The Role of Helicobacter Pylori Virulence and Host Genetic Factors in Gastroduodenal Disease

The role of Helicobacter pylori virulence and host genetic factors in gastroduodenal disease

Heather-Marie Ann Schmidt

A thesis submitted as fulfilment of the requirement for the degree of

Doctor of Philosophy (PhD)

School of Biotechnology and Biomolecular Sciences

University of New South Wales

2010 Abstract

Helicobacter pylori, a gastric bacterial pathogen, colonises >50% of the world’s population. Colonisation invariably leads to chronic gastritis, with a small but significant proportion progressing to peptic ulcer disease and gastric cancer (GC). The reason why some individuals progress to severe disease is unknown, but is likely the result of an as yet unidentified combination of bacterial, host and environmental factors. This thesis aimed to elucidate the contribution of host genetics and H. pylori virulence to severe disease, particularly GC, in patients from different ethnicities.

Papers I, II and III investigated a comprehensive set of H. pylori virulence factors for their association with pro-inflammatory signalling in gastric epithelial cells in vitro and disease development. The major H. pylori virulence factors, including the cag PAI, vacA, babA and oipA were highly prevalent in all three ethnic groups resident in Malaysia and Singapore irrespective of the disease state of the host, suggesting all strains from this region are highly virulent. There was significant ethnic variation in the CagA EPIYA-motifs, HP0521 alleles and dupA, confirming that distinct H. pylori strains circulate within particular populations.

Paper IV assessed strains isolated from patients with different levels of inflammation for their ability to cause dendritic cell maturation and activation, and induce proinflammatory signalling in gastric epithelial cells in vitro. Strains from a higher inflammatory background induced higher levels of pro-inflammatory cytokine secretion in vitro.

Paper V examined a novel range of host genetic polymorphisms from genes in immune and cell signalling pathways for associations with susceptibility to H. pylori infection or GC. The prevalence of many polymorphisms varied significantly between the ethnic groups (Chinese, Indian and Malay), and several polymorphisms were associated with H. pylori infection (EBBR2+1963G) or GC (PTGS2-1195G, IL1B-1473C) in ethnic Chinese resident in Malaysia and Singapore.

In conclusion, substantial variation exists in the prevalence of genetic polymorphisms and the diversity of circulating strains among ethnic groups, as well as the ability of strains to induce inflammation; these findings may underlie differences in GC development and H. pylori infection. A combination of these factors in the same individual may provide a more complete picture.

Statement of Originality

‘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 acknowledgment 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……………………………………………………………………………………….

‘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 aproprietary rights, such as patent rights. I also retain the right to use in future works (such as articles or books) all or part of this thesis or dissertation. I also authorise University Microfilms to use the 350 word abstract of my thesis in Dissertation Abstract International (this is applicable to doctoral theses only). I have either used no substantial portions of copyright material in my thesis or I have obtained permission to use copyright material; where permission has not been granted I have applied/will apply for a partial restriction of the digital copy of my thesis or dissertation.'

Signed…………………………………………………………………………………….

Date……………………………………………………………………………………….

Authenticity Statement

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

Signed…………………………………………………………………………………….

Date……………………………………………………………………………………….

Acknowledgements

This thesis would not have been possible without the help and support of so many amazing people. I would especially like to thank the following:

Hazel Mitchell, truly the most wonderful supervisor I could have wished for. Thank you for so many things, personally and professionally; the opportunities to travel, to work in different labs, to present at conferences in Australia and abroad; the encouragement to get things done and to do things better – lab work, papers, thesis; the cakes and wine; the valuable discussions about where to go next with my research; your formidable editing skills. I am grateful for your continual support, though I must have frustrated you immensely sometimes. Thank you for your kindness and your friendship throughout these past few years. Thank you for so much more than I can ever really put down on paper. I have so many happy memories and I am so glad that I stayed to do a PhD with you.

Lars Engstrand, the best sort of co-supervisor! Thank you adopting me wholeheartedly into Bakteriella Tarmpatogener, I loved being a member of your group as much for the excellent science as for the social program.

Sönke Andres, my lab-buddy and co-author extraordinaire. Thank you for being such a great friend, inside the lab and out, and it didn’t hurt that you are a superb scientist too.

Christina Nilsson, a brilliant scientist, mentor and co-author. I am grateful for all of your help, advice and patience over the years, and I feel very privileged to call you a friend.

Nadeem Kaakoush, thank you for the discussions that were all too often about anything but science, and for all of your help in the lab, especially towards the end. You’ve been a super co-author and unexpected friend.

Zsuzsanna Kovach, thank you for sharing some of the burden in the lab, for being a supportive co-author and for just being there when I needed a chat.

KL Goh and KM Fock, thank you for the specimens. Faye Taylor, Jenny Barrett and David Forman, thank you for your help with statistics. Mai Dung Ha, thank you for everything you did in the lab. Thank you to all my co-authors for your helpful suggestions and support.

The members of Lab 301 (UNSW, Sydney, Australia), past and present, thank you for the good times and for never making it feel like I’ve been away for very long. Jani O’Rourke, thank you for your support, your interest in what I was up to, your friendship and, of course, the teaching opportunities. Si Ming Man, thank you for being a good friend and a talented scientist, I really enjoyed spending time in and out of the lab with you. Li Zhang, thank you for your help in the lab when I first started so many years ago and for your support and friendship ever since. John Wilson, thank you for passion for microbiology, you always made it fun to be both a student and a tutor. Ruiting, Jacob, Sophie, Alfred, Johnny, Natalia, Alex, Ping and all the other lab members, thank you also for the help in the lab, the chats, the advice and the fun we’ve had.

The members of the pylori pinglorna (Karolinksa Institutet, Stockholm and Smittskyddsinstitutet, Solna, Sweden) past, present and honorary, thank you for making me feel at home at SMI. The one and a half years I spent in Stockholm was one of the best times of my life, I have so many happy memories, and it was mostly because you are all such amazing people. Tack så jätte mycket. Mathilda Lindberg, thank you for being a friend and a great Swedish tutor. Karin Wreiber and Roger Lundberg, thank you for being so absolutely kind, I cannot tell you how much your friendship means to me. Helene Kling-Bäckhead and Anna Skoglund, thank you for the Swedish lessons and generally being fun to be around. Sandra Rodin, thank you for being such a sweetie. Laura Plant, thank you for all your help in the lab. Annelie Lundin, thank you for your friendship. Thank you Lena, Kristina, Marianne, Britta, Cia, Hedvig, Valteri, Anders, Wilhelm, Britt-Marie and Zong-Li, and everyone else at SMI and KI for making me feel welcome and a part of the group.

Thank you to the members of the Thomas Borén group at Umeå Universitet for for taking me under your wing(s) for those few weeks in the freezing north.

All my girls here in Oz and mina kompisar i Stockholm, thank you for your friendship, your distractions, your support, your advice and for so often not mentioning I was in the middle of a PhD. Julie Wood thank you for understanding what I was going through.

And finally, my family, thank you for simply putting up with me sometimes and for being proud of me always, I love you. Thank you especially to my wonderful Mum, words cannot express the gratitude I feel for everything you have done for me. Your

support, encouragement, and unconditional love mean the world to me. Thank you for your faith in me, for forcing me to sit down and work, for sharing the ups and downs and for being there when I just needed someone to listen. This would not have been possible without you.

List of publications

The thesis is based on the following papers, which will be referred to by their Roman numerals:

I. Schmidt H. –M. A., Goh K. L., Fock K. M., Hilmi I., Dhamodaran S., Forman D. and Mitchell H. Distinct cagA EPIYA motifs are associated with ethnic diversity in Malaysia and Singapore. Helicobacter, 2009 Aug; 14(4): 256-63.

II. Schmidt H. –M. A., Andres S., Kaakoush N. O., Engstrand L., Eriksson L., Goh K. L., Fock K. M., Hilmi I., Dhamodaran S., Forman D. and Mitchell H. The prevalence of the duodenal ulcer promoting gene (dupA) in Helicobacter pylori isolates varies by ethnic group and is not universally associated with disease development: a case- control study. Gut Pathogens, 2009 Mar; 11; 1(1):5.

III. Schmidt H. –M. A., Andres S., Kovach Z., Nilsson C., Kaakoush N. O., Engstrand L., Goh K. L., Fock K. M., Forman D. and Mitchell H. The cag PAI is intact and functional but HP0521 varies significantly in Helicobacter pylori isolates from Malaysia and Singapore. Eur J Clin Microbiol Infect Dis, 2010 Apr; 29(4):439-51.

IV. Andres S., Schmidt H. –M. A., Mitchell H., Rhen M., Maeurer M., Engstrand L. Helicobacter pylori defines local immune response through interaction with dendritic cells. Accepted FEMS Immunol Med Microbiol. DOI: 10.1111/j.1574- 695X.2010.00761.x

V. Schmidt H. –M. A., Ha M. D., Taylor F., Kovach Z., Goh K. L., Fock K. M., Barret J. H., Forman D. and Mitchell H. Variation in human genetic polymorphisms, their association with H pylori acquisition and gastric cancer in a multi-ethnic country. Submitted to J Gastroenterol Hepatol.

