulence proteins expressed by C. concisus strain P6CDO-S1 cultured under AnaeroH2+ and AnaeroH2- conditions*

Protein name Locus tag SC SC (AnaeroH2+) (AnaeroH2-) Flagellin B CCC13826_2297 16.9 ± 3.59 15.1 ± 3.96 Fibronectin-binding protein CCC13826_0739 12.8 ± 3.43 12.8 ± 1.98 Protease htpx CCC13826_1039 8.44 ± 0.990 10.5 ± 1.93 Omp18 CCC13826_0923 7.49 ± 4.48 7.80 ± 3.51 S-layer-RTX protein CCC13826_1838 8.15 ± 3.74 4.00 ± 1.62 CjaC CCC13826_0963 4.93 ± 2.90 6.33 ± 0.768 EvpB family type VI secretion CCC13826_1182 4.96 ± 0.555 5.33 ± 0.722 protein Hemagglutinin/hemolysin-related CCC13826_0009 4.39 ± 1.77 2.33 ± 0.667 protein * Proteins were identified using mass spectrometry analysis SC: The value of the mean spectral counts from four replicates with standard deviation. The SC values of virulence proteins in C. concisus strain P6CDO-S1 cultured under H2+ H2- anaerobic condition with H2 (Anaero ) and anaerobic condition without H2 (Anaero ) conditions were not significantly different (P > 0.05).

Table 3-4. A list of four proteins in which a significant change in protein expression was found via mass spectrometry analysis

Protein name Locus tag SC# SC# Fold T test@ (AnaeroH2+) (AnaeroH2-) increase^

Chemotaxis protein CCC13826_1580 8.37 ± 3.58 3.60 ± 0.975 2.33 * CheA

Flavocytochrome c CCC13826_0270 17.3 ± 5.79 7.40 ± 0.867 2.34 * flavin subunit

Succinate dehydrogenase CCC13826_1283 10.8 ± 2.07 4.13 ± 1.10 2.62 ** flavoprotein subunit

Formate CCC13826_0761 7.63 ± 3.26 2.27 ± 0.162 3.36 * dehydrogenase

# SC: spectral counts ^ Fold increase is the relative ratio of SC values under anaerobic condition with hydrogen (AnaeroH2+) to those under anaerobic condition without hydrogen (AnaeroH2-) @ Statistically significance was analysed by student's t test. One asterisk symbol (*) indicates P < 0.05, and two asterisk symbols indicate P < 0.01.

3.4. Discussion

In this study, the growth of C. concisus strains under different atmospheric conditions was examined. It was previously described that C. concisus is a bacterium which requires H2- enriched microaerobic conditions for growth and some C. concisus strains may grow under anaerobic conditions if fumarate and formate are present in the culture plates (8, 143, 145). In this study, it was found that under anaerobic conditions the majority of oral C. concisus strains (91%, 52/57) grew on HBA plates containing no formate or fumarate without the presence of H2, suggesting that oral C. concisus is an anaerobic bacterium and that H2 gas, formate and fumarate are not essential requirements for the anaerobic growth of oral C. concisus strains. None of the 57 oral C. concisus strains grew under microaerobic conditions without H2, suggesting that microaerobic growth of C. concisus requires the presence of H2, which is consistent with previous findings [9, 11]. The solubility of H2 gas in H2O is extremely low, thus liquid culture methods are not suitable for assessing the impact of H2 gas on C. concisus growth (149). Given this, in this study, the CFU numbers of C. concisus strains were determined using a plate culture method.

Under anaerobic conditions, the presence of H2 greatly increased the growth of C. concisus, demonstrated by the increased colony sizes observed macroscopically and the increased CFU numbers of the same strain cultured under AnaerH2+ and AnaerH2- conditions (Table 3-2). These results suggest that under anaerobic conditions C. concisus have different respiration pathways in generating energy for growth and oxidisation of H2 is a pathway generating high energy for a rapid growth.

Mass spectrometry analysis showed that three metabolic proteins, including flavocytochrome c flavin subunit, succinate dehydrogenase flavoprotein subunit and formate dehydrogenase, significantly increased under AnaeroH2+ condition in comparison with that under AnaeroH2- condition (Table 3-4). These proteins were known to generate protons from hydrogen sulfide, succinate and formate, and in turn provide energy (150- 152). This result can support the previous result showing the increased growth rate of C. concisus under AnaeroH2+ condition. Interestingly, those all three proteins are membrane- bounded enzymes. Especially, succinate dehydrogenase was the only membrane bound

enzyme involved in tricarboxylic acid cycle, and might play a role in C. concisus respiration (153).

In humans, H2 is produced by anaerobic bacteria predominantly in the colon (154, 155). H2 generated in the intestine is disposed by H2 consuming bacteria such as methanogenic bacteria, sulfate-reducing bacteria and acetogenic bacteria (156). Some H2 is diffused into blood and this H2 can be measured by breath testing (157). Dietary factors and the composition of an individual’s intestinal microbiota affect intestinal H2 production and consumption (158-160). The natural host of C. concisus is humans and the primary colonisation site is the human oral cavity (8, 43). The concentration of excreted H2 in the oral cavity is extremely low. The basal level of hydrogen in healthy individuals is usually less than 10 ppm, thus a H2 level of less than 0.001 % (1 ppm = 0.0001 %) (146). In addition to anaerobic bacteria in the intestine, oral anaerobic bacteria may also produce H2 by fermentation of carbohydrate residues from food. However, the level of H2 produced by oral anaerobes is very low. Mastropaolo and Rees showed that following a solid meal, the

H2 produced by oral anaerobes was 25 ppm (0.0025 %) and this level was retained for only 73 minutes (161). Given this, C. concisus colonising the oral cavity is unlikely to have constantly available H2 for growth. The finding in this study that oral C. concisus strains were able to grow without the presence of H2 under anaerobic conditions helps to explain why C. concisus is able to colonise the human oral cavity.

Despite the fact that H2 dramatically increases the growth of C. concisus and the intestine is the dominant place for H2 production in humans, it is interesting to note that C. concisus has selected the oral cavity, rather than the intestinal environment as its natural colonisation site. This suggests that in healthy individuals there are some factors in the GIT that inhibit C. concisus intestinal colonisation. It is likely that such inhibitory factors are low or lacking in patients with IBD, which contributes to the higher intestinal prevalence of C. concisus in these patients. One of such factors may be methanogenic bacteria; the dominant H2 consuming bacteria in the human intestine that produce methane. It is possible that methanogenic bacteria in the intestine compete with C. concisus for use of H2.

A study by Mckay et al. examining hydrogen and methane excretion in patients with IBD and controls showed that the prevalence of methane excretion was 13 % in patients with CD and 15 % in patients with UC, which was significantly lower than that in healthy controls (54 %) (162). This observation was supported by a study from Pimentel et al., which showed that 97 % of patients with IBD (75/78), who had a predominant diarrheal condition, excreted H2 only (no methane) (163). These results suggest that there is a low level of methanogenic bacteria in patients with IBD. Indeed, a study conducted by Scanlan et al. detected a low prevalence of intestinal methanogenic bacteria in patients with IBD in comparison to healthy controls and other disease groups (164). Methanogenic bacteria play a predominant role in disposing intestinal H2 in humans (160). The lack of sufficient intestinal methanogenic bacteria in patients with IBD may have generated an intestinal environment that allows C. concisus to use H2 for a rapid growth.

Considering that luminal environment is more likely to be anaerobic, in this study, enteric bacteria including zoonotic pathogen C. jejuni and commensal bacteria E. coli K12 displayed H2-independent growth under anaerobic condition unlikely to C. concisus growth. Only C. ureolyticus showed H2-dependent growth under anaerobic condition. C. ureolyticus is reported to be a potential opportunistic pathogen in human gut. In 2011, Bullman et al. examined that the abundance of C. ureolyticus in patients presenting with diarrhoea based on age and showed the predominance at extreme age (< 5 and > 70 year old) (165). The same author stated that C. ureolyticus is unlikely to be a commensal bacteria in human intestine as the detection rate is only 1.15 % of 7194 samples (166). In 2012, Burgo-Portugal examined the pathogenic potential of C. ureolyticus UNSWCD to Caco2 cells and showed the significant increase of cytokines production (TNF-α and IFN-

γ) (167). This finding that both C. concisus and C. ureolyticus exhibited the similar H2- dependent anaerobic growth pattern may reveal the important characteristic of opportunistic pathogen leading to chronic inflammation in human gut.

It was previously showed that some oral C. concisus strains were able to colonise the intestinal tract and have the potential to cause enteric disease [16, 33]. In this study, it was found that strain P6CDO-S1, an oral C. concisus strain isolated from a patient with CD, expressed a number of putative virulence proteins. These proteins were previously reported to contribute to the virulence of other bacterial species (88, 92, 94, 168-172).

However, their roles in C. concisus virulence remain to be characterised. If indeed these putative proteins play a role in C. concisus virulence, the finding in this study that the expression levels of these proteins remain similar when strain P6CDO-S1 cultured under anaerobic conditions with and without H2 suggests that the impact of H2 on C. concisus virulence is unlikely through affecting these proteins. It is likely that H2 may impact on C. concisus virulence through increasing the growth of C. concisus to a disease-causing threshold.

This study also found that under anaerobic conditions, strain P3UCO-S1, an oral strain isolated from a patient with UC, had a significantly higher CFU in the presence of 5 % H2, as compared with 2.5 % H2 and 10 % H2. Under microaerobic conditions, this strain had a significantly higher CFU in the presence of 10 % H2 compared with 2.5 % and 5 % H2. It appeared that the concentrations of H2 supplied in bacterial cultivation affect the optimal growth of C. concisus. This aspect should be further investigated by examining more C. concisus strains using systems that are able to supply fixed concentrations of CO2, N2 and

H2, which will provide useful information to clinical laboratories in isolation of C. concisus from clinical samples.

In addition to the 57 oral C. concisus strains, six enteric strains were included, with five strains being isolated from patients with IBD, into this study. These enteric strains showed an anaerobic growth pattern that was similar to oral C. concisus strains.

In summary, this study found that oral C. concisus strains were able to grow under anaerobic conditions without H2, formate or fumarate and that these strains did not grow under microaerobic conditions without H2, suggesting that they are anaerobes. The presence of H2 in the anaerobic conditions greatly increased the growth of oral C. concisus strains. Using mass spectrometry analysis, an oral C. concisus strain isolated from a patient with CD was found to express a number of putative virulence proteins and the expression levels of these proteins under anaerobic conditions with and without H2 remained similar. While the numbers of enteric C. concisus strains included in this study were small, these enteric strains and oral C. concisus strains had a similar anaerobic growth pattern. This study provides useful information in understanding the natural colonisation site and pathogenicity of C. concisus.

Chapter 4: The effects of formate and fumarate on Campylobacter concisus hydrogen sulfide production and growth

4.1. Introduction

A major function of gut microbiota in the colon is to ferment the dietary components that are not absorbed in the small intestine into short-chain fatty acids such as acetate, propionate and butyrate and gases such as hydrogen gas (H2) and carbon dioxide (173). These fermented products can have an impact on the host as well as the microbiota (173).

For instance of H2, these microbes compete for traces of H2 to maximise the ATP yield, but high levels of H2 can also prevent re-oxidation of pyridine nucleotides (174). Therefore, the H2-oxidising species are critical in maintaining the homeostasis of hydrogen level in the intestine (175). Most of the H2-oxidising bacteria grow best under microaerobic conditions, and use oxygen (O2) as the terminal electron acceptor. Some H2-oxidising bacteria utilise sulfate as terminal electron acceptor under anaerobic conditions (176).

Sulfate-reducing bacteria (SRB) are one of H2-oxidising bacteria that obtain energy by reducing sulfate to hydrogen sulfide (H2S) (177). Most of the SRB are anaerobes, with some being O2 tolerant. In anaerobic respiration, these bacteria use sulfate, instead of oxygen, as the terminal electron acceptor. In addition to sulfate, some SRB may reduce other oxidised inorganic sulfuric compounds such as thiosulfate (177). Various studies showed that SRB are associated with ulcerative colitis (UC), one form of inflammatory bowel disease (IBD). H2S excreted in faecal samples from patients with UC was significantly higher than that collected from controls (111, 112, 178).

Campylobacter concisus is a Gram-negative bacterium living in the human oral cavity (10), and currently is recognised as a possible pathogenic cause of IBD as discussed in the Chapter 1. However, the exact pathogenicity of C. concisus in IBD is still unclear. It was previously reported that less than 11 % of Campylobacter concisus strains produced trace

amount of H2S when tested using the triple sugar iron (TSI) agar (179). TSI is a differential medium that is mainly used to identify members of the Enterobacteriaceae. TSI differentiates bacteria based on their fermentation of lactose, glucose and sucrose and the production of H2S. Currently, no studies have examined the growth of C. concisus on

TSI agar. It is possible that the low production of H2S by C. concisus on TSI agar was due to poor growth of C. concisus on TSI agar.

Therefore, in this chapter, it was examined whether the growth and H2S production by C. concisus can be improved by the modification and supplementations on TSI agar plate. In fact that C. concisus grows well on HBA plates, the production of H2S by oral C. concisus strains was examined using modified HBA plates instead of TSI plates. For the most preferential supplementation to the agar plate, formate and fumarate were considered.

Formic acid is a carboxylic acid with the chemical formula HCOOH. Salts derived from formic acid are referred to as formate. Formate is a key metabolite in the energy metabolism of many bacteria. Formate acts as an electron donor and is converted into CO2 and H2 (180, 181). Fumarate is one of the intermediates of the Krebs cycle. Under aerobic conditions, succinate dehydrogenase catalyses the oxidation of succinate to fumarate and consequently generates electrons that can be used by the electron transport chain for energy production. Under anaerobic conditions, fumarate reductase mediates the reduction of fumarate to succinate and acts as an electron acceptor (182). Previously, it was observed that when formate and fumarate were both supplemented together in the culture media under anaerobic conditions, the growth of C. concisus increased (8).

In the previous chapter, it was found that the growth pattern and protein expression in C. concisus were significantly mediated by atmospheric conditions. In this chapter, the investigation whether intestine-like atmospheric conditions have effects on the production of H2S in oral C. concisus strains was also included.

Earlier study by Mahendran et al. from our research group reported that C. concisus strains consisted of five subpopulations through a study of the housekeeping gene sequences

(183). Based on this finding, the differential H2S positivity in oral C. concisus strains of different subpopulation groups was compared.

4.2. Materials and methods

4.2.1. Examination of H2S production in C. concisus strains using modified HBA plates and the effects of supplementation of both formate and fumarate in culture media as well as H2 gas under atmospheric conditions on the production of H2S by C. concisus strains

A total of 57 oral C. concisus strains were used in this study. These included 19 oral strains collected from patients with CD, 14 oral strains collected from patients with UC and 24 oral strains collected from healthy controls. These C. concisus strains were isolated in our previous studies (43, 89)

HBA plates were supplemented with 0.03 % (w/v) sodium thiosulfate (Sigma-Aldrich) and 0.02 % (w/v) ferric (II) sulfate (Sigma-Aldrich). This modified HBA plate (HBAFe/S) was used for detection of H2S production by C. concisus strains. The method was adopted from TSI agar method (184). The pH of HBAFe/S plates was 7 by a pH test strip.

All C. concisus strains were incubated on HBA plate for 48 hours at 37 °C under atmospheric conditions generated by placing BR56A gas pack into a 2.5 L jar, which was the original atmospheric condition used for isolation of C. concisus from clinical samples (OriCON) (2). C. concisus strains were collected from HBA plates and washed with 1 mL of PBS twice. The bacterial suspension with OD reading of 0.05 at 600 nm was prepared for inoculation. Each bacterial strain was inoculated onto two sets of two different types of plates, including HBAFe/S plates and HBAFe/S plates supplemented with 0.2 % (w/v) sodium formate (Sigma-Aldrich) and 0.4 % (w/v) sodium fumarate dibasic (Sigma- Aldrich) by streaking. One set of the plates was incubated under anaerobic conditions with H2+ H2 (Anaero ), and another set of plates was incubated under anaerobic conditions H2- without H2 (Anaero ). The atmospheric conditions were generated as previously described (2). Following incubation of the plates for 48 hours at 37°C, bacterial colonies were observed using stereo microscope. Production of H2S was determined by the appearance of black pigmented colonies. The experiment was carried out in triplicates.

4.2.2. Examination of individual effect of formate and fumarate on the production of H2S in C. concisus strains

The individual effect of formate or fumarate on the production of H2S in C. concisus was further investigated. Six strains were examined, of which two strains were randomly chosen from each disease group among the 57 strains, including two C. concisus strains isolated from patients with CD (P10CDO-S1 and P10CDO-S2), two strains isolated from patients with UC (P15UCO-S1 and P15UCO-S2) and two strains isolated from healthy individuals (H11O-S1 and H17O-S1). Bacterial suspensions were prepared as described above. Each C. concisus strain was inoculated onto two sets of four different plates, including HBAFe/S plates, HBAFe/S supplemented with 0.2 % (w/v) sodium formate, HBAFe/S supplemented with 0.4 % (w/v) sodium fumarate dibasic, and HBAFe/S supplemented with both 0.2 % (w/v) sodium formate and 0.4 % (w/v) sodium fumarate dibasic. The plates were incubated under AnaeroH2+ and AnaeroH2- conditions respectively for 48 hours at 37 °C, and the production of H2S recorded as described above

4.2.3. Quantification of C. concisus growth with and without supplementation of formate and fumarate

The effects of formate and fumarate on the growth of C. concisus were examined. Briefly, C. concisus UNSWCD strain was cultured on HBA plates for 48 hours under OriCON conditions. C. concisus bacteria were collected from HBA plates and washed with 1mL of PBS twice. The bacterial suspension (50 µL) with 0.05 of OD reading at 600 nm was inoculated onto two sets of four different types of plates including HBAFe/S plates, HBAFe/S supplemented with 0.2 % (w/v) sodium formate, HBAFe/S supplemented with 0.4 % (w/v) sodium fumarate dibasic, and HBAFe/S supplemented with both 0.2 % (w/v) sodium formate and 0.4 % (w/v) sodium fumarate dibasic by spreading using L-shaped glass rod. One set of plates was incubated under AnaeroH2+ and the other set was incubated under AnaeroH2- condition at 37 °C. After continuous incubation for 48 hours, the growth of C. concisus was measured by counting CFU on each plate and then converting to CFU/mL bacterial suspension. This experiment was carried out in triplicate.

4.2.4. Statistical analysis

Fisher’s exact test was performed to determine the effects of hydrogen, or supplementation of formate and fumarate on H2S positivity in different C. concisus strains. Unpaired t test was used to compare CFU counts of C. concisus cultured in media with different supplement. P value less than 0.05 was considered as statistically significant. All statistical analysis was performed by Excel 2010 (Microsoft, WA, US)

4.3. Results

Fe/S 4.3.1. H2S positivity in oral C. concisus strains cultured on HBA plates under AnaeroH2+ and AnaeroH2- conditions

H2S production in C. concisus strains was determined by the presence of black pigmented colonies on HBAFe/S plates (Table 4-1). Of the 57 oral C. concisus strains examined, 23 H2+ strains (40 %) produced H2S when cultured under Anaero conditions. None of these H2- strains produced H2S when cultured under Anaero conditions (Table 4-1).

4.3.2. H2S positivity in C. concisus strains isolated from patients with IBD and controls

H2- As mentioned above, none of these C. concisus strains produced H2S under Anaero conditions. For the HBAFe/S plates incubated under AnaeroH2+ conditions, nine of the 19 strains collected from patients with CD were positive for H2S production (47 %), which was not statistically different from that in C. concisus strains collected from healthy controls (12/24, 50 %). Of the 14 strains isolated from patients with UC, only two strains produced H2S (14 %). The H2S positivity in strains collected from patients with UC was significantly lower than that in strains collected from patients with CD and healthy controls, with P values being 0.043 and 0.025, respectively.

In culture media supplemented with formate and fumarate supplementation, the H2S productivity in C. concisus strains isolated from patients with CD, UC and healthy controls were 79 % (15/19), 57 % (8/14) and 79 % (19/24), respectively, which were not significantly different from each other (CD vs UC, P = 0.4769; UC vs healthy control, P = 0.2657; healthy controls vs CD, P = 0.7302).

4.3.3. The effects of supplementation of both fumarate and formate into culture media on the production of H2S in C. concisus strains

Whether supplementation of formate and fumarate in the culture media affects the production of H2S was examined. All C. concisus strains were cultured under both AnaeroH2+ and AnaeroH2- conditions (Table 4-1). Under AnaeroH2+ conditions, 41 of 57

Fe/S strains (72 %) produced H2S. The H2S positivity in C. concisus strains cultured on HBA plates supplemented with both formate and fumarate under AnaeroH2+ conditions were significantly higher than that in the same strains grown on HBAFe/S plates with no supplement (P < 0.0005). Interestingly, under AnaeroH2- conditions, 54 % of C. concisus strains (31/57) were able to produce H2S with supplementation of formate and fumarate in the culture media, whereas none of these strains produced H2S under the same atmospheric conditions without the supplements of formate and fumarate (Table 4-1).

H2+ H2- Table 4-1. H2S positivity of oral C. concisus strains under Anaero and Anaero conditions with or without supplementation#

HBAFe/S supplemented with Group HBAFe/S formate and fumarate

AnaeroH2+ AnaeroH2- AnaeroH2+ AnaeroH2-

CD 47 % (9/19) 0 % (0/19) 79 % (15/19) 53 % (10/19)

UC 14 % (2/14)^ 0 % (0/14) 57 % (8/14) 43 % (6/14)

Controls 50 % (12/24) 0 % (0/24) 79 % (19/24) 63 % (15/24)

Total 40 % (23/57) 0 % (0/57) 72 % (41/57)* 54 % (31/57)*

# Fifty-seven oral C. concisus strains were cultured under anaerobic conditions with hydrogen (AnaeroH2+) and anaerobic condition without hydrogen (AnaeroH2-) for 48 hours at 37°C on HBAFe/S plates and HBAFe/S supplemented with both formate and fumarate. ^ C. concisus strains collected from patients with UC shows a significantly lower rate of H2S positivity as compared with those strains collected from patients with CD and healthy controls. P values were 0.043 and 0.025, respectively.

* The supplementation of both formate and fumarate significantly increases H2S positivity in C. concisus strains cultured under AnaeroH2+ and AnaeroH2- conditions, as compared with that in the same strains cultured without supplementation.

4.3.4. H2S positivity in C. concisus strains of different subpopulations

In the previous study performed by Mahendran et al. (183), oral C. concisus strains were divided into five different subpopulation groups. In this study, out of 57 strains used in this study, 45 C. concisus strains were previously subjected to the population structure analysis.

These 45 strains were re-categorised into the five subpopulations, and their H2S productivity was compared.

When grown on HBAFe/S plates under AnaeroH2+ conditions, 18 of the 45 strains (40 %) were positive for H2S production (Table 4-2), which was not significantly different from that of the total 57 C. concisus strains (Table 4-1). None of the strains incubated under H2- Anaero conditions were positive for H2S positivity.

When these 45 C. concisus strains grew on HBAFe/S supplemented with formate and fumarate under AnaeroH2+ and AnaeroH2- conditions, 33 (73 %) and 29 (64 %) strains were positive for H2S production respectively (Table 4-2), which was not significantly different from those prior to the recategorisation of subpopulations using 57 strains (Table 4-1).

Among five subpopulation groups, the H2S positivity of strains from the subpopulation II appeared low. Under AnaeroH2+ conditions on HBAFe/S plates, none of the five strains in this subpopulation group were positive for H2S, which was significantly lower than that from subpopulations I with P value of 0.048. The H2S positivity among other subpopulation groups were not statistically different.

Under AnaeroH2+ conditions on HBAFe/S plates supplemented with formate and fumarate, the H2S positivity in strains of subpopulation group II was 20 % (1/5), which was significantly lower than that in strains from subpopulation group III, IV and V with P values of 0.033, 0.040 and 0.034, respectively. The H2S positivity among other subpopulation groups were not statistically different.

Under AnaeroH2- conditions on HBAFe/S plates supplemented with formate and fumarate, the H2S positivity in strains from different subpopulation groups was not statistically different (Table 4-2).

Among five subpopulation groups, the H2S positivity of strains from the subpopulation group III and V significantly increased with formate and fumarate under AnaeroH2+ condition, as comparing to that without the formate and fumarate under AnaeroH2+ conditions. In the other subpopulations, including group I, II and IV, there were no significant effects of supplementation of formate and fumarate on the H2S positivity, possibly due to the small number of strains.

^ Table 4-2. H2S positivity in C. concisus strains from five subpopulation groups

Subpopulation HBAFe/S supplemented with HBAFe/S group Formate and Fumarate AnaeroH2+ AnaeroH2- AnaeroH2+ AnaeroH2-

I 75 % (3/4) 0 % (0/4) 75 % (3/4) 75 % (3/4)

II 0 % (0/5)# 0 % (0/3) 20 % (1/5)* 40 % (2/5)

III 43 % (9/21) 0 % (0/21) 76 % (16/21) 57 % (12/21)

IV 50 % (2/4) 0 % (0/4) 100 % (4/4) 100 % (4/4)

V 36 % (4/11) 0 % (0/11) 82 % (9/11) 64 % (7/11)

Total 40 % (18/45) 0 % (0/41) 73 % (33/45) 64 % (29/45) ^ Five subpopulation groups were obtained from Mahendran et al. (183) * The significantly lower number of oral C. concisus in the subpopulation II produces H2S when formate and fumarate are available as compared with that in the subpopulation III, IV, and V. P values are 0.033, 0.040, and 0.034, respectively. # None of the oral C. concisus strains in subpopulation group II grown under AnaeroH2+ conditions without supplementation produces H2S, which is significantly lower than that in subpopulation group I under the same condition with P value of 0.048.

4.3.5. Individual effect of formate or fumarate on H2S production in oral C. concisus

Whether individual supplementation of formate or fumarate has any effect on the H2S positivity was investigated using six C. concisus strains.

Under AnaeroH2+ conditions, four of six strains (67 %) cultured on HBAFe/S plates were positive for H2S production.

When these six strains were grown on HBAFe/S plates supplemented with formate, five strains (83 %) were positive for H2S production. The colour of the colonies appeared much darker than those grown on HBAFe/S plates. The colonies of the remaining strain did not show any black colouration, inactive of no H2S production.

Of these strains cultured on HBAFe/S plates supplemented with fumarate, five strains (83 %) Fe/S produced H2S. The colour of the colonies was similar to that grown on HBA plates.

All of strains (6/6) produced H2S, when both formate and fumarate were provided to the HBA media.

The colony sizes of these six strains, as judged by visual inspection, appeared different when grown on HBAFe/S plates with and without supplementation of formate or fumarate. The colony sizes on HBAFe/S plates supplemented with fumarate were the largest. The colony sizes on HBAFe/S plates and HBAFe/S plates supplemented with both formate and fumarate were similar. The colony sizes on HBAFe/S plates supplemented with formate were the smallest.

Table 4-3. H2S positivity of six C. concisus strains with supplementation of formate or fumarate@

Plates AnaeroH2+ AnaeroH2-

HBAFe/S 67 % (4/6) 0 % (0/6)

HBAFe/S supplemented with formate 83 % (5/6) 0 % (0/6)

HBAFe/S supplemented with fumarate 83 % (5/6) 0 % (0/6)

HBAFe/S supplemented with formate and fumarate 100 % (6/6) 50 % (3/6)

@ Six oral C. concisus strains were culture under AnaeroH2+ and AnaeroH2- condition for 48 hours at 37°C on four different plates, which included HBAFe/S, HBAFe/S supplemented with formate, HBAFe/S supplemented with fumarate and HBAFe/S supplemented with formate. The black-pigmented colonies were observed using optical microscopy.

4.3.6. Quantitative examination of formate and fumarate supplementation on C. concisus growth

As mentioned above, the supplementation of formate and fumarate had an impact on the size of colonies. This may reflect an altered growth, which was further examined. The C. concisus UNSWCD strain was cultured on four different HBA plates at 37 °C under AnaeroH2+ and AnaeroH2- conditions. The CFU value was calculated.

Under AnaeroH2+ conditions, the CFU/mL from the HBA plates was (5.02 ± 0.59) ×109. From the HBA plate supplemented with formate, the CFU/mL was (3.50 ± 0.50) ×108, which was significantly lower than that on the HBA plates (P value = 0.0006). Fumarate, or both formate and fumarate supplementation significantly increased the CFU/mL (1.45 ± 0.10) ×1010 and (6.17 ± 2.47) ×109, respectively, as compared with the CFU count obtained from the HBA plate (P value < 0.001 and P value = 0.036 respectively).

When C. concisus strain was cultured under AnaeroH2- conditions on HBA plate, the CFU/mL was (2.25 ± 0.48) ×109, which was significantly lower as compared with that under AnaeroH2+ conditions, which was consistent with the previous finding [3]. When formate was added into the media, the CFU/mL of C. concisus decreased to (2.60 ± 0.74) ×108, which was significantly lower as compared to that grown on HBA plate under AnaeroH2- conditions, with a P value of 0.0020. The supplementation of fumarate significantly increased the CFU/mL of C. concisus to (1.82 ± 0.51) ×1010 with P value of 0.0056. However, when both formate and fumarate were supplemented in HBA plates, the CFU/mL was not significantly different as compared with that from HBA plates with no supplementation (P value = 0.49).

Figure 4-1. Quantification of the growth of C. concisus strain UNSWCD with and without the supplementation of formate and fumarate C. concisus strain UNSWCD was grown on HBA plates for 48 hours at 37°C with different supplementations, including HBA plates (None), HBA plates with formate (Formate), HBA plates with fumarate (Fumarate) and HBA plates with both formate and fumarate (Both). Anaerobic conditions with hydrogen (AnaeroH2+) and anaerobic conditions without hydrogen (AnaeroH2-) were used. The CFU numbers from different culture plates were compared to those from the HBA plates without formate and fumarate. NS indicates no significance. One asterisk indicates P value smaller than 0.05. Two and three asterisks indicate P value smaller than 0.01, and 0.001 respectively.

4.4. Discussion

This chapter firstly, examined the production of H2S by C. concisus strains under AnaeroH2+ and AnaeroH2- conditions using HBAFe/S plates. It was found that the production of H2S in oral C. concisus strains only occurred when H2 gas was supplied (Table 4-1), suggesting that H2 as an electron donor is an essential requirement for H2S production by C. concisus. The human oral cavity is known to be a natural reservoir for C. concisus (6) and hydrogen availability in human oral cavity is very poor. This suggests that oral C. concisus strains may not be able to produce H2S when they colonise the human oral cavity.

On the other hand, the human intestine provides H2 gas, mainly produced by bacteria residing the intestinal tract (185). The H2 levels in the intestinal tract vary from individual to individual, which are determined by diet and community of gut microbiota (186). Given this, oral C. concisus may be able to produce H2S in some individual’s intestinal tract during bacteria transition to intestine or colonise the intestine where H2 is available.