Contents

1 Introduction ...... 1 1.1 History...... 1 1.2 Microbiology ...... 2 1.3 Epidemiology...... 3 1.3.1 Prevalence ...... 3 1.3.2 Transmission...... 4 1.3.3 Human migrations and H. pylori...... 5 1.3.4 Low Prevalence populations ...... 6 1.4 Diagnosis and Treatment...... 6 1.4.1 Diagnosis of H. pylori infection ...... 6 1.4.2 H. pylori eradication therapy ...... 7 1.4.3 Mass screening programs ...... 8 1.5 H. pylori–associated diseases of the stomach...... 8 1.5.1 Gastritis...... 8 1.5.2 Functional dyspepsia...... 9 1.5.3 Peptic ulcer disease ...... 9 1.5.4 Gastric cancer...... 12 1.5.5 The enigmas ...... 14 1.6 Immune response to H. pylori ...... 14 1.7 Host genetics and H. pylori-associated disease...... 17 1.8 Environmental factors and H. pylori-associated disease...... 20 1.9 H. pylori genes and virulence factors...... 21 1.9.1 The H. pylori genome ...... 21 1.9.2 Colonisation ...... 22 1.9.3 Attachment and outer membrane proteins ...... 22 1.9.4 Vacuolating cytotoxin...... 24 1.9.5 CagA and the cag PAI...... 25 1.9.6 Duodenal ulcer promoting gene ...... 31 1.10 Aims...... 32

1.11 References...... 34 2 Publications...... 65 2.1 Paper I...... 66 2.2 Paper II ...... 74 2.3 Paper III...... 82 2.4 Paper IV...... 95 2.5 Paper V ...... 107 3 Concluding Remarks and Recommendations ...... 119 4 Appendix...... 123 4.1 Author contributions to publications included in this thesis...... 123

List of Abbreviations

AGS cells human gastric adenocarcinoma cells

ASR age standardised rate bab blood group antigen binding adhesion bp base pairs cag cytotoxin associated gene

CFU colony forming units

Csk C-terminal Src kinase

DC dendritic cell

DNA deoxyribonucleic acid

DU duodenal ulcer dup duodenal ulcer promoting gene

ELISA enzyme-linked immunosorbant assay

EPIYA glutamic acid-proline-isoleucine-tyrosine-alanine

FD functional dyspepsia

G+C guanine plus cytosine

GC gastric cancer

GU gastric ulcer

Ig immunoglobulin

IL interleukin

IFN interferon

IS insertion sequence

Kb kilo base pairs

LPS lipopolysaccharides

Le Lewis

MOI multiplicity of infection

NFκB nuclear factor kappa-light-chain-enhancer of activated B cells oip outer membrane inflammatory protein

OMP outer membrane protein

ORF open reading frames

PAI pathogenicity island

PCR polymerase chain reaction

PUD peptic ulcer disease sab sialic acid binding

SH2 Src homology 2

SHP2 SH2 domain-containing tyrosine phosphatase

TFSS type four secretion system

Th T helper

TLR Toll-like receptor

TNF tumor necrosis factor

UBT urea breath test vac vacuolating cytotoxin

1 Introduction 1.1 History

The human stomach was long believed to be a sterile environment, due primarily to the low pH of gastric acid. Although some isolated reports had been made regarding the existence of bacteria in the stomach, dating back to the 1800’s, a few even proposing tentative links to the development of gastric disease (143), it was not until 1982 that the major breakthrough in the understanding gastric disease occurred (64, 87).

In an observational ‘letter to the editor’ and followed up in a landmark study, Barry Marshall and J. Robin Warren, reported the presence of ‘unidentified curved bacilli’ associated with antral gastritis (190, 297). In their study of 100 consecutive patients undergoing endoscopy, they successfully isolated this novel bacterium, now known as Helicobacter pylori, from 11 patients with chronic gastritis and identified it by histology in 58 patients, including almost all with chronic gastritis, as well as peptic ulcer disease (PUD) (which includes duodenal ulcer (DU) and gastric ulcer (GU)) (190). Their study demonstrated a clear association between this novel bacterium and antral gastritis, the presence of inflammation and the novel bacterium almost always being coincident, further suggesting that the novel bacterium was not a commensal microorganism (190). With enormous forethought, the authors proposed that this bacterium may have a causal role in other gastritis-associated clinical outcomes such as PUD and gastric cancer (GC) (189-190, 297). The novel bacterium, they noted, was present in close association with the gastric epithelium, possibly being protected from the harsh gastric acid by the mucous layer, suggesting that it colonized the gastric epithelium, rather than existing in the stomach transiently, although how this was accomplished remained very much a mystery (297).

These propositions, though compelling, were in stark contrast to the established wisdom of the day and were met with intense scepticism, particularly from the medical profession. As validation, Marshall successfully fulfilled Koch’s postulates by ingesting 109 CFU of the so-called pyloric Campylobacters, originally isolated from a patient with gastritis (187). Within 10 days, Marshall, who prior to ingestion had normal histology and no gastric symptoms, developed acute gastritis, confirmed by histology, and importantly these pyloric Campylobacters were re-isolated from the affected areas of the stomach (187).

1 Through their vision in recognizing the importance of what they had discovered and their perseverance in the face of forceful opposition in promoting their discovery, Marshall and Warren were awarded the Nobel Prize in Physiology and Medicine in 2005.

1.2 Microbiology

Although H. pylori was initially classified as a member of the Campylobacter genus, due to their similarity in morphology, DNA base composition and atmospheric requirements, differences from estavlished Campylobacter spp. in terms of flagellar morphology were noted (190, 297). While established Campylobacters possessed a single unsheathed polar flagellum at one or both ends of the cell, C. pylori was found to possess up to 5 flagellae at one end of the cell (297). The fatty acids and major protein bands were also found to differ from other Campylobacters. In 1989, H. pylori was reclassified into a new genus, Helicobacter, primarily based on sequence analysis of the 16s rRNA (1).

Helicobacter belongs to the family Helicobacteraceae, order Campylobacterales and the Epsilonproteobacteria class. Several key features were ascribed to this new genus to allow differentiation from the Campylobacteraceae, namely, motility by means of sheathed flagellae, 35-44% G+C content, in vitro production of external glycocalyx in liquid media, and menaquinone-6 present as the major isoprenoid quinone (219).

Helicobacters are slow-growing microaerophiles, growing optimally at 37°C on a rich medium containing blood (219). These motile Gram-negative bacteria are non-spore forming. H. pylori is oxidase, catalase and urease positive, measuring approximately 0.5um in diameter and 2.5um in length, in the form of smooth coated short spirals of 1 to 2 wavelengths (190, 297).

Currently, at least 32 Helicobacter species have been identified in humans and animals (http://www.bacterio.cict.fr/h/helicobacter.html). These are generally divided into two groups based on the niche they colonise: ‘gastric’ Helicobacters primarily colonise the stomach whilst ‘enterohepatic’ Helicobacters colonise the intestine and hepatobiliary system.

H. pylori is the only known bacterium to persistently colonise the normal human stomach. This bacterium is not an acidophile, but is able to withstand the low pH of

2 gastric acid (pH 2) only through the production urease. The enzyme urease converts urea to ammonia and carbon dioxide, which buffers the acidic gastric environment. Colonisation is assisted by 4-8 bipolar flagellae which produce a corkscrew-like motility enabling H. pylori to penetrate the viscous mucous layer and move toward the gastric epithelium where the pH is closer to neutral (261).

1.3 Epidemiology

1.3.1 Prevalence

Current figures show that H. pylori infects more than half of the world’s population (16, 65). However, the local prevalence can vary considerably (Figure 1.1). While, 70-90% of developing world is infected, the prevalence drops to 25-50% in many developed countries (149). The prevalence in Australia, for example, has been reported to be as low as 21% (200), while in India, the prevalence rates are as high as 87% (205). H. pylori prevalence may also vary significantly within a single country (105, 202). For example, H. pylori prevalence varies considerably between the three primary ethnic groups resident in Malaysia, with Malays, Chinese and Indians reported to have prevalence rates of 11.9-29.2%, 26.7-57.5% and 49.4-52.3% respectively (97).

Numerous studies have shown an inverse correlation between socioeconomic factors such as education, density of living, household hygiene and income, and the seroprevalence of H. pylori (40, 69, 174, 181, 299-300)(230). Indeed, the incidence in most developing countries has been steadily declining over the past few decades as such factors have improved (60). For example, a recent study in Malaysia demonstrated a reduction in the prevalence of H. pylori in dyspeptic patients (12.1% Malay, 21.7% Chinese and 30.7% Indian) (184) as compared with a previous study at the same centre 8 years previously (16.4% Malay, 48.5% Chinese and 61.8% Indians) (95).

3 10% 30% 30% 70% 50% 50% 40% 70% 90% 70% 70% 70% 90% 80% 20% 80%

Figure 1.1: Schematic representation of the world-wide distribution of H. pylori infection.