In this study, it was found that the overall H2S positivity of C. concisus strains was 40 % Fe/S when tested on HBA plates, whilst the previously reported H2S positivity in C. concisus strains was less than 11 % (179). It appears that HBAFe/S plate is a better choice for testing

H2S production in C. concisus strains.

Another finding of this study was that the formate and fumarate supplementation H2+ remarkably increased the H2S positivity in oral C. concisus strains under both Anaero and AnaeroH2- conditions. Formate is an electron donor that is as efficient as hydrogen in terms of redox potential (Table 4-4), and can be oxidised to H2 and CO2. Given that H2 is an essential requirement for H2S production by C. concisus, formate would have stimulated the production of H2S through the production of H2, which is served as an electron donor.

Very recently, Mahendran et al. from our research group reported that there were five subpopulation groups among oral C. concisus strains (183). Of the 57 strains examined in this chapter, 45 strains were also used in the analysis by Mahendran et al. This has provided an opportunity to examine whether C. concisus strains from different subpopulation groups differ in their production ability of H2S. Interestingly, it was found

that the H2S productivity in the strains from subpopulation group II was significantly lower than that in the strains from the subpopulation group I on the HBAFe/S plates, and also other subpopulation groups III, IV and V on the HBAFe/S plates supplemented with formate and fumarate both. In the genome of C. concisus strain 13826, which belongs to subpopulation group III, one gene (CCC13826_2016) encoded for enzyme, reducing sulfite to hydrogen sulfide, was found. The genomes of C. concisus strains from different subpopulation groups have been sequenced recently by our group, which will provide further understanding of why subpopulation group II has less H2S production strains compared with other subpopulation groups.

Results from this chapter showed that supplementation of formate into HBAFe/S plates increased the production of H2S in C. concisus strains. The colonies of H2S positive strains were much darker compared with the colonies of the same strains grown on HBAFe/S plates (Table 4-3). However, it inhibited the growth of C. concisus at the same time (Figure 4-1). The inhibition to bacterial growth by formate was previously observed in other bacterial species, such as Bacillus cereus (187), Pseudomonas oxalaticus (188) and obligate methylotrophs (189). Chu et al. speculated that this was because bacterial cells spent biosynthetic energy for restoring impaired pH homeostasis upon formate transport (189).

Despite the inhibitory effect of formate to C. concisus growth observed in this chapter, a previous study showed that C. concisus was chemoattracted to formate with the optimal concentration of 50 mM under aerobic conditions without hydrogen gas in a micro- capillary tube (190). This was possibly because in that experimental system, no hydrogen gas or other electron donors were available; formate was the only available electron donor, which attracted C. concisus. Nevertheless, the impact of formate on C. concisus growth was examined in this chapter using only one strain. Given the great diversity of C. concisus strains, formate may only inhibit the growth of some C. concisus strains.

With fumarate supplementation, the H2S positivity in C. concisus strains was increased, although this was not significant. It suggests that C. concisus has a preference to undergo anaerobic respiration in which fumarate is involved rather than using sulfite as electron acceptor. Theoretically, it is possible because fumarate (+0.031V) has higher redox potential than sulfite (-0.11V) according to the standard redox potential at pH 7 (Table 4-4).

Table 4-4. List of the redox potential at pH 7.0

# Redox couple E`0 (volts) References + - CO2 + H + e / Formate -0.42 (181) + - 2H + 2e / H2 -0.41 (191) 2- HSO3 / H2S -0.11 (192) Fumarate2- + 2H+ + 2e- / Succinate2- 0.031 (193) - - O2 + 4H + 4e / 2H2O 0.815 (193)

# E`0 is the standard reduction potential at pH 7.0

Another finding from this chapter was that the fumarate supplementation significantly increased the growth of C. concisus. In bacterial metabolism, fumarate can play a role as an electron donor or an electron acceptor. C. concisus strain 13826 seems to have incomplete aerobic Krebs cycle with the missing gene encoded for succinate thiokinase enzyme. Therefore, fumarate respiration may be critical in some C. concisus strains electron acceptor. The finding that C. concisus had the best growth on HBA media supplemented with fumarate alone suggests that HBAfumarate plates should be used in the future for isolation of C. concisus from clinical samples.

In this study, it was also found that a significantly fewer numbers of oral C. concisus strains isolated from patients with UC produced H2S as compared with the strains isolated from CD patients and controls, suggesting that C. concisus is not responsible for the increased production of H2S previously observed in faecal samples collected from patients with UC (111, 112, 178). Thus, if C. concisus contributes to the development of UC, other pathogenic mechanisms may be involved. Further investigation may be required.

This study can be linked to IBD. A previous study showed an association between the consumption of fumarate and the severity of the IBD. In 2005, Erichsen et al. found that after the oral administration of ferrous fumarate (120 mg daily) in 19 patients with IBD,

the clinical disease activity, general well-being score and abdominal pain score were significantly deteriorated (194). Given that C. concisus was previously found to be highly prevalent in the intestinal tract of patients with IBD and the finding in this chapter that fumarate stimulated the growth of C. concisus, it is possible that fumarate has stimulated the growth of C. concisus in the intestinal tract of these patients, which contributed to the deteriorated clinical symptoms in these patients. Therefore, it may be better to use ferrous sulfate rather than ferrous fumarate to treat iron deficiency in patients with IBD (195-197).

In this study, it was found that supplementation of sodium fumarate increased the growth of C. concisus and supplementation of formate and fumarate together increased the production of H2S by C. concisus. Further investigation should be followed using an animal model. Many foods such as fruit juices contain fumaric acid and formic acid (198). If such supplementation can consistently aggravate the pathogenicity of C. concisus indeed inside the intestinal tract, patients with IBD should avoid these foods to prevent relapse.

In summary, studies in this chapter showed that 40 % C. concisus produced H2S on Fe/S HBA plates. The presence of H2 under atmospheric cultivation conditions is essential for detection of H2S production in C. concisus. Supplementation of formate and fumarate together significantly increased the H2S positivity of C. concisus strains with and without the presence of H2. Supplementation of fumarate alone increased the growth of C. concisus, suggesting that this chemical can be included in HBA plates in future isolation of C. concisus from clinical samples. C. concisus strains isolated from patients with UC had a significantly lower H2S positivity as compared with strains from patients with CD and controls, suggesting that if C. concisus contributes to the pathogenesis of IBD, other pathogenic mechanisms may be involved.

Chapter 5: In silico identification analysis of Campylobacter concisus prophages

5.1. Introduction

Prophages are temperate bacteriophage genomes that are integrated into bacterial chromosomal DNA or plasmid. These results in a latent form of a phage, in which the viral genes are preserved in the host bacterium. Prophages are commonly found in a bacterial genome, which contributes to various aspects of bacterial life such as bacterial fitness, genetic diversity and evolution (199). Prophages also play an important role in bacterial virulence. For example, the Escherichia coli strains O157:H7 causing epidemic haemorrhagic diarrhoea have acquired two Shiga toxin encoding prophages (200), and virulent Vibrio cholera strains have acquired a CTX prophage that encodes the cholera toxin (201).

A number of studies have reported an association between Campylobacter concisus and patients with inflammatory bowel disease (IBD). This is derived from a significantly higher prevalence of C. concisus in intestinal biopsy samples and faecal samples from patients with IBD as compared with the controls detected (43, 78, 81, 82).

Further analysis of housekeeping genes showed that C. concisus strains colonising the intestinal tract of patients with IBD originated from oral C. concisus strains, either from the patient’s own oral C. concisus or oral C. concisus strains from the others (89). This view was further supported by a recent study analysing the population structure of C. concisus, which showed that there was no distinct enteric C. concisus strain clusters (183).

A number of studies have examined the enteric virulence of C. concisus strains. It was found that C. concisus upregulated the level of TLR4 and induced IL-8 production in human intestinal epithelial cell line, HT-29 cells (139). C. concisus strains were also

shown to adhere and invade the human intestinal epithelial cell line, Caco2 cells. Furthermore, C. concisus strains were shown to induce apoptosis in HT-29 cells and caused damage to the intestinal epithelial barrier (45, 89, 139). These studies all found that the virulence abilities varied between strains.

C. concisus is a commensal oral bacterium (2). It is unclear why some oral C. concisus strains have acquired the potential for enteric pathogenicity. One possibility is that some oral strains have acquired virulence factors from bacteriophages. In order to investigate this hypothesis, in silico analysis of the genome of C. concisus strain 13826 to search for prophage acquisition was performed in this chapter.

5.2. Materials and methods

5.2.1. Identification of prophage in C. concisus 13826 reference strain

The PHAge Search Tool (PHAST) is a web server designed to rapidly and accurately identify, annotate and graphically display prophage sequences within bacterial genomes or plasmids (202). This web-based tool was used to identify any potential prophages within C. concisus strain 13826 (Gene bank accession number NC_009802). The genome of C. concisus strain 13826 is available in National Center for Biotechnology Information (NCBI) database. Based on the criteria of PHAST software for scoring prophage regions, a prophage region with a score less than 70 is marked as an incomplete phage, a score between 70 to 90 as questionable and or greater than 90 as intact (202). The PHAST software automatically identifies phage-like proteins and putative attachment sites. In addition, the genes and proteins within the resulted prophage region were manually inspected using the NCBI database.

5.2.2. Re-annotation of some proteins in the prophage region

Genes and proteins within the PHAST identified prophage region were compared using MEGA-5 software and Clustal Omega respectively (203, 204). Genes and proteins with identical sequences but different names in the NCBI database were re-annotated.

The protein basic local alignment search tool (BLAST) was used to identify the hypothetical proteins within the prophage region against virus taxonomy (tax id 10239) (205). Any missing protein sequences from auto-annotation in the NCBI database was manually translated by the publicly available software ExPASy translate tool (http://web.expasy.org/translate/) (206). The proteins were re-annotated with the result of BLAST.

The protein identity between proteins in the prophage found in C. concisus strain 13826 was calculated by the protein alignment using Clustal Omega. The proteins with identity more than 40 % were re-annotated (207).

5.2.3. Examination of secretion proteins

Putative secreted proteins encoded by genes in the identified prophage regions in the genome of C. concisus strain 13826 were predicted using SignalP 4.1 and SecretomeP 2.0 software (208). SignalP 4.1 predicts secreted proteins based on the presence of the traditional N-terminal signal peptides (208, 209). SecretomeP 2.0 predicts non-classical (not signal peptide triggered) protein secretion based on analysis of post-translational and sub-localisational aspects of the proteins (209).

5.2.4. Examination of operon in prophage

Putative operons were predicted using previously published software ProOpDB [23].

5.2.5. Examination of paralogue proteins in prophage

KEGG Sequence Similarity Data Base (SSDB) was used to determine the paralogue proteins in CON_phi prophage (210). Only proteins with a Smith-Waterman (SW) similarity score greater than 100 was considered as a paralogue in this study. The prophage protein having the largest amino acid length was used for the comparison.

Five paralogue proteins were identified within each prophage element in the genome of C. concisus strain 13826. These included phage integrase, bacteriophage replication gene A protein, putative phage assembly protein, DicA domain-containing sensory box protein and envelope glycoprotein.

The phylogenetic relationship of each paralogue protein between prophage elements were examined at nucleotide level. For this analysis, MEGA5 was used with the maximum- likelihood tree and a bootstrap value of 1,000.

In addition, multi-locus sequence typing (MLST) analysis was performed using the five paralogue proteins. Based on the nucleotide alignments, the longest matched nucleotide sequence of each paralogue protein among prophage elements was selected. The phylogenetic tree was generated using MEGA5. The full sequences used for the MLST analysis were shown in the appendix 1.

5.2.6. Consensus sequence of XerC or XerD binding sites of prophages

Four prophages were known to employ the XerC/D system for phage integration, which include Ypfφ (211), VGJφ (212), CTXφ (212), and φXacF1 (213). In this analysis, φXacF1 was excluded due to the distinct host specie (Xanthomonas axonopodis pv. citri), which only infects plants. The nucleotide sequences of XerC and XerD binding sites from 6 organisms are shown in Table 5-1.

Consensus sequence of XerC or XerD binding sites were analysed by WebLogo 2.8.2 with the default setting (214).

Table 5-1. DNA binding sites to the XerC or XerD site-specific recombinase

Organisms attP/attB/attL* XerC binding site XerD binding site References

Ypfφ attP GGTGCGCATAA TTATGTTGAAA (211) Y. pestis attB GGTGCACATAA TTATGTTAAAT (211) Ypfφ attL GGTGCACATAA TTATGTTGAAAA (211) VGJφ attP TTTTACCATAA TAATGCGAAGT (212) CTXφ attP1 AGTGCGTATTA TTATGTTGAGG (212) CTXφ attP2 AATGCGTATTA TTATGTTACGG (212) * attP, attB and attL indicate the attachment sites in phage, bacteria, and prophage, respectively. CTXφ phage contained two attP sites in the genome.

5.2.7. Protein identity of phage integrases in CON_phi with XerC, XerD, XerH and XerT

Protein sequences of three phage integrases encoded by CCC13826_0638, CCC13826_2077 and CCC13826_1213 in the CON_phi prophage and two bacterial Xer recombinases encoded by CCC13826_2090 and CCC13826_0105 in C. concisus strain 13826 were compared with that of XerC and XerD from E. coli strain K12, XerH from H. pylori strain 26695 and XerT from H. pylori strain J99.

All protein sequences were obtained from the NCBI database and aligned by Clustal Omega. The proteins with identity greater than 40 % was considered as significant (207).

5.2.8. Characteristics of phage integrase in prophage in C. concisus strain 13826

The InterPro tool was used to identify the conserved motif in phage integrase (215). Conserved domain database (CDD) was used to determine the potential function for full- length phage integrase (216).

The Phyre2 tool was used to predict the three-dimensional structure of phage integrase from C. concisus strain 13826 (217) and the predicted model was analysed by using Jmol (218).

5.2.9. Identification of CON_phi-like prophages in Campylobacter genus

The UniProtKB (219) protein database was used to search for the zot protein. This resulted in more than 2,000 protein sequences. Among those proteins, bacterial species belonging to the genus Campylobacter were selectively screened. These included C. concisus, Campylobacter ureolyticus, Campylobacter gracilis. Campylobacter hyointestinalis, Campylobacter corcagiensis, Campylobacter jejuni and other Campylobacter species. The genome structure was manually inspected using the NCBI database. The protein identity (%) was compared by Clustal Omega (204). A protein identity greater than 40 % was considered as significant (207). Since most of proteins identified were hypothetical proteins, the locus tag of the genes encoding for those proteins was labelled instead of the

name of genes in the Figures and Tables. In addition, any missing proteins from the genome in the NCBI database were manually detected by ExPASy translate tool.

5.2.10. Phylogenetic tree construction of the zot gene and phage maturation protein of Campylobacter species

The zot protein and phage maturation protein were used in this study. A total of thirteen Campylobacter species were subjected to this analysis. The phylogenetic tree was constructed based on nucleotide levels. A maximum-likelihood tree was generated by MEGA5 with a bootstrap setting of 1,000.

The phylogenetic tree of 16S rRNA from the thirteen Campylobacter spp. was included as controls. In this tree, eleven 16S rRNA sequences were used, including C. hyointestinalis subsp. hyointestinalis, C. jejuni subsp. jejuni 86605, C. jejuni subsp. jejuni 6004, C. jejuni subsp. doylei 269.97, C. ureolyticus ACS-301-V-Sch3b, C. corcagiensis CIT045, C. gracilis RM3268, Campylobacter sp. FOBRC14, C. concisus UNSWCS, C. concisus UNSW3 and C. concisus 13826

5.3. Results

5.3.1. Identification of prophage in C. concisus strain 13826

PHAST search was used to predict the prophage existence in the genome of C. concisus strain 13826. As a result, a prophage in C. concisus strain 13826 was identified in the nucleotide position 1,576,683 to 1,615,449 (38,767 bp) in the genome (Figure 5-1). The prophage identified in the genome of C. concisus strain 13826 had a score of 80, being classified as a “questionable” prophage according to the PHAST criteria. The identified prophage had a GC content of 32.6 %, which is lower than that in the whole genome of C. concisus strain 13826 (39.4 %). In this region, 39 genes with open reading frames were identified. Four of these genes encoded phage integrases and 10 genes encoded phage-like proteins. In addition to the PHAST result, two more phage-like genes (CCC13826_0198 and CCC13826_0199) followed at the nucleotide position from 1,615,472 to 1,616,148.

Questionable prophage (17,453bp)

Figure 5-1. Prophage identified by PHAST search in the genome map of C. concisus strain 13826 C. concisus strain 13826 contained one chromosome (size 2.05 Mbps), in which the questionable prophage was detected by PHAST. The prophage region is clearly indicated as a grey box on the bacterial genome map. The length of the box corresponds to the relative sizes of the prophage DNA with respect to the bacterial chromosome. The identified prophage is located at nucleotide position between 1,576,686 and 1,614,075.

5.3.1.1. Re-annotation of proteins in the putative prophage identified by PHAST

Comparison of genes and proteins in the identified prophage region revealed that some identical genes and proteins were annotated with different protein names in the NCBI database. A number of genes that have identical nucleotide sequences were re-annotated with identical protein names. The tandem repeat was also observed. Furthermore, phage- like proteins identified by PHAST or BLAST in this study were also annotated as such (Table 5-2). A protein with less than an E-value of 1 from the BLAST result was considered as a phage-like protein.

Table 5-2. Phage-like proteins in the genome of C. concisus strain 13826 Gene (locus tag) Matched phage proteins E value Identity* CCC13826_0568 Aotine herpesvirus 1 membrane protein A23 2.0×10-5 20.0 % CCC13826_0638 Bordetella phage BMP-1 integrase 1.0 ×10-21 18.0 % CCC13826_2272 Mycobacterium phage U2 Putative DNA methylase 3.0 ×10-6 20.6 % CCC13826_2273 @ Bacteriophage replication gene A protein <1 27 % CCC13826_2274 Enterobacteria phage M13 phage assembly protein 1.0 ×10-4 8.9 % CCC13826_2275 @ Human immunodeficiency virus 1 envelope glycoprotein <1 34 % CCC13826_2082 Clostridium phage 3626 putative integrase 5.0×10-21 19.0 % CCC13826_1099 Erwinia phage ENT90 replication protein A 2.0×10-7 9.2 % CCC13826_1100 Enterobacteria phage I2-2 gene IV product 2.0×10-6 26.0 % CCC13826_2276 Vibrio phage fs1 hypothetical protein fs1p06 7.0×10-6 19.5 % CCC13826_0183 Synechococcus phage S-IOM18 baseplate wedge component 3.0×10-5 24.6 % CCC13826_2077 Clostridium phage 3626 putative integrase 5.0×10-21 21.8 % CCC13826_0020 Erwinia phage ENT90 replication protein A 2.0×10-7 9.2 % CCC13826_0019 Vibrio phage fs2 putative maturation protein 1.0×10-8 14.0 % CCC13826_2075 Vibrio phage fs1 hypothetical protein fs1p06 7.0×10-6 19.5 % CCC13826_1299 Synechococcus phage S-IOM18 baseplate wedge component 3.0×10-5 24.6 % CCC13826_0706 Bordetella phage BMP-1 integrase 1.0×10-21 18.0 % CCC13826_0188 Mycobacterium phage U2 Putative DNA methylase 3.0 ×10-6 20.6 % CCC13826_0189 @ Bacteriophage replication gene A protein <1 27 % CCC13826_0190 Vibrio phage VFJ phage morphogenesis protein 3×10-13 18.1 % CCC13826_0198 @ Staphylococcus phage StauST398-3 Cro-like protein <1 48 % * Identity is the percentage of the number of identical amino acids in the protein alignment to the total number of amino acids in the matched phage protein. @ Phage-like proteins were identified by the NCBI BLAST search.

The protein encoded by CCC13826_0568 was annotated as a hypothetical protein in the NCBI database. This study showed that it was a phage-like protein, with 20 % identity with membrane protein A23 of Aotine herpesvirus 1. Thus, it was re-annotated as a putative membrane protein.

Gene sequence comparison showed that CCC13826_0638 and CCC13826_0706 were identical. In the NCBI database, while the protein encoded by CCC13826_0638 was annotated as lipoprotein, the protein encoded by CCC13826_0706 was annotated as site- specific recombinase (phage integrase family protein). As a result, both proteins were re- annotated as phage integrase.

The protein encoded by CCC13826_2272 was annotated as a glutathionylspermidine synthase family protein in the NBCI database. Gene sequence comparison showed that CCC13826_2272 was identical to CCC13826_0188. In this study, the protein encoded by CCC13826_2272 and CCC13826_0188 showed a 20.6 % identity to the putative DNA methylase of Mycobacterium phage U2, indicating that it is a phage-like proteins. The protein encoded by CCC13826_2272 was re-annotated as putative DNA methylase.

The protein encoded by CCC13826_2274 was annotated as protein Yitk in the NBCI database. This study showed that it was a phage-like protein, with 8.9 % identity with the phage assembly protein of enterobacteria phage M13. The protein encoded by CCC13826_2274 was re-annotated as putative phage assembly protein.

The protein encoded by CCC13826_2275 was annotated as hypothetical protein in the NCBI database. The gene comparison showed that the gene sequence of CCC13826_2275, CCC13826_2078 and CCC13826_0186 were identical. Additionally, the protein encoded by CCC13826_2275, CCC13826_2078 and CCC13826_0186 showed 34 % identity with the envelope glycoprotein of Human immunodeficiency virus 1, indicating that it is a phage-like protein. The protein encoded by CCC13826_2272 was re-annotated as an envelope glycoprotein.

Gene sequence comparison showed that CCC13826_1099 and CCC13826_0020 were identical. In the NCBI database, while the protein encoded by CCC13826_1099 was annotated as radical SAM domain-containing protein, the protein encoded by CCC13826_0020 was re-annotated as bacteriophage replication gene A protein. In this study, both proteins were annotated as bacteriophage replication gene A protein.

Gene sequence comparison showed that CCC13826_1102, CCC13826_0164 and CCC13826_0185 were identical. In the NCBI database, while the proteins encoded by CCC13826_1102 and CCC13826_0185 were annotated as sensory box protein, the protein encoded by CCC13826_0164 was annotated as a hypothetical protein. In this study, all three proteins were annotated as sensory box protein. These proteins were then found to contain the transcriptional repressor DicA domain (Accession Number PRK09706) in the NCBI conserved domain database. Therefore, they were re-named as DicA domain- containing sensory box protein.

Gene comparison showed that CCC13826_2277 and CCC13826_2074 were identical. In the NCBI database, these two proteins were labelled as hypothetical protein. This study showed that the protein encoded by CCC13826_2277 and CCC13826_2074 was a phage- like protein, with 27 % identity with Cordyline virus 3 polyprotein. The proteins encoded by CCC13826_2277 and CCC13826_2074 were annotated as a polyprotein.

The genes CCC13826_2082 and CCC13826_2077 have identical nucleotide sequences. In the NCBI database, CCC13826_2082 was annotated as site-specific recombinase (phage integrase family protein) with a length of 260 amino acids and CCC13826_2077 was annotated as phage integrase with a length of 302 amino acids. In this study, the protein encoded by CCC13826_2082 was re-annotated as phage integrase with a length of 302 amino acids.

The nucleotide sequence for a CCC13826_1100 and CCC13826_0019 were identical. In the NCBI database, the protein encoded by CCC13826_1100 was annotated as an ABC transporter ATP-binding protein, and the protein encoded by CCC13826_0019 was annotated as a hypothetical protein. This study showed that the protein encoded by CCC13826_1100 and CCC13826_0019 was a phage-like protein, with 14 % identity with

Vibrio phage fs2 putative maturation protein. The proteins encoded by CCC13826_1100 and CCC13826_0019 were re-annotated as putative phage maturation proteins.

Genes CCC13826_0183 and CCC13826_1299 have identical nucleotide sequences. In the NCBI database, the proteins encoded by CCC13826_0183 and CCC13826_1299 were annotated as hypothetical proteins. PHAST search revealed it to be a phage-like protein, with 24.6 % identity with the Synechococcus phage S-IOM18 baseplate wedge component. In this study, the protein encoded by CCC13826_0183 and CCC13826_1299 was re- annotated as putative Synechococcus phage S-IOM18 baseplate wedge component.

The protein encoded by CCC13826_0190 was annotated as type II and III secretion system protein in the NCBI database. This study showed that it was a phage-like protein, with 18.1 % identity with the phage morphogenesis protein of Vibrio phage VFJ. The protein encoded by CCC13826_0190 was then annotated as putative phage morphogenesis protein.

The protein encoded by CCC13826_0191 was annotated as hypothetical protein in the NCBI database. Comparison of this protein with the Zot protein encoded by CCC13826_2276 and CCC13826_2075 revealed 27.85 % identity. Given this, the protein encoded by CCC13826_0191 was annotated as Zot-like protein.

The protein encoded by CCC13826_0198 was annotated as hypothetical protein in the NCBI database. The protein blast result showed that it was a phage-like protein, with 48 % identity with Cro-like protein of Staphylococcus phage StauST398-3. Cro protein family is known to repress the gene transcription during the lytic cycle of phage. Thus, the protein encoded by CCC13826_0198 was annotated as Cro-like protein.

A total of 21 phage-like proteins in the phage island were identified by PHAST and BLAST, which were shown in Table 5-2. All the proteins in this region were listed in the Table 5-3.

Table 5-3. Proteins encoded by putative prophage island in C. concisus strain 13826

Nucleotide position Locus tag Encoded proteins# Size (aa) 1576686..1581986 CCC13826_0568 membrane protein 1766 1582376..1583269 CCC13826_0638 phage integrase 297 1583269..1583880 CCC13826_2272 putative DNA methylase 203 1583895..1585235 CCC13826_2273 bacteriophage replication gene A protein 446 1585467..1585853 CCC13826_2274 putative phage assembly protein 128 1585894..1586115 CCC13826_1101 phosphonate uptake transporter 73 1586807..1587043 CCC13826_1102 DicA domain-containing sensory box protein 78 1587044..1587496 CCC13826_2275 envelope glycoprotein 150 1587598..1588506 CCC13826_2082 phage integrase 302 1588755..1589966 CCC13826_1099 bacteriophage replication gene A protein 403 1590015..1591184 CCC13826_1100 putative phage maturation protein 389 1591228..1592352 CCC13826_2276 zonular occludens toxin (Zot) 374 1592354..1592800 CCC13826_2277 polyprotein 148 1592797..1594923 CCC13826_0183 putative baseplate wedge component 708 1594955..1595095 CCC13826_0184 hypothetical protein 46 1595477..1595698 CCC13826_2278 phosphonate uptake transporter 73 1596412..1596648 CCC13826_0164 DicA domain-containing sensory box protein 78 1596649..1597101 CCC13826_2078 envelope glycoprotein 150 1597203..1598111 CCC13826_2077 phage integrase 302 1598360..1599571 CCC13826_0020 bacteriophage replication gene A protein 403 1599620..1600789 CCC13826_0019 putative phage maturation protein 389 1600833..1601957 CCC13826_2075 Zot 374 1601959..1602405 CCC13826_2074 polyprotein 148 1602402..1604528 CCC13826_1299 putative baseplate wedge component 708 1604560..1604700 CCC13826_1298 hypothetical protein 46 1605082..1605303 CCC13826_2279 phosphonate uptake transporter 73 1606017..1606253 CCC13826_0185 DicA domain-containing sensory box protein 78 1606254..1606706 CCC13826_0186 envelope glycoprotein 150 * aa (amino acids) # Protein names with new annotation

Table 5-3. (Continued)

Nucleotide Locus tag Encoded proteins# Size position (aa) 1606808..1607701 CCC13826_0706 phage integrase 297 1607701..1608312 CCC13826_0188 putative DNA methylase 203 1608327..1609667 CCC13826_0189 bacteriophage replication gene A protein 446 1609899..1611041 CCC13826_0190 putative morphogenesis protein 380 1610992..1612116 CCC13826_0191 Zot-like protein 374 1612118..1612438 CCC13826_0192 hypothetical protein 106 1612537..1613946 CCC13826_0193 hypothetical protein 469 1613956..1614075 CCC13826_0194 alkyl hydroperoxide reductase 39 1614090..1614209 CCC13826_0195 hypothetical protein 39 1614489..1614707 CCC13826_0196 hypothetical protein 72 1614801..1615178 CCC13826_0197 hypothetical protein 125 1615472..1615681 CCC13826_0198 Cro-like repressor protein 69 1615681..1616148 CCC13826_0199 hypothetical protein 155 * aa (amino acid) # Protein names with new annotation

5.3.1.2. Multiple prophages within the prophage

Following re-annotation of the proteins, it became apparent that the prophage identified in the C. concisus 13826 genome contained four prophage elements, each beginning with a phage integrase (Table 5-3). The overall genetic structure of the identified prophage island in C. concisus strain 13826 genome is shown in Figure 5-3

It was found that the prophage island contained four prophages (Figure 5-3 A). The first prophage, CON_phi1, had a genome size of 5.2 kb, which contained seven protein encoding genes. This prophage was inserted at the 3 prime end of a gene encoding a tRNA-Ser. The second and third prophage appeared to be a tandem repeat at the nucleotide level. Each identical tandem repeat had a genome size of 9.6 kb consisting of 10 protein- encoding genes. This prophage displaying a tandem repeat was named CON_phi2. The fourth prophage named CONphi_3 contained 12 protein-encoding genes with a genome size of 8.6 kb. Moreover, the sequences of last three genes in CON_phi1, CON_phi2 and CON_phi3 were identical.

Another prophage which was apart from the phage island detected by PHAST search was found by manual inspection in the genome of C. concisus strain 13826, which was named CON_phi4. This CON_phi4 was located at position from 938,750 to 946,901 (8,151 bp) (Figure 5-2) with the similar gene structure to CON_phi3 (Figure 5-3 B). This region contained 12 genes with an open reading frame. This CON_phi4 prophage was inserted at the 3 prime end of the gene encoding a tRNA-Met.

Figure 5-2. The location of prophages in C. concisus strain 13826 on the genome The prophages determined by PHAST and manual inspections were indicated as grey boxes on the bacterial genome map. The lengths of the boxes correspond to the relative sizes of the prophage DNA with respect to the bacterial chromosome. The pointed end of each box represents the C-terminal position of the phage integrase. The prophage island containing CON_phi1, CON_phi2 and CON_phi3 is located at the nucleotide position between 1,576,686 and 1,614,075, while CON_phi4 prophage is located at the nucleotide position between 939,158 and 946,901.

Figure 5-3. Genetic structures of the prophages identified in C. concisus strain 13826 Four prophages were identified in the genome of C. concisus strain 13826, which includes CON_phi1, CON_phi2, CON_phi3 and CON_phi4. Gene arrangement of the prophage island containing multiple prophages, including CON_phi1, CON_phi2 and CON_phi3 is shown in A. The gene structure in the CON_phi4, is shown in B. Identical genes are highlighted with the same colour. int gene encoding phage integrase, and zot gene encoding zonula occludens toxin. The number above each gene indicates the locus tag IDs according to the NCBI.

5.3.1.3. Identification of CON_phi4 in C. concisus strain 13826 and re-annotation of proteins

Gene arrangement in the CON_phi4 prophage was very similar to the CON_phi3. The protein identity (%) was compared between CON_phi4 and CON_phi3 using Clustal Omega (Table 5-4). Most of the protein identities of the two prophages were more than 40 % except CCC13826_1213 and CCC13826_1204. Given this, the protein names in the CON_phi4 prophage was re-annotated according to the proteins listed in Table 5-5.

The protein encoded by CCC13826_1213 was annotated as phage integrase in the NCBI database, and the BLAST search result that was the best matched protein was phage integrase from Clostridium phage phi CD211 with 29 % protein identity and 2×10-15 of E- value. Therefore, the protein encoded by CCC13826_1213 remained as phage integrase.

The protein encoded by CCC13826_1204 was annotated as a SNF7 family protein in the NCBI database. The blast result showed that this protein was a phage-like protein with 35 % identity to the Porcine torovirus replicase. Given this, the protein encoded by CCC13826_1204 was annotated as replicase.

There was no gene in CON_phi4 corresponding to the CCC13826_0188 gene encoded for the putative DNA methylase (Table 5-3).

A total of 12 proteins in the CON_phi4 were identified and are listed in the Table 5-5.

Table 5-4. The identity (%) of proteins in CON_phi3 to that in the CON_phi4

CON_phi3 CON_phi4 Identity (%)@ CCC13826_0706 CCC13826_1213 23.2 CCC13826_0188 NA*

CCC13826_0189 CCC13826_1212 93.7 CCC13826_0190 CCC13826_1211 99.0 CCC13826_0191 CCC13826_1210 92.3 CCC13826_0192 CCC13826_1209 92.0 CCC13826_0193 CCC13826_1208 58.9 CCC13826_0194 CCC13826_2249 100.0 CCC13826_0195 CCC13826_2248 97.4 CCC13826_0196 CCC13826_1206 61.1 CCC13826_0197 CCC13826_1205 95.2 CCC13826_0198 CCC13826_1204 37.5 CCC13826_0199 CCC13826_1203 85.8 * The corresponding gene to the gene with locus tag of CCC13826_0188 is absent in the CON_phi4 @ The protein identity greater than 40 % is labelled in bold

Table 5-5. A list of phage-like proteins in the CON_phi4

Nucleotide Locus tag Encoded proteins# Size* position 946035..946901 CCC13826_1213 phage integrase 288 944687..946027 CCC13826_1212 bacteriophage replication gene A protein 446 943316..944458 CCC13826_1211 putative morphogenesis protein 380 942241..943365 CCC13826_1210 Zot-like protein 374 941821..942237 CCC13826_1209 hypothetical protein 138 941167..941817 CCC13826_1208 hypothetical protein 216 940312..940431 CCC13826_2249 alkyl hydroperoxide reductase 39 940179..940298 CCC13826_2248 hypothetical protein 39 939681..939902 CCC13826_1206 hypothetical protein 73 939158..939586 CCC13826_1205 hypothetical protein 142 938747..938983 CCC13826_1204 replicase 78 938283..938750 CCC13826_1203 hypothetical protein 155 * Amino acid size # Protein names with new annotation

5.3.1.4. Paralogue search and MLST

The genetic comparison between four prophages in C. concisus strain 13826 was performed based on a paralogue search. The possible paralogous proteins among the phage-like proteins were detected in the four phage elements using KEGG Sequence Similarity Data Base (SSDB) (210).

Only proteins with Smith-Waterman (SW) similarity score greater than 100 was considered as paralogue. For this search, the prophage protein having the longest amino acid length was chosen. From this study, it was found that each prophage element contained five paralogous proteins (Table 5-6). In addition, MEGA5 was used to generate the phylogenetic tree.

The first protein with 100 SW-score between phage elements was phage integrase. Based on the phage integrase from CON_phi2 (CCC13826_2077), the phage integrase from CON_phi1 is a paralogues of it with a SW-score of 1854. The phage integrase from CON_phi3 was the identical to that of CON_phi1 and had the same SW-score. Those three phage integrases were from the prophage island. The phage integrase from CON_phi4 was the least similar paralogue to phage integrase encoded by CCC13826_0638 with a SW- score of 231. The phylogenetic relationship between phage integrases displayed the similar result to paralogue analysis using SW-score (Figure 5-4 A1).

Bacteriophage replication gene A protein encoded by CCC13826_2273 from CON_phi1 was very similar to that encoded by CCC13826_0189 from CON_phi3 and CCC13826_1212 from CON_phi4 with a SW-score of 2991 and 2817, whereas the protein encoded by CCC13826_1099 from CON_phi2 had the lowest SW-score compared that from CON_phi1, which was 867 (Figure 5-4 A2). The phylogenetic relationship obtained from the MLST study showed that three bacteriophage replication gene A proteins from CON_phi1, CON_phi3 and CON_phi4 were in the same cluster.

Based on both phylogenetic analyses (Figure 5-4 A1 and Figure 5-4 A2), the phage integrase was shown to be closely related to bacteriophage replication gene A protein, as both trees were very alike.

The putative morphogenesis protein encoded by CCC13826_0190 from CON_phi3 was used for a paralogue search (Figure 5-4 A3). The paralogue encoded by CCC13826_1211 was found in CON_phi4 with a SW-score of 2357. The putative phage assembly protein encoded by CCC13826_2274 from CON_phi1 and putative phage maturation protein encoded by CCC13826_1100 from CON_phi2 followed with a score of 545 and 402. The phylogenetic tree for those proteins showed that CON_phi1, CON_phi3 and CON_phi4 were more genetically related than CON_phi2, which were shown through the phylogenetic relationship between the bacteriophage replication gene A proteins of each.

KEGG-SSDB paralog search suggested that the replicase encoded by CCC13826_1204 from CON_phi4 and DicA domain-containing sensory box protein encoded by CCC13826_1102 from CON_phi1 were paralogous to each other with a score of 219 (Figure 5-4 A4). The DicA domain-containing sensory box protein encoded by CCC13826_0164 from CON_phi2 and Cro-like repressor protein encoded by CCC13826_0198 from CON_phi3 had a SW-score of 218 and 128 based on the replicase encoded by CCC13826_1204.

The hypothetical protein encoded by CCC13826_1203 from CON_phi4 was used to search for paralogous proteins in C. concisus 13826 (Figure 5-4 A5). The hypothetical protein encoded by CCC13826_0199 from CON_phi3 was the most similar paralogue, with a score of 908, whereas the envelope glycoprotein encoded by CCC13826_2275 from CON_phi1 showed less paralogous similarity with a score of 153. The envelope glycoprotein from CON_phi2 was identical to that from CON_phi1. Moreover, the phylogenetic relationship showed that two hypothetical protein from CON_phi3 and CON_phi4 fell into one cluster and the other envelope glycoprotein from CON_phi1 and CON_phi2 formed another cluster.

Using the five paralogues, the phylogenetic tree between prophage elements was generated by MLST analysis at the nucleotide level (Figure 5-4 B). The gene sequence used for MLST analysis is shown in the Appendix 1. Two distinct clusters were identified. The major difference between the clusters was that one cluster consisting of CON_phi1 and CON_phi2 contained the phosphonate uptake transporter, whereas another cluster

consisting of CON_phi3 and CON_phi4 contained alkyl hydroperoxide reductase and three hypothetical proteins instead.

Table 5-6. Paralogous proteins in four prophage elements*

Phage elements Locus tag Protein name

CON_phi1 1 CCC13826_0638 Phage integrase 2 CCC13826_2273 Bacteriophage replication gene A protein 3 CCC13826_2274 Putative phage assembly protein 4 CCC13826_1102 DicA domain-containing sensory box protein 5 CCC13826_2275 Envelope glycoprotein

CON_phi2 1 CCC13826_2082 Phage integrase 2 CCC13826_1099 Bacteriophage replication gene A protein 3 CCC13826_1100 Putative phage maturation protein 4 CCC13826_0164 DicA domain-containing sensory box protein 5 CCC13826_2078 Envelope glycoprotein

CON_phi3 1 CCC13826_0706 Phage integrase 2 CCC13826_0189 Bacteriophage replication gene A protein 3 CCC13826_0190 Putative morphogenesis protein 4 CCC13826_0198 Cro-like repressor protein 5 CCC13826_0199 Hypothetical protein

CON_phi4 1 CCC13826_1213 Phage integrase 2 CCC13826_1212 Bacteriophage replication gene A protein 3 CCC13826_1211 Putative morphogenesis protein 4 CCC13826_1204 Replicase 5 CCC13826_1203 Hypothetical protein * The paralogue proteins in each prophage element were determined by KEGG SSDB. Only SW scores greater than 100 were included.

100

Figure 5-4. Phylogenetic relationship between genes encoding paralogous protein in four prophages. The five paralogous phage proteins in each CON_phi were shown in Table 5-6. A. Phylogenetic tree (maximum-likelyhood) of each paralogous protein between prophage elements at the nucleotide level, which included (A1) phage integrase, (A2) bacteriophage replication gene A protein, (A3) maturation or morphogenesis protein, (A4) DicA domain- containing sensory box protein, Cro-like protein or replicase, and (A5) envelope protein. The gene name is denoted as the locus tag, and the name of prophages is shown in brackets. B. The MLST analysis was performed on these five paralogues.

101

5.3.1.5. Structure of integrase in C. concisus prophage

From the identified prophages, three different phage integrases were found, including CCC13826_0638 from CON_phi1 (identical to CCC13826_0706 from CON_phi3), CCC13826_2082 from CON_phi2 (identical to CCC13826_2077) and CCC13826_1213 from CON_phi4.

All three integrase contained two identical protein motifs; analysis by InterPro showed there was a sterile alpha motif (SAM)-like N-terminal (IPR004107) motif at the N- terminal residue having approximately 100 amino acids in length, which is associated with DNA binding, and a catalytic motif (IPR002104) at the C-terminal, which is related to Cre recombinase.

The conserved domain database (CDD) was used to predict a multi-domain. All phage integrases from CON_phi1, CON_phi2 and CON_phi3 contained a XerC multi-domain (Accession No. COG 0582). The phage integrase from CON_phi4 contained a XerD multi- domain (Accession No. COG4974).

Secondary structure and tertiary structure were predicted by Phyre2 and Jmol (Figure 5-5). This showed that CON_phi prophage integrase consisted of two domains, a N-terminal α- helix domain and a C-terminal α-helix and β-sheet domain. All three integrases contained two α-helix hairpins arranged at 90° to each other at the N-terminal residue. The integrase encoded by CCC13826_2082 had one more extra helix at the beginning of the N-terminal (Figure 5-5 A, mid).

At the C-terminal domains, all three integrases contained predominantly α-helices with a few β-sheets. Interestingly, the integrase encoded by CCC13826_0638 and CCC13826_2077 (both integrases from the prophage island) contained three sequential anti-parallel β-sheets, while the integrase encoded by CCC13826_1213 (from the CON_phi4) contained three sequential anti-parallel β-sheets followed by two sequential anti-parallel β-sheets (Figure 5-5 B)

102

Figure 5-5. The 3D structure of three integrase found in CON_phi prophage. Phyre2 and Jmol were used to analyse the 3D structure of each phage integrase from CON_phi1, CON_phi2 and CON_phi4. The integrase from CON_phi3, which was the identical to that from CON_phi1 was eliminated. Amino acid sequences were obtained from the NCBI database. In A, the ribbon ending with the dark-blue colour indicated the N-termini. In B, the α-helices and β-sheets are highlighted in pink and orange colour respectively.

103

5.3.1.6. Consensus sequence of XerC or XerD binding sites in various organisms

The nucleotide sequences binding to XerC or XerD for bacterial cell division is highly conserved between bacterial species. However, the conservation of XerC or XerD binding sites for phage insertion has not been compared. In this study, using seven XerC and XerD binding sites from known prophages (Table 5-1), the nucleotide consequences was compared by WebLogo (Figure 5-6). Interestingly, the sequences of last four nucleotides of the XerC binding site (ATAA) and first four nucleotides for XerD binding site (TTAT) were reverse complementary.

Figure 5-6. Frequency of nucleotide of XerC or XerD binding sites The nucleotide sequences binding to XerC or XerD recombinase were obtained from the Table 5-1. The nucleotide consensus sequence binding to each recombinase was analysed by WebLogo 2.8.2 (214). The left end of the X-axis indicates the five prime end of each nucleotide sequence.

104

5.3.1.7. Protein identity (%) of phage integrase in CON_phi comparing with XerC, XerD, XerH and XerT

There are four types of Xer recombinases, which includes XerC, XerD, XerH and XerT. It was shown that the phage integrases in the CON_phi prophage are Xer recombinases. However, it is unknown what types of Xer recombinases they are. In this study, the amino acid sequences of phage integrases were compared with various Xer type recombinases previously known. Two bacterial Xer recombinases (CCC13826_2090 and CCC13826_0105) found in C. concisus strain 13826 were included as control. Only protein identities (%) greater than 40 % were considered as significant (207) (Table 5-7).

As a result, four phage integrases found in the CON_phi prophages had no significant result, with all protein identities lower than 22 %. Conversely, the two Xer recombinase controls, CCC13826_2090 and CCC13826_0105, showed significant protein identity with XerH type recombinase, which were 47.6 % and 47.5 % respectively.

Table 5-7. Protein identity (%) of Xer type recombinases in C. concisus strain 13826 with XerC, XerD, XerH and XerT@

Xer type recombinases Position* XerC XerD XerH XerT CCC13826_0638 CON_phi1 17.6% 18.2% 20.6% 20.6% CCC13826_2077 CON_phi2 16.6% 18.9% 19.5% 17.9% CCC13826_1213 CON_phi4 18.1% 16.0% 21.5% 21.5% CCC13826_2090 Host 23.8% 23.2% 47.6% 17.3% CCC13826_0105 Host 23.7% 23.2% 47.5% 26.3% * Four phage integrases (Xer type recombinases) were found in the C. concisus prophage elements. Two Xer type recombinases encoded by CCC13826_2090 and CCC13826_0105 are from the host C. concisus strain 13826. The phage integrase CCC13826_0706 from CON_phi3 was excluded as its amino acid sequence is identical to CCC13826_0638 from CON_phi1. @ Amino acid sequences of XerC (B3811) from E. coli strain K12, XerD (B2894) from E. coli strain K12, XerH (HP0675) from H. pylori strain 26695 and XerT (JHP_RS04925) from H. pylori strain J99 were used.

105

5.3.1.8. Putative attachment sites

The upstream and downstream nucleotide sequences of CON_phi were compared. From this analysis, seven identical attachment sites were recognised. The sequences of the putative attachments and their positions in C. concisus strain 13826 genome were shown in Table 5-8. In addition, the alignment result between putative attachment sites identified by comparative gene analysis was shown in Table 5-8.

The first attachment site having the sequence of TTC AAA TCC CTC TCT GTC CGC CAC CA was located at -90 to -65 bp from of CON_phi1. This sequence is overlapping the 3-prime terminus of tRNA-Ser. The other attachment sites were also located at -90 to - 65 bp from each first copy of CON_phi2, second copy of CON_phi2 and CON_phi3. The last attachment site was located at +12 and +37 bp from CON_phi3 prophage.

Two identical putative attachment sites were identified around the CON_phi4. This nucleotide sequence was CTC ATA ACC CGA AGG TCG GCG GTT CAA ATC CGT CCT CCG CAA CCA AAT ACC GA. Moreover, this sequence was overlapping with the 3-prime terminus of tRNA-Met. The location was +92 bp and +40 bp from the CON_phi4. Another attachment site was found -992 to -940 bp from CON_phi4. There was a putative phage integrase, which was found at -1013 bp from CON_phi4.

There were two conserved nucleotide sites between seven attachment sites, which were TTC AAA TCC NTC and TCC GCN ACC A. No reverse-complement relationship was found (Table 5-8, the underlined sequence).

Furthermore, the sequences of the putative attachments and their positions in C. concisus strain 13826 were predicted using PHAST and shown in the Table 5-8 with information of an attL or attR. The attachment site AAAATGCCAAAT was at the beginning of the identified CON_phi1. The other attachment sites were within the identified prophage region. Four putative attachment sites identified from PHAST were overlapped with the putative attachment sites as identified above (Table 5-8, the sequence in green)

106

Table 5-8. The predicted attachment sites of CON_phi prophage

Nucleotide Position Attachment gene sequences@ L/R* Manual search# 1582286…1582311 TTCAAATCCCTCTCTGTCCGCCACCA n/a 1587508…1587533 TTCAAATCCCTCTCTGTCCGCCACCA n/a 1597113…1597138 TTCAAATCCCTCTCTGTCCGCCACCA n/a 1606718…1606743 TTCAAATCCCTCTCTGTCCGCCACCA n/a 1616160…1616185 TTCAAATCCCTCTCTGTCCGCCACCA n/a 946941…946993 CTCATAACCCGAAGGTCGGCGGTTCAAAT n/a CCGTCCTCCGCAACCAAATACCGA 937290…937342 CTCATAACCCGAAGGTCGGCGGTTCAAAT n/a CCGTCCTCCGCAACCAAATACCGA PHAST search 1576683..1576694 AAAATGCCAAAT attL 1582301..1582318 GTCCGCCACCAACCCCAA attL 1587523..1587540 GTCCGCCACCAACCCCAA attL 1596357..1596405 GATTATTGCATAATAATTATAAAATAAATC attL TTAAATATTGGAATAATTT 1605962..1606010 GATTATTGCATAATAATTATAAAATAAATC attR TTAAATATTGGAATAATTT 1606733..1606750 GTCCGCCACCAACCCCAA attR 1606733..1606750 GTCCGCCACCAACCCCAA attR 1615438..1615449 AAAATGCCAAAT attR * Attachment left (attL) or attachment right (attR) were determined by PHAST search # The nucleotide sequences between attachment sites determined by manual search were aligned by Clustal Omega. The identically matched sequences are indicated in bold with underline. @ The identically matched nucleotide sequences between the manual search and PHAST search are highlighted in green colour.

107

5.3.1.9. Putative operons

Twelve putative operons were identified. There were three operons in CONphi_1, four operons in CONphi_2, two operons in CONphi_3 and three operons in CONphi_4 (Table 5-9). The zot gene was organised into an operon (Table 5-9).

Table 5-9. Predicted operons in the PHAST identified prophage region of C. concisus strain 13826 genome Prophage Prophage genes organised into operons CON_phi1 Operon 1: CCC13826_0638, CCC13826_2272, and CCC13826_2273 Operon 2: CCC13826_2274 and CCC13826_1101 Operon 3: CCC13826_1102 and CCC13826_2275 CON_phi2 Operon 4: CCC13826_1099, CCC13826_1100, CCC13826_2276 (zot), CCC13826_2277 and CCC13826_0183 (first copy) Operon 5: CCC13826_0164 and CCC13826_2078 CON_phi2 Operon 6: CCC13826_0020, CCC13826_0019, CCC13826_2075 (zot), CCC13826_2074 and CCC13826_1299 (second copy) Operon 7: CCC13826_0185 and CCC13826_0186 CON_phi3 Operon 8: CCC13826_0190, CCC13826_0191 and CCC13826_0192 Operon 9: CCC13826_0198 and CCC13826_0199 CON_phi4 Operon 10: CCC13826_1203 and CCC13826_1204 Operon 11: CCC13826_1208, CCC13826_1209, CCC13826_1210 and CCC13826_1211 Operon 12: CCC13826_1212 and CCC13826_1213 Zot proteins are labelled in bold

108

5.3.1.10. Putative secreted proteins encoded by genes in the prophage region

Seven types of protein were predicted to be secreted proteins by SignalP 4.1 software and seven types of protein were predicted to be secreted proteins by SecretomeP 2.0. The Zot protein was not predicted to be a secreted protein by SignalP 4.1 nor SecretomeP 2.0 (Table 5-10).

SignalP 4.1 revealed that protein encoded by CCC13826_1101, CCC13826_2278, CCC13826_2279, CCC13826_0190, CCC13826_1211 and CCC13826_1206 is secreted in a classical protein secretion pathway, while SecretomeP 2.0 showed that protein encoded by CCC13826_0568, CCC13826_0197, CCC13826_1205 and CCC13826_1208 is secreted in a non-classical protein secretion pathway.

Three protein types encoded by CCC13826_0183, CCC13826_1299, CCC13826_0193 and CCC13826_0196 were predicted to be secreted in both signalling pathway (Table 5- 10).

109

Table 5-10. Predicted secreted proteins encoded by prophage genes in the genome of C. concisus strain 13826

Software Predicted secreted proteins* SignalP 4.1 1. Phosphonate uptake transporter (CCC13826_1101, CCC13826_2278, CCC13826_2279) 2. Putative baseplate wedge component (CCC13826_0183, CCC13826_1299) 3. Putative morphogenesis protein (CCC13826_0190) 4. Hypothetical protein (CCC13826_0193) 5. Hypothetical protein (CCC13826_0196) 6. Putative morphogenesis protein (CCC13826_1211) 7. Hypothetical protein (CCC13826_1206)

SecretomeP 2.0 1. Putative membrane protein (CCC13826_0568) 2. Putative baseplate wedge component (CCC13826_0183, CCC13826_1299) 3. Hypothetical protein (CCC13826_0193) 4. Hypothetical protein (CCC13826_0196) 5. Hypothetical protein (CCC13826_0197) 6. Hypothetical protein (CCC13826_1205) 7. Hypothetical protein (CCC13826_1208) *The proteins predicted by both SignalP 4.1 and SecretomeP 2.0 are in bold

110

5.3.2. Identification of CON_phi2-like prophage and CON_phi3-like prophage in Campylobacter genus other than C. concisus

It was shown that C. concisus strain 13826 contained two major prophages, CON_phi2 and CON_phi3. Given these two C. concisus prophages, it was examined whether the orthologous prophages exist in other Campylobacter species. The zot protein was used, of which DNA sequence is highly conserved (220). UniProt database was used and only Campylobacter spp. were screened. Consequently, eleven Campylobacter strains other than C. concisus strain 13826 were found. These included 1 C. ureolyticus, 2 C. concisus, 2 C. corcagiensis, 3 C. jejuni, 1 C. gracilis, 1 C. hyointestinalis and 1 Campylobacter spp. strain. It was found that all of the detected prophages in the Campylobacter genus began with either tRNA-Ser or tRNA-Met (Table 5-11) except C. concisus UNSWCS and C. jejuni subsp. jejuni 86605, whose nucleotide sequence was unavailable.

These 11 prophages were divided into two groups by protein identity in a comparison to CON_phi2 and CON_phi3 at the amino acid level (Table 5-11, 5-12 and 5-13).

111

Table 5-11. List of tRNA types at upstream of prophage in Campylobacter species

Bacteria tRNA type Locus tag (amino acid) CON_phi2-like prophages C. concisus 13826 Ser CCC13826_2271 C. ureolyticus ACS-301-V-Sch3b Ser HMPREF9309_01752 C. concisus UNSWCS N/A* C. concisus UNSW3 Ser UNSW3_RS00825 C. corcagiensis CIT045 Met BG71_RS0106460

CON_phi3-like prophages C. concisus 13826 Met CCC13826_2250 C. jejuni subsp. doylei 269.97 Ser JJD26997_0343 C. gracilis RM3268 Met CAMGR0001_2931 Campylobacter sp. FOBRC14 Met HMPREF1139_RS06525 C. jejuni subsp. jejuni 60004 Ser CJE11_RS03945 C. jejuni subsp. jejuni 86605 N/A* C. corcagiensis CIT045 Ser BG71_RS0104640 C. hyointestinalis subsp. Met CR67_01850 hyointestinalis DSM 19053 * Only partial shotgun genome sequence was available for C. concisus UNSWCS and C. jejuni subsp. jejuni 86605. Thus, the nucleotide information of tRNA of those bacteria was not known.

112

Four Campylobacter spp. having prophage island similar to CON_phi2 were identified, which included C. ureolyticus ACS-301-V-Sch3b, C. concisus UNSWCS, C. concisus UNSW3 and C. corcagiensis CIT045. Clustal Omega was used to align the protein sequences between the corresponding and CON_phi2 protein. Only protein identity with more than 40 % in all bacterial strains were considered as a conserved protein. Only partial sequences for the C. concisus UNSWCS strain in the NCBI database was available (Table 5-12). Thus, C. concisus UNSWCS strain was excluded from this conservation analysis. Subsequently, six conserved proteins among the species were detected. These 6 proteins were putative phage replication protein A (CCC13826_1099), putative phage maturation protein (CCC13826_1100), Zot (CCC13826_2276), hypothetical protein (CCC13826_2277), hypothetical protein (CCC13826_0183), and hypothetical protein (CCC13826_2278). Interestingly, first five proteins were from the operon 4 described in Table 5-9.

Seven Campylobacter spp. which have prophages similar to CON_phi3 were identified: these included C. jejuni subsp. doylei 269.97, C. gracilis RM3268, Campylobacter sp. FOBRC14, C. jejuni subsp. jejuni 60004, C. jejuni subsp. jejuni 86605, C. corcagiensis CIT045, C. hyointestinalis subsp. hyointestinalis DSM 19053.

In this type of CON_phi3-like prophages, only two conserved proteins were identified among ten phage proteins. These included putative morphogenesis protein (CCC13826_0190) and hypothetical protein (CCC13826_0191). The latter protein (CCC13826_0191) was shown to be Zot-like protein previously (6).

By amino acid comparison, it was observed that there were two proteins were highly conserved between Campylobacter species in both CON_phi2 and CON_phi3 as following:

 Zot protein (CCC13826_2276) in CON_phi2 or Zot-like protein in CON_phi3 (CCC13826_0190).  Putative phage maturation protein (CCC13826_1100) in CON_phi2 or putative morphogenesis protein (CCC13826_0191) in CON_phi3.

113

Table 5-12. Protein identity percentage between CON_phi2 proteins and corresponding phage-like proteins in C. ureolyticus ACS-301- V-Sch3b, C. concisus UNSWCS, C. concisus UNSW3, and C. corcagiensis CIT045

CON_phi2 C. ureolyticus ACS-301-V-Sch3b C. concisus UNSWCS C. concisus UNSW3 C. corcagiensis CIT045 CCC13826_2082 HMPREF9309_00764 71% UNSW3_599 95% BG71_RS0106465 25%

n/c@ BG71_RS0106470 -

CCC13826_1099 HMPREF9309_00765 60% UNSW3_569 83% BG71_RS0106475 62%

CCC13826_1100 HMPREF9309_00766 73% UNSW3_633 97% BG71_RS0106480 66%

CCC13826_2276 HMPREF9309_00767 59% UNSWCS_2123 89% UNSW3_703 89% BG71_RS0106485 56% CCC13826_2277 HMPREF9309_00768 70% UNSWCS_2127 53% Negative1* 97% BG71_RS0106490 52% CCC13826_0183 HMPREF9309_00769 47% UNSWCS_2125 62% UNSW3_712 73% BG71_RS0106495 45% CCC13826_0184 HMPREF9309_00770 9% UNSWCS_2126 26% Negative2* 13% Positive6* 17% n/c HMPREF9309_00771 - UNSWCS_2124 - Negative3* - BG71_RS0106510 - CCC13826_2278 HMPREF9309_00772 67% UNSWCS_2122 93% UNSW3_681 90% BG71_RS0106515 73% n/c Negative4* - BG71_RS0106520 -

CCC13826_0164 HMPREF9309_00773 17% Positive5* 51% BG71_RS0106525 19%

CCC13826_2078 UNSW3_691 75% BG71_RS0106530 13%

Protein identity percentage was estimated by Clustal Omega alignment and any protein identity larger than 40 % percentage compared to CON_phi2 was shaded in grey. @ No corresponding proteins in CON_phi2 are labelled as n/c. * Negative and Positive indicates a missing protein sequence from auto-annotation in the NCBI database. The protein labelled as positive is the missing protein sequence from NCBI database in the positive strand of C. concisus strain 13826 genome, while the negative-labelled protein is in the negative strand of C. concisus strain 13826 genome.

114

Figure 5-7. Schematic view of the potential CON_phi2-like prophage genome structure among the Campylobacter genus. Four Campylobacter species contained the potential CON_phi2-like prophage. The genome structure of CON_phi2 is shown on the top of the figure. CON_phi2 prophage was highly conserved between species and strains based on the protein identity (shown in Table 5-12). The genes in the same colour shared more than 40 % protein identity with genes in CON_phi2. The number below each gene indicates the locus tag ID according to the NCBI database. The missing gene sequences from NCBI database that has an open reading frame are indicated with black stripe without any locus tag.

115

CON_phi3 C. jejuni subsp. C. gracilis Campylobacter C. jejuni subsp. C. jejuni subsp. C. corcagiensis C. hyointestinalis subsp. (CCC13826 doylei 269.97 RM3268 sp. FOBRC14 jejuni 60004 jejuni 86605 CIT045 hyointestinalis DSM ) (JJD26997) (CAMGR0001) (HMPREF1139) (CJE11) (CJE13) (BG71) 19053 (CR67) pseudo 0706 0344 56% 2460 25% 1485 23% 51% RS01165 56% RS0104635 60% 01855 25% integrase 0188 2459 14%

0189 0345 38% 2458 34% 1484 61% RS08070 45% RS01160 45% RS0104630 25% 01860 41% n/c@ 0346 -

0190 0347 43% 2457 45% 1483 74% RS08065 44% RS01155 44% RS0104625 46% 01865 45% 0191 0348 40% 2456 44% 1482 51% RS08060 40% RS01150 40% RS0104620 41% 01870 43% 0192 0349 26% 2455 28% 1481 27% RS08055 25% RS01145 25% RS0104615 25% 01875 37% 0193 0350 24% 2454 20% 1480 27% RS08050 8% RS01140 9% RS0104610 22% 01880 18% n/c Negative9* -

0194 0351 46% 2453 13% 1479 10% Positive7* 8% Negative10* 41%

0195 0352 10% 2452 18% 1478 31% Positive8* 38% Negative11* 36%

n/c RS0104590 - Negative12* -

0196 0353 36% 2451 50% RS0104585 17% Negative13* 38%

0197 0354 47% 2450 28% RS0104580 22% 01900 30%

0198 0355 41% RS0104575 73% 01905 25% 0199 0356 18% RS0104570 27% 01910 14% Protein identity percentage was estimated and any protein with the identity greater than 40 % percentage compared to CON_phi3 was shaded in grey. @ No corresponding proteins in CON_phi3 were recorded as n/c. Prefix of each locus tag is written in the bracket below the strain name. * Negative and Positive indicates a missing protein sequence from auto-annotation in NCBI database. The proteins labelled as positive are the missing protein sequence from NCBI database on the positive strand of C. concisus strain 13826 genome, while proteins labelled as negative are those on the negative strand of C. concisus strain 13826 genome.