1.3.2 Transmission

Despite more than two decades of research, the transmission and spread of H. pylori is poorly understood, partially because of the difficulty of obtaining isolates and the extensive genetic diversity between isolates, which is suggestive of non-epidemic spread. The majority of H. pylori infections are acquired in early childhood. Compelling evidence of intrafamilial transmission, particularly parent to child, has been provided by numerous studies showing an intrafamilial clustering of infection (68, 148, 204, 210, 230, 246, 276, 294). Indeed, a study involving DNA fingerprinting in Japan demonstrated that the majority of mothers and their children were infected with the same strain (151). A study of families from rural South Africa and urban United Kingdom, United States, Columbia and Korea by Schwarz et al. also supported a central role for intrafamilial transmission within urban families who had restricted physical contact outside the immediate family as well as good sanitation (246). Based on their study, the authors proposed that in high prevalence countries interfamilial horizontal transmission played a greater role (246). Horizontal transmission has a been observed in institutionalized patients (32) and via endoscopes (262), however, infection of adults probably plays a minor role in the incidence of H. pylori.

H. pylori is likely spread by a combination of the faecal-oral and oral-oral routes, yet a stable niche outside the human stomach has not been found. While in vivo animal

4 experiments have supported oral-oral over faecal-oral transmission (159-160, 259), studies in humans have demonstrated that both transmission routes are possible. Megraud et al. found a higher risk of H. pylori infection associated with premastication of food when mothers’ fed their children (193). There is also suggestion that gastric juice may also play a role, potentially through intermittent reflux which may introduce H. pylori into the mouth (315) has been supported by the detection of transcriptionally active H. pylori and successful isolation of a viable H. pylori strain from vomitus (136, 162). Perhaps the strongest evidence for the faecal-oral transmission route comes from a study by Parsonnet et al. who found that while H. pylori could not be cultured from normal stools of H. pylori-positive patients, it could be isolated from 50% of patient’s stools who were given a cathartic to induce diarrhoea (224).

1.3.3 Human migrations and H. pylori

Due to a high mutation and recombination frequency, and natural competence H. pylori is genetically diverse (3, 265-266). The genetic diversity decreases with increasing geographic distance from East Africa, suggesting that anatomically modern humans became infected with H. pylori before the migrations out of Africa and that this bacterium has remained intimately associated with humans ever since (169). H. pylori can be divided into genotypes with distinct geographic distributions and can therefore be used to identify the region of origin of the host even after centuries of coexistence in individual geographic locations (50, 95, 181).

Based on MLST of seven housekeeping genes and the virulence factor vacA, Falush et al. demonstrated that there are four major modern populations of H. pylori, hpAfrica 1, hpAfrica 2, hpEurope and hpEastAsia (78), which further subdivided. The hpEastAsia contains hspAmerInd, hspMaori and hspAsia, and hpEurope includes strains from Europe, Turkey, Bangladesh, Ladakha (India), Sudan and Israel (78). Interestingly, isolates from Europe form a heterogeneous population with the modern hpEurope population being mostly a combination between the two ancestral European populations which likely arrived in Europe in different waves (78). The agreement between known models of human migration and H. pylori evolution, as well as agreement between H. pylori population trees and language evolution, for example (234), suggests that H. pylori has the potential to resolve questions on contentious aspects of human migrations (78).

5 1.3.4 Low Prevalence populations

One of the enigmas in the study of H. pylori is the existence of populations which have unexpectedly low prevalence of H. pylori, such as Pemba Island residents in Zanzibar (79), ethnic Malays resident in Malaysia (287) and ethnic Javanese living in Indonesia (277). It has been proposed that such populations passed through bottlenecks with H. pylori-free founding populations and that acquisition of H. pylori has been a recent event (107).

In a study of women living on Pemba Island, Zanzibar Farag et al. reported an H. pylori prevalence of 17.5%, which is both unexpectedly low for a developing country and considerably lower than nearby countries such as Tanzania, Zambia, Congo and Kenya (79). Interestingly, the investigators noted a spatial distribution in prevalence with the highest rates of infection closest to the port and that interestingly the prevalence of infection decreased with age, suggesting that H. pylori infection is a recent phenomenon in this population and is likely being imported from the Tanzanian mainland (79).

The prevalence of H. pylori in Malays has also sparked enormous interest. Initially, it was noted that Malays had a very low rate of both PUD and GC (101). Goh demonstrated that there was a significant difference in the prevalence of H. pylori between Indians, Chinese and ethnic Malays, with ethnic Malays having an unusually low prevalence (95). The ‘racial cohort phenomenon’ is a term first used by Goh and Parasakthi to describe the racial differences in Malaysia, attributing the differences in part to the low rate of intermarriage which resulted in H. pylori strains remaining confined to single racial groups (97). Indeed, investigations by our group have demonstrated that strains carried by Chinese and Indians are distinct and that the strains carried by Malays as a race are typically a mixture of Indian and Chinese type strains, indicating that these strains are introduced (245, 274).

1.4 Diagnosis and Treatment

1.4.1 Diagnosis of H. pylori infection

The vast majority of H. pylori infections are both asymptomatic and occur in early childhood, which means the infection is often undiagnosed and rarely diagnosed during the acute phase. The gold standard of diagnosis is culture, however, while 6 culture is 100% specific, it lacks sensitivity and requires the collection of a biopsy through endoscopy which is invasive and expensive, and thus is usually reserved for symptomatic patients (43). In the clinical setting, diagnosis is usually made on the basis of the non-invasive ‘gold standard’ urea breath test (UBT), developed by Graham et al. (103) which exploits the endogenous urease activity of H. pylori (84).

For epidemiological studies it is very difficult to obtain both ethics approval and volunteer patients for endoscopy-related procedures. The UBT is also unsuitable in many cases for epidemiological studies as it is difficult to administer and is relatively expensive. In these settings, both commercial and in-house enzyme-linked immunosorbant assays (ELISAs) and immunoblots (western blots) are used frequently (84), as sample collection (sera) is less invasive, the assays can be performed rapidly, easily and inexpensively, and with high throughput, enabling the processing of large numbers of specimens (43). However, limitations exist. A study by Everhart et al. questioned the interassay reproducibility of ELISAs for the diagnosis of H. pylori in a study of patients from different ethnic groups in the US, this study showing that the results of repeat testing of sera from a single patient could differ by as much as 30% (76). Furthermore, as these assays usually rely on the presence of IgG, they are unable to differentiate between previous and current infection (43). However, given that H. pylori infection is usually life-long in the absence of specific treatment, this is not a major draw back. Such assays however are unsuitable to assess the efficacy of H. pylori eradication therapy as antibody titres may remain elevated for a considerable time following treatment, the UBT is used in this context (84).

1.4.2 H. pylori eradication therapy

Once established, H. pylori infection is generally life-long without antibiotic intervention. H. pylori is intrinsically resistant to only a few antimicrobials. Indeed, in 1985 Marshall et al. reported that all isolates were sensitive to penicillin, erythromycin, tetracycline, cephalosporins, gentamicin and bismuth citrate, and that 80% were sensitive to metronidazole or tinidazole (189). Treatment of H. pylori by proton-pump inhibitor (PPI)-based triple therapy, the combinational use of a PPI such as omeprazole with two antibiotics (amoxicillin, clarithromycin and metranidazole), was first proposed almost simultaneously in 1993 by two groups (30, 157). This is still a commonly used first-line treatment regime. Treatment failure occurs in up to 20% of

7 cases (269) and is increasing due to increasing resistance to both clarithromycin and metranidazole (2, 84). Alternative first-line treatment regimes are emerging, including sequential therapy regimes where a PPI and amoxicillin are used for the first 5 days, metronidazole and clarithromycin replacing amoxicillin for a further 5 days (192). This regime is intended to reduce the bacterial load prior to metronidazole and clarithromycin treatment, to reduce the likelihood of resistance developing (192). Second-line therapies may be PPI and/or bismuth-based and usually include a combination of two of the following antibiotics: amoxicillin, metranidazole, tetracycline, rifabutin and furazolidone (192).

1.4.3 Mass screening programs

The ‘Second Asia-Pacific consensus guidelines for Helicobacter pylori infection’ recommend the screening and treatment of H. pylori in communities with a high risk of GC as an effective way to prevent GC (84). However, the majority of infections occur in developing countries that may not be able to afford a comprehensive screening program. There is also controversial evidence on the cost-effectiveness of any such program. A study of ethnic Chinese resident in Singapore, representing a high prevalence, high GC risk group, did not find it cost effective to screen adults aged 40 year using either serology or the UBT (303). However, a study in mainland china demonstrated that screening of adults aged 20 years was cost effective and was likely to prevent 1 in every 4 to 6 cases of GC (313). Studies in countries with a low H. pylori prevalence and low GC rates have also demonstrated a lack of cost effectiveness. For example, a study in Denmark observed only a modest though insignificant effect on the rate of dyspepsia although a modest, significant, effect of the consultation rate and the number of sick leave days taken for dyspepsia was also observed (114).