116

Figure 5-8. Schematic view of the potential CON_phi3-like prophage genome structure among the Campylobacter genus. Seven Campylobacter species contained potential CON_phi3-like prophage. The genome structure of CON_phi3 is shown on the top of the figure. CON_phi3 prophage was highly conserved between species and strains based on the protein identity (shown in the Table 5-13). The genes in the same colour shared more than 40 % identity in amino acid sequence compared to CON_phi3. The number below each gene indicates the locus tag ID according to the NCBI database. The missing gene sequences from the NCBI database that has an open reading frame are indicated with black stripe without any locus tag.

117

5.3.3. Phylogenetic relationship of Zot and phage maturation protein identified in Campylobacter spp.

Previously it has been shown that the Zot protein and phage maturation protein were highly conserved among the Campylobacter species. Given that the CON_phi prophage exists in multiple hosts within the Campylobacter genus, it is unclear whether the CON_phi-like prophages among Campylobacter spp. were inherited from the same ancestor or horizontally transferred. The 16S rRNA gene tree was compared with the phylogenetic tree based on gene sequences encoded for Zot and phage maturation proteins.

The 16S rRNA phylogenetic tree was generated based on total of eleven gene sequences (Figure 5-9). C. concisus strains and C. jejuni strains were formed a monophyletic group in the 16S rRNA phylogenetic tree, which were well-matched with the one previously reported by the SILVA server (221) except C. ureolyticus strain.

A phylogenetic tree based on the gene sequence encoding for the phage maturation protein of eight Campylobacter species containing the CON_phi3 type prophages and five Campylobacter species containing CON_phi2 type prophages were shown in Figure 5-10A. It showed that the genes encoding for the phage maturation protein were closely related depending on the type of CON_phi prophages. Eight Campylobacter species containing CON_phi3 type prophage were separated from the group of five Campylobacter species containing CON_phi2 type prophage (Figure 5- 10A).

In addition, some Campylobacter species containing CON_phi3 type were in good agreement with the 16S rRNA phylogenetic tree. The phage maturation protein of C. concisus 13826 and Campylobacter sp. FOBRC14 were in one cluster in both phylogenetic trees. Three C. jejuni strains formed one cluster in both 16S rRNA tree and phage maturation protein tree. This may suggest that some CON_phi3 type prophages were inherited from the same ancestor rather than horizontal transfer.

In contrast, the phylogenetic relationship based on the gene sequence of phage maturation protein from CON_phi2 type prophages was different from that based on the 16S rRNA gene in Campylobacter sp., which may confer to horizontal acquisition of the CON_phi2

118

type prophage among those species. Only C. concisus UNSWCS and C. concisus 13826 formed one cluster for both trees.

A phylogenetic tree based on the gene sequence encoding the Zot-like protein from eight Campylobacter species containing the CON_phi3 prophage type and Zot protein from five Campylobacter species containing CON_phi2 prophage type were shown in Figure 5-10B. There was a major division in the tree, splitting into the CON_phi2 cluster and CON_phi3 type cluster, which was the similar arrangement seen in the phylogenetic tree for phage maturation protein (Figure 5-10A).

The phylogenetic relationship based on the gene sequence encoding the Zot-like protein from the CON_phi3 prophage type cluster was in a good agreement with the 16S rRNA phylogenetic relationship (Figure 5-9 and Figure 5-10B). All three C. jejuni strains formed one cluster in both trees. Moreover, C. concisus strain 13826, C. gracilis strain RM3268 and Campylobacter sp. FOBRC14 had a similar relationship in both Zot tree and 16S rRNA tree, which may suggest the possible evidence of inheritance from the same ancestor.

The phylogenetic tree based on Zot gene (Figure 5-10B) was compared with the 16S rRNA phylogenetic tree for both Zot and the Zot-like protein. Only a C. concisus group consisting of C. concisus strain 13826, C. concisus strain UNSWCS and C. concisus strain UNSW3 had a similar relationship in both Zot tree and 16S rRNA tree (Figure 5-9 and Figure 5-10B), which may suggest the horizontal gene transfer.

119

Figure 5-9. Phylogenetic tree of 16S rRNA in Campylobacter genus. The maximum-likelihood phylogenetic tree was constructed using eleven Campylobacter strains. These included C. hyointestinalis subsp. hyointestinalis, C. jejuni subsp. jejuni 86605, C. jejuni subsp. jejuni 6004, C. jejuni subsp. doylei 269.97, C. ureolyticus ACS- 301-V-Sch3b, C. corcagiensis CIT045, C. gracilis RM3268, Campylobacter sp. FOBRC14, C. concisus UNSWCS, C. concisus UNSW3 and C. concisus 13826. The tree shown represents a consensus tree evaluated by 1,000 bootstrap replicates. Branch length values are indicated. All the steps involved in tree construction were conducted in MEGA5.

120

Figure 5-10. The phylogenetic trees based on the gene sequence encoding the phage maturation protein and Zot protein in the Campylobacter genus. Thirteen Campylobacter species were used in these trees. Phage maturation proteins from either CON_phi2 or CON_phi3 type are indicated in brackets. The Zot proteins from CON_phi3 prophage type were named as Zot-like protein. All the gene sequence data was delivered from NCBI database. The maximum-likelihood phylogenetic tree was constructed based on the phage maturation protein (A) or Zot protein (B). The tree shown represents a consensus tree evaluated by 1,000 bootstrap replicates. Branch length values are indicated. All the steps involved in tree construction were conducted in MEGA5

121

5.4. Discussion

In the genome of C. concisus strain 13826, multiple prophage elements were identified, consisting of CON_phi1, CON_phi2, CON_phi3 and CON_phi4. The proteins in each prophage did not show high similarities to a specific phage, suggesting that the prophages identified in the genome of C. concisus strain 13826 may be novel prophages.

All C. concisus prophages identified contained a phage integrase. It was reported that some phages such as Yersinia pestis phage Ypfφ (211), V. cholerae phage VGJφ (212), V. cholerae phage CTXφ (212), and Xanthomonas citri phage φXacF1 (213) use bacterial host recombinases for their integration, which is known as the Xer recombinase system. In this study, it was found that the identified C. concisus prophages carried their own gene encoding for the Xer integrase.

The 3-dimensional (3D) structure of the phage integrase in CON_phi was also predicted. Two phage integrases from CON_phi1 (identical to phage integrase in CON_phi3) and CON_phi2 contained three parallel β-sheets at the C-terminus (Figure 5-5B), which is consistent with the previously known 3D structure of XerD by Subramanya et al. (222) and XerC structure in Escherichia coli (Entry No. P0A8P6) predicted by the SWISS- MODEL (223). In contrast, the phage integrase in CON_phi4 had extra two β-sheets (Figure 5-5B), which suggests that Xer integrase in CON_phi4 is different from the typical XerC or XerD recombinases.

Given this, the Xer types of the phage integrases in the CON_phi were further examined by protein sequences comparison (Table 5-7). In E. coli, XerC or XerD recombinases are known to participate in the separation of circular replicated chromosomal DNA at the specific site named dif, prior to cell division (224) with bacterial DNA translocase FtsK (225). For such separation, E. coli requires both the XerC and XerD proteins. Previous studies showed that a number of bacterial species possess XerC and XerD recombinase system in their genome (226). In this study, the results from protein comparison showed that the phage integrase in CON_phi had low similarities to XerC or XerD types, suggesting that they do not belong to either XerC or XerD type recombinases (Table 5-7).

122

A previous study reported that all species belonging to the ɛ-proteobacteria including C. concisus had only one type of Xer recombinase in their genome called XerH, which is a homologue to the XerD recombinase in E. coli (226). In this study, it was determined that the phage integrases from CON_phi are not of the XerH type either (Table 5-7).

In addition, XerT recombinase was found in a transposable element integrated in the H. pylori genome (227). The protein similarities of the four prophage integrases found in CON_phi and XerT were less than 22 %. Based on the results of protein comparative analysis, the phage integrases in CON_phi prophages are most likely new types of phage integrases and not of the XerC, XerD, XerH, or XerT types.

In addition, the multiple putative attachment sites were found (Table 5-8). This study showed that these putative attachment sites in the CON_phi prophages had no reverse- complement consensus sequences of XerC or XerD binding sites found in other prophages (Figure 5-6), which further supports that CON_phi prophages carried previously not reported integrases.

At this point, it is uncertain whether CON_phi prophages can be induced to release phage particles. From the protein re-annotation, the putative phage assembly protein in CON_phi was found. This protein contained the bacterial type II and III secretion system motif (pfam00263). This type of protein is required for phage extrusion for a filamentous phage (228). For instance, some filamentous phages such as E. coli phage f1, E. coli phage I2-2, E. coli phage IF1 and E. coli phage Ike contained type II secretion system protein D (EpsD)-like homologous protein encoded by phage gene IV (229). However, V. cholerae CTXφ prophage, despite being a filamentous phage, has no gene encoding for secretion, but relies on the host’s secretion protein EpsD, which is a component of the general type II secretion system that opens the outer membrane translocation pore (230).

Moreover, the CON_phi prophage contains the Zot protein. Previously, the Zot protein was discovered in some V. cholerae strains carrying the CTXφ prophage. Zot has a conserved homology domain at the N-terminus with gene product I in the E. coli filamentous phage M13 (231). Gene product I is known to be involved in filamentous phage morphogenesis by formation of channel in the bacterial inner membrane. Indeed, the

123

CTXφ phage virion was not observed in V. cholerae strains with zot gene mutation (101). Therefore, Zot in CON_phi may also contribute to the phage morphogenesis.

Interestingly, the previous study showed that V. cholerae strains containing the CTXφ prophage with Zot toxin were related to the epidemic and pandemic of cholera (102). In addition, Burkholderia pseudomallei strains containing a possible prophage was found in clinical isolates from patients with melioidosis, while the environmental isolates did not possess this possible prophage (232). This possible prophage contained one gene encoding for a putative Zot toxin as well (GI locus tag BPSL3348).

Additionally, the Zot toxin in C. concisus may have clinical importance in initiation of IBD. It was shown that the Zot protein from V. cholerae phage CTXφ increased intestinal permeability (100, 233). Increased intestinal permeability is commonly observed in patients with IBD and is considered as a risk factor to develop the disease (234-237). Moreover, it was reported that 54.4 % of patients with active IBD and 40 % healthy controls had zot-positive C. concisus strains colonising their oral cavity (105). However, C. concisus Zot with polymorphic forms resulting in substitution of valine at position 270 were only found in patients with active IBD (P=0.011, as compared with controls) (105). This suggests that C. concisus Zot with specific polymorphism is associated with active IBD.

Currently, it is not unknown whether CON_phi prophages can be induced or release phage particles. Generally, prophage induction occurs when host bacterial cells are subjected to stress. The human intestinal tract is a less optimal environment for C. concisus as compared with the human oral cavity (46, 238). Given this, C. concisus prophage induction is likely to occur more often in the intestinal tract. Lepage et al. detected a significantly higher number of virus-like particles in intestinal biopsies collected from patients with Crohn’s disease as compared with controls (239). Whether C. concisus prophages contributed to the increased phage particles in the intestinal tract of patients with IBD remains to be investigated.

This study found that four CON_phi prophage elements contained the DicA domain- containing sensory box protein, which is a helix-turn-helix (HTH) type transcriptional regulator. This protein is known to regulate cell division by expressing a division

124

inhibition gene dicB in E. coli at the low temperatures (240). Some bacteriophage also contains phage DNA replication repressors. V. cholerae prophage CTXφ contains the RstR repressor, which contains an HTH homology domain within the DicA protein. The RstR repressor is able to bind to DNA to form tetrameric repressor-operator complexes, preventing the transcription of phage structural proteins (241). Likely, the transcription of CON_phi prophage may be prevented by the DicA domain-containing sensory box protein under normal conditions.

Other phage proteins in the CON_phi prophage that are possibly involved in phage replication are bacteriophage replication gene A protein and putative DNA methylase. Previously, it was shown that bacteriophage replication gene A protein functioned as a DNA endonuclease during phage DNA replication (242), and phage DNA methylase provided immunity to the host restriction enzyme attack (243).

A further interesting finding is that both CON_phi3 and CON_phi4 contained a gene encoding alkyl hydroperoxide reductase (CCC13826_0194 and CCC13826_2249). Alkyl hydroperoxide reductase is an enzyme that reduces peroxides, which have been demonstrated to confer aerotolerance and oxidative stress resistance in microaerophilic bacterial species such as the human enteric pathogen Campylobacter jejuni (244). The fact that CON_phi3 and CON_phi4 carry a gene encoding alkyl hydroperoxide reductase suggests that prophage elements may contribute to the ability of C. concisus to survive in different natural environments such as different parts of the human gastrointestinal tract.

Moreover, it seemed that CON_phi-like prophage in 10 Campylobacter species possessed either tRNA-Met or tRNA-Ser, which was also found in the CON_phi in C. concisus strain 13826. Some phage, mobile elements and pathogenicity islands contained tRNA sequence in their genome as an integration site (245-247). Therefore, the tRNA-Met or tRNA-Ser are likely to be associated with the phage integration.

Based on the comparative analysis in protein sequences (Table 5-12 and Table 5-13), two conserved proteins among CON_phi-like prophage were determined, which were Zot and a putative maturation protein. Both proteins were located in one operon so both proteins are likely to be inserted and transcribed together.

125

Based on the phylogenetic tree of Zot or putative maturation proteins, there was a low similarity with the 16S rRNA phylogenetic tree (Figure 5-9 and Figure 5-10), suggesting that CON_phi prophage is likely to be transferred horizontally rather than inherited from the same ancestor.

In summary, using bioinformatics tools, this study has identified multiple prophages including CON_phi1, CON_phi2, CON_phi3 and CON_phi4. Based on the protein alignments, these prophages were divided into two categories, which were the CON_phi2 type with Zot protein and CON_phi3 type with a Zot-like protein. The phage integrases possessed by every CON_ phi prophage elements were also analysed. These data have laid a foundation for future studies in investigating the possible role of C. concisus and Zot toxin in initiation of IBD.

126

Chapter 6: The release of Zot protein from Campylobacter concisus in the presence of bile

6.1. Introduction

The bile is a yellow-green coloured solution, which is secreted from the liver to the small intestine. The bile contains bile salts, bile acids, phospholipids, cholesterol, pigments and water. Bile salts are bile acids conjugated with either glycine or taurine (248). The four different kinds of bile acids in humans are cholic acid, chenodeoxycholic acid, deoxycholic acid and lithocholic acid (249). The concentration of bile salts in the small intestine ranges from approximately 0.2 to 2 % in human, depending upon the individual and the type and amount of food ingested (249). The primary role of bile salts is to emulsify ingested fats in the intestinal contents. Furthermore, bile salts are bactericidal. Previous studies reported that bile salts enter bacterial cells, cause oxidative stress and damage to DNA in bacteria (250, 251).

In addition, bile salts were shown to be associated with the bacterial pathogenicity. It was reported that bile salts in bacteria stimulated the expression of virulence factors in Campylobacter jejuni (252) and Helicobacter pylori (253). Moreover, bile salts induced prophage release in C. jejuni (254), and Salmonella Typhimurium (255). In Campylobacter concisus, it was found that 2 % bile salts inhibited the growth of 26 out of 58 strains in laboratory culture on horse blood agar plates (251). However, it was not tested whether bile altered the expression of virulence factors in C. concisus.

C. concisus is a commensal bacterium inhabiting in human oral cavity (8, 81). Recently, their association with inflammatory bowel disease (IBD) including Crohn’s disease (CD) and ulcerative colitis (UC) was found (79, 82, 86, 139). However, it is still unclear how oral C. concisus strains exhibits enteric virulence in the intestinal tract.

127

Studies in previous chapters showed that some environmental factors such as hydrogen gas and supplements in culture media may affect the enteric virulence of oral C. concisus strains by affecting their growth or production of H2S. Studies in this chapter will examine whether bile, an environmental factor in the intestine, affect the expression and release of zonula occludens toxins (Zot).

As shown in Chapter 5, the zot gene in C. concisus is a component of prophage CON_phi. (6). The Zot protein produced by Vibrio cholera is known to increase the intestinal permeability (100, 256-258). Mahendran et al. noted that C. concisus strains isolated from IBD patients with active disease have specific polymorphisms (105). They found a significantly higher prevalence of substitutional mutation of valine at the position of 270 Zot in C. concisus strains collected from patients with active IBD was found, as compared with C. concisus strains isolated from the healthy controls (105).

However, it was also not tested whether the expression of Zot in C. concisus was indeed inducible. In this chapter, it was examined whether the Zot protein in C. concisus was inducible by bile. Moreover, it was tested if Zot in C. concisus has harmful effects on intestinal permeability on Caco2 cells.

128

6.2. Materials and methods

6.2.1. Development of a monoclonal antibody targeting C. concisus Zot

The protein sequences of Zot encoded by the gene with locus tag CCC13826_2276 in C. concisus strain 13826 was used for selection of epitopes for monoclonal antibody (mAb) development. The mAb targeting C. concisus Zot was purchased (AbMART, Shanghai, China). This antibody was developed through the surface epitope antibody library technology. The 1-4C381 mAb was provided by this company. The epitope peptides (QEYTLSSQYENG) are located from 68 to 79 amino acid position (12 amino acids in length) of the target protein (total 374 amino acids in length). The specificity of purified antibody was tested using enzyme-linked immunosorbent assay by the manufacturer.

6.2.2. Cloning of zot gene into pETBlue2 E. coli expression system

The zot gene coding for Zot in C. concisus strain 13826 was genetically inserted into pETBlue2 vector (Novagen, Darmstadt, Germany) according to the company manual. The nucleotide sequence of zot in full length was obtained from the NCBI database.

C. concisus strain 13826 (also known as strain BAA-1457) was purchased from American type culture collection (ATCC, VA, US). The commercially available Genomic DNA purification kit (Qiagen, Venlo, Netherlands) was used to isolate the bacterial genomic DNA. This kit contains cell lysis solution, RNase A solution, protein precipitation solution and DNA hydration solution. Briefly, C. concisus strain 13826 was cultured on the blood agar base No.2 (Oxoid, Hampshire, UK) supplemented with 6 % (v/v) defibrinated horse blood (Oxoid) and 10 µg/mL of vancomycin (Sigma-Aldrich, MO, US) at 37 °C for 48 hours under anaerobic condition with hydrogen gas generated by AN25A gas-generation- system (Oxoid) including 0.042 g of sodium borohydride and 10 ml of H2O in a container placed in a 2.5 L incubation jar, which generated 5.0 % (v/v) of H2. The bacteria were harvested and washed with 1 mL of PBS. After removal of the supernatant, the bacterial pellet was resuspended with cell lysis solution, and then treated with RNase A solution at room temperature for two minutes. Protein precipitation solution was then added. The supernatant was removed by centrifugation and isopropanol (100 %) was added into the solution. After centrifuging and removing the supernatant, ethanol (70 %, v/v) was added

129

into the tube. One more centrifugation and removal of the supernatant was performed. DNA hybridisation solution was used to resuspend the pellet. The extracted DNA was stored at -80 °C and used for polymerase chain reaction (PCR) as the DNA template.

The primers were designed using Primer-Blast (259). Forward and reverse primers were designed to have approximately the same melting temperature. In this study, NcoI and XhoI restriction sites were placed at the beginning of the forward primer and reverse primer, respectively. The 5-prime ends of both forward and reverse primer were extended with six bases (CAC ATG and CAA TAA, respectively). The stop codon at the reverse primer was eliminated. The sequence of forward primer was CAC ATG CCA TGG CGA TGC TTA GTT TGA TTA TCG GTC CT and reverse primer was CAA TAA CTC GAG CTT GTG AGT AGG AAA CAT AGA. The pair of primers were synthesised by the manufacturer (Sigma-Aldrich).

To amplify the zot gene in full length, PCR was performed. One PCR reaction (20 µL) contained 2.5 µL of 10X PCR buffer (Thermo Fisher Scientific, MA, US), 2.5 µL of 20 mM dNTP (Thermo Fisher Scientific), 1.5 µL of 25 mM magnesium chloride (Thermo Fisher Scientific), 1 µL of each primer (10 pmol/µL), 2 µL of DNA template (10 pmol/µL), 2 µL of Thermo-Start Taq DNA polymerase (Thermo Fisher Scientific) and nuclease free water (Gibco) to make 25 µL of final volume. The thermal cycling conditions were 94°C for 10 minutes, followed by 35 cycles of 94 °C for 10 seconds, 53 °C for 15 seconds and 72 °C for 1 minute. The size of resulted PCR product was confirmed by agarose gel (Bio-Rad).

The PCR product was sequenced to confirm the gene sequence. Gene sequencing of PCR products was performed using Big Dye terminator chemistry (Applied Biosystems, Foster City, CA) on an ABI Capillary DNA sequencer (Applied Biosystems). Sequencing of both 5 prime and 3 prime ends of the amplicons were performed consisting 1 μl ABI PRISM Big Dye Terminator (Applied Biosystems). The sequencing data were analysed by FinchTV Ver. 1.4.0 (Geospiza, WA, US).

Prior to the enzyme digestion, the PCR product was purified by conventional DNA ethanol purification method. Briefly, ethanol (80 %, v/v) was added into PCR product and gently mixed by flicking. After 30 minute incubation at room temperature, the tube was

130

centrifuged at 16,100 xg for 20 minutes at 4 °C. The supernatant was removed carefully and the tube was incubated in 78 °C water bath for 5 minutes with the cap open. PCR product was dissolved in 10 µL of nuclease free water.

Double-digestion was followed on the PCR product using NcoI R0193T and XhoI R0146S restriction enzymes (New England Biolab, MA, US) at 37 °C for 2 hours. The pETBlue2 (Novagen) was also digested in the same way. The double digested PCR product and vector was ligased by T4 DNA ligase M0202S (New England Biolab) overnight at room temperature. The ligation was checked on the agarose gel.

The recombinant plasmid was transformed into Novablue competent E. coli cell 70181 (Novagen). The transformed cell was grown aerobically at 37 °C overnight on the selective nutrient agar plate (Oxoid) containing 12.5 μg/mL tetracycline (Sigma-Aldrich), 70 μg/mL X-gal (Sigma-Aldrich), 100 μg/mL ampicillin (Sigma-Aldrich) and 1.0 mM IPTG (Sigma- Aldrich). Colony PCR and gene sequencing were followed to confirm the transformation and complete ligation.

The recombinant plasmid containing Zot gene was then extracted using alkaline extraction method (260) with 5 M potassium acetate. The purified recombinant plasmid was transformed into Tuner ™ E. coli cell BL21(DE3) pLacI (Novagen). This recombinant E. coli was grown aerobically at 37 °C on the selective NA media containing 34 µg/ml chloramphenicol (Sigma-Aldrich), 50 µg/ml ampicillin (Sigma-Aldrich) and 1 % (w/v) glucose and stored in a liquid nitrogen tank.

6.2.3. Induction of Zot protein in E. coli

Briefly, the frozen recombinant Tuner E. coli cell prepared above was thawed and cultured overnight on the NA media containing chloramphenicol, ampicillin and glucose. This cultural mixture was used as an inoculum into sterile conical flask containing nutrient broths with antibiotic. The OD600 was monitored every an hour until it reached 0.5. Isopropyl β-D-1-thiogalactopyranoside (IPTG) (Sigma-Aldrich) was added to the flask at the concentration of 1.0 mM and incubated for further 3 hours.

131

Bacteria cells were harvested and lysed by three time-freezing and thaw cycle followed by sonication (89). The detail method was described in the Section 2.2.7.

The whole E. coli lysate was denatured and separated by 12 % SDS-PAGE. The detailed method for SDS-PAGE was described in the Section 2.5. The bacterial proteins were then transferred to the PVDF membranes (Bio-Rad, CA, US) at 100 volts for 1.5 hours for western blot analysis. The PVDF membrane was incubated with blocking solution overnight according to the manufacturer’s manual. The transferring method was described in the previous Section 2.6. The antibodies used in this study included anti penta-His antibody 34660 (Qiagen, Limburg, Netherlands), goat anti-mouse IgG conjugated with horseradish peroxidase SC2031 (Santa Cruz Biotech, Texas, US). Clarity western blot ECL substrate (Bio-Rad) was used. The western blot picture was generated by LAS-3000 imaging system (Fujifilm).

After the confirmation of induction by anti penta-His antibody, the SDS-PAGE and western blot was repeated but using the 1-4C381 mAb.

6.2.4. Detection of secreted Zot protein

Two sets of media containing 50 mL of MEM media 1143-0030 (Gibco) supplemented with 10 mM sodium pyruvate (Invitrogen), 0.15 % sodium bicarbonate (Invitrogen), 1x non-essential amino acids (Invitrogen) and 4 mM glutamine (Invitrogen), and the other media containing extra 0.1 % (w/v) bile bovine B3883 (Sigma-Aldrich). This MEM media was designed based on the media used to culture C. jejuni (261, 262).

C. concisus strain 13826 was cultured on the HBA plates (prepared as described in the general method) for 48 hours at 37 °C. The bacteria was harvested and washed with 1 mL of PBS. The bacterial suspension was adjusted to OD600 2.5. This was used for following cultures as the initial inoculum. One millilitre of this inoculum was added into one set of media placed in a conical flask. The other media set without bacteria was transferred to conical flasks too, which was included as controls. Four flasks were incubated with agitation at 180 rpm for 24 hours at 37 °C under anaerobic condition with 10 % (v/v) hydrogen (2). Hydrogen gas was generated by the reaction of 0.084 g of sodium borohydride with 10 mL of water in a container.

132

After 24 hour incubation, the cultural mixtures from the flasks were divided by centrifugation (5 minutes at 9,000 g) into two parts, supernatant part and bacterial pellet.

The whole bacterial lysate was prepared from the bacterial pellet as described in the general method section of this thesis. This bacterial lysate was stored at -80 °C freezer

The supernatant was filtrated through 0.45 µm filter (Millipore). Protein contents in the filtrate were concentrated using Amicon ultra filter 10 kDa UFC501008 (Millipore). This concentrated filtrate was stored at -80 °C freezer.

The proteins in both the whole bacterial lysate (15 µg) and concentrated filtrate (50 µL) were separated by 12 % SDS-PAGE (as described in the general method). The Zot protein was detected by immunoblotting using the 1-4C381 antibody with 1 in 1000 dilution.

6.2.5. The measurement of mRNA level of zot in C. concisus strain 13826

The mRNA level of zot in C. concisus strain 13826 treated with bile and without bile was measured by real time PCR (RT-PCR).

6.2.5.1. List of primers

All PCR primer pairs were designed using Primer-BLAST (http://www.ncbi.nlm.nih.gov/tools/primer-blast/) (259). The optimal melting temperature was set to 62°C. All the primer was synthesised by Sigma-Aldrich. Three housekeeping genes including rpoA, 16S rRNA and flagellin B were used as reference gene for RT-PCR analysis.

133

Table 6-1. List of primer pairs used in this study

Target Name F/R# Sequence (5 prime to 3 prime end) Product protein * size^ RpoA F_rt_conc_rpoA F GCTCGCTGCCGCTTTTTAAT 82 bp RpoA R_rt_conc_rpoA R ATGCGTGGGATGCTTGAAGA

16S rRNA F_rt_conc_16S F GGTAATGGCTCACCAAGGCT 77 bp 16S rRNA R_rt_conc_16S R GGACCGTGTCTCAGTTCCAG

FlaB F_rt_conc_FlaB F GTCGTGCGTTTAGCGTTTCA 101 bp FlaB R_rt_conc_FlaB R ACCATCAAGGCGAACAAGGT

Zot F_rt_conc_zot2276 F AAGCGTCAGACAAATAGAGAGT 73 bp Zot R_rt_conc_zot2276 R GCCTTTTACCGCTAGGCTGA * RNA polymerase α subunit (RpoA) encoded by CCC13826_1751, flagellin B (FlaB) encoded by CCC13826_2297, zonula occludens toxin (Zot) encoded by CCC13826_2276 and 16S ribosomal RNA (16S rRNA) encoded by CCC13826_2336 # A pair of forward primer (F) and reverse primer (R) ^ Product size is in base pair (bp)

6.2.5.2. Efficiency test of each primer pairs

The PCR efficiency was calculated for four pairs of primers listed in the Table 6-1.

Genomic DNA from C. concisus strain 13826 was extracted using Genomic DNA purification kit (Qiagen). The concentration of the genomic DNA was adjusted by Nanodrop (Thermo Fisher Scientific) to 30 ng/µL. By 1 in 10 serial dilutions, five samples containing 30 ng/µL, 3 ng/µL, 300 pg/µL, 30 pg/µL and 3 pg/µL of genomic DNA were prepared.

The RT-PCR reaction mix was prepared for each gene and for each dilution with the total volume of 60 µL consisting of 1X Sensi-FAST SYBR No-ROX (Bioline), 400 nM forward primer, 400 nM reverse primer and 3 µL of diluted DNA template. Overall, 20 tubes were prepared (four genes x five standard DNA), and then transferred to three clear PCR tubes

134

(Qiagen) in triplicates (total 60 tubes). Rotor Gene 6000 (Qiagen) was used to quantify the PCR. The setting information was shown in the Section 6.2.5.7. No template control was included.

The cycle threshold (Ct) values for each dilution was plotted in linear scale against DNA amount in log scale. From this linear graph, the slope and simple linear regression were calculated by LINEST and RSQ formulas, respectively, using the software named Microsoft Excel 2010 (Microsoft, WA, US). The efficiency for each primer pairs was calculated using the following equation;

[Efficiency]= -1 + 10( -1/[slope] )

6.2.5.3. Total RNA preparation for RT-PCR

C. concisus strain 13826 was cultured on the HBAV plates for 48 hours at 37 °C under anaerobic conditions with 5 % (v/v) hydrogen. The bacteria were harvested and washed with 1 mL of PBS. The OD600 was adjusted to 2.5. One millilitre of this bacterial suspension was inoculated into a conical flask containing 50 mL of MEM with 0.1 % (w/v) bile and into another flask containing 50 mL of MEM without bile respectively, which was shown in the Section 6.2.4. The flasks were incubated under anaerobic condition with hydrogen gas for 24 hours.

After incubations, the bacterial pellets were prepared by centrifugation for 5 minutes at 9,000 g. The pellet was washed with 1 mL of PBS. ISOLATE Ⅱ RNA mini kit (Bioline, London, UK) was used to extract the total RNA. This kit contained a violet filter column, blue silica-membrane column, lysis buffer, membrane desalting buffer, DNase I, washing buffer and nuclease free water. Briefly, the bacterial pellet was resuspended with TE buffer containing lysozyme (Sigma-Aldrich) and incubated for 10 minutes at 37 °C. Lysis buffer with β-mercaptoethanol (Sigma-Aldrich) was added and then filter through the violet filter column. The ethanol (70 %, v/v) was added to the resulted filtrate, which was consequently loaded to the blue silica membrane column. After centrifugation, the silica based membrane was washed with membrane desalting buffer. The working DNase I solution was applied onto the membrane. The column was washed three times with wash

135

buffers. The purified RNA from the silica membrane in the column was eluted by centrifugation with 60 µL of nuclease free water.

6.2.5.4. Removal of bacterial DNA

The purified RNA was treated with DNase I (NEB). DNase I reaction solution (100 µL) was prepared in 1.5 mL eppendorf tubes, which contain 10 µL of 10X DNase I reaction buffer, 1 µL of DNase I, 60 µL of the bacterial total RNA solution, and 29 µL of nuclease free water. The reaction proceeded at 37 °C for 15 minutes. The reaction was terminated by adding 4 µL of ethylenediaminetetraacetic acid (EDTA, 125 mM) and heating for 10 minutes at 70 °C. The tubes were then immediately kept on ice.

6.2.5.5. RNA re-purification

Bacterial total RNA was re-purified by conventional precipitation method using sodium acetate. Ten microliters of 3 M sodium acetate (pH 5.2) and 300 µL of 100 % ice-cold ethanol were added. The mixture was incubated overnight at -20 °C. After centrifugation for 15 minutes at 16,100 xg at 4°C, the supernatant was removed carefully. The RNA pellet was mixed with 100 µL of 70 % (v/v) ethanol. After another centrifugation, the RNA pellet was air-dried for two minutes with the cap open, and then resuspended with 30 µL of nuclease free water. All reagents used in this step were kept in the ice.

6.2.5.6. Synthesis of cDNA

Tetro-cDNA Synthesis Kit (Bioline) was used in this study. Briefly, total RNA (180 ng) was mixed with the cDNA master mix, which contained 1 µL random hexamer, 0.5 mM dNTP, 1× RT buffer, 0.5 unit/µL ribosafe RNase inhibitor, 10 units/µL Tetro reverse transcriptase in a final volume of 20 µL. The solution was incubated at 25 °C for 10 minutes, and then at 45 °C for 30 minutes. The reaction was terminated by incubating at 80 °C for 5 minutes. After the heat inactivation, the tubes were kept at 4 °C.

6.2.5.7. Quantification of RT-PCR

SYBR green dye was used to quantify the double strand DNA product in the RT-PCR reaction mix. RT-PCR reaction master mix with the total volume of 20 µL was prepared

136

with 1X Sensi-FAST SYBR No-ROX, 400 nM forward primer, 400 nM reverse primer. In total, 1.33 μL of the cDNA template were added into 20 µL reaction volume. Rotor gene 6000 (Qiagen) was used for the quantification of mRNA. The thermal cycle for RT-PCR was achieved with following settings;

 1 cycle of hot start at 95°C for 2 minutes  40 cycles of denaturation at 95°C for 10 seconds, annealing at 62°C for 10 seconds, and elongation at 72°C for 10 seconds  Green channel activated for the elongation step for SYBR detection

6.2.5.8. Analysis of RT-PCR

The expression of mRNA level of three rpoA, flaB or 16S rRNA in C. concisus, were first compared between C. concisus cultured in media with and without bile. These three genes were used as housekeeping genes in previous studies (263-265). Whether these three genes can be used as housekeeping genes in assessing the change of mRNA levels C. concisus zot was assessed by comparing the Ct values in C. concisus cultured in media with and without the presence of bile.

The Pfaffl method was used to evaluate RT-PCR result (266). The efficiency for each primer pairs was obtained (referring to Section 6.2.5.2). The relative expression ratio of target gene normalised to reference gene in mRNA level under ‘sample’ condition over ‘control’ condition can be expressed by following two formulas;

Amplification rate Efficiency 1

Amplification rate of target geneCt value of control-Ct value of sample Fold change Amplification rate of reference geneCt value of control-Ct value of sample

6.2.6. Effect of the C. concisus culture supernatant on intestinal epithelial barrier

C. concisus strain P14UCO-S1, an oral C. concisus strain isolated from a patient with active IBD (105), was used in part of the study. The C. concisus strain was cultured in

137

MEM media with 0.1 % (w/v) bile and without bile for 24 hours under anaerobic conditions with 10 % (v/v) hydrogen gas. After 24-hour incubation, the supernatants were collected by centrifugation. The supernatants were filtered through 0.45 μm filter to remove any bacteria. These filtrates with 0.1 % (w/v) bile and without bile were labelled as the culture supernatant with bile and culture supernatant, respectively.

Two MEM media-only controls containing 0.1 % bile and 0 % bile respectively were incubated in the same manner for 24 hours under anaerobic conditions with 10 % hydrogen gas, followed by centrifugation and filtration through 0.45 μm filter. These filtrates with 0.1 % bile and without bile were labelled as media with bile and media-only.

These four supernatants were used as stimuli to the confluent monolayer of Caco2 cells with a stable TEER. Caco2 cells were maintained in the MEM media (Invitrogen) with the supplements as described above (referring to Section 2.3). Cells were harvested at 80 % confluence using 0.25 % trypsin-EDTA (Invitrogen). Caco2 cells were then seeded to the transwell insert (Corning) at a cell density of 5×105 cells/well and grown for 18 days. The formation of monolayer with electrically stable tight junction was monitored by measuring trans forward primer, 400 nM epithelial electrical resistance (TEER) every 5 to 7 days. Media was changed every two days. STX2 electrode (World precision instruments, FL, US) and EVOM2 (World precision instruments) were used.

After 18 days, the media from the apical and basolateral compartment was removed and washed with DPBS (Invitrogen) three times. The both compartments were filled with DPBS and incubated for 30 minutes at 37 °C. Immediately, the TEER was measured, which was considered as time 0. The DPBS from the apical compartment was removed. 100 µL of four different stimuli were added, which were the media-only, media with bile, culture supernatant with bile and culture supernatant. Another apical compartment was filled with 10 % (v/v) dimethyl sulfoxide (DMSO) as a positive control. TEER values were measured after zero (immediate measurement after adding the stimuli), 15 minutes, 30 minutes, one hour, and two hours. The experiment was performed in duplicate.

TEER values (Ω·cm2) across the cell monolayers were calculated by multiplication of area (cm2) of the apical compartment by TEER value subtracted by TEER in a blank well, and

138

then normalised to the initial TEER value prior to the stimulation by the supernatant (Time = 0).

The t test was performed to compare two TEER values between samples and time points.

6.2.7. Molecular weight prediction of Zot protein in C. concisus

The molecular weight of Zot protein encoded by CCC13826_2276 in C. concisus strain 13826 was predicted by Compute pl/Mw tool via ExPASy bioinformatics resource portal (267).

139

6.3. Results

6.3.1. Expression of Zot using E. coli expression system

For the validation of the 1-4C381 mAb targeting Zot protein encoded by CC13826_2276 in C. concisus strain 13826, the E. coli expression system, pETBlue-2, was used. The histidine tag (His-His-His-His-His-His) was added at the C-termini of the recombinant protein. The expression of Zot in E. coli was first confirmed by anti penta-His antibody. The recognition of Zot by antibody 1-4C381 mAb was then assessed.

By using the anti penta-His antibody (Figure 6-1A), the protein band with 45 kDa was detected in the transformed E. coli with the vector having zot gene after IPTG induction, whereas no protein band was observed in the transformed E. coli with the vector only after IPTG induction. Given that the size of C. concisus Zot was predicted to be 43.42 kDa, this protein band with 45 kDa was the induced C. concisus Zot in full length with the histidine tag. In addition, multiple cleaved Zot fragments were found in E. coli.

The same whole E. coli lysate was subjected to the western blot using the 1-4C381 mAb developed in this study. From the transformed E. coli with the vector having C. concisus Zot, the protein band with the size of 45 kDa was observed, while no protein band were appeared in the transformed E. coli with the vector only (Figure 6-1B), which indicated that the 1-4C381 mAb has an specificity to C. concisus Zot protein. No cleaved Zot fragments were detected in E. coli by the 1-4C381mAb.

140

Figure 6-1. Expressed C. concisus Zot in the transformed E. coli was detected by anti penta-His antibody and 1-4C381 mAb. The zot gene (CCC13826_2276) from C. concisus strain 13826 was inserted into pETBlue2 vector with the histidine (His) tag, which was subsequently introduced to the E. coli. The E. coli with intact original vector was included as negative control. After induction with 1.0 M IPTG, the bacterial lysate from the transformed E. coli with pETBlue2-zot vector was loaded into the lane 1, and the negative control was loaded into the lane 2 of the gel. Western blot was performed using anti-penta His antibody (A) or 1- 4C381 mAb (B).

141

6.3.2. Secretion of C. concisus Zot protein

In this study, the secretion of Zot protein in C. concisus by bile was examined (Figure 6-2). C. concisus strain 13826 was cultured in the liquid media supplemented with 0.1% bile for 24 hours. The Zot molecules in the whole bacterial lysates and supernatants separately were detected by western blot using the 1-4C381 mAb.

In the bacterial lysates, (Figure 6-2B), two protein bands with 45 and 17 kDa were observed when C. concisus were grown with no bile, whereas no protein band was detected in bacterial lysates prepared from C. concisus grown with 0.1 % bile.

In the supernatant obtained from the C. concisus culture media containing 0.1% bile (Figure 6-2A), the 1-4C381 mAb detected two protein bands, which had 45 and 17 kDa in size respectively. In contrast, no protein bands were observed in the supernatant prepared from the C. concisus culture media without bile.

Figure 6-2. The Zot protein in C. concisus strain 13826 released into supernatant in the presence of bile. C. concisus strain 13826 was cultured in the liquid media with 0.1 % bile and 0 % bile. After 24 hour incubation, the supernatant and bacteria were separated by centrifugation. In turn, the supernatant (A) and C. concisus bacterial lysates (B) were separated by 12% SDS-PAGE. Zot was detected by Western blot using the 1-4C381 mAb. The black arrows indicate the full length of Zot protein. M: molecular weight marker (113, 79, 47, 34, 27, and 17 kDa from the top).

142

6.3.3. Selection of housekeeping genes that can be used as internal control for examination of the effects of bile on Zot expression in C. concisus by RT-PCR

The RT-PCR efficiency of each primer pair was obtained by the standard curve of Ct value against the number of bacterial DNA copy (Figure 6-3). The linear regression coefficient for all primer pair was greater than 0.9800.

Using the equation mentioned in the materials and method, the efficiencies of primer pairs for 16S rRNA, flgB, rpoA and zot gene were calculated as 0.86, 0.84, 0.88, and 0.80, respectively (Table 6-2).

Figure 6-3. The standard curve of Ct value vs genomic DNA for four genes, 16S rRNA, flgB, rpoA and zot. The standard curve was generated to calculate the efficiency for primer pairs targeting 16S rRNA, rpoA, flgB, and zot. The mean Ct value of each gene are plotted against genomic DNA amount ng/μL in log scale. The simple linear regressions (r2) were analysed for four genes, which were greater than 0.9800.

143

Table 6-2. Efficiency of four genes used in this study

Genes RT-PCR efficiency 16S rRNA 0.86 flgB 0.84 rpoA 0.88 zot 0.80

In addition, genes for flgB, 16S rRNA and rpoA were examined for their potential use as the internal control housekeeping genes in assessing the impact of bile on zot mRNA expression in C. concisus. The Ct values of cDNA of these three genes in C. concisus strain 13826 cultured with and without bile were compared (Table 6-3). Pfaffl method was used to analysis Ct values between two conditions.

The mean Ct value for flgB cDNA in C. concisus grown without bile were 27.78 ± 0.06, whereas the mean Ct value for flgB cDNA in C. concisus grown with bile was 29.17 ± 0.05. The difference in Ct values was 1.42.

The mean Ct value of 16S rRNA cDNA in C. concisus grown without bile was 19.40 ± 0.06, whereas the mean Ct value of 16S rRNA cDNA in C. concisus induced by bile was 22.03 ± 0.21. The difference in Ct values for 16S rRNA was 2.53.

When C. concisus was cultured without bile, the mean Ct value of rpoA cDNA was 28.62 ± 0.06. The C. concisus strain cultured in the presence of 0.1% bile had 29.98 ± 0.11 Ct value for rpoA cDNA. The difference of Ct values for rpoA was 1.36 (Table 6-3). Given that the Ct value of rpoA cDNA changed least when cultured with and without bile among the three genes examined, it was decided that this gene will be used as the internal control housekeeping gene to examine the impact of bile on C. concisus zot expression.

144

Table 6-3. The expression level difference of three candidate housekeeping genes in C. concisus strain 13826 cultured in media with bile versus without bile#

Time Genes Difference* 24h flgB 1.42 24h 16S rRNA 2.53 24h rpoA 1.36 # Three candidate housekeeping genes (flgB, 16S rRNA and rpoA) were tested between C. concisus strain 13826 grown with 0.1% bile and the strain grown without bile. The same amount of mRNA was subjected to cDNA synthesis. * The difference of mean Ct values between C. concisus strain 13826 cultured in media with bile and without bile for each candidate housekeeping gene.

145

6.3.4. The mRNA level of zot in C. concisus strain 13826 cultured in media with and without bile

The expression of mRNA levels for zot in C. concisus strain 13826 cultured for two hours and 24 hours in media with and without the presence of 0.1% bile were examined using RT-PCR. The Ct values for zot cDNA obtained from RT-PCR were normalised to that for rpoA gene (Table 6-4).

After two hours, the relative zot mRNA level in the strain grown with 0.1% bile was 0.4849, whereas that in the strain without bile was 0.5178. The fold increase of zot mRNA in the strain with bile versus without bile was 1.068. When the strain was stimulated for 24 hours by 0.1% bile, the normalised zot mRNA level was increased by 0.8048-fold in comparison with the non-induced bacteria.

Table 6-4. The fold increase of zot mRNA induced by bile in C. concisus strain 13826

Time Genes Mean Ct value Normalised Ct value* Fold increase#

No bile 0.1% Bile No bile 0.1% Bile

2h zot 16.89 16.80 0.4849 0.5178 1.068

rpoA 14.58 14.60

24h zot 21.45 23.28 234.9 189.0 0.8048

rpoA 28.62 28.98

* The normalised Zot was calculated by Pfaffl method described in the materials and methods. In this study, the mRNA level of zot was normalised to that of rpoA gene. # The fold increase was calculated by the normalised Ct value at 0.1% bile divided by that at 0 % bile.

146

6.3.5. The effects of C. concisus Zot released in the presence of bile on intestinal epithelial permeability

The effect of culture supernatant collected from the media where C. concisus strain 13826 had grown for 24 hours on the integrity of tight junction in intestinal epithelial cells was examined by the measurement of TEER (Figure 6-4). In this study, the confluent monolayer of Caco2 cells with stable TEER was used. The culture supernatant was applied for two hours to the 100% confluent monolayer of Caco2 cells. TEER was monitored for two hours.

Caco2 cells incubated with 10 % DMSO were as positive controls (Figure 6-4B). TEER values were reduced to (59.2 ± 4.0) % after one hour incubation, as comparing to the initial TEER value (P=0.0005). After two hour incubation, TEER values decreased further to (55.3 ± 2.2) %.

Caco2 cells were incubated with the media only (Figure 6-4A). It was shown that the TEER values significantly increased to (111.1 ± 1.7) % at 30 minutes as comparing to the initial (P=0.011). In turn, the TEER values decreased down to (83.4 ± 1.9) % after two hours.

It was found that the TEER values after 30 minute incubation with the culture supernatant collected from the media where C. concisus strain had grown for 24 hours were (127.1 ± 0.5) % (in the Figure 6-4A), which were significantly higher as comparing to the initial record (P<0.0005). This relative TEER value was also significantly higher than that with the media only at 30 minutes (P=0.006).

Caco2 cells were incubated with the media with bile (Figure 6-4A). TEER values significantly decreased to (59.0 ± 5.3) % after 30 minutes (P=0.0008). After two hour incubation with the media with bile, TEER values in Caco2 cells decreased to (24.1 ± 0.1) %.

Furthermore, it was examined whether the culture supernatant with bile collected from the media supplemented with 0.1% bile, where C. concisus grew for 24 hours, had impact on TEER value of the monolayer (Figure 6-4A). The relative TEER value after 30 minute

147

incubation was (104.2 ± 8.0) %, which was not significantly higher than the initial TEER value with P value of 0.53, but significantly increased when it was compared with that in the media with bile (P=0.011).

148

A Media only Culture supernatant Media with bile Culture supernatant with bile 140

120

100

The normalised TEER (%) TEER normalised The 20

0 0 0.5 1 1.5 2Time (hour)

B DMSO 120

100

The normalised TEER (%) TEER normalised The 20

0 0 0.5 1 1.5 2 Time (hour) Figure 6-4. The TEER in Caco2 cells treated with the C. concisus culture supernatant. The monolayer of Caco2 cells was incubated with four different stimuli for two hours. These included the media only (♦), the culture supernatant from the media where C. concisus strain P14UCO-S1 grew for 24 hours (Supernatant ◊), the media with 0.1 % bile (▲), and the culture supernatant from the media supplemented with 0.1% bile where the strain was cultured for 24 hours (∆). The TEER values were recorded at time 0, 0.5, 1 and 2 hours and shown in A. 10 % DMSO (■) was included as positive control (shown in B). The normalised TEER value shown indicates the percentage of TEER to the initial TEER value. The error bars indicate the standard deviation out of duplicate.

149

6.4. Discussion

The role of tight junction is to maintain the cell polarity from the apical region and cell integrity between neighbouring cells and to transfer water through the paracellular region (268). Thus, the impaired integrity of tight junction can lead to the penetration of luminal antigens into the deeper intestinal layer, and inflammation. In V. cholerae, Zot toxin was shown to modulate the integrity of tight junction (269), and lead to increase in intestinal paracellular permeability (270). Some C. concisus strains were found to contain the zot gene (105, 271). However, the expression and release of Zot toxin in C. concisus have not been previously studied.

In this study, it was found that C. concisus expressed Zot when cultured in a liquid media. However, without the presence of bile in the cultured media, the Zot toxin expressed in C. concisus was retained in the bacterial lysates, but not released into the supernatant. The bile induced the release of C. concisus Zot into the supernatant. These results suggest that some potential C. concisus virulent proteins may be only released in the intestinal environment, as bile is present in the intestinal tract. These data in part explain why some C. concisus strains may behave as a commensal oral bacterium, but also as an opportunistic enteric pathogen.

The specificity of mAb 1-4C381 targeting C. concisus Zot was validated using E. coli expression system. Only the full-length recombinant C. concisus Zot protein with molecular weight of 45 kDa was clearly observed in E. coli lysate, as compared with the negative control.

In C. concisus, it was found that both the full-length C. concisus Zot and a Zot fragment of about 17 kDa were released by C. concisus in the presence of bile using the same mAb. Previous studies showed that in V. cholera, the N-terminal fragment of Zot with 33 kDa molecular weight remained inside of V. cholerae and a C-terminal fragment (about 12 kDa) was secreted (258). In addition, this cleaved C-terminal fragments of V. cholerae Zot contained AT-1002 active domain (FCIGRL in amino acids). AT-1002 peptide was shown to have an activity increasing intestinal permeability both in vivo and in vitro (272). A previous study showed that C. concisus Zot has a fragment with protein sequences of GRFLSYHG that had similarities to AT-1002 (6). This suggests that the secreted N-

150

terminal fragment of C. concisus Zot toxin may also increase intestinal permeability, which should be investigated in future studies. The mechanisms by which the bile induced the release of Zot toxin by C. concisus are not clear. Bile is known to damage the both Gram-negative and Gram-positive bacterial membrane (273, 274). Given that Zot is a transmembrane protein (unpublished results from a PhD student in our research group), it is possible that bile has induced the release of Zot through damage C. concisus membranes.

In addition to showing that bile induced the release of Zot toxin, this chapter also attempted to investigate whether bile has affected the synthesis of Zot at an mRNA level. In order to quantify the expression of Zot in C. concisus cultured in media with and without bile, an internal housekeeping gene that has the stable expression under these two culture media is needed. Assessment of the expression of three potential internal control genes showed that rpoA was had the most stable expression among the three genes examined. RpoA gene was therefore used as the internal control. The results showed that the bile did not seem to increase the synthesis of Zot in C. concisus under the experimental conditions used in this study.

The zot gene in C. concisus is a component of prophage CON_phi2 (6). In the previous studies, it was shown that bile induced phage synthesis in Salmonella enterica subsp. enterica serovar Typhimurium (255) and C. jejuni (254). Currently, it is unknown whether bile has induced phage production in C. concisus. It would be interesting to further investigate this aspect in future studies.

This chapter also attempted to examine whether Zot released into the C. concisus culture supernatant in the presence of bile affects intestinal epithelial barrier using the Caco2 cell culture model. The supernatant collected from the C. concisus cultured in media without bile, which does not contain Zot protein, slightly increased TEER compared with Caco2 cells treated with culture media alone, showing that without the release of Zot, C. concisus culture supernatant did not damage the intestinal permeability. It was not clear why this C. concisus culture supernatant caused an increase in TEER as compared with media alone. It is possible that some of the proteins released by C. concisus without the presence of bile may have the abilities to increase intestinal permeability, which may be one of the reasons

151

why C. concisus behave as a commensal bacterium in oral cavity where there is no bile present.

Media containing 0.1 % bile caused a rapid decrease of TEER in Caco2 cells, showing the bile damaged the intestinal barrier. Indeed, previous studies showed similar findings (275, 276). Given this, this chapter was unable to answer the question, whether the released Zot induced by bile has affected an intestinal barrier. Future studies should clone and express C. concisus Zot in E. coli. The expressed C. concisus Zot proteins can be purified and used to assess the effects of Zot on intestinal barrier function. Alternatively, a method should be developed to remove bile from C. concisus culture supernatant in order to investigate the effects of bile induced C. concisus proteins, including Zot on intestinal barrier function.

In summary, it was found that in the presence of bile, C. concisus strain 13826 released Zot proteins, both the full length and a cleaved fragment, into the culture supernatant. The bile does not appear to affect Zot expression in C. concisus. Given that 0.1% bile in the media showed damage to the intestinal barrier, it was unable to determine whether the Zot protein released into the C. concisus culture supernatant in the presence of bile has the impact on an intestinal barrier.

152

Chapter 7: Investigation of the possible effects of C. concisus on actin in Caco2 cells

7.1. Introduction

Actin is an abundant protein found in all eukaryotic cells, constituting about 10 % of total cell protein of muscle cells and 1 to 5 % of total cell protein in other cells (277). Actin has three types of isoforms. The α-actin isoforms are expressed primarily in skeletal muscle. The β-actin and γ-actin are ubiquitously expressed. The protein sequences between the isoforms are highly conserved with the amino acid similarity of more than 93% (278). These actin monomers, which are also called globular (G) actin, polymerise to form filamentous (F) actin (279). Actin polymerisation is reversible with dynamic equilibrium between G-actin and F-actin, and this process is regulated by a number of actin-binding proteins (279).

Individual actin filaments are further organised into actin bundles or actin networks. In the intestinal epithelial cells, some actin bundles are participated in the fundamental assembly of tight junction, which is a belt-like structure around each cell attaching neighbouring cells. This cell-to-cell contact is mediated by tight junction transmembrane proteins such as claudins, occludin and JAM. These transmembrane proteins cytosolic tails bind to the Zonula Occludens (ZO) family of proteins, which are anchored to the actin bundles (280). This underlying structure of tight junction provides the intestinal epithelial barrier against thousands of antigens. Some enteric pathogens such as Clostridium difficile, Clostridium perfringens, Bacteroides fragilis, enteropathogenic Escherichia coli (EPEC), and Vibrio cholerae disrupt the structure of tight junction (281). Some pathogens directly induce actin rearrangement to facilitate their invasion to host cells such as C. difficile, B. fragilis, EPEC and Salmonella species (282-288) (281).

153

Campylobacter concisus is a Gram-negative oral bacterium (10). Recent studies suggest that some C. concisus strain may be involved in enteric diseases such as diarrheal disease and inflammatory bowel disease (IBD). Previous studies investigating the enteric pathogenic mechanisms of C. concisus showed that C. concisus infection altered the structure of tight junction in intestinal epithelial cells. In 2010, Man et al. reported that C. concisus strain UNSWCD significantly increased the intestinal permeability in Caco2 cells by measurement of the transepithelial electrical resistance, as compared with the E. coli strain K12 infected control, and that C. concisus strain UNSWCD caused degradation and internalisation of ZO-1 at the cell boundaries (86). In 2011, Nielsen et al. showed that C. concisus infection significantly decreased the protein expression of claudin-5 in HT-29/B6 cells by western blot technique (45). They also showed that there was a significantly higher intestinal permeability of 332 Da fluorescein in HT-29/B5 cells after infection by C. concisus infection, as compared with the control (45). Nonetheless, no study has determined the changes of actin bundles in the host cell by the infection of C. concisus.

In this chapter, to further demonstrate the pathogenic effects of oral C. concisus strains in the tight junction, whether the β-actin expression and F-actin arrangement in intestinal epithelial cells were affected by C. concisus strains was investigated.

154

7.2. Materials and methods

7.2.1. Detection of β-actin and α-tubulin by flow cytometry analysis

The protein levels of β-actin and α-tubulin, another major cytoskeleton protein, in Caco2 cells infected by C. concisus were first examined by using flow cytometry analysis. Inclusion of α-tubulin (289) in this study was to see whether this protein can be used as an internal control for analysis of β-actin using western-blot analysis in later studies.

In this study, a total of five C. concisus strains were used. These included three C. concisus strains P2CDO4, P12CDO-S2, and P4CDO-S1 isolated from patients with CD, one C. concisus strain P3UCO-S1 isolated from patients with UC, and one C. concisus strain H1O1 from healthy individual. Among these, C. concisus strains P2CDO4, P12CDO-S2 and H1O1 were previously characterised to contain the gene encoding for Zonula occludens toxin (Zot) in their genome (105). All C. concisus strains were incubated for 48 hours on HBAV plates under anaerobic conditions with 5 % (v/v) hydrogen gas. Bacterial suspensions with OD600 of 1.0 for each strain were prepared as an inoculum. The methods for C. concisus cultivation and quantification of bacterial numbers were described in the general materials and methods in detail.

The colorectal epithelial cell line Caco2 cells were maintained in T75cm2 tissue culture flasks (Sigma-Aldrich, MO, US) at 37 °C. Caco2 cells were harvested and washed. Detailed methods were described in the general materials and methods (Section 2.3).

Caco2 cells were seeded at 2.0×106 cells/well in 6-well tissue culture flask, followed by incubation for two days prior to co-culture. The monolayer of Caco2 cells was treated with the five C. concisus strains with a multiplicity of infection (MOI) of 25.

After 24 hours of co-incubation, the monolayer was washed five times with DPBS. The Caco2 cells were treated with 0.25 % trypsin (Invitrogen) for 8 minutes. MEM media (10 mL) containing 10 % (v/v) FBS was added into the trypsinised cell suspension. The supernatant was removed by centrifugation at 300 ×g for 5 minutes at 4 °C. The cell pellet was resuspended in 1 mL of PBS, and then transferred two 1.5 mL eppendorf tubes with the equal volume of 500 μL. Both tubes were centrifuged at 300 ×g for 5 minutes at 4 °C.

155

The supernatants were removed. The Caco2 cell pellets in both tubes were treated with three solutions at room temperature, which were 3.6 % (v/v) paraformaldehyde-PBS solution for 15 minutes, 0.1 % (v/v) Triton X-100 (Sigma-Aldrich) for 10 minutes and 1 % (w/v) BSA (Sigma-Aldrich) in PBS for 30 minutes in this order. After each treatment, the supernatant was removed by the centrifugation.

Anti-β-actin antibody SC-47778 (Santa Cruz Biotech) was added into one of two tubes at a 1:40 dilution and incubated for 45 minutes. The treatment of anti-mouse IgG conjugated with Alexa Fluor 647 (Invitrogen) followed, at a 1:400 dilution. Between treatments, the cells was washed twice with 1 mL of 1 % (w/v) BSA solution. The isotype control and non-infected cells were included in this study as negative controls.

In the other tube, anti-α-tubulin antibody (Santa Cruz Biotech) and anti-goat IgG conjugated with Alexa Fluor 488 (Invitrogen) were treated in the same manner.

BD LSRFortessa SORP cell analyser (BD Biosciences, San Jose, US) was used and at least 100,000 cells were counted. The data collected was analysed in Flow Jo software (Tree Star Inc., OR, US).

This experiment was repeated twice.

7.2.2. Detection of β-actin in Caco2 cells by western blot

The protein levels of β-actin in Caco2 cells in response to C. concisus infection were examined by Western blot technique. In this study, five C. concisus strains, P2CDO4, P12CDO-S2, P4CDO-S1, H1O1 and P3UCO-S1 were used in this study.

Caco2 cells were seeded onto 6-well tissue culture plates (Sigma-Aldrich) in MEM media at 2×106 cells/well, followed by further incubation for two days at 37 °C. The monolayer of Caco2 cells were washed with DPBS five times carefully.

Each strain with 25 MOI was added to each chamber of the 6-well tissue culture plates. A no bacteria control was included. The co-culture was incubated for 24 hours.

156

The monolayer was harvested with PBS, washed with PBS and lysed using RIPA buffer (Thermofisher). SDS-PAGE was used to separate the protein and western blot was performed to detect β-actin and α-tubulin. The detailed methods for SDS-PAGE and western blot were described in Section 2.5. Briefly, whole cell proteins of each strain (20 µg) were separated by 12 % SDS-PAGE. The separated proteins were transferred to PVDF membranes at 100 volts for 2.5 hours. The PVDF membrane was blocked with blocking buffer overnight at 4 °C, and then probed with mouse anti-β-actin antibody conjugated with HRP SC-4778HRP (Santa Cruz Biotech). The protein band was detected using a Immuno-StarTM WesterCTM Chemiluminescence Kit (Bio-Rad) and a LAS-3000 imaging system (Fujifilm).

The PVDF membranes were treated with 50 mL of stripping buffer containing 0.1 M β- mercaptoethanol (Sigma-Aldrich), 2 % (w/v) SDS, 62.5 mM Tris-HCl pH 6.8 for 30 minutes at 50 °C with agitation. The PVDF membranes were washed with the wash buffer five times and blocked with the blocking buffer overnight at 4 °C. The PVDF membranes were treated with goat anti-α-tubulin antibody SC-31779 (Santa Cruz Biotech) and bovine anti-goat-IgG conjugated with HRP SC-2353 (Santa Cruz Biotech). The PVDF membranes were washed three times with the wash buffer between each step. The protein bands were visualised in the same manner.

The experiment was repeated in triplicate.

7.2.3. Examination of effects of C. concisus on F-actin arrangement in Caco2 cells using confocal microscopy

In this study, C. concisus strain P2CDO4 was chosen based on the western blot result. The bacterial strain was maintained in the same manner as the previous experiment.