1.5 H. pylori–associated diseases of the stomach

1.5.1 Gastritis

H. pylori is not found in histologically normal mucosa (189), rather its presence is associated with inflammation of the gastric mucosa (gastritis). The acute phase of infection is subclinical making it difficult to determine exactly when the infection was acquired, however, an acute gastritis which affects the entire stomach manifests within 8 a few weeks. Adult volunteers experimentally infected with H. pylori have been reported to experience fever, vomiting, nausea and mucosal inflammation, followed by cramping, gastric pain, and the development of hypochlorhydria (106, 187, 207, 258). Although spontaneous eradication has been observed in children, the majority of individuals fail to eradicate the infection due to an ineffective immune response. If H. pylori is not eliminated, the acute gastritis progresses to chronic gastritis. The majority of H. pylori infected individuals are asymptomatic despite the ongoing gastritis (108). Of those infected, 10-20% will go on to develop severe gastroduodenal disease, which is determined by the pattern and severity of gastritis (Figure 1.2).

1.5.2 Functional dyspepsia

Functional dyspepsia (FD), also called non-ulcer dyspepsia, has been defined as the presence of symptoms such as pain, discomfort, nausea, vomiting and bloating, originating in the gastroduodenal region in the absence of any structural focus or lesion likely to be responsible for it such as malignancy, ulcer or gastroespohageal reflux, as confirmed by endoscopy (27, 63, 218). Between 30% and 93% of people diagnosed with FD have H. pylori-associated gastritis, and although no causal link has been reliably established between H. pylori infection and symptomatology, studies have shown that H. pylori eradication is associated with a reduction or elimination of dyspeptic symptoms in a percentage of patients (27, 206, 254-255, 271).

1.5.3 Peptic ulcer disease

The stomach and intestines are protected from damage by gastric acid, microbes and toxic compounds by a mucous layer and the secretion of bicarbonate. When these protective factors are impaired, gastric acid can access the underlying mucosa predisposing to peptic ulcer disease (PUD) which may involve the stomach (GU) or the duodenum (DU). H. pylori is associated with approximately 95% of DU cases and 80% of GU cases (154, 214) and there is strong evidence for a causal role of H. pylori in PUD, based on the finding that the use of antibiotics to eradicate H. pylori dramatically reduces the recurrence rate of PUD (85, 104).

9 H. Pylori infection

Acute Gastritis Spontaneous Clearance

Chronic Gastritis

Antrum Corpus predominant gastritis No Symptoms predominant gastritis

High acid production Suppressed acid production

Gastric Ulcer Gastric metaplasia Atrophy in duodenum

Duodenal ulcer Intestinal metaplasia in stomach

Dysplasia

Gastric adenocarcinoma

Figure 1.2: Classical pathways for the development of H. pylori-associated diseases of the stomach

The level of acid secretion by parietal cells in the corpus of the stomach has been reported to be a determining factor in the development of PUD (158). In DU, individuals have intrinsically high acid production and the acute stage pan-gastritis gradually shifts to an antral-predominant gastritis, whereas GU and GC are associated with lower acid production and corpus predominant gastritis (Figure 1.3) (250). These two pathways are mutually exclusive.

10 Cardiac Oesophagus Sphincter

Fundus

Pyloric Sphincter

Duodenum

Corpus

Pylorus

Antrum

Figure 1.3: A schematic illustration of the human stomach.

In high acid secretors there is an increased output of acid to the duodenum, which triggers the replacement of intestinal cells with gastric epithelial cells, known as gastric metaplasia. H. pylori is unable to colonise the normal duodenal mucosa, but areas of gastric metaplasia enables H. pylori to establish an infection leading to further inflammation that may result in duodenal ulcer (37, 112). Intriguingly, being of the Indian race is an independent risk factor for DU development (95). However, the reasons for this predisposition, beyond a higher level of acid secretion, have not been adequately explained given that a minority of H. pylori infected individuals develop DU.

In individuals with a lower acid output H. pylori is able to colonise the antrum and corpus, resulting in a corpus-predominant gastritis or diffuse pan-gastritis. It has been shown that lower acid output is partially a response to H. pylori induced inflammatory mediators such as the cytokine IL-1β which has an inhibitory effects on acid secretion by parietal cells (71). H. pylori colonization damages the gastric mucosa directly through bacterial products such as cytotoxins and indirectly through an elevated immune response including the production of reactive nitrogen and oxygen species. This damage increases the susceptibility of the gastric mucosa to the effects of gastric acid and thus predisposes to gastric ulceration.

11 1.5.4 Gastric cancer

GC is a significant public health burden. Although there has been a notable decline, GC is the fourth most common cancer by incidence with 900,000 new cases in 2002; only lung, breast, colon and rectal cancers having higher rates (221). Due to the poor prognosis, GC is the second most common cause of cancer-related mortality, after lung cancer, with approximately 700 000 deaths annually (80, 221). Furthermore, the anticipated growth in world population combined with increased longevity is predicted to result in a net increase in the overall number of GC cases for several decades to come (6).

There is convincing evidence that H. pylori infection is the primary predisposing factor in non-cardia GC. In 1994, the World Health Organisation’s international agency for research on cancer (IARC) classified H. pylori as a class-one, i.e. definite, carcinogen, based on the strength of sero-epidemiological studies (222). In a study of 145 people, Blaser et al. reported odds ratios (OR) of 2.77 (95% CI 1.04 -7.97), whilst in a study of 287 people, Talley et al. reported an OR 2.67 (95% CI 1.01-7.06) (38, 272). In 2001, Helicobacter and Cancer Collaborative Group reported an estimated six-fold increase in the risk of non-cardia GC in H. pylori infected individuals based on a meta-analysis of 12 case-control studies (118). However, many believe that the OR is falsely low, in part owing to the high prevalence of H. pylori infection in certain countries (102). The interval between diagnosis and sample collection is also likely to have played an important role in negatively influencing the incidence of H. pylori in GC patients. H. pylori may disappear spontaneously from patients progressing to GC due altered physiology and intestinal metaplasia. This may result in false-negatives by Enzyme- Linked Immunosorbant Assay (ELISA) which is the most common form of analysis in epidemiological studies (28). Evidence to support this has provided by Elkstrom et al. who demonstrated a significant increase in the odds ratio from 2.2 (95% CI 1.4 - 3.6) to 21.0 (95% CI 8.3-53.4) (69), using the more sensitive technique of immunoblotting rather than the conventional ELISA. Immunoblotting, unlike ELISA, detects responses to specific H. pylori antigens and the use of this technique has shown that many ELISA H. pylori negative GC patients were previously infected with H. pylori (69, 201). H. pylori infection is now considered the primary cause of non-cardia GC (102). Conservative estimates suggest that H. pylori is responsible at least 65% and 80% of non-cardia GC cases in developing and developed countries respectively, but recent

12 evidence from a Japanese survey suggests it may be responsible for between 89.4-98% of non-cardia GC cases (118, 141). Further evidence for the role of H. pylori in gastric carcinogenesis comes from in vivo animal models. For example, an elegant study by Romero-Gallo et al. found that antibiotic-based eradication of H. pylori in a Mongolian Gerbil model could prevent GC, but reported that the timing of treatment in relation to infection was critical for preventing the formation of pre-neoplastic lesions (240).

H. pylori-associated GC, an adenocarcinoma derived from the glandular epithelial cells, typically develops after a lifetime of H. pylori infection. There are two histologically distinct variants, intestinal and diffuse, characterized by the presence of well and poorly differentiated cells. Progression to intestinal type GC occurs through a multistep pathway that was comprehensively characterized by Correa et al. (52). During this process superficial gastritis progresses into a chronic corpus-predominant gastritis associated with persistent H. pylori infection. This may progress to atrophic gastritis, which presents with gland loss and infiltration of inflammatory cells into the glandular zones. Hypochlorhydria results from an accompanying loss of parietal cell function. Intestinal metaplasia may then develop with the normal gastric mucosa being replaced by intestinal-type epithelial cells. The final step before GC is dysplasia, a neoplasia where the cells have not yet gained the ability to invade and metastasize.

The incidence of GC varies significantly between countries, with less developed countries bearing the highest burden (80). However, the distribution of GC does not follow clear geographic trends and within the same country the prevalence can vary significantly between populations. In Malaysia, there is enormous disparity in the relative rates of GC between the three primary ethnic groups. Ethnic Chinese experience a high rate of GC (males: 11.3/100,000, females: 9.1/100,000) (166). The rate for ethnic Indians is approximately half that of Chinese (males: 5.5/100,000, females: 5.3) and for ethnic Malays it is almost one-fifth that of the Chinese (males: 2.1/100,000, females: 1.4/100,000) (166). These rates strongly reflect the trends evident with respect to the mother countries,for example in mainland China the ASR for GC in males is 41.4 while a substantially lower ASR (5.7) is reported on the Indian subcontinent (80).