Caco2 cells were maintained in T75cm2 tissue culture flask (Sigma-Aldrich) at 37 °C. For detailed information refer to Section 2.2.1. Two sets of Caco2 cells were prepared on cover slips. For each cover slip, Caco2 cells were seeded at a concentration of 1×105 cells/well in 6-well tissue culture plates (Corning). One of two sets was incubated for two days at 37 °C, while the other one was cultured for ten days.

157

After the incubation, the monolayers of Caco2 cells were washed five times gently without detachment of the monolayer. MEM medium (3 mL) without the antibiotics were added into each well. The monolayers of Caco2 cells were co-cultured with C. concisus strain P2CDO4 at a MOI of 25 for 24 hours.

Without detaching the Caco2 cells, the cover slips were washed carefully with DPBS four times, followed by fixation with 3.6 % (v/v) paraformaldehyde for 15 minutes, permeabilisation with 0.1 % (v/v) Triton X-100 (Sigma-Aldrich) for 10 minutes and then blocking with 1 % (w/v) BSA-PBS solution (Sigma-Aldrich) for 1 hour at room temperature. The F-actin in Caco2 cells was stained with a phalloidin labelled with Alexa Fluor 488 Cat. No. 8878S (Cell signalling technology) for 30 minutes at room temperature. The nucleus was counter-stained with Hoechst 33342 (Invitrogen). The cover slips were washed three times with the blocking solution for each treatment.

The cover slips were mounted with ProLong Gold Antifade Mountant (Invitrogen). Olympus FluoView FV1000 confocal laser scanning microscope (Olympus, Tokyo, Japan) was used. The laser output intensity and detector sensitivity was adjusted to the same level for all confocal laser image acquisition.

7.2.4. Examination of F-actin in the Caco2 cells treated with different concentrations of C. concisus using confocal microscopy

Whether the F-actin alteration in the Caco2 cells induced by C. concisus strain P2CDO4 was dose-dependent was examined. F-actin was visualised by the confocal microscopy as described above. In this study, the Caco2 cells were cultured for ten days prior to bacterial infection. Cells were then incubated with different concentrations of C. concisus strain including MOIs of 0.2, 1, 5 and 125.

7.2.5. Statistical analysis

The intensity of protein bands for β-actin and α-tubulin on the PVDF membrane from the western blot was densitometrically analysed by ImageJ software (National Institute of Mental Health, Maryland, US). The relative β-actin protein levels in Caco2 cells were normalised against the α-tubulin protein level. An unpaired t test was used to compare the

158

intensity of protein band from the result of western blot between samples and non-infected Caco2 control using Excel (Microsoft, WA, US)

159

7.3. Results

7.3.1. Decreased protein level of β-actin in Caco2 cells in the presence of C. concisus detected using flow cytometry

The effect of C. concisus infection on the β-actin and α-tubulin expression in Caco2 cells was examined using flow cytometry. The level of α-tubulin was also assessed to see whether this protein can be used as an internal control for western blot analysis of β-actin in responses to C. concisus infection.

The histogram showed that the peak of intensity of Alexa Fluor 647 dye (β-actin) in Caco2 cells infected by all five C. concisus strains shifted towards left, suggesting reduced protein levels of β-actin in these cells (Figure 7-1), as compared with the non-infected Caco2 cells (grey background with a black solid line).

In addition, it was found that the peak intensity of Alexa Fluor 488 dye (α-tubulin) in Caco2 cells infected by C. concisus strains seemed to be similar to that in the non-infected control (Figure 7-2).

160

Figure 7-1. The flow cytometry analysis of β-actin in Caco2 cells in response to oral C. concisus strains Caco2 cells were infected by five C. concisus strains P2CDO4, P12CDO-S2, P4CDO-S1, H1O1 and P3UCO-S1 for 24 hours and subjected to flow cytometry analysis. The β-actin was labelled with Alexa Fluor 647. Isotype control (purple solid line without background), unstained control (black dot line with grey background) and non-infected control (None, black solid line with grey background) were included.

161

Figure 7-2. The flow cytometry analysis of α-tubulin in Caco2 cells in response to oral C. concisus strains Caco2 cells were infected by five C. concisus strains P2CDO4, P12CDO-S2, P4CDO-S1, H1O1 and P3UCO-S1 for 24 hours and subjected to flow cytometry analysis. The α- tubulin were labelled with Alexa Fluor 488. Isotype control (purple solid line without background), unstained control (black dot line with grey background) and non-infected control (None, black solid line with grey background) were included.

162

7.3.2. Decreased protein level of β-actin in Caco2 cells in the presence of C. concisus detected using western blot

In this study, a western blot technique was used to examine the effect of C. concisus strains on the expression of β-actin in Caco2 cells. One representative western blot of the triplicates is shown in Figure 7-3.

It was shown that the protein level of β-actin in Caco2 cells was decreased when C. concisus was co-incubated, as compared with the non-infected control (Figure 7-3A), while the protein levels of α-tubulin in Caco2 cells infected by C. concisus strains were consistent with that in Caco2 cells without bacterial infection (Figure 7-3A).

Densitometric analysis was conducted to quantify the intensity of each protein bands. The protein level of β-actin in Caco2 cells is normalised to that of α-tubulin, and the ratio of the expression of β-actin with the infection to that without the infection is shown in the bar chart (Figure 7-3B). The statistical analysis showed that the decrease of the relative β-actin level in Caco2 cells was induced by all five C. concisus strains (Figure 7-3B), as compared to the non-infected control.

The lowest relative expression level of β-actin was recorded in Caco2 cells infected by C. concisus strain P2CDO4 (0.31 ± 0.17), which is significantly lower than that in the non- infected control (P value < 0.05). The relative expression of β-actin in Caco2 cells infected by C. concisus strains P12CDO-S2, P4CDO-S1, and H1O1 were 0.53 ± 0.16, 0.62 ± 0.29, and 0.58 ± 0.10 respectively, which was significant (all P values < 0.05). However, it was found that the normalised β-actin level in Caco2 cells infected by C. concisus strain P3UCO-S1 was 0.78 ± 0.26, was not significantly different from that in the non-infected control (P value = 0.21).

163

* * * *

Figure 7-3. Western blot analysis of β-Actin level in Caco2 cells in response to oral C. concisus strains The monolayer of Caco2 cells was infected by C. concisus strains P2CDO4, P12CDO-S2, P4CDO-S1, H1O1 and P3UCO-S1 with 25 MOI for 24 hours. The β-actin levels were determined by Western blot. α-tubulin was included as the internal control. The representative image of triplicate is shown (A). The densitometric quantification (B) of β- actin protein level normalised to the internal control is represented as mean with standard deviation. None indicates non-infected Caco2 cells. Asterisk (*) and NS indicate significant (P < 0.05) and not significant as compared to the non-infected Caco2 cell control, respectively (t test).

164

7.3.3. F-actin alteration induced by C. concisus

In this study, whether the infection of C. concisus strain was able to alter F-actin was examined. C. concisus strain P2CDO4 was chosen, which decreased β-actin at the most based on the result of western blot (Figure 7-3B).

Caco2 cells grown for two and ten days on cover slips were used in this experiment. F- actin in C. concisus infected Caco2 cells appeared more concentrated in the cell-to-cell boundary region as compared with the non-infected cells. Furthermore, F-actin appeared more aggregated in C. concisus infected Caco2 cells (Figure 7-4). These results suggest that C. concisus infection may promote actin polymerisation.

165

Figure 7-4. Effects of C. concisus infection on F-actin in Caco2 cells A monolayer of Caco2 cells was infected by C. concisus strain P2CDO4 for 24 hours and subjected to immune- fluorescence imaging using confocal microscopy. F-actin (red) was labelled with phalloidin conjugated with Alexa Fluor 488, and the nuclei (blue) was counter-stained. Non-infected Caco2 cells were included as a control. The scale bars represent 20 μm.

166

7.3.4. Dose-dependent effects of C. concisus infection on F-actin in Caco2 cells

It was examined whether F-actin in Caco2 cells could be altered by C. concisus strain P2CDO4 in a dose-dependent manner. MOIs of 0 to 125 were examined (Figure 7-5). It was shown that the infection of C. concisus strain P2CDO4 with 0.2 and 1 MOI on Caco2 cells caused no visual changes of F-actin distribution as compared with the non-infected Caco2 cells. F-actin in Caco2 cells with five and 125 MOI of C. concisus formed aggregations (Figure 7-5).

167

Figure 7-5. Differential F-actin morphology in Caco2 cells infected by C. concisus strain P2CDO4 with various MOI. Caco2 cells were cultured for ten days, followed by infection with C. concisus strain P2CDO4 for 24 hours at various MOIs, including 0.2, 1, 5 and 125 MOI. Confocal microscopy was used to visualise F-actin (red, stained with phalloidin-Alexa Fluor 488) and cell nuclei (blue, stained with Hoechst) of Caco2 cells.

168

7.4. Discussion

This chapter examined the impact of C. concisus on β-actin expression level in intestinal epithelial cell line Caco2 cells. Flow cytometry was used first. As the result, it was shown that all five examined C. concisus strains induced a decrease of β-actin in Caco2 cells after 24 hours of infection (Figure 7-1), whereas these C. concisus strains did not reduce the levels of α-tubulin in Caco2 (Figure 7-2). Based on these results, when the effects of C. concisus on actin level in Caco2 cells were examined using Western-blot, α-tubulin was used as the internal control.

Of the five C. concisus strains examined, four strains induced a significantly decrease of actin in Caco2 cells on Western-blot analysis (Figure 7-2). The five C. concisus strains showed consistent results when examined using two different methods. However, on Western-blot analysis, although strain P3UCO-S1 reduced the level of β-actin, the reduction was not statistically significant as compared to the β-actin level in the non- infected Caco2 cells. These results suggest that different C. concisus strains may have varied abilities in reduction of β-actin.

The β-actin isoform in epithelial cells play a role in regulating structure and intestinal permeability. In 2012, Baranwal et al. has found that β-actin in SK-CO15 human colonic epithelial cells is recruited to the region of newly formed epithelial junctions after calcium repletion and the down-regulation of β-actin leads to attenuated reassembly of ZO-1 and F- actin structure (290). In addition, the down-regulation of β-actin enhances the intestinal permeability in SK-CO15 cells and diminishes the apicobasal cell polarity in Caco2 cells (290). In this study, the irregular scaffolding of the tight junction was demonstrated by a dose-dependent C. concisus rearrangement of the F-actin in Caco2 cells (Figure 7-4 and 7- 5). If C. concisus is able to directly reduce the level of β-actin expression, this data can be linked to the previous observation that the C. concisus induces the increased permeability and the reduced amount of tight junction molecules (45, 86).

However, the mechanisms by which C. concisus strains reduced the level of β-actin and F- actin rearrangement in Caco2 cells are unknown at this stage. One possibility is that C. concisus strains have secreted proteinases that have degraded actin in Caco2 cells (291,

169

292). Another possibility is that C. concisus strains have promoted actin polymerisation and therefore reduced the level of actin monomers including β-actin. The F-actin in C. concisus treated Caco2 cells had irregular aggregation at the cell boundaries, showing that C. concisus has affected F-actin arrangement and promoted a focal actin polymerisation. The other possible mechanisms could be the programmed cell death, apoptosis, and Zot secretion.

In the previous study, the reduction of β-actin mRNA was observed during the apoptosis (293). However, the reduction of β-actin level in Caco2 in this study is unlikely to result from the apoptosis as no F-actin purse-string formation, a sign of apoptosis, observed in this study in both non-infected and infected Caco2 cells. To date, there has been no experimental data to demonstrate the apoptotic effects of oral C. concisus in Caco2 cells. In 2011, the previous study by Nielsen et al. showed that prolonged infection by C. concisus in HT-29/B6 cells had apoptotic effects inferred by the impaired paracellular permeability, high proportion of TUNEL-positive cells, activation of caspase-3, and F- actin purse string formation (45). The prolonged infection of C. concisus in Caco2 cells may remain for further investigation.

Furthermore, this β-actin decrease is unlikely to be zot dependent. Zot is known to regulate the tight junction through PKC-dependent F-actin rearrangement (256). Of the five strains examined, three strains P2CDO4, P4CDO-S1 and H1O1 have the zot gene (105). In this chapter, it was found that not only zot-positive C. concisus strains reduced the level of β- actin in Caco2 cells, but also the zot-negative strain P12CDO-S1 exhibited the same effect. These results suggest that other virulence factors in C. concisus strains have contributed to the change of β-actin in Caco2 cells. Although, this cannot rule out the possibility that Zot may affect actin in Caco2 cells. Based on the observations in Chapter 6, without the presence of bile, Zot was not released out from the C. concisus bacteria.

Many bacterial pathogens, such as Salmonella enterica subsp. enterica serovar Typhimurim and Chlamydia trachomatis, produce virulence factors that change the actin arrangement in host cells, which facilitate their intracellular survival or spreading to adjacent cells (294, 295). In intestinal epithelial cells, actin molecules are known to maintain the microvilli and tight junctions. The finding in this study that C. concisus

170

strains affect actin arrangement suggests that C. concisus may interfere with the barrier function of intestinal epithelial cells if colonising the intestinal tract.

In this study, given that both C. concisus strains isolated from patients with IBD and the strain isolated from a healthy control reduced the β-actin level, oral C. concisus strains are not likely to be different between disease groups and healthy group, which were shown in the previous studies (183). Therefore, the significance of C. concisus, affecting intestinal epithelial actin remains to be further investigated in the pathogenesis of IBD in conjunction with other pathogenic properties of individual C. concisus strains. For example, oral C. concisus strains were shown to have different abilities to resistance low pH and bile (251). Therefore, only those strains that are able to arrive at the lower intestinal tract alive can exacerbate the barrier of intestinal epithelial cells. Another virulence factor that would be combined with β-actin reduction in Caco2 cells is invasiveness. Ismail et al. reported that C. concisus strain P2CDO4 was invasive to Caco2 cells with an invasive index of 6.5 ± 1.4, while the other strains were non-invasive (89, 296). This invasive strain may alter the β-actin more effectively and directly in the host cell. Indeed, C. concisus strain P2CDO4 in this study was the most virulent strain decreasing β-actin in Caco2 cells the most as compared with the other strains. In addition, the great genetic diversity between C. concisus strains (183) should be taken into consideration.

In summary, it was demonstrated that the infection of C. concisus strains on Caco2 cells reduced β-actin expression significantly as compared with the non-infected cells. Moreover, it was shown that the altered F-actin organisation in Caco2 cells was induced by C. concisus strains in a dose-dependent manner. However, the detailed molecular mechanisms involved in the F-actin and β-actin change are still unclear. Further experiments will be required to identify key proteins which were involved in alteration of actin and tight junction integrity.

171

General discussion and future directions

General discussion

Campylobacter concisus is a human oral bacterium that has been shown to be associated with inflammatory bowel disease (IBD). Previous studies of genetic relatedness of C. concisus strains isolated from oral and enteric samples of patients with IBD and controls showed that C. concisus strains found in enteric samples of patients with IBD originated from oral C. concisus strains, either from the patient’s own oral C. concisus strains or oral C. concisus strains from other individuals (89).

C. concisus is present in nearly every human individual’s oral cavity, thus should be considered as a commensal oral bacterium. Despite that, the finding that C. concisus is associated with IBD, which suggests that this oral bacterium may contribute to the development of IBD in human. They may have enteric virulence. However, what determines the enteric virulence of C. concisus is not clear. Studies conducted in this PhD project showed that the enteric virulence of C. concisus may be determined by factors from both the human host and the heterogeneity of each individual strain.

Firstly, an individual’s intestinal environmental factors may affect the enteric virulence of

C. concisus. This PhD project found that without the presence of hydrogen gas (H2), C. concisus strains do not grow under microaerobic conditions. The human intestinal tract has microaerobic to anaerobic environments, and the areas close to the epithelial cells are microaerobic due to the diffusion of oxygen from the intestinal tissues (297). Therefore, without H2, C. concisus is unlikely to have the opportunity to invade epithelial cells under microaerobic condition. H2 also affects the production of H2S by C. concisus.

H2 levels in the intestinal tract vary greatly between individuals due to diverse dietary factors and the resident bacterial species. Likely, the individual’s intestinal environments are controlled by the host dietary factors. Both fumarate and formate are used as food additives (198). Indeed, it was shown that the H2S positivity rate significantly increased by

172

the addition of both fumarate and formate, and the growth rate significantly increased by the addition of fumarate.

Secondly, specific virulence genes carried by individual strains may affect their enteric virulence. This PhD project identified a number of prophages in the genome of C. concisus strain 13826. One of the prophages, CON_Phi2 contains the gene encoding the zonula occludens toxin (Zot). It was found that the zot gene was only expressed and released in the presence of bile. The biological activities of C. concisus Zot was investigated by another PhD student in our laboratory, which showed that C. concicus Zot caused prolonged intestinal barrier defect and induced production of cytokines in intestinal epithelial cell line HT-29 cells and macrophages. Our lab previously found that about 30 % of oral C. concisus strains have the zot gene. These results suggest that the C. concisus strains that possess the zot gene may be more pathogenic in the gastrointestinal tract.

Furthermore, the host factors may also have an impact on the expression of some of these virulence factors. For example, in this project, it was found that bile induced the release of Zot from C. concisus. These results show that the virulence of Zot is more likely seen when the bacterium colonises the intestinal tract.

Taking together, results from this PhD project show that C. concisus is an opportunistic enteric pathogen and its enteric virulence is determined by multiple factors such as the properties of individual strains, prophages and host intestinal factors.

Future directions

From this research, future investigations should follow to expand the enteric pathogenic mechanisms in oral C. concisus.

In this study, the possession of multiple CON_phi prophages containing zot gene, the transcription of zot, and the release of Zot protein by bile were found. Despite these findings, the formation of actual virion particles remains to be investigated in the future studies. The plaque assay may be effective. Due to the possession of the gene encoding for Zot protein, CON_phi prophages are more likely to belong to inovirus. The most well-known inovirus

173

species are Escherichia coli phage M13 and Vibrio cholerae phage CTXφ, and both contain a homologue protein to Zot (101). Although inovirus species are secreted without cell lysis of the host, plaques can be seen due to slower-growing cells on faster-growing cells of the bacterial host (298). However, it is uncertain whether CON_phi prophage is able to form a plaque over slow-growing C. concisus lawn. Another useful tool is a transmission electron microscopy.

Moreover, the impacts of C. concisus on the F-actin structure in Caco2 cells were examined. Given that F-actin distribution is affected by the differentiation of Caco2 cells (299), it is suggested that the altered F-actin distribution by C. concisus strains may be incorporated with the differentiation mechanism in Caco2 cells. The future examination of the altered differentiation by C. concisus in Caco2 cells can be linked to the IBD. It was shown that the ileum mucosa biopsy from patients with CD has a significantly reduced mRNA expression of Wnt-signalling pathway transcription factor Tcf-4, regulating Paneth cell differentiation as compared to the healthy controls (300). Moreover, the proportion of mature goblet cell was reported to be significantly lower in patients with CD and UC than the controls (P values were 0.034 and 0.003 respectively) (301).

In this study, the effect of hydrogen gas on the growth and H2S production in oral C. concisus strains were examined. Besides hydrogen gas, the human flatus contains methane and carbon dioxide (302), H2S, methyl mercaptan and dimethyl sulfide (303) in healthy individuals. The examination of whether other intestinal gases under anaerobic conditions have an impact on the growth of C. concisus or not, should be followed to provide fundamental growth information.

Lastly, Caco2 cells were used to observe the pathogenic effects of C. concisus. Caco2 cells are colorectal epithelial adenocarcinoma cells, in which the cellular responses are different from the primary epithelial cells in terms of differentiation, aneuploidy and undefined DNA mutations (304). To obtain more convincing results reflecting the clinical aspect of IBD, an in vivo animal test may be required, which is likely to be more challenging. Currently, a number of animal models were established to study IBD (305, 306). However to date, there is no report isolating C. concisus from faecal and saliva samples in healthy animals, as mentioned in Chapter 1. This indicates that the current animal models for IBD or any healthy animal are not suitable for in vivo colonisation of C. concisus. Moreover, IBD is a chronic inflammatory

174

disorder in humans. To date, the aetiology of IBD is not fully understood, and thought to be caused by a number of factors, including environmental exposures, diets, genetic susceptibility and microbiota (307), which are different factors in animals. Finding a suitable animal model to study the pathogenicity of C. concisus is essential. In addition, choosing C. concisus strains for administration is another consideration due to the great heterogeneity characterised in this thesis and also other papers (89, 105, 183, 251).

175

Appendix 1. The alignment of nucleotide sequences of five phage paralogue proteins used in MLST analysis

Prophage elements 1 No.# CON_phi1 A T C T A C T A G A T A T A A C G A T A A A T T T C A A A T T T T T T T A T A T G T T G G T T T C T 50 CON_phi2 ...... T . CON_phi3 ...... CON_phi4 . G . . . T . T . C G . A . . . A C . G . T G A . A . C T C . G A . A . . . T . T . . T T C . . A C

Prophage elements 1 No. CON_phi1 T T A C T G G C A T G C G G A C T G G T G A G A T A C T A T C T T T A A A G A T G A A A G A T A T T 100 CON_phi2 ...... C ...... T ...... A ...... CON_phi3 ...... CON_phi4 . . C A C . . . . G . . . T . A A A A . . . A G . . T . . A G C C . T . . A T . T . G C . . . . . A

Prophage elements 1 2 No. CON_phi1 G A T T T A G A A A A T G A A G A T G C A A G C G A T G A A AT A A A G A A T T T T G G C A A T A C 150 CON_phi2 ...... C . . A . C . T . . . T A G C . . G ...... G . C ...... G . . . . CON_phi3 ...... CON_phi4 A . C . . T A ...... G ...... C T .

Prophage elements 2 No. CON_phi1 T C T A A A A C A A A A G C A G G T A T T G A A A A T T T G C G A T G A G C T G T T T A A A A T T G 200 CON_phi2 C T ...... G . . . A A . T . . A ...... T . . C . . . . . T ...... G C . . CON_phi3 ...... CON_phi4 C T ...... G . . . A A . T . . A ...... T . . C . . . . . T ...... C . .

Prophage elements 2 No. CON_phi1 A A A C A C G A G T T A A G C C A G T A A G T A A A A T G C G A G A T T A C G A G C T A G T A A A T 250 CON_phi2 . . . A G A A ...... T ...... T . . . CON_phi3 ...... CON_phi4 . . . A G A A ...... T . . .

Prophage elements 2 No. CON_phi1 T A C T A T C A A A G C T T G G G C G G T G A T G T A A A T G T T C A A C A T T T G G C T T A T G T 300 CON_phi2 . . T ...... C ...... CON_phi3 ...... CON_phi4 . . T ...... T ...... C ...... * Five genes encoding for phage integrase (1), bacteriophage replication gene A protein (2), maturation or morphogenesis protein (3), DicA domain-containing sensory box protein, Cro-like protein or replicase (4) and envelope glycoprotein (5) of each CON_phi prophage elements were analysed. Dot indicates that the base is the same as the consensus. # The number indicates the nucleotide position

176

Appendix 1. (Continue)

Prophage elements 2 No. CON_phi1 C G A A A A T T T A A T G C T CG A T A G G G A G C T A G A T A A C T T C A C A C A T T A T C A C A 350 CON_phi2 T ...... A. . . C . A A . C T ...... T ...... G CON_phi3 ...... CON_phi4 T ...... T . . . . . A ...... C . . . . . T . . . . . C ...... G

Prophage elements 2 No. CON_phi1 A A A A G C A C G A T T T A A A T G C C C C T G A T A T T G A T A G T T T T G T A G A T A G A T T T 400 CON_phi2 . . . G ...... C . T ...... C . . . T . . . . C ...... CON_phi3 ...... CON_phi4 ...... A . . C . T . . . . . T ...... C ......

Prophage elements 2 3 No. CON_phi1 T T G A T T T G T A A T G A G T T T T A A G T G G C G A C G T T C T A G T T C T T A G C G G T A T C 450 CON_phi2 ...... A ...... A . A C A A . T T C A T . T C . . A . A G ...... T CON_phi3 ...... CON_phi4 ...... C ...... T ...... C ......

Prophage elements 3 No. CON_phi1 A A C A A A A A A A C T A A T G C T A A G C A A C G T A A C G G C G T G C C A G T T T T A A A A G A 500 CON_phi2 . . T . G C . . . G A A . C A A T . . . A T C . A C A . . G A C T A . T . . . T . . A . T G . . A . CON_phi3 ...... CON_phi4 . . T . . . . . G . . C . C ...... T ......

Prophage elements 3 4 No. CON_phi1 T A T T T G G C T T C T T A A A T G A A C A A A A G A G A A G T A G C C G A A T T C A T C G G A A A 600 CON_phi2 . . . . C C T A . . . . . G G ...... CON_phi3 ...... C T . G G C A . . . . T . . . . A . . T A A A T . A A A T . T CON_phi4 ...... G . A . . . . A . . . G A . . . . A A . T . . A C . T . A . . T

Prophage elements 4 5 No. CON_phi1 A G A T A T A A A A A C T A T C T A T A A C T G G G A A A A A A A T A A C C C A A A C C T A T A A A 650 CON_phi2 ...... CON_phi3 C A C . . G . . . C . . G C . . A C A ...... G . A . . A . . . G . A . . . C . . . CON_phi4 C . . A C . . . G . . . A C . A ...... T C A . G A . . T . . G . . . C . . .

Prophage elements 5 No. CON_phi1 T T T T A T A C T C A C A G C A G G C A G T A A C T G G A A T A T A A G T T T T G A A A A A A A T C 700 CON_phi2 ...... CON_phi3 G A . G . C G G A T T T . T T T A T G . T C T G G A A . . T G G C . . A . . . A T . C . C G . . A A CON_phi4 G A . G . . G G . T T . . T . T A T G . T . T . G A T A . T G G C . . A . . . A T . C . C G . . A A

Prophage elements 5 No. CON_phi1 A C G A A A T T T T A C T T G A T A A T C A 722 CON_phi2 ...... CON_phi3 . T . G . G C . A C . . A A . . G . G C A . CON_phi4 . T . G . G C . A . . . A A . . G . . C A G

177

References

1. van Vliet AH, Ketley JM. 2001. Pathogenesis of enteric Campylobacter infection. Symp Ser Soc Appl Microbiol:45S-56S.

2. Lee H, Ma R, Grimm MC, Riordan SM, Lan R, Zhong L, Raftery M, Zhang L. 2014. Examination of the anaerobic growth of Campylobacter concisus strains. Int J Microbiol 2014:476047.

3. Lastovica AJ, On SLW, Zhang L. 2014. The Family Campylobacteraceae, p 307- 335. In Rosenberg E, DeLong EF, Lory S, Stackebrandt E, Thompson F (ed), The Prokaryotes: Deltaproteobacteria and Epsilonproteobacteria doi:10.1007/978-3- 642-39044-9_274. Springer Berlin Heidelberg, Berlin, Heidelberg.

4. On SL. 2001. Taxonomy of Campylobacter, Arcobacter, Helicobacter and related bacteria: current status, future prospects and immediate concerns. Symp Ser Soc Appl Microbiol:1S-15S.

5. Lawson AJ, On SL, Logan JM, Stanley J. 2001. Campylobacter hominis sp. nov., from the human gastrointestinal tract. Int J Syst Evol Microbiol 51:651-660.

6. Zhang L, Lee H, Grimm MC, Riordan SM, Day AS, Lemberg DA. 2014. Campylobacter concisus and inflammatory bowel disease. World J Gastroenterol 20:1259-1267.

7. Moore JE, Corcoran D, Dooley JS, Fanning S, Lucey B, Matsuda M, McDowell DA, Mégraud F, Millar BC, O’Mahony R. 2005. Campylobacter. Vet Res 36:351-382.

8. Tanner ACR, Badger S, Lai CH, Listgarten MA, Visconti RA, Socransky SS. 1981. 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 31:432-445.

9. Man SM. 2011. The clinical importance of emerging Campylobacter species. Nat Rev Gastroenterol Hepatol 8:669-685.

10. Zhang L. 2015. Oral Campylobacter species: Initiators of a subgroup of inflammatory bowel disease? World J Gastroenterol 21:9239-9244.

11. Johnson CC, Finegold SM. 1987. Uncommonly encountered, motile, anaerobic gram-negative bacilli associated with infection. Rev Infect Dis 9:1150-1162.

12. Long SS, Pickering LK, Prober CG. 2012. Principles and practice of pediatric infectious disease, 3rd ed. Elsevier Health Sciences.

178

13. Macuch P, Tanner A. 2000. Campylobacter species in health, gingivitis, and periodontitis. J Dent Res 79:785-792.

14. Lastovica A, Allos B, Nachamkin I, Szymanski C, Blaser M. 2008. Clinical significance of Campylobacter and related species other than Campylobacteri jejuni and Campylobacter coli. Campylobacter:123-149.

15. Abbott SL, Waddington M, Lindquist D, Ware J, Cheung W, Ely J, Janda JM. 2005. Description of Campylobacter curvus and C. curvus-like strains associated with sporadic episodes of bloody gastroenteritis and Brainerd's diarrhea. J Clin Microbiol 43:585-588.

16. Wetsch NM, Somani K, Tyrrell GJ, Gebhart C, Bailey RJ, Taylor DE. 2006. Campylobacter curvus-associated hepatic abscesses: a case report. J Clin Microbiol 44:1909-1911.

17. Koga M, Yuki N, Takahashi M, Saito K, Hirata K. 1999. Are Campylobacter curvus and Campylobacter upsaliensis antecedent infectious agents in Guillain– Barré and Fisher’s syndromes? J Neurol Sci 163:53-57.

18. Shinha T. 2015. Fatal bacteremia caused by Campylobacter gracilis, United States. Emerg Infect Dis 21:1084-1085.

19. Suzuki J, Sugiyama T, Ito K, Hadano Y, Kawamura I, Okinaka K, Kurai H, Ohkusu K, Ohmagari N. 2013. Campylobacter showae bacteremia with cholangitis. J Infect Chemother 19:960-963.

20. Quinn PJ, Markey BK, Leonard FC, FitzPatrick ES, Fanning S, Hartigan P. 2011. Campylobacter and Helicobacter species, Vet Microbiol Microb D, 2nd ed. Wiley-Blackwell.

21. Koziel M, Lucid A, Bullman S, Corcoran GD, Lucey B, Sleator RD. 2014. Draft genome sequence of Campylobacter corcagiensis strain CIT045T, a representative of a novel Campylobacter species isolated from lion-tailed macaques (Macaca silenus). Genome Announc 2:e00248-00214.

22. Zanoni RG, Debruyne L, Rossi M, Revez J, Vandamme P. 2009. Campylobacter cuniculorum sp. nov., from rabbits. Int J Syst Evol Microbiol 59:1666-1671.

23. Grogono-Thomas R, Dworkin J, Blaser M, Newell D. 2000. Roles of the surface layer proteins of Campylobacter fetus subsp. fetus in ovine abortion. Infect Immun 68:1687-1691.