13 1.5.5 The enigmas

Several investigators have reported geographical or population-associated enigmas, such as the African, Asian and Costa Rican enigmas, where the observations of a particular population or geographic region do not correlate with the expected outcomes (96, 124, 205, 256, 282). The Asian enigma refers to the phenomenon where the prevalence of H. pylori infection in countries such as Bangladesh and Pakistan is high and the rate of GC is low, whereas in countries such as Korea, Japan and China where H. pylori is also prevalent, GC is common (205). The Indian enigma, part of the Asian enigma, was first described in ethnic Indians resident in Malaysia and relates to the low prevalence of GC despite a high prevalence of H. pylori infection (96). DU is common in many regions in India and continues to be so among migrants from the Indian subcontinent (280). It is not unexpected that GU and GC are comparatively rare in India given that these outcomes are mutually exclusive. It has been suggested that differences in the pathogenesis of H. pylori infection, particularly underlying the degree and patterns of mucosal inflammation, should be sufficient to explain these enigmas (101). For example, it is known that gastric acid secretion is low in the Japanese population which may go some way in explaining the high rate of GC in comparison to Caucasian populations (205). Interestingly, the level of gastric acid secretion has been increasing over the past few decades in Japan, however, it is unclear whether this is attributable to the lower prevalence of H. pylori infection or to other factors (205). While the enigmas have been attributed to differences in traditional environmental factors such as diet (96), substantial differences in host genetics, including polymorphisms, have been postulated to account for these apparent enigmas (92).

1.6 Immune response to H. pylori

H. pylori infection invariably triggers an immune response characterised by inflammation of the gastric mucosa, however, the mechanisms by which H. pylori induces and maintains the immune response, whilst evading clearance are poorly understood. H. pylori attachment to epithelial cells induces the production of the proinflammatory chemokine interleukin(IL)-8, which triggers the infiltration and activation of lymphocytes, plasma cells, granulocytes and macrophages into the gastric mucosa (57, 293). In H. pylori infected patients elevated levels of an array of cytokines,

14 including TNF-α, IFN-γ, IL-1β, IL-2, IL-6, IL-8 and IL-12, have been described (28, 57, 168, 267).

The Toll-like receptors (TLRs) are pattern recognition receptors important in initiating the innate immune response by recognition of specific ligands such as peptidoglycan and lipoteichoic acid from Gram positives (TLR2), lipopolysaccharide (LPS) from Gram negatives (TLR4), bacterial flagella (TLR5) and unmethylated CpG oligonucleotides in bacterial DNA (TLR9) (73, 119, 291). The pattern of TLR expression varies between cell types. For example, TLR9 expression is largely restricted to monocytes, plasma cells and dendritic cells (DC), where it has a role in DC maturation (152). TLR2 is reported to be absent in gastric epithelial cells, but present in infiltrating immune cells including dendritic cells where it is the major surface receptor (186, 233). Activation of TLRs results in a signal transduction pathway involving the activation and translocation of the transcription factor nuclear factor (NF)-κB to the nucleus and the transcription of antimicrobial and pro-inflammatory cytokine and chemokine genes, leading to acute inflammation (225).

The role of specific TLRs in the recognition of H. pylori has been controversial. Smith et al. initially demonstrated that TLR2 and TLR5, but not TLR4, were required for the activation of NFκB and the secretion of cytokines in vitro (257). Mandell et al. further showed that it was TLR2 that is a critical receptor for the recognition of intact H. pylori (186). Although capable of activating TLR4 and TLR5, in comparison with other Gram- negative bacteria, H. pylori has weak agonist activity (25, 91, 186). Since H. pylori cells may be only weakly recognised by these TLRs, they are unlikely to be the major driver of the response to H. pylori and as gastric epithelial cells do not express TLR2, the substantial inflammatory cytokine response to H. pylori may only develop after the recruitment and accumulation of TLR2-expressing cells to the stomach (186).

Murine models have provided strong evidence that H. pylori induced inflammation is T-cell dependent (213). The processing and presentation of H. pylori antigen by professional antigen presenting cells such as DCs and macrophages to T helper (Th) 0 cells stimulates differentiation into Th1 or Th2 cells depending on the array of stimulating cytokines. Th2 cells secrete the anti-inflammatory and immunomodulatory cytokines IL-4, IL-5, IL-6, IL-10 and transforming growth factor(TGF)-β and are required for stimulation of B cells for antibody production, which is necessary for the eradication of extracellular pathogens. In contrast, the Th1

15 response is characterised by the production of pro-inflammatory cytokines including interferon(IFN)-γ, tumour necrosis factor(TNF)-α, IL-12 and IL-2, which serve to continuously recruit leukocytes to the gastric mucosa and activate Th1 lymphocytes, driving cell mediated immunity important in response to intracellular pathogens (208, 227). It is generally accepted that the response to H. pylori infection is skewed towards a Th1-type response which promotes gastritis and is largely ineffective at eradicating H. pylori. It has been proposed that an abnormal pattern of DC maturation could promote chronic inflammation through inappropriate activation of T cells (296). In vitro experiments have shown that stimulation of human-monocyte derived DCs by H. pylori generates a Th1-biased response (33, 111). Generation of immunoregulatory cytokines by DCs such as IL-10 may also activate regulatory T cells, suppressing Th17- specific anti-H. pylori immunity helping to mediate immune escape (140). Indeed, there is increasing evidence that the immune response may be more heterogeneous than originally believed. A recent study by Goll et al. demonstrated that there is upregulation of both Th1 and Th2 cytokines in chronic H. pylori infection and that the immune response shifted toward Th2 in the antral mucosa of patients with DU (98).

H. pylori has recently been demonstrated to induce gastric epithelial cells to secrete thymic stromal lymphopoietin, which mediates DC activation and Th2 responses, triggering B cell activation (144). However, although H. pylori is also capable of inducing a robust antibody mediated response, this does not result in eradication or confer protective immunity to prevent reinfection following successful treatment (208). The gastric niche occupied by H. pylori has been suggested to be relatively inaccessible to specific antibodies and there is evidence that such antibodies may cross react with the gastric epithelium contributing to gastritis (227). Indeed, Arnold et al. elegantly demonstrated that tolerance to H. pylori infection, mediated by regulatory T cells, rather than immunity per se, protects against gastric pre-neoplastic lesions in a murine model (12).

The chronic inflammation induced in response to H. pylori has negative effects. The continued presence of activated neutrophils, macrophages and monocytes in chronic gastritis results in the excessive production of reactive oxygen species (ROS) and the expression of inducible nitric oxide synthase (iNOS) produces reactive nitrogen species (RNS) (126, 195). Both ROS and RNS mediate damage to DNA, RNA, proteins and lipids, and activate various signal transduction pathways including those involving the

16 induction of proto-oncogenes and cell cycle regulation, contributing to carcinogenesis (126). RNS’ additional roles include immunosuppression (161), prevention of apoptosis (150), p53 accumulation and post-transciptional modification (123), and activation of cyclooxygenase(COX)-2 (147).

Alterations in genes related to the immune response to H. pylori could predispose individuals to H. pylori infection and promote carcinogenesis.

1.7 Host genetics and H. pylori-associated disease

Despite the high prevalence of H. pylori, only a fraction of those infected will develop disease, with different mutually exclusive outcomes (115, 158, 285). Thus, while H. pylori infection is established as a necessary cause of almost all cases of GC, it is not a sufficient cause per se (298). Diversity in H. pylori strains has been proposed to have a role; however, the high frequency of virulence factors in strains in some populations means that this cannot be the exclusive explanation. Indeed, a family history of GC is reported to be associated with a high risk of GC. In family studies, first degree relatives of patients with GC have a two to three fold increased risk of GC not explained by familial clustering of H. pylori (238), suggesting that both host genetic polymorphisms and environmental factors may have a role in predisposition to disease development.

The possibility that genetic polymorphisms could influence GC risk was first proposed by Ishizaki et al. who reported correlation of L-myc genotypes with respect to both GC and breast cancer (132). Since that report the interest in the association between genetic polymorphisms and the risk of GC has increased exponentially, and polymorphisms in genes encoding primary cytokine mediators of the immune response, including interleukin(IL)-1β and tumour necrosis factor (TNF)-α, have been shown to influence the production of other cytokines, and have been tentatively linked with the eventual development of GC (62, 71-72). El-Omar et al. first described that polymorphisms in IL- 1β may predispose to atrophic gastritis and thereby increase the risk for GC development (72), findings supported by some investigators, but not others (81, 180). IL-1β, a Th1-type pro-inflammatory cytokine, is triggered in response to H. pylori infection, and is believed to play a major role in the inflammatory response through the stimulation of other cytokines and inducing the proliferation of epithelial cells. IL-1β is

17 also a potent acid suppressor which acts on parietal cells directly and indirectly to inhibit gastric acid secretion facilitating the spread of H. pylori from the antrum to the corpus (72, 237). Variation in the levels of IL-1β in the gastric mucosa may in this way contribute to the differences in acid secretion between infected individuals, and thereby influence the disease development pathway toward GC (low acid secretion) or DU (high acid secretion). Three polymorphisms in the IL1B gene have been observed to increase expression of IL-1β, IL1B-31C, -511T and +3954T (93). In subjects from Scotland and Poland, El-Omar et al. reported that the IL1B-31 C allele and homozygous carriers of the IL1RN*2 allele were associated with an increased risk of GC (72). A subsequent study by this group demonstrated a greater than 20-fold increased risk of GC associated with the presence of three or more pro-inflammatory polymorphisms in the genes of cytokines IL-1β, IL-1RN, IL-10 and TNF-α (70). Similarly and almost simultaneously, Machado et al. reported an OR of 5.8 and 9.7 for the association between the presence of three or more pro-inflammatory cytokine polymorphisms and the GC-precursor chronic atrophic gastritis and GC respectively (179).

TNF-α is upregulated in the gastric mucosa of H. pylori infected individuals being involved in the immune response to H. pylori infection. This cytokine has contradictory effects on gastric acid secretion. Gastric acid suppression is achieved through the induction of apoptosis in enterochomaffin like cells resulting in decreased histamine release, whilst induction of gastrin release stimulates gastric acid secretion by parietal cells. Several polymorphisms in the promoter region of TNFA have been described, the -308 G/A polymorphism is the most intensively studied. The -308AA genotype has been associated with a statistically significant net increase in the risk of GC in two recent meta-analyses, despite a degree of heterogeneity in individual studies (100, 316). A study by Canedo et al. sought to explain this heterogeneity, proposing that the association between the TNFA-308A haplotype and GC was dependent on linkage disequilibrium with an as yet unidentified locus (46).

Polymorphisms in several other interleukin genes have been associated with GC risk. IL-8 is important in the initiation and amplification of acute inflammation through recruitment of immune cells, including polymorphonuclearcytes, which sustain chronic inflammation (117). Fifteen polymorphisms in IL8 have been characterised (93). The -251 T/A polymorphism in the transcription start site increases promoter activity and resultant production of IL-8 (125). While Garza-Gonzalez et al. have reported an

18 association between this polymorphism and distal GC in a Mexican population (90), Canedo et al. found no significant association with GC in a Portuguese population (44). Based on a retrospective analysis of published data, they further reported that the association between this polymorphism and disease is likely to be ethnicity specific (44). In agreement with this, a recent meta-analysis by Liu et al., has also demonstrated that this association was not consistent across studies (173). The anti-inflammatory Th2-type cytokine gene polymorphisms of IL4-168C and -590T alleles have been associated with a decreased risk of GC and the IL6-174GG genotype was associated with an increased risk of GC, while the IL2-330G and IL4-590T alleles have been associated with a decreased risk of PUD (268). Polymorphisms in immune regulatory cytokines such as IFN-γ and IL-10, cytokines which have opposite effects in response to H. pylori infection, may also have impact upon H. pylori-associated disease development. The finding that the IFNGR1-56T allele results in a 10-fold increase in gene expression in vitro adds biological plausibility to the association between this polymorphism and early onset GC (45). Alleles in genes coding for both cytokines have been associated with higher transcription levels of the cytokines and may thereby affect the nature of the inflammatory response and the Th skew (28).

Polymorphisms in many other genes, including those involved in the immune response, oxidative stress, cell-cycle regulation and cellular receptor and signalling molecules, have been implicated in GC development including PTGS2-1195A (171, 317), SOD2+47 (Val16Ala) (284), AGER+242 (Gly82Ser) (109), NFkB-94 ATTG (174), TP53 +215 (Arg72Pro) (89), ERBB2+1963 (Ile655Val) (155) and multiple TLR genes (73).

Although investigators have found significant relationships between host genetic polymorphisms and H. pylori-associated disease development, many others have obtained conflicting data that suggest no effect of the proposed polymorphism. It is likely that this is due to several factors involving the multi-factorial nature of H. pylori related disease development. Indeed, there is strong evidence from meta-analyses that the risk associated with a specific polymorphism may vary by ethnicity (89, 100, 173, 268). It is possible therefore that such genetic differences could partially explain differences in GC development between ethnic groups, such as those resident within Malaysia and Singapore. From a practical view point, host genetic polymorphisms may be used to identify individuals and subpopulations at higher risk of GC and other H. pylori-associated disease, which could eventually lead to more effective primary

19 intervention including targeted screening and eradication therapy programs. However, to date, no polymorphisms have been found which are specific or sensitive enough to form the basis of a screening program (47). However host genetic polymorphisms in combination with H. pylori virulence factors may provide a more complete screening tool. In a study of IL1B polymorphisms and H. pylori virulence factors, Figeuiredo et al. demonstrated that the odds for GC increased dramatically in carriers of both bacterial- and host-high risk factors (cagA and vacAs1m1, and IL1B-511T and IL1RN*2) (81). For example, carriers of IL1B-511T and vacAs1 had an OR of 87 (95% CI: 11, 679), considerably higher than either IL1B-511T (OR 3.3, 95% CI: 1.3,8.2) or vacAs1 (OR 17, 95% CI: 7.8,38) assessed independently (81).

1.8 Environmental factors and H. pylori-associated disease

The consumption of alcohol, highly salted, pickled, fermented or smoked foods, exposure to metal and cement dust and cigarette smoking have all been attributed to an increased risk of GC development (129, 175). A recent in vivo study demonstrated convincingly that GC was induced in H. pylori-infected rhesus macaques in association with the consumption of nitrosamines, such as those commonly found in pickled vegetables (172). N-nitroso compounds including nitrosamines are probable human carcinogens (130, 197) and may also be found in nitrate cured meat, alcoholic beverages, tobacco products and created in the human stomach (199). High salt intake may also predispose to GC directly damaging the gastric mucus, improving temporary epithelial cell proliferatuo and the incidence of endogenous mutations (295).

Intake of fresh fruit and vegetables, particularly fruit, has been found to be inversely correlated with GC risk (178, 281). For example, a case-control study in Hawaii found a significant trend between GC and yellow vegetables in the presence of H. pylori infection, and with total vegetable intake when the presence of the H. pylori virulence factor, CagA, was considered (75). While the constituents in fruit and vegetables specifically conferring a protective effect against GC are unclear, there is increasing evidence in favour of several compounds including vitamin C playing a role (281). Vitamin C is known to inhibit the formation and action of N-nitroso compounds in the stomach (197-198) and the antioxidant capacity of this compound may be beneficial in neutralising free radicals such as the ROS formed during chronic inflammation.

1.9 H. pylori genes and virulence factors

The different disease outcomes of H. pylori infection and the fact that only a minority of individuals progress to severe disease imply that disease development is a multifactorial process. A number of virulence factors of H. pylori have been proposed to contribute to colonisation and persistent infection, and to modulate the risk of severe disease development. However, despite promising initial associations, to date none have been exclusively associated with disease development. It has been proposed that any virulence factor must not only have a correlation with an in vivo disease, but must also be biologically plausible and be epidemiologically consistent across regions and populations (177).

1.9.1 The H. pylori genome

It is thought that the presence of strain-specific genes may account for the observed variability in the clinical manifestations of H. pylori infection as has been shown in the case of pathogenic Escherichia coli strains (16, 138, 164). The first complete genome sequence of H. pylori was published in 1997, more than 10 years after H. pylori’s initial isolation (278). The release of the second complete genome in 1999, made H. pylori the first bacterial species for which two complete genomes were available for genetic comparison (4). Comparison the genomes of strains J99, isolated from a DU patient, and 26695, an extensively subcultured isolate from a GU patient, demonstrated the while the overall genome organization was similar there was pronounced inter-strain variability (4, 278). Microarray studies demonstrated that up to 18% of the H. pylori genome is composed of strain specific genes, which are predominantly located in regions such as the plasticity zones and the cag pathogenicity island (PAI) (36, 241, 243). The majority of these are unique to H. pylori coding hypothetical proteins of unknown function, some of which may have roles in virulence.

H. pylori is panmitic, having a much higher mutation and recombination rate than observed for most other bacteria (231, 266, 292). The enormous genetic diversity is in part attributable to a very high mutation rate (6.2-9.2x10-7), which has been proposed to be almost deleterious (74, 170). The high mutation rate is due to the lack of an efficient DNA mismatch repair system (35). Furthermore, horizontal gene transfer and free

21 recombination is exceptionally frequent due to an effective DNA uptake system with the result that different loci and polymorphisms within each locus are at linkage equilibrium (35, 266).

Most isolates from an individual are genetically identical even if found in different gastric locations (19, 212), although co-infection with multiple strains has been observed (300). More frequently genetic diversity within a host takes the form of closely related quasispecies, which are suggested to have been the result of evolution or genetic exchange with an unrelated strain persistently or transiently colonizing the same gastric niche due to H. pylori’s natural competence (34, 133, 153, 212).