24. Hum S, Stephens L, Quinn C. 1991. Diagnosis by ELISA of bovine abortion due to Campylobacter fetus. Aust Vet J 68:272-275.

179

25. Hum S, Hornitzky M, Berg T. 2009. Bovine genital Campylobacteriosis. Australia and New Zealand standard diagnostic procedures, SCAHLS, ed. (Department of Agriculture, Fisheries and Forestry).

26. Patrick ME, Gilbert MJ, Blaser MJ, Tauxe RV, Wagenaar JA, Fitzgerald C. 2013. Human infections with new subspecies of Campylobacter fetus. Emerg Infect Dis 19:1678-1680.

27. Hänninen ML, Sarelli L, Sukura A, On SL, Harrington C, Matero P, Hirvelä- Koski V. 2002. Campylobacter hyointestinalis subsp. hyointestinalis, a common Campylobacter species in reindeer. J Appl Microbiol 92:717-723.

28. On SL, Bloch B, Holmes B, Hoste B, Vandamme P. 1995. Campylobacter hyointestinalis subsp. lawsonii subsp. nov., isolated from the porcine stomach, and an emended description of Campylobacter hyointestinalis. Int J Syst Bacteriol 45:767-774.

29. Chua K, Gürtler V, Montgomery J, Fraenkel M, Mayall BC, Grayson ML. 2007. Campylobacter insulaenigrae causing septicaemia and enteritis. J Med Microbiol 56:1565-1567.

30. Ketley JM, Konkel ME. 2005. Campylobacter: molecular and cellular biology. Horizon Scientific Press.

31. Hermans D, Pasmans F, Messens W, Martel A, Van Immerseel F, Rasschaert G, Heyndrickx M, Van Deun K, Haesebrouck F. 2012. Poultry as a host for the zoonotic pathogen Campylobacter jejuni. Vector Borne Zoonotic Dis 12:89-98.

32. Weis AM, Miller WA, Byrne BA, Chouicha N, Boyce WM, Townsend AK. 2014. Prevalence and pathogenic potential of Campylobacter isolates from free- living, human-commensal american crows. Appl Environ Microbiol 80:1639-1644.

33. Schweitzer N, Damjanova I, Kaszanyitzky É, Ursu K, Samu P, Tóth ÁG, Varga J, Dán Á. 2011. Molecular characterization of Campylobacter lanienae strains isolated from food-producing animals. Foodborne Pathog Dis 8:615-621.

34. Debruyne L, On SL, De Brandt E, Vandamme P. 2009. Novel Campylobacter lari-like bacteria from humans and molluscs: description of Campylobacter peloridis sp. nov., Campylobacter lari subsp. concheus subsp. nov. and Campylobacter lari subsp. lari subsp. nov. Int J Syst Evol Microbiol 59:1126- 1132.

35. Gurgan T, Diker KS. 1994. Abortion associated with Campylobacter upsaliensis. J Clin Microbiol 32:3093-3094.

36. Ketley JM. 1997. Pathogenesis of enteric infection by Campylobacter. Microbiol 143 ( Pt 1):5-21.

180

37. Blaser MJ, Taylor DN, Feldman RA. 1983. Epidemiology of Campylobacter jejuni infections. Epidemiol Rev 5:157-176.

38. Rees JH, Soudain SE, Gregson NA, Hughes RA. 1995. Campylobacter jejuni infection and Guillain–Barré syndrome. N Engl J Med 333:1374-1379.

39. Kuroki S, Saida T, Nukina M, Haruta T, Yoshioka M, Kobayashi Y, Nakanishi H. 1993. Campylobacter jejuni strains from patients with Guillain- Barré syndrome belong mostly to penner serogrpup 19 and contain β-N- acetylglucosamine residues. Annals of neurology 33:243-247.

40. Nachamkin I, Allos BM, Ho T. 1998. Campylobacter species and Guillain-Barre syndrome. Clini Microbiol Rev 11:555-567.

41. Tanner AC, Dzink JL, Ebersole JL, Socransky SS. 1987. Wolinella recta, Campylobacter concisus, Bacteroides gracilis, and Eikenella corrodens from periodontal lesions. J Periodontal Res 22:327-330.

42. Kamma JJ, Diamanti-Kipioti A, Nakou M, Mitsis FJ. 2000. Profile of subgingival microbiota in children with mixed dentition. Oral Microbiol Immunol 15:103-111.

43. Zhang L, Budiman V, Day AS, Mitchell H, Lemberg DA, Riordan SM, Grimm M, Leach ST, Ismail Y. 2010. Isolation and detection of Campylobacter concisus from saliva of healthy individuals and patients with inflammatory bowel disease. J Clin Microbiol 48:2965-2967.

44. Petersen RF, Harrington CS, Kortegaard HE, On S. 2007. A PCR-DGGE method for detection and identification of Campylobacter, Helicobacter, Arcobacter and related Epsilobacteria and its application to saliva samples from humans and domestic pets. J Appl Microbiol 103:2601-2615.

45. Nielsen HL, Nielsen H, Ejlertsen T, Engberg J, Gunzel D, Zeitz M, Hering NA, Fromm M, Schulzke JD, Bucker R. 2011. Oral and fecal Campylobacter concisus strains perturb barrier function by apoptosis induction in HT-29/B6 intestinal epithelial cells. PLoS One 6:e23858.

46. Engberg J, On SL, Harrington CS, Gerner-Smidt P. 2000. 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 38:286-291.

47. Nielsen HL, Ejlertsen T, Engberg J, Nielsen H. 2013. High incidence of Campylobacter concisus in gastroenteritis in North Jutland, Denmark: a population-based study. Clin Microbiol Infect 19:445-450.

48. Chaban B, Musil KM, Himsworth CG, Hill JE. 2009. Development of cpn60- based real-time quantitative PCR assays for the detection of 14 Campylobacter

181

species and application to screening of canine fecal samples. Appl Environ Microbiol 75:3055-3061.

49. Chaban B, Ngeleka M, Hill JE. 2010. Detection and quantification of 14 Campylobacter species in pet dogs reveals an increase in species richness in feces of diarrheic animals. BMC Microbiol 10:73.

50. Lynch OA, Cagney C, McDowell DA, Duffy G. 2011. Occurrence of fastidious Campylobacter spp. in fresh meat and poultry using an adapted cultural protocol. Int J Food Microbiol 150:171-177.

51. Moore WE, Holdeman LV, Smibert RM, Good IJ, Burmeister JA, Palcanis KG, Ranney RR. 1982. Bacteriology of experimental gingivitis in young adult humans. Infect Immun 38:651-667.

52. Tyrrell KL, Citron DM, Warren YA, Nachnani S, Goldstein EJ. 2003. Anaerobic bacteria cultured from the tongue dorsum of subjects with oral malodor. Anaerobe 9:243-246.

53. Moore LV, Moore WE, Cato EP, Smibert RM, Burmeister JA, Best AM, Ranney RR. 1987. Bacteriology of human gingivitis. J Dent Res 66:989-995.

54. Lastovica A, Le Roux E, Warren R, Klump H. 1993. Clinical isolates of Campylobacter mucosalis. J Clin Microbiol 31:2835-2836.

55. Lauwers SD, T.;Van Etterijck, R.;Breynaert, J.;Van Zeebroeck, A.;Smekens, L.;Kersters, K.;Vandamme, P. 1991. Isolation of Campylobacter concisus from human faeces. Microb Ecol Health D 4.

56. Van Etterijck R, Breynaert J, Revets H, Devreker T, Vandenplas Y, Vandamme P, Lauwers S. 1996. Isolation of Campylobacter concisus from feces of children with and without diarrhea. J Clin Microbiol 34:2304-2306.

57. Zhi N, Revets H, Van Zeebroek A, Lauwers S. 1996. Serological Response to Campylobacter concisus Infection, p 673-678. In Newell D, Ketley J, Feldman R (ed), Campylobacters, Helicobacters, and Related Organisms doi:10.1007/978-1- 4757-9558-5_127. Springer US.

58. Sartor RB. 2008. Microbial influences in inflammatory bowel diseases. Gastroenterol 134:577-594.

59. Loftus EV, Jr. 2004. Clinical epidemiology of inflammatory bowel disease: Incidence, prevalence, and environmental influences. Gastroenterol 126:1504- 1517.

60. Logan RF. 1998. Inflammatory bowel disease incidence: up, down or unchanged? Gut 42:309-311.

182

61. Hendrickson BA, Gokhale R, Cho JH. 2002. Clinical aspects and pathophysiology of inflammatory bowel disease. Clin Microbiol Rev 15:79-94.

62. Strober W, Fuss I, Mannon P. 2007. The fundamental basis of inflammatory bowel disease. J Clin Invest 117:514-521.

63. M.Yates V, Watkinson G, Kelman A. 1982. Further evidence for an association between psoriasis, Crohn's disease and ulcerative colitis. Brit J Dermatol 106:323- 330.

64. Di MA. 2000. Therapy of digestive disorders: A companion to sleisenger and Fordtran's gastrointestinal and liver disease. Gastroenterol 118:1275-1276.

65. Munkholm P. 2003. Review article: the incidence and prevalence of colorectal cancer in inflammatory bowel disease. Aliment Pharmacol Ther 18 Suppl 2:1-5.

66. Solem CA, Harmsen WS, Zinsmeister AR, Loftus EV, Jr. 2004. Small intestinal adenocarcinoma in Crohn's disease: a case-control study. Inflamm Bowel Dis 10:32-35.

67. Mombaerts P, Mizoguchi E, Grusby MJ, Glimcher LH, Bhan AK, Tonegawa S. 1993. Spontaneous development of inflammatory bowel disease in T cell receptor mutant mice. Cell 75:274-282.

68. Shull MM, Ormsby I, Kier AB, Pawlowski S, Diebold RJ, Yin M, Allen R, Sidman C, Proetzel G, Calvin D, et al. 1992. Targeted disruption of the mouse transforming growth factor-beta 1 gene results in multifocal inflammatory disease. Nat 359:693-699.

69. Kuhn R, Lohler J, Rennick D, Rajewsky K, Muller W. 1993. Interleukin-10- deficient mice develop chronic enterocolitis. Cell 75:263-274.

70. Sadlack B, Merz H, Schorle H, Schimpl A, Feller AC, Horak I. 1993. Ulcerative colitis-like disease in mice with a disrupted interleukin-2 gene. Cell 75:253-261.

71. Dianda L, Hanby AM, Wright NA, Sebesteny A, Hayday AC, Owen MJ. 1997. T cell receptor-alpha beta-deficient mice fail to develop colitis in the absence of a microbial environment. Am J Pathol 150:91-97.

72. Sellon RK, Tonkonogy S, Schultz M, Dieleman LA, Grenther W, Balish E, Rennick DM, Sartor RB. 1998. Resident enteric bacteria are necessary for development of spontaneous colitis and immune system activation in interleukin- 10-deficient mice. Infect Immun 66:5224-5231.

73. Engle SJ, Ormsby I, Pawlowski S, Boivin GP, Croft J, Balish E, Doetschman T. 2002. Elimination of colon cancer in germ-free transforming growth factor beta 1-deficient mice. Cancer Res 62:6362-6366.

183

74. Rutgeerts P, Goboes K, Peeters M, Hiele M, Penninckx F, Aerts R, Kerremans R, Vantrappen G. 1991. Effect of faecal stream diversion on recurrence of Crohn's disease in the neoterminal ileum. Lancet 338:771-774.

75. Jonkers D, Stockbrügger R. 2003. Probiotics and inflammatory bowel disease. J R Soc Med 96:167-171.

76. Prantera C, Lochs H, Campieri M, Scribano ML, Sturniolo GC, Castiglione F, Cottone M. 2006. Antibiotic treatment of Crohn's disease: results of a multicentre, double blind, randomized, placebo-controlled trial with rifaximin. Aliment Pharmacol Ther 23:1117-1125.

77. Dejaco C, Harrer M, Waldhoer T, Miehsler W, Vogelsang H, Reinisch W. 2003. Antibiotics and azathioprine for the treatment of perianal fistulas in Crohn's disease. Aliment Pharmacol Ther 18:1113-1120.

78. Zhang L, Man SM, Day AS, Leach ST, Lemberg DA, Dutt S, Stormon M, Otley A, O'Loughlin EV, Magoffin A, Ng PH, Mitchell H. 2009. Detection and isolation of Campylobacter species other than C. jejuni from children with Crohn's disease. J Clin Microbiol 47:453-455.

79. Man SM, Zhang L, Day AS, Leach ST, Lemberg DA, Mitchell H. 2010. Campylobacter concisus and other Campylobacter species in children with newly diagnosed Crohn's disease. Inflamm Bowel Dis 16:1008-1016.

80. Hansen R, Mukhopadhya I, Russell R, Bisset W, Berry S, Thomson J, El- Omar E, Hold G. 2011. The role of microaerophilic colonic mucosal bacteria in de novo paediatric inflammatory bowel disease. Gut 60:A147-A147.

81. Mahendran V, Riordan SM, Grimm MC, Tran TA, Major J, Kaakoush NO, Mitchell H, Zhang L. 2011. Prevalence of Campylobacter species in adult Crohn's disease and the preferential colonization sites of Campylobacter species in the human intestine. PLoS One 6:e25417.

82. Mukhopadhya I, Thomson JM, Hansen R, Berry SH, El-Omar EM, Hold GL. 2011. Detection of Campylobacter concisus and other Campylobacter species in colonic biopsies from adults with ulcerative colitis. PLoS One 6:e21490.

83. Guslandi M, Zhang L, Man SM, Day AS, Leach ST, Lemberg DA, Dutt S, Stormon M, Otley A, O'Loughlin EV, Magoffin A, Ng PH, Mitchell H. 2009. Campylobacter concisus: a new character in the Crohn's disease story? J Clin Microbiol 47:1614-1615.

84. Lavrencic P, Kaakoush NO, Huinao KD, Kain N, Mitchell HM. 2012. Investigation of motility and biofilm formation by intestinal Campylobacter concisus strains. Gut Pathog 4:22.

184

85. Shigematsu M, Umeda A, Fujimoto S, Amako K. 1998. Spirochaete-like swimming mode of Campylobacter jejuni in a viscous environment. J Med Microbiol 47:521-526.

86. Man SM, Kaakoush NO, Leach ST, Nahidi L, Lu HK, Norman J, Day AS, Zhang L, Mitchell HM. 2010. Host attachment, invasion, and stimulation of proinflammatory cytokines by Campylobacter concisus and other non- Campylobacter jejuni Campylobacter species. J Infect Dis 202:1855-1865.

87. Kaakoush NO, Deshpande NP, Wilkins MR, Tan CG, Burgos-Portugal JA, Raftery MJ, Day AS, Lemberg DA, Mitchell H. 2011. The pathogenic potential of Campylobacter concisus strains associated with chronic intestinal diseases. PLoS One 6:e29045.

88. Kalischuk LD, Inglis GD. 2011. Comparative genotypic and pathogenic examination of Campylobacter concisus isolates from diarrheic and non-diarrheic humans. BMC Microbiol 11:53.

89. Ismail Y, Mahendran V, Octavia S, Day AS, Riordan SM, Grimm MC, Lan R, Lemberg D, Tran TA, Zhang L. 2012. Investigation of the enteric pathogenic potential of oral Campylobacter concisus strains isolated from patients with inflammatory bowel disease. PLoS One 7:e38217.

90. Songer JG. 1997. Bacterial phospholipases and their role in virulence. Trends Microbiol 5:156-161.

91. Istivan TS, Coloe PJ, Fry BN, Ward P, Smith SC. 2004. Characterization of a haemolytic phospholipase A(2) activity in clinical isolates of Campylobacter concisus. J Med Microbiol 53:483-493.

92. Istivan TS, Smith SC, Fry BN, Coloe PJ. 2008. Characterization of Campylobacter concisus hemolysins. FEMS Immunol Med Microbiol 54:224-235.

93. Fagan RP, Fairweather NF. 2014. Biogenesis and functions of bacterial S-layers. Nat Rev Microbiol 12:211-222.

94. Thompson SA. 2002. Campylobacter surface-layers (S-layers) and immune evasion. Ann Periodontol 7:43-53.

95. Ausiello CM, Cerquetti M, Fedele G, Spensieri F, Palazzo R, Nasso M, Frezza S, Mastrantonio P. 2006. Surface layer proteins from Clostridium difficile induce inflammatory and regulatory cytokines in human monocytes and dendritic cells. Microbes Infect 8:2640-2646.

96. Gillespie MJ, Smutko J, Haraszthy GG, Zambon JJ. 1993. Isolation and partial characterization of the Campylobacter rectus cytotoxin. Microb Pathog 14:203- 215.

185

97. Miyamoto M, Maeda H, Kitanaka M, Kokeguchi S, Takashiba S, Murayama Y. 1998. The S-layer protein from Campylobacter rectus: sequence determination and function of the recombinant protein. FEMS Microbiol Lett 166:275-281.

98. Linhartova I, Bumba L, Masin J, Basler M, Osicka R, Kamanova J, Prochazkova K, Adkins I, Hejnova-Holubova J, Sadilkova L, Morova J, Sebo P. 2010. RTX proteins: a highly diverse family secreted by a common mechanism. FEMS Microbiol Rev 34:1076-1112.

99. Kaakoush NO, Man SM, Lamb S, Raftery MJ, Wilkins MR, Kovach Z, Mitchell H. 2010. The secretome of Campylobacter concisus. FEBS J 277:1606- 1617.

100. Fasano A, Baudry B, Pumplin DW, Wasserman SS, Tall BD, Ketley JM, Kaper JB. 1991. Vibrio cholerae produces a second enterotoxin, which affects intestinal tight junctions. Proc Natl Acad Sci U S A 88:5242-5246.

101. Waldor MK, Mekalanos JJ. 1996. Lysogenic conversion by a filamentous phage encoding cholera toxin. Science 272:1910-1914.

102. Faruque SM, Mekalanos JJ. 2012. Phage-bacterial interactions in the evolution of toxigenic Vibrio cholerae. Virulence 3:556-565.

103. Wang W, Uzzau S, Goldblum SE, Fasano A. 2000. Human zonulin, a potential modulator of intestinal tight junctions. J Cell Sci 113 Pt 24:4435-4440.

104. Fasano A. 2008. Physiological, pathological, and therapeutic implications of zonulin-mediated intestinal barrier modulation: living life on the edge of the wall. Am J Pathol 173:1243-1252.

105. Mahendran V, Tan YS, Riordan SM, Grimm MC, Day AS, Lemberg DA, Octavia S, Lan R, Zhang L. 2013. The prevalence and polymorphisms of zonula occluden toxin gene in multiple Campylobacter concisus strains isolated from saliva of patients with inflammatory bowel disease and controls. PLoS One 8:e75525.

106. On SL. 1994. Confirmation of human Campylobacter concisus isolates misidentified as Campylobacter mucosalis and suggestions for improved differentiation between the two species. J Clin Microbiol 32:2305-2306.

107. Gibson GR, Macfarlane GT, Cummings JH. 1993. Sulphate reducing bacteria and hydrogen metabolism in the human large intestine. Gut 34:437-439.

108. Deplancke B, Finster K, Graham WV, Collier CT, Thurmond JE, Gaskins HR. 2003. Gastrointestinal and microbial responses to sulfate-supplemented drinking water in mice. Exp Biol Med (Maywood) 228:424-433.

109. Florin T, Neale G, Gibson GR, Christl SU, Cummings JH. 1991. Metabolism of dietary sulphate: absorption and excretion in humans. Gut 32:766-773.

186

110. Macfarlane G, Gibson G, Cummings J. 1992. Comparison of fermentation reactions in different regions of the human colon. J Appl Microbiol 72:57-64.

111. Gibson GR, Cummings JH, Macfarlane GT. 1991. Growth and activities of sulphate-reducing bacteria in gut contents of healthy subjects and patients with ulcerative colitis, vol 86.

112. Levine J, Ellis CJ, Furne JK, Springfield J, Levitt MD. 1998. Fecal hydrogen sulfide production in ulcerative colitis. Am J Gastroenterol 93:83-87.

113. Mills DJ, Tuohy KM, Booth J, Buck M, Crabbe MJ, Gibson GR, Ames JM. 2008. Dietary glycated protein modulates the colonic microbiota towards a more detrimental composition in ulcerative colitis patients and non-ulcerative colitis subjects. J Appl Microbiol 105:706-714.

114. Zinkevich VV, Beech IB. 2000. Screening of sulfate-reducing bacteria in colonoscopy samples from healthy and colitic human gut mucosa. FEMS Microbiol Ecol 34:147-155.

115. Loubinoux J, Bronowicki JP, Pereira IA, Mougenel JL, Faou AE. 2002. Sulfate-reducing bacteria in human feces and their association with inflammatory bowel diseases. FEMS Microbiol Ecol 40:107-112.

116. Pitcher MC, Beatty ER, Cummings JH. 2000. The contribution of sulphate reducing bacteria and 5-aminosalicylic acid to faecal sulphide in patients with ulcerative colitis. Gut 46:64-72.

117. Fite A, Macfarlane GT, Cummings JH, Hopkins MJ, Kong SC, Furrie E, Macfarlane S. 2004. Identification and quantitation of mucosal and faecal desulfovibrios using real time polymerase chain reaction. Gut 53:523-529.

118. Reiffenstein RJ, Hulbert WC, Roth SH. 1992. Toxicology of hydrogen sulfide. Annu Rev Pharmacol Toxicol 32:109-134.

119. Beauchamp RO, Jr., Bus JS, Popp JA, Boreiko CJ, Andjelkovich DA. 1984. A critical review of the literature on hydrogen sulfide toxicity. Crit Rev Toxicol 13:25-97.

120. Medani M, Collins D, Docherty NG, Baird AW, O'Connell PR, Winter DC. 2011. Emerging role of hydrogen sulfide in colonic physiology and pathophysiology. Inflamm Bowel Dis 17:1620-1625.

121. Roediger WE, Duncan A, Kapaniris O, Millard S. 1993. Sulphide impairment of substrate oxidation in rat colonocytes: a biochemical basis for ulcerative colitis? Clin Sci (Lond) 85:623-627.

122. Christl SU, Eisner HD, Dusel G, Kasper H, Scheppach W. 1996. Antagonistic effects of sulfide and butyrate on proliferation of colonic mucosa: a potential role for these agents in the pathogenesis of ulcerative colitis. Dig Dis Sci 41:2477-2481.

187

123. Attene-Ramos MS, Wagner ED, Plewa MJ, Gaskins HR. 2006. Evidence that hydrogen sulfide is a genotoxic agent. Mol Cancer Res 4:9-14.

124. Attene-Ramos MS, Wagner ED, Gaskins HR, Plewa MJ. 2007. Hydrogen sulfide induces direct radical-associated DNA damage. Mol Cancer Res 5:455-459.

125. Attene-Ramos MS, Nava GM, Muellner MG, Wagner ED, Plewa MJ, Gaskins HR. 2010. DNA damage and toxicogenomic analyses of hydrogen sulfide in human intestinal epithelial FHs 74 Int cells. Environ Mol Mutagen 51:304-314.

126. Li J, Kong X, Zhang J, Luo Q, Li X, Fang L. 2013. MiRNA-26b inhibits proliferation by targeting PTGS2 in breast cancer. Cancer Cell Int 13:7.

127. Park JK, Song JH, He TC, Nam SW, Lee JY, Park WS. 2009. Overexpression of Wnt-2 in colorectal cancers. Neoplasma 56:119-123.

128. Takeuchi H, Setoguchi T, Machigashira M, Kanbara K, Izumi Y. 2008. Hydrogen sulfide inhibits cell proliferation and induces cell cycle arrest via an elevated p21 Cip1 level in Ca9-22 cells. J Periodontal Res 43:90-95.

129. Chen W, Kajiya M, Giro G, Ouhara K, Mackler HE, Mawardi H, Boisvert H, Duncan MJ, Sato K, Kawai T. 2010. Bacteria-derived hydrogen sulfide promotes IL-8 production from epithelial cells. Biochem Biophys Res Commun 391:645- 650.

130. Distrutti E, Sediari L, Mencarelli A, Renga B, Orlandi S, Antonelli E, Roviezzo F, Morelli A, Cirino G, Wallace JL. 2006. Evidence that hydrogen sulfide exerts antinociceptive effects in the gastrointestinal tract by activating KATP channels. J Pharmacol Exp Ther 316:325-335.

131. Zanardo RCO, Brancaleone V, Distrutti E, Fiorucci S, Cirino G, Wallace JL. 2006. Hydrogen sulfide is an endogenous modulator of leukocyte-mediated inflammation. FASEB J 20:2118-2120.

132. Abe K, Kimura H. 1996. The possible role of hydrogen sulfide as an endogenous neuromodulator. J Neurosci 16:1066-1071.

133. Schulzke JD, Ploeger S, Amasheh M, Fromm A, Zeissig S, Troeger H, Richter J, Bojarski C, Schumann M, Fromm M. 2009. Epithelial tight junctions in intestinal inflammation. Ann N Y Acad Sci 1165:294-300.

134. Zeissig S, Burgel N, Gunzel D, Richter J, Mankertz J, Wahnschaffe U, Kroesen AJ, Zeitz M, Fromm M, Schulzke JD. 2007. Changes in expression and distribution of claudin 2, 5 and 8 lead to discontinuous tight junctions and barrier dysfunction in active Crohn's disease. Gut 56:61-72.

135. Yamamoto-Furusho JK, Mendivil-Rangel EJ, Fonseca-Camarillo G. 2012. Differential expression of occludin in patients with ulcerative colitis and healthy controls. Inflamm Bowel Dis 18:E1999.

188

136. Alto NM, Orth K. 2012. Subversion of cell signaling by pathogens. Cold Spring Harb Perspect Biol 4:a006114.

137. Stevens JM, Galyov EE, Stevens MP. 2006. Actin-dependent movement of bacterial pathogens. Nat Rev Microbiol 4:91-101.

138. Bonavia A, Thompson M, Schryver B, Ehrhardt R. 2012. Bacteria cryopreservation protocol. Nat. Protoc. doi:10.1038/protex.2012.006

139. Ismail Y, Lee H, Riordan SM, Grimm MC, Zhang L. 2013. The effects of oral and enteric Campylobacter concisus strains on expression of TLR4, MD-2, TLR2, TLR5 and COX-2 in HT-29 cells. PLoS One 8:e56888.

140. Mukhopadhya I, Thomson JM, Hansen R, Alotaibi B, Berry SH, El-Omar EM, Hold GL. 2011. Detection of campylobacter concisus in colonic biopsies from adult patients with ulcerative colitis. Gut 60:A211.

141. Sartor RB. 2006. Mechanisms of disease: pathogenesis of Crohn's disease and ulcerative colitis. Nat Clin Pract Gastroenterol Hepatol 3:390-407.

142. Lastovica AJ. 2009. Clinical relevance of Campylobacter concisus isolated from pediatric patients. J Clin Microbiol 47:2360.

143. Vandamme P, Dewhirst FE, Paster BJ, W. OSL. 2005. Genus 1. Campylobacter. In Brenner DJ, Krieg NR, Staley JT, Garrity GM (ed), Bergey's Manual® of Systematic Bacteriology, 2nd ed, vol 2. Springer-Verlag, NY.

144. Matsheka MI, Lastovica AJ, Zappe H, Elisha BG. 2006. The use of (GTG)5 oligonucleotide as an RAPD primer to type Campylobacter concisus. Lett Appl Microbiol 42:600-605.

145. Lastovica AJ. 2006. Emerging Campylobacter spp.: The tip of the iceberg. Clin Microbiol Newsletter 28:49-56.

146. Simren M, Stotzer PO. 2006. Use and abuse of hydrogen breath tests. Gut 55:297-303.

147. Gatlin CL, Kleemann GR, Hays LG, Link AJ, Yates JR, 3rd. 1998. Protein identification at the low femtomole level from silver-stained gels using a new fritless electrospray interface for liquid chromatography-microspray and nanospray mass spectrometry. Anal Biochem 263:93-101.

148. Searle BC. 2010. Scaffold: a bioinformatic tool for validating MS/MS-based proteomic studies. Proteomics 10:1265-1269.

149. Pray HA, Schweickert C, Minnich B. 1952. Solubility of hydrogen, oxygen, nitrogen, and helium in water at elevated temperatures. Ind Eng Chem 44:1146- 1151.

189

150. Xie W, Wang F, Guo L, Chen Z, Sievert SM, Meng J, Huang G, Li Y, Yan Q, Wu S, Wang X, Chen S, He G, Xiao X, Xu A. 2011. Comparative metagenomics of microbial communities inhabiting deep-sea hydrothermal vent chimneys with contrasting chemistries. ISME J 5:414-426.

151. Shen J, Gunsalus RP. 1997. Role of multiple ArcA recognition sites in anaerobic regulation of succinate dehydrogenase (sdhCDAB) gene expression in Escherichia coli. Mol Microbiol 26:223-236.

152. Wood GE, Haydock AK, Leigh JA. 2003. Function and regulation of the formate dehydrogenase genes of the methanogenic archaeon Methanococcus maripaludis. J Bacteriol 185:2548-2554.

153. Ruiz-Herrera J, Garcia LG. 1972. Regulation of succinate dehydrogenase in Escherichia coli. J Gen Microbiol 72:29-35.

154. Levitt MD. 1969. Production and excretion of hydrogen gas in man. N Engl J Med 281:122-127.

155. Cummings J. 1983. Fermentation in the human large intestine: evidence and implications for health. Lancet 321:1206-1209.

156. Nakamura N, Lin HC, McSweeney CS, Mackie RI, Gaskins HR. 2010. Mechanisms of microbial hydrogen disposal in the human colon and implications for health and disease. Food Sci Technol 1.

157. Levitt MD, Donaldson RM. 1970. Use of respiratory hydrogen (H2) excretion to detect carbohydrate malabsorption. J Lab Clin Med 75:937-945.

158. Bond JH, Levitt MD. 1978. Effect of dietary fiber on intestinal gas production and small bowel transit time in man. Am J Clin Nutr 31:S169-S174.

159. Gibson GR, Cummings JH, Macfarlane GT, Allison C, Segal I, Vorster HH, Walker AR. 1990. Alternative pathways for hydrogen disposal during fermentation in the human colon. Gut 31:679-683.

160. Strocchi A, Furne J, Ellis C, Levitt M. 1991. Competition for hydrogen by human faecal bacteria: evidence for the predominance of methane producing bacteria. Gut 32:1498-1501.

161. Mastropaolo G, Rees W. 1987. Evaluation of the hydrogen breath test in man: definition and elimination of the early hydrogen peak. Gut 28:721-725.

162. McKay LF, Eastwood M, Brydon W. 1985. Methane excretion in man-a study of breath, flatus, and faeces. Gut 26:69-74.

163. Pimentel M, Mayer AG, Park S, Chow EJ, Hasan A, Kong Y. 2003. Methane production during lactulose breath test is associated with gastrointestinal disease presentation. Dig Dis Sci 48:86-92.