1.9.2 Colonisation

H. pylori is the only known bacterium able to survive the harsh environment of the normal human stomach and establish a persistent infection. Factors important in this initial stage of infection include production of the enzyme urease that neutralizes the acidic pH of the stomach by converting urea to ammonia and carbon dioxide, with the subsequent uptake of H+ ions by ammonia to form ammonium. The importance of this enzyme is underlined by the fact that 15% of the protein content of H. pylori is comprised of preformed urease (188, 299). Unipolar flagella allow chemotaxis driven motility to propel the bacterium through the gastric mucous layer towards the higher pH and protection of the gastric epithelial surface (314). Colonisation is the necessary first step in H. pylori-associated disease development, but as discussed above, it does not always lead to more serious disease outcomes.

1.9.3 Attachment and outer membrane proteins

The majority of colonisating H. pylori are free living in the gastric mucus layer, with less than 1% attaching to the gastric epithelium (77). Attachment of H. pylori to the gastric epithelium is thought to be a major contributor to H. pylori persistence by providing access to nutrients and protection from gastric acid and mucus turnover (41). H. pylori outer membrane proteins (OMPs) have been proposed to be critical for adaptation to the host and persistent colonisation. Indeed, attachment is mediated by several OMPs, the best characterised of which are the Lewis(Le)b blood-group antigen binding adhesin (BabA), the outer membrane inflammatory protein (OipA) and the

22 sialic acid binding adhesin (SabA). BabA mediates strong binding of H. pylori to the fucolsylated Leb blood group antigen structure (Fucα1,2Galβ1,3[Fucα1,4]GlcNAcβ) found on the surface of gastric epithelial cells and in the gastric mucus (131, 167). OipA (HopH) and SabA (HopP) are members of the Helicobacter outer membrane porins (Hop) family of OMPs. The receptor for OipA adhesin is unknown, but in vitro mutagenesis studies have shown that OipA is involved in attachment to gastric epithelial cells (66). SabA (HopP) recognises NeuAcα2,3Galβ1,4 epitopes including sialyl-Lewis(Le)x and sialyl-Lewis(Le)a which are scarce in the normal mucosa but abundant in inflamed tissue which permits H. pylori to maintain adherence during inflammation as well as to activate neutrophils, which may contribute to H. pylori persistence (185, 270, 286).

Attachment of H. pylori is subject to sophisticated regulation via phase variation and allele switching. Phase variation is regulated by a slipped strand repair mechanism based on a number of dinucleotide CT repeats in oipA and sabA (185, 309). BabA is reported to be regulated by homologous recombination between babA and babB loci resulting in chimera formation, which can restore function to non-Leb binding strains, as well as by frame shift based phase variation (26, 232, 260). Although adherence facilitates better access to nutrients and delivery of effector molecules, the ability of H. pylori to rapidly modulate attachment through phase variation may aid H. pylori to escape from a vigorous host response as well as to adapt to the changing gastric mucosa over the lifetime of infection (50-51, 310).

The association between strains possessing these OMPs and disease development is controversial, largely due to a wide variability in the prevalence of these adhesins between populations. Surprisingly few studies have looked for an association between OipA and disease development. OipA has been positively linked with the presence of DU and GC, high H. pylori density and severe neutrophil infiltration (310). However, this association has been unable to be consistently replicated by other groups in human and animal studies (31, 59, 61, 86). There is also evidence that OipA is involved in the induction of IL-8 in gastric epithelial cells in vitro, as well as increased mucosal IL-8 in vivo, however this remains controversial (6, 66, 309-310). Similarly, for babA there is contradictory evidence for an association between strains possessing this adhesin and disease development, particularly in East Asia where babA is highly prevalent (304). A recent study by Fujimoto et al., however, found that a lack of BabA was associated with

23 milder gastric injury and lower bacterial density in isolates from Caucasians (88). This might suggest that investigators have been examining associations with BabA in the wrong way and that it is the absence of BabA that has a protective effect in H. pylori related disease development. Investigation of the role of SabA in disease development has gained momentum over the last few years, with evidence suggesting that SabA not only plays a role in disease development but that this role may be consistent across populations. For example, a study of a developing Hispanic country (Columbia) and a developed Caucasian country (USA) found a similarly increased prevalence of SabA in DU and GC in both populations(310). It has also been reported that the OFF status of sabA is associated with DU, but not GU, suggesting that SabA may be a reliable marker for specific disease outcome (61). However, a more recent study in Taiwan found no association with disease (252).

1.9.4 Vacuolating cytotoxin

The vacuolating cytotoxin, the first virulence factor identified for H. pylori, causes vacuolation of gastric epithelial cells (54, 163). A range of further effects on gastric epithelial and immune cells have been attributed to VacA, including the induction of apoptosis, alteration of antigen presentation by B cells and inhibition of T cell activation (50). Thus vacA is likely to play an important role in H. pylori persistence and pathology.

Although the vacA gene is found in virtually all H. pylori strains, only half express an active toxin and this is related to the alleles present (55, 163, 275). vacA is composed of three primary allelic segments, the signal sequence (s1 and s2), middle region (m1 and m2) and the recently identified intermediate region (i1 and i2) (13, 236). Mature VacA from toxigenic s1-type strains has a hydrophobic N terminal region which is able to insert into the host cell membrane leading to vacuolation, whereas non-toxigenic s2- type strains have an N-terminus preceded by a hydrophilic region that blocks vacuolation (191). Mosaicism in the m-region of vacA confers cell specificity (137). The s1m1 variants are the most virulent and are associated with high vacuolating activity, whereas s1m2 strains have a lower though variable vacuolating activity and s2m2 lack activity (13). The i region influences toxin binding and internalization with i1-type strains having a higher activity (236). Rhead et al. found that among Western H. pylori strains, all s1m2 vacA alleles possessed vacuolating activity and were i1-type, whereas

24 all s2m2 alleles were non-vacuolating and i2-type (236). In addition, they reported that the i-region determined the activity of the toxin in s1m2-type strains, with approximately 50% of western strains being i1-type (236).

An association between high virulence vacA type s1m1 and severe clinical outcome has been proposed (13-14, 289), yet as is so common with H. pylori virulence factors, no significant association has been observed in numerous other studies (113, 134, 308, 320). Indeed, studies in East Asia commonly report high virulence vacA alleles to be common regardless of the clinical out come of the patient (134-135). More recently several studies have reported i1 strains to be strongly associated with GC and PUD (29, 127, 217, 236), however, almost as soon as such associations were proposed, studies from Asia reported a high prevalence of i1 strains in their population, thus precluding any association with specific disease (135, 216).

In addition to the established alleles, Ogiwara et al. recently reported the variable presence an 81bp deletion between the m- and i-regions, termed the d-region (217). In their analysis, the presence of d1 (no deletion) was correlated with GC in western countries as well as neutrophil infiltration and gastric atrophy. They proposed that the d-region could prove a more useful indicator of histological inflammation and atrophy than the s, m or i-regions in western countries. The usefulness of this indicator has not yet been evaluated in East Asian countries which usually have a high prevalence of virulent genotypes.

1.9.5 CagA and the cag PAI

One of the most intensively studied of the H. pylori virulence factors is the 40kb cag pathogenicity island (PAI). PAIs are genetic elements containing virulence- associated genes. The different G+C content and codon usage to the rest of the genome, the flanking by direct repeats, the frequent presence of mobility genes and their genetic instability suggest that PAIs are acquired by horizontal transfer (110). The chromosomally encoded cag PAI fulfils most of these criteria. It encodes novel insertion sequences, such as IS605, is integrated into the glutamate racemase gene and is flanked by a 31bp direct repeat, and has a G+C content of 35%, 3-10% lower than the rest of the genome (3-4, 48, 278).

25 An intact cag PAI containing 27 to 32 genes encodes the effector protein CagA, a type IV secretion system (TFSS) and several hypothetical proteins with unknown functions (Figure 1.4) (18, 40, 48, 212). The CagA protein was the first component of the cag PAI to be discovered, it’s identification and characterisation in the early 1990’s by three independent groups pre-dated the cag PAI itself (53, 56, 58). Early studies observed that the highly immunogenic CagA was present in only 60% of isolates, that antibodies to CagA were higher in patients with PUD and that a correlation existed with vacuolating activity (53, 56, 58). Subsequent analysis of the latter observation confirmed an association with cytotoxin production, but found that it was not mediated by the gene producted itself, leading to the name cytotoxin-associated gene A (cagA). H. pylori strains may be classified into two major types based on this association; type I strains possess cagA and vacuolating cytotoxin activity whereas type II strains do not (302). Strong associations between type I strains and more severe inflammation, enhanced mucosal IL-8 and the development GC and PUD in H. pylori infected patients were reported early on (39, 153, 223, 228, 279). When investigators began to examine non-Western populations, however, many studies failed to replicate the association between cagA and gastroduodenal disease, finding that more than 80% of strains in these countries may carry cagA, irrespective of the disease state of the host (181-182, 194, 203, 220, 253, 312).

520 521 522 523 524 525 526 527 528 529 530 531 532 533+34 535 536 537 538 539 540 541 542 543 544 545 546 547 virD4 virB11 virB10 virB9virB8 virB7 cagL virB4 cagA cagT cagE Figure 1.4: A schematic diagram of the 40 kb cag PAI showing 27 genes including cagA (striped) and the genes essential for a functional TFSS (grey).

CagA is an effector protein that is translocated into the host cell by a functional TFSS encoded on the cag PAI (24, 215, 264). TFSSs are syringe-like structures ancestrally related to conjugation systems which facilitate the delivery of virulence factors into host cells, DNA uptake and release, and the conjugative transfer of DNA (22, 49). Several cag PAI genes show homology to TFSS components of other bacterial species, especially the well characterised VirB/D system of Agrobacterium tumefaciens which injects oncogenic T-DNA into host plant cells. Fischer et al. demonstrated that 19 genes

26 on the cag PAI, including 12 virB and virD genes and cagA, were essential for the translocation of cagA and 14 of these, but not cagA, were essential for IL-8 induction in vitro (83). Some of the functions of these genes are known. For example, cagL encodes a pilus covering protein which acts as a specialised adhesin to connect the TFSS to the host cell via integron receptors, cagF encodes a chaperon-like protein required for CagA translocation, cagE (virB4) encodes an ATPase, and the lipoprotein encoded by cagT (virB7) in combination with gene product of virB9 is postulated to play a role in stabilising the VirB proteins (24, 156, 226). In addition, cagY which has been found to produce different sized proteins as a result of frequent recombination, has been posited to have a role in immune evasion for the otherwise conserved TFSS (8).

After attachment of H. pylori to the gastric epithelial cell and TFSS-dependent translocation, CagA associates with the cell membrane where it may become phosphorylated by the Src family kinases or Abl kinases (248, 263, 273). The in vitro targets of tyrosine phosphorylation are the five amino-acid glutamic acid-proline- isoleucine-tyrosine-alanine (EPIYA) motifs (21, 248, 263). The phosphorylated EPIYA- motifs of CagA interact with the Src homology 2 (SH2) domain-containing tyrosine phosphatase (SHP-2) that is activated and deregulated (120, 122, 311). SHP-2, an intracellular signalling molecule involved in the regulation of motility, cellular proliferation and morphogenesis, is the first phosphatase found to act as a human oncoprotein (116). Deregulation of SHP-2 by CagA leads to disruption of host cell function, including the induction of a growth factor-like response, cytoskeletal rearrangements, cell scattering, cell elongation and increased cellular motility characteristic of the hummingbird phenotype in vitro (11, 37, 263). Binding of phosphorylated CagA to the SH2 domain of the intracellular signalling molecule C- terminal Src kinase (Csk) induces down-regulation of the Src family kinases, but not Abl, as well as triggering the dephosphorylation of cortacin affecting actin polymerisation (249, 273, 283). There is evidence that CagA interacts with the growth factor receptor-bound protein 2 (Grb2) which activates the Ras/MEK/ERK cellular pathway resulting in similar disruptions and morphological changes to that observed from the interaction with SHP-2 (196). Although, this is independent of the phosphorylated state of CagA, the proline-tyrosine residue region of the EPIYA-motifs was proposed to be indispensable in Grb2 binding and induction of the aforementioned cellular responses (196). CagA interacts with at least 20 cellular proteins and has several other effects on the cell (23). CagA may disrupt intercellular

27 junction functions through interaction with the epithelial tight-junction scaffolding protein ZO-1 and E-cadherin, promotes the loss of epithelial cell polarity and causes increased proliferation without a corresponding increase in apoptosis in vitro (5, 209, 229, 239, 242). There is also new evidence of a role for CagA in IL-8 induction (10, 42, 146). In an elegant study, Brandt et al. demonstrated that transfected and translocated CagA is capable of activating NFκB to induce IL-8 secretion in vitro in a time- and strain-dependent manner (42). By selectively exchanging cagA genes between high and low IL-8 inducing strains, the authors further showed that the level of IL-8 induction was determined by the cagA gene (42). Thus, CagA is a multi-functional effector protein, eminently capable of deregulating several host cell signalling pathways and disrupting normal cell function.

During the initial characterisation of CagA, Covacci et al. noted that the gene’s 3’ region was exceptionally variable, containing several repeated motifs including the EPIYA- motifs (53). There are four classes of EPIYA motif, defined by their accompanying amino-acid sequences (Figure 1.5). EPIYA-A (EPIYA(Q/K) VNKKK(T/A)GQ) and EPIYA-B (EPIY(A/T)QVAKKV) have been reported to be present in almost all CagA proteins (16). EPIYA-C (EPIYATIDDLGGPFPL) which may be absent or present in up to three copies in isolates from most western countries including Australia, is the major if not exclusive site of tyrosine phosphorylation within CagA and is required for interaction with SHP-2 (20-21, 121, 321). Greater numbers of EPIYA-C result in a higher degree of phosphorylation of CagA, stronger binding to SHP-2 and more pronounced cellular disruption and morphological changes in vitro (11, 121, 306, 318). In isolates from East Asia EPIYA-C is replaced by EPIYA-D (EPIYATIDFDEANQAG), which is the major site of phosphorylation and is necessary for interaction with SHP-2 (20, 121). EPIYA-D shows a stronger affinity for SHP-2 than does EPIYA-C and a comparison with the biological activity of these motifs demonstrated that the degree of binding to SHP-2 was greater in strains possessing EPIYA-D in comparison to those possessing three EPIYA-C motifs (121, 211). A recent study by Argent et al. further demonstrated that the EPIYA-D and greater numbers of EPIYA-C repeats significantly increased IL-8 induction in vitro (10).

28 5’ 3’

A BC

5’ 3’

A BC CC

5’ 3’

A BD

Figure 1.5: A schematic representation of cagA, illustrating examples of three variations of the EPIYA motifs. A. Classes A, B and C, 3 motifs in western ABC pattern, B. Classes A, B and C, 5 motifs in ABCCC pattern, C. Classes A, B and D, 3 motifs in East Asian ABD pattern.

In addition, a recent study by Reyes-Leon et al. reported that there was a considerable heterogeneity in the activity of strains with the same EPIYA-motifs (235). They found that an absent or modified EPIYA-B was an important contributor to the biological activity of CagA in their population (235). Indeed, the results of a data mining study by Xia et al. suggest that there is considerable variability within the EPIYA-motifs identifying several minor sequence types (variations on the four established EPIYA- motifs) (301).

Although the association in the number of EPIYA motifs and disease has been somewhat controversial, the majority of studies, including a comprehensive data- mining study, support the notion that a greater number of EPIYA motifs are more likely to be associated with enhanced histological injury and disease development (17, 29, 139, 244, 306-307). Thus, the functional differences between CagA in terms of the EPIYA-motifs could exert a considerable effect on the virulence of CagA translocating strains, and may contribute to the high prevalence of GC in countries where the virulent EPIYA-D motif is most common, such as Japan and China (11, 121, 306).

29 In addition to CagA-dependent functions, early reports noted that the cag PAI was be able to promote inflammation through enhancing the transcription of IL-8 through activation of NFκB in a CagA-independent manner (3, 48, 94, 165, 247, 251). In 2004, Viala et al. finally identified that peptidoglycan was the effector molecule responsible (290). Nod-1, an intracellular pathogen recognition molecule, was reported to be stimulated via the intracellular delivery of peptidoglycan mediated by the cag PAI encoded TFSS resulting in activation of NFκB (290). The authors further reported that Nod1 deficient mice were more susceptible to infection by strains possessing the cag PAI than wild type mice, suggesting that Nod1 is required for the control of H. pylori (290).

Not all the genes on the cag PAI have an assigned function. HP0521 is located near the 5’ end of the cag PAI, but unlike the majority of the genes in the cag PAI, this gene is neither involved in translocation of CagA nor in the induction of IL-8 (3, 83). Thus, the function of HP0521 and its role in virulence remains unknown.

There are two distinct alleles for HP0521 (40). The novel HP0521B allele, found in 54% of Swedish isolates, does not bear significant sequence similarity to any entries in the GenBank database (40). The sequence variation between HP0521 and the novel HP0521B allele is therefore likely to account for the apparent absence of HP0521 in many studies, especially in micro-array based studies where only the originally described HP0521 sequence was present on the array (34, 145). In addition to the described allelic variation of HP0521, diversity within the originally described HP0521 was also observed, when Blomstergren et al. (40) compared this region in the two completely sequenced strains, J99 and 26695. In 26695 there is a frame shift in HP0521 which is likely to disrupt translation and in J99 it contains large gaps resulting in a smaller version of the HP0521 protein (40). No association with virulence or disease has yet been reported for either HP0521 allele.

Though undeniably a virulence determinant, the cag PAI is unstable. A complete cag PAI can be inserted into or deleted from the H. pylori genome resulting in the presence of both cag PAI positive and negative isolates within a host (34, 82, 288). In addition, intermediate forms exist resulting from insertions and deletions in the cag PAI suggesting that rearrangement has occurred in strains after cag PAI acquisition (48, 142, 183). As a result of this, it has been reported