190

164. Scanlan PD, Shanahan F, Marchesi JR. 2008. Human methanogen diversity and incidence in healthy and diseased colonic groups using mcrA gene analysis. BMC Microbiol 8:79.

165. Bullman S, Corcoran D, O'Leary J, O'Hare D, Lucey B, Sleator RD. 2011. Emerging dynamics of human Campylobacteriosis in Southern Ireland. FEMS Immunol Med Microbiol 63:248-253.

166. Bullman S, Corcoran D, O'Leary J, Lucey B, Byrne D, Sleator RD. 2011. Campylobacter ureolyticus: an emerging gastrointestinal pathogen? FEMS Immunol Med Microbiol 61:228-230.

167. Burgos-Portugal JA, Kaakoush NO, Raftery MJ, Mitchell HM. 2012. Pathogenic potential of Campylobacter ureolyticus. Infect Immun 80:883-890.

168. Lahteenmaki K, Virkola R, Pouttu R, Kuusela P, Kukkonen M, Korhonen T. 1995. Bacterial plasminogen receptors: in vitro evidence for a role in degradation of the 24 mammalian extracellular matrix. Infect Immun 63:3659-3664.

169. Jönsson K, Guo BP, Monstein H-J, Mekalanos JJ, Kronvall G. 2004. Molecular cloning and characterization of two Helicobacter pylori genes coding for plasminogen-binding proteins. Proc Natl Acad Sci U S A 101:1852-1857.

170. Sodeinde OA, Subrahmanyam Y, Stark K, Quan T, Bao Y, Goguen JD. 1992. A surface protease and the invasive character of plague. Science 258:1004-1007.

171. Lertpiriyapong K, Gamazon ER, Feng Y, Park DS, Pang J, Botka G, Graffam ME, Ge Z, Fox JG. 2012. Campylobacter jejuni type VI secretion system: roles in adaptation to deoxycholic acid, host cell adherence, invasion, and in vivo colonization. PLoS One 7:e42842.

172. Josenhans C, Suerbaum S. 2002. The role of motility as a virulence factor in bacteria. Int J Med Microbiol 291:605-614.

173. Fischbach MA, Sonnenburg JL. 2011. Eating for two: how metabolism establishes interspecies interactions in the gut. Cell host & microbe 10:336-347.

174. Wolin MJ, Miller TL. Interactions of microbial populations in cellulose fermentation, p 109-113. In (ed),

175. Olson JW, Maier RJ. 2002. Molecular hydrogen as an energy source for Helicobacter pylori. Science 298:1788-1790.

176. Carbonero F, Gaskins HR. 2014. Sulfate-reducing bacteria in the human gut microbiome, p 1-3, Encyclopedia of Metagenomics. Springer.

177. Barton LL, Fauque GD. 2009. Biochemistry, physiology and biotechnology of sulfate-reducing bacteria. Adv Appl Microbiol 68:41-98.

191

178. Pitcher M, Beatty E, Gibson G, Cummings J. 1995. Incidence and activities of sulphate-reducing bacteria in patients with ulcerative colitis. Gut 36:A63.

179. Garrity GM, Bell JA, Lilburn T. 2005. Class V. Epsilonproteobacteria class. nov, p 1145-1194, Bergey’s Manual® of Systematic Bacteriology. Springer.

180. Leonhartsberger S, Korsa I, Bock A. 2002. The molecular biology of formate metabolism in enterobacteria. J Mol Microbiol Biotechnol 4:269-276.

181. Ruschig U, Muller U, Willnow P, Hopner T. 1976. CO2 reduction to formate by NADH catalysed by formate dehydrogenase from Pseudomonas oxalaticus. Eur J Biochem 70:325-330.

182. Unden G, Bongaerts J. 1997. Alternative respiratory pathways of Escherichia coli: energetics and transcriptional regulation in response to electron acceptors. Biochimica et Biophysica Acta (BBA)-Bioenergetics 1320:217-234.

183. Mahendran V, Octavia S, Faruk Demirbas O, Sabrina S, Ma R, Lan R, S MR, Grimm MC, Zhang L. 2015. Delineation of genetic relatedness and population structure of oral and enteric Campylobacter concisus strains by analysis of housekeeping genes. Microbiol doi:10.1099/mic.0.000112.

184. Hajna AA. 1945. Triple-sugar iron agar medium for the identification of the intestinal group of bacteria. J Bacteriol 49:516-517.

185. Levitt MD, Bond JH, Jr. 1970. Volume, composition, and source of intestinal gas. Gastroenterol 59:921-929.

186. Ono H, Nishijima Y, Adachi N, Sakamoto M, Kudo Y, Kaneko K, Nakao A, Imaoka T. 2012. A basic study on molecular hydrogen (H2) inhalation in acute cerebral ischemia patients for safety check with physiological parameters and measurement of blood H2 level. Med Gas Res 2:21.

187. Wong HC, Chen YL. 1988. Effects of lactic acid bacteria and organic acids on growth and germination of Bacillus cereus. Appl Environ Microbiol 54:2179-2184.

188. Dijkhuizen L, Harder W. 1975. Substrate inhibition in Pseudomonas oxalaticus OX1: a kinetic study of growth inhibition by oxalate and formate using extended cultures. Antonie Van Leeuwenhoek 41:135-146.

189. Chu IM, Keuer TA, Papoutsakis ET. 1987. Formate transport, growth-inhibition and the membrane protonmotive force in 2 Methylotrophs (T15 and L3). Appl Microbiol Biotechnol 26:70-77.

190. Paster BJ, Gibbons RJ. 1986. Chemotactic response to formate by Campylobacter concisus and its potential role in gingival colonization. Infect Immun 52:378-383.

192

191. Indrakanti VP, Kubicki JD, Schobert HH. 2009. Photoinduced activation of CO2 on Ti-based heterogeneous catalysts: Current state, chemical physics-based insights and outlook. Energy Environ Sci 2:745-758.

192. Liebensteiner MG, Tsesmetzis N, Stams AJ, Lomans BP. 2014. Microbial redox processes in deep subsurface environments and the potential application of (per) chlorate in oil reservoirs. Front Microbiol 5.

193. Willey JM, Sherwood L, Woolverton CJ, Prescott LM. 2008. Prescott, Harley, and Klein's microbiology, 7th ed. McGraw-Hill Higher Education, New York.

194. Erichsen K, Ulvik RJ, Nysaeter G, Johansen J, Ostborg J, Berstad A, Berge RK, Hausken T. 2005. Oral ferrous fumarate or intravenous iron sucrose for patients with inflammatory bowel disease. Scand J Gastroenterol 40:1058-1065.

195. Swan HT, Jowett GH. 1959. Treatment of iron deficiency with ferrous fumarate. Assessment by a statistically accurate method. Br Med J 2:782-787.

196. Hyder SZ, Persson LÅ, Chowdhury A, Ekström E-C. 2002. Do side-effects reduce compliance to iron supplementation? A study of daily-and weekly-dose regimens in pregnancy. J Health Popul Nutr:175-179.

197. al-Momen AK, al-Meshari A, al-Nuaim L, Saddique A, Abotalib Z, Khashogji T, Abbas M. 1996. Intravenous iron sucrose complex in the treatment of iron deficiency anemia during pregnancy. Eur J Obstet Gynecol Reprod Biol 69:121- 124.

198. Lü ck E, Jager M. 1997. Antimicrobial food additives : characteristics, uses, effects, 2nd rev. and enl. ed. Springer, Berlin ; New York.

199. Fortier LC, Sekulovic O. 2013. Importance of prophages to evolution and virulence of bacterial pathogens. Virulence 4:354-365.

200. Aertsen A, Faster D, Michiels CW. 2005. Induction of Shiga toxin-converting prophage in Escherichia coli by high hydrostatic pressure. Appl Environ Microbiol 71:1155-1162.

201. Rubin EJ, Lin W, Mekalanos JJ, Waldor MK. 1998. Replication and integration of a Vibrio cholerae cryptic plasmid linked to the CTX prophage. Mol Microbiol 28:1247-1254.

202. Zhou Y, Liang Y, Lynch KH, Dennis JJ, Wishart DS. 2011. PHAST: a fast phage search tool. Nucleic Acids Res 39:W347-352.

203. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S. 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol Biol Evol 28:2731- 2739.

193

204. Sievers F, Wilm A, Dineen D, Gibson TJ, Karplus K, Li W, Lopez R, McWilliam H, Remmert M, Soding J, Thompson JD, Higgins DG. 2011. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol Syst Biol 7:539.

205. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ. 1990. Basic local alignment search tool. J Mol Biol 215:403-410.

206. Artimo P, Jonnalagedda M, Arnold K, Baratin D, Csardi G, de Castro E, Duvaud S, Flegel V, Fortier A, Gasteiger E, Grosdidier A, Hernandez C, Ioannidis V, Kuznetsov D, Liechti R, Moretti S, Mostaguir K, Redaschi N, Rossier G, Xenarios I, Stockinger H. 2012. ExPASy: SIB bioinformatics resource portal. Nucleic Acids Res 40:W597-603.

207. Pearson WR. 2013. An introduction to sequence similarity ("homology") searching. Curr Protoc Bioinformatics Chapter 3:Unit3 1.

208. Petersen TN, Brunak S, von Heijne G, Nielsen H. 2011. SignalP 4.0: discriminating signal peptides from transmembrane regions. Nat Methods 8:785- 786.

209. Bendtsen JD, Kiemer L, Fausbøll A, Brunak S. 2005. Non-classical protein secretion in bacteria. BMC Microbiol 5:58.

210. Kanehisa M. 2002. The KEGG database. Novartis Found. Symp. 247:91-101; discussion 101-103, 119-128, 244-152.

211. Chouikha I, Charrier L, Filali S, Derbise A, Carniel E. 2010. Insights into the infective properties of YpfΦ, the Yersinia pestis filamentous phage. Virol 407:43- 52.

212. Das B, Bischerour J, Barre F-X. 2011. VGJɸ integration and excision mechanisms contribute to the genetic diversity of Vibrio cholerae epidemic strains. Proc Natl Acad Sci U S A 108:2516-2521.

213. Ahmad AA, Askora A, Kawasaki T, Fujie M, Yamada T. 2014. The filamentous phage XacF1 causes loss of virulence in Xanthomonas axonopodis pv. citri, the causative agent of citrus canker disease. Front Microbiol 5:321.

214. Crooks GE, Hon G, Chandonia JM, Brenner SE. 2004. WebLogo: a sequence logo generator. Genome Res 14:1188-1190.

215. Mitchell A, Chang HY, Daugherty L, Fraser M, Hunter S, Lopez R, McAnulla C, McMenamin C, Nuka G, Pesseat S, Sangrador-Vegas A, Scheremetjew M, Rato C, Yong SY, Bateman A, Punta M, Attwood TK, Sigrist CJ, Redaschi N, Rivoire C, Xenarios I, Kahn D, Guyot D, Bork P, Letunic I, Gough J, Oates M, Haft D, Huang H, Natale DA, Wu CH, Orengo C, Sillitoe I, Mi H, Thomas PD, Finn RD. 2015. The InterPro protein families database: the classification resource after 15 years. Nucleic Acids Res 43:D213-221.

194

216. Marchler-Bauer A, Derbyshire MK, Gonzales NR, Lu S, Chitsaz F, Geer LY, Geer RC, He J, Gwadz M, Hurwitz DI, Lanczycki CJ, Lu F, Marchler GH, Song JS, Thanki N, Wang Z, Yamashita RA, Zhang D, Zheng C, Bryant SH. 2015. CDD: NCBI's conserved domain database. Nucleic Acids Res 43:D222-226.

217. Kelley LA, Mezulis S, Yates CM, Wass MN, Sternberg MJ. 2015. The Phyre2 web portal for protein modeling, prediction and analysis. Nat Protoc 10:845-858.

218. Herraez A. 2006. Biomolecules in the computer: Jmol to the rescue. Biochem Mol Biol Educ 34:255-261.

219. Boutet E, Lieberherr D, Tognolli M, Schneider M, Bairoch A. 2007. UniProtKB/Swiss-Prot. Methods Mol Biol 406:89-112.

220. Moss J, Iglewski B, Vaughan M, Tu AT. 1995. Handbook of natural toxins. 8. Bacterial toxins and virulence factors in disease, vol 8. Marcel Dekker Inc., NY.

221. Quast C, Pruesse E, Yilmaz P, Gerken J, Schweer T, Yarza P, Peplies J, Glockner FO. 2013. The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Res 41:D590-596.

222. Subramanya HS, Arciszewska LK, Baker RA, Bird LE, Sherratt DJ, Wigley DB. 1997. Crystal structure of the site-specific recombinase, XerD. EMBO J 16:5178-5187.

223. Kiefer F, Arnold K, Kunzli M, Bordoli L, Schwede T. 2009. The SWISS- MODEL repository and associated resources. Nucleic Acids Res 37:D387-392.

224. Sherratt DJ, Arciszewska LK, Blakely G, Colloms S, Grant K, Leslie N, McCulloch R. 1995. Site-specific recombination and circular chromosome segregation. Philos Trans R Soc Lond B Biol Sci 347:37-42.

225. Aussel L, Barre FX, Aroyo M, Stasiak A, Stasiak AZ, Sherratt D. 2002. FtsK Is a DNA motor protein that activates chromosome dimer resolution by switching the catalytic state of the XerC and XerD recombinases. Cell 108:195-205.

226. Carnoy C, Roten CA. 2009. The dif/Xer recombination systems in proteobacteria. PLoS One 4:e6531.

227. Kersulyte D, Lee W, Subramaniam D, Anant S, Herrera P, Cabrera L, Balqui J, Barabas O, Kalia A, Gilman RH, Berg DE. 2009. Helicobacter pylori's plasticity zones are novel transposable elements. PLoS One 4:e6859.

228. Genin S, Boucher CA. 1994. A superfamily of proteins involved in different secretion pathways in gram-negative bacteria: modular structure and specificity of the N-terminal domain. Mol Gen Genet 243:112-118.

229. Russel M, Linderoth NA, Sali A. 1997. Filamentous phage assembly: variation on a protein export theme. Gene 192:23-32.

195

230. Davis BM, Lawson EH, Sandkvist M, Ali A, Sozhamannan S, Waldor MK. 2000. Convergence of the secretory pathways for cholera toxin and the filamentous phage, CTXphi. Science 288:333-335.

231. Koonin EV. 1992. The second cholera toxin, Zot, and its plasmid-encoded and phage-encoded homologues constitute a group of putative ATPases with an altered purine NTP-binding motif. FEBS Lett 312:3-6.

232. Bartpho T, Wongsurawat T, Wongratanacheewin S, Talaat AM, Karoonuthaisiri N, Sermswan RW. 2012. Genomic islands as a marker to differentiate between clinical and environmental Burkholderia pseudomallei. PloS one 7:e37762.

233. Levine MM, Kaper JB, Herrington D, Losonsky G, Morris JG, Clements ML, Black RE, Tall B, Hall R. 1988. Volunteer studies of deletion mutants of Vibrio cholerae O1 prepared by recombinant techniques. Infect Immun 56:161-167.

234. Hollander D, Vadheim CM, Brettholz E, Petersen GM, Delahunty T, Rotter JI. 1986. Increased intestinal permeability in patients with Crohn's disease and their relatives. A possible etiologic factor. Ann Intern Med 105:883-885.

235. May GR, Sutherland LR, Meddings JB. 1993. Is small intestinal permeability really increased in relatives of patients with Crohn's disease? Gastroenterol 104:1627-1632.

236. Irvine EJ, Marshall JK. 2000. Increased intestinal permeability precedes the onset of Crohn's disease in a subject with familial risk. Gastroenterol 119:1740- 1744.

237. Welcker K, Martin A, Kolle P, Siebeck M, Gross M. 2004. Increased intestinal permeability in patients with inflammatory bowel disease. Eur J Med Res 9:456- 460.

238. Ye Y, Xie X, Yu J, Zhou L, Xie H, Jiang G, Yu X, Zhang W, Wu J, Zheng S. 2010. Involvement of Th17 and Th1 effector responses in patients with Hepatitis B. J Clin Immunol 30:546-555.

239. Lepage P, Colombet J, Marteau P, Sime-Ngando T, Dore J, Leclerc M. 2008. Dysbiosis in inflammatory bowel disease: a role for bacteriophages? Gut 57:424- 425.

240. Bejar S, Cam K, Bouche JP. 1986. Control of cell division in Escherichia coli. DNA sequence of dicA and of a second gene complementing mutation dicA1, dicC. Nucleic Acids Res 14:6821-6833.

241. Kimsey HH, Waldor MK. 2004. The CTXphi repressor RstR binds DNA cooperatively to form tetrameric repressor-operator complexes. J Biol Chem 279:2640-2647.

196

242. Liu Y, Haggard-Ljungquist E. 1994. Studies of bacteriophage P2 DNA replication: localization of the cleavage site of the A protein. Nucleic Acids Res 22:5204-5210.

243. Kruger DH, Bickle TA. 1983. Bacteriophage survival: multiple mechanisms for avoiding the deoxyribonucleic acid restriction systems of their hosts. Microbiol Rev 47:345-360.

244. Baillon ML, van Vliet AH, Ketley JM, Constantinidou C, Penn CW. 1999. An iron-regulated alkyl hydroperoxide reductase (AhpC) confers aerotolerance and oxidative stress resistance to the microaerophilic pathogen Campylobacter jejuni. J Bacteriol 181:4798-4804.

245. Wang L, Chen S, Deng Z. 2011. Phosphorothioation: an unusual post-replicative modification on the dna backbone. In Seligmann DH (ed), DNA Replication- Current Advances. INTECH Open Access Publisher.

246. Che D, Hasan MS, Chen B. 2014. Identifying pathogenicity islands in bacterial pathogenomics using computational approaches. Pathog 3:36-56.

247. Williams KP. 2002. Integration sites for genetic elements in prokaryotic tRNA and tmRNA genes: sublocation preference of integrase subfamilies. Nucleic Acids Res 30:866-875.

248. Khurana I. 1981. Textbook of medical physiology. Elsevier, New Delhi.

249. Kristoffersen SM, Ravnum S, Tourasse NJ, Okstad OA, Kolsto AB, Davies W. 2007. Low concentrations of bile salts induce stress responses and reduce motility in Bacillus cereus ATCC 14579 [corrected]. J Bacteriol 189:5302-5313.

250. Bernstein C, Bernstein H, Payne CM, Beard SE, Schneider J. 1999. Bile salt activation of stress response promoters in Escherichia coli. Curr Microbiol 39:68- 72.

251. Ma R, Sapwell N, Chung HK, Lee H, Mahendran V, Leong RW, Riordan SM, Grimm MC, Zhang L. 2015. Investigation of the effects of pH and bile on the growth of oral Campylobacter concisus strains isolated from patients with inflammatory bowel disease and controls. J Med Microbiol 64:438-445.

252. Malik-Kale P, Parker CT, Konkel ME. 2008. Culture of Campylobacter jejuni with sodium deoxycholate induces virulence gene expression. J Bacteriol 190:2286-2297.

253. Shao C, Zhang Q, Sun Y, Liu Z, Zeng J, Zhou Y, Yu X, Jia J. 2008. Helicobacter pylori protein response to human bile stress. J Med Microbiol 57:151-158.

254. Clark CG, Chong PM, McCorrister SJ, Simon P, Walker M, Lee DM, Nguy K, Cheng K, Gilmour MW, Westmacott GR. 2014. The CJIE1 prophage of

197

Campylobacter jejuni affects protein expression in growth media with and without bile salts. BMC Microbiol 14:70.

255. Kim S, Ryu K, Biswas D, Ahn J. 2014. Survival, prophage induction, and invasive properties of lysogenic Salmonella Typhimurium exposed to simulated gastrointestinal conditions. Arch Microbiol 196:655-659.

256. Fasano A, Fiorentini C, Donelli G, Uzzau S, Kaper JB, Margaretten K, Ding X, Guandalini S, Comstock L, Goldblum SE. 1995. Zonula occludens toxin modulates tight junctions through protein kinase C-dependent actin reorganization, in vitro. J Clin Invest 96:710-720.

257. Finkelstein RA. 1996. Cholera, Vibrio cholerae O1 and O139, and other pathogenic Vibrios. In Baron S (ed), Med Microbiol, 4th ed, Galveston (TX).

258. Uzzau S, Cappuccinelli P, Fasano A. 1999. Expression of Vibrio cholerae zonula occludens toxin and analysis of its subcellular localization. Microb Pathog 27:377- 385.

259. Ye J, Coulouris G, Zaretskaya I, Cutcutache I, Rozen S, Madden TL. 2012. Primer-BLAST: a tool to design target-specific primers for polymerase chain reaction. BMC Bioinformatics 13:134.

260. Birnboim HC, Doly J. 1979. A rapid alkaline extraction procedure for screening recombinant plasmid DNA. Nucleic Acids Res 7:1513-1523.

261. Alazzam B, Bonnassie-Rouxin S, Dufour V, Ermel G. 2011. MCLMAN, a new minimal medium for Campylobacter jejuni NCTC 11168. Res Microbiol 162:173- 179.

262. Palyada K, Threadgill D, Stintzi A. 2004. Iron acquisition and regulation in Campylobacter jejuni. J Bacteriol 186:4714-4729.

263. Phongsisay V. 2006. Campylobacter jejuni and the Guillain-Barré syndromeRMIT University.

264. Taylor D, Eaton M, Yan W, Chang N. 1992. Genome maps of Campylobacter jejuni and Campylobacter coli. J Bacteriol 174:2332-2337.

265. Carrillo-Casas EM, Hernandez-Castro R, Suarez-Guemes F, de la Pena- Moctezuma A. 2008. Selection of the internal control gene for real-time quantitative rt-PCR assays in temperature treated Leptospira. Curr Microbiol 56:539-546.

266. Pfaffl MW. 2001. A new mathematical model for relative quantification in real- time RT-PCR. Nucleic Acids Res 29:e45.

267. Gasteiger E, Hoogland C, Gattiker A, Duvaud Se, Wilkins MR, Appel RD, Bairoch A. 2005. Protein identification and analysis tools on the ExPASy server, p

198

571-607. In Walker JM (ed), The Proteomics Protocols Handbook doi:10.1385/1- 59259-890-0:571. Humana Press.

268. Sawada N. 2013. Tight junction-related human diseases. Pathol Int 63:1-12.

269. Fasano A. 2011. Zonulin and its regulation of intestinal barrier function: the biological door to inflammation, autoimmunity, and cancer. Physiol Rev 91:151- 175.

270. Salama NN, Fasano A, Lu R, Eddington ND. 2003. Effect of the biologically active fragment of zonula occludens toxin, ΔG, on the intestinal paracellular transport and oral absorption of mannitol. Int J Pharm 251:113-121.

271. Fouts DE, Mongodin,E.F., Puiu,D., Sebastian,Y., Miller,W.G., Mandrell,R.E., On,S. and Nelson,K.E. 2007. Genome Sequence of Campylobacter concisus 13826 isolated from human feces. National Center for Biotechnology Information, NIH, Bethesda, MD 20894, USA.

272. Goldblum SE, Rai U, Tripathi A, Thakar M, De Leo L, Di Toro N, Not T, Ramachandran R, Puche AC, Hollenberg MD, Fasano A. 2011. The active Zot domain (aa 288-293) increases ZO-1 and myosin 1C serine/threonine phosphorylation, alters interaction between ZO-1 and its binding partners, and induces tight junction disassembly through proteinase activated receptor 2 activation. FASEB J 25:144-158.

273. Merritt ME, Donaldson JR. 2009. Effect of bile salts on the DNA and membrane integrity of enteric bacteria. J Med Microbiol 58:1533-1541.

274. Taranto MP, Perez-Martinez G, Font de Valdez G. 2006. Effect of bile acid on the cell membrane functionality of lactic acid bacteria for oral administration. Res Microbiol 157:720-725.

275. Catalioto RM, Triolo A, Giuliani S, Altamura M, Evangelista S, Maggi CA. 2008. Increased paracellular absorption by bile salts and P-glycoprotein stimulated efflux of otilonium bromide in Caco-2 cells monolayers as a model of intestinal barrier. J Pharm Sci 97:4087-4100.

276. Anderberg EK, Nystrom C, Artursson P. 1992. Epithelial transport of drugs in cell culture. VII: Effects of pharmaceutical surfactant excipients and bile acids on transepithelial permeability in monolayers of human intestinal epithelial (Caco-2) cells. J Pharm Sci 81:879-887.

277. Lodish HF, Berk A, Zipursky SL, Matsudaira P, Baltimore D, Darnell J. 2000. Molecular cell biology, 4th ed. W. H. Freeman.

278. Perrin BJ, Ervasti JM. 2010. The actin gene family: function follows isoform. Cytoskeleton (Hoboken) 67:630-634.

199

279. Cooper GM. 2000. Structure and organization of actin filaments, The Cell: A Molecular Approach, 2nd ed. Sinauer Associates, Sunderland (MA).

280. Van Itallie CM, Anderson JM. 2014. Architecture of tight junctions and principles of molecular composition. Semin Cell Dev Biol 36:157-165.

281. Berkes J, Viswanathan VK, Savkovic SD, Hecht G. 2003. Intestinal epithelial responses to enteric pathogens: effects on the tight junction barrier, ion transport, and inflammation. Gut 52:439-451.

282. Li Q, Zhang Q, Wang C, Li N, Li J. 2008. Invasion of enteropathogenic Escherichia coli into host cells through epithelial tight junctions. FEBS J 275:6022-6032.

283. Patel JC, Galan JE. 2005. Manipulation of the host actin cytoskeleton by Salmonella-all in the name of entry. Curr Opin Microbiol 8:10-15.

284. Robbins JR, Barth AI, Marquis H, de Hostos EL, Nelson WJ, Theriot JA. 1999. Listeria monocytogenes exploits normal host cell processes to spread from cell to cell. J Cell Biol 146:1333-1350.

285. Ma AT, Mekalanos JJ. 2010. In vivo actin cross-linking induced by Vibrio cholerae type VI secretion system is associated with intestinal inflammation. Proc Natl Acad Sci U S A 107:4365-4370.

286. Luo Y, Frey EA, Pfuetzner RA, Creagh AL, Knoechel DG, Haynes CA, Finlay BB, Strynadka NC. 2000. Crystal structure of enteropathogenic Escherichia coli intimin-receptor complex. Nat 405:1073-1077.

287. Donnenberg MS. 2000. Pathogenic strategies of enteric bacteria. Nat Clin Pract Gastroenterol Hepatol 406:768-774.

288. Cossart P, Sansonetti PJ. 2004. Bacterial invasion: the paradigms of enteroinvasive pathogens. Science 304:242-248.

289. Loren V, Cabre E, Ojanguren I, Domenech E, Pedrosa E, Garcia- Jaraquemada A, Manosa M, Manye J. 2015. Interleukin-10 enhances the intestinal epithelial barrier in the presence of corticosteroids through p38 MAPK activity in Caco-2 monolayers: a possible mechanism for steroid responsiveness in ulcerative colitis. PLoS One 10:e0130921.

290. Baranwal S, Naydenov NG, Harris G, Dugina V, Morgan KG, Chaponnier C, Ivanov AI. 2012. Nonredundant roles of cytoplasmic beta- and gamma-actin isoforms in regulation of epithelial apical junctions. Mol Biol Cell 23:3542-3553.

291. Higashi-Fujime S, Suzuki M, Titani K, Hozumi T. 1992. Muscle actin cleaved by proteinase K: its polymerization and in vitro motility. J Biochem 112:568-572.

200

292. Mangel WF, Baniecki ML, McGrath WJ. 2003. Specific interactions of the adenovirus proteinase with the viral DNA, an 11-amino-acid viral peptide, and the cellular protein actin. Cell Mol Life Sci 60:2347-2355.

293. Naora H, Naora H. 1995. Differential expression patterns of beta-actin mRNA in cells undergoing apoptosis. Biochem Biophys Res Commun 211:491-496.

294. Meresse S, Unsworth KE, Habermann A, Griffiths G, Fang F, Martínez- Lorenzo MJ, Waterman SR, Gorvel JP, Holden DW. 2001. Remodelling of the actin cytoskeleton is essential for replication of intravacuolar Salmonella. Cell Microbiol 3:567-577.

295. Kumar Y, Valdivia RH. 2008. Reorganization of the host cytoskeleton by the intracellular pathogen Chlamydia trachomatis. Commun Integr Biol 1:175-177.

296. Ismail Y. 2012. Campylobacter concisus and its possible role in inflammatory bowel disease. PhD's thesis. the Univeristy of New South Wales.

297. Marteyn B, Scorza FB, Sansonetti PJ, Tang C. 2011. Breathing life into pathogens: the influence of oxygen on bacterial virulence and host responses in the gastrointestinal tract. Cell Microbiol 13:171-176.

298. Brown LR, Dowell C. 1968. Replication of coliphage M-13 I. effects on host cells aftersynchronized infection. J Virol 2:1290-1295.

299. Draaijer M, Koninkx J, Hendriks H, Kik M, Van Dijk J, Mouwen J. 1989. Actin cytoskeletal lesions in differentiated human colon carcinoma Caco-2 cells after exposure to soybean agglutinin. Biol Cell 65:29-35.

300. Wehkamp J, Wang G, Kubler I, Nuding S, Gregorieff A, Schnabel A, Kays RJ, Fellermann K, Burk O, Schwab M, Clevers H, Bevins CL, Stange EF. 2007. The Paneth cell alpha-defensin deficiency of ileal Crohn's disease is linked to Wnt/Tcf-4. J Immunol 179:3109-3118.

301. Gersemann M, Becker S, Kubler I, Koslowski M, Wang G, Herrlinger KR, Griger J, Fritz P, Fellermann K, Schwab M, Wehkamp J, Stange EF. 2009. Differences in goblet cell differentiation between Crohn's disease and ulcerative colitis. Differentiation 77:84-94.

302. Tomlin J, Lowis C, Read NW. 1991. Investigation of normal flatus production in healthy volunteers. Gut 32:665-669.

303. Suarez F, Springfield J, Levitt M. 1998. Identification of gases responsible for the odour of human flatus and evaluation of a device purported to reduce this odour. Gut 43:100-104.

304. Moon C, VanDussen KL, Miyoshi H, Stappenbeck TS. 2014. Development of a primary mouse intestinal epithelial cell monolayer culture system to evaluate factors that modulate IgA transcytosis. Mucosal Immunol 7:818-828.

201

305. Mizoguchi A. 2012. Animal models of inflammatory bowel disease. Prog Mol Biol Transl Sci 105:263-320.

306. Goyal N, Rana A, Ahlawat A, Bijjem KR, Kumar P. 2014. Animal models of inflammatory bowel disease: a review. Inflammopharmacology 22:219-233.

307. Hansen JJ. 2015. Immune responses to intestinal microbes in inflammatory bowel diseases. Curr Allergy Asthma Rep 15:562.

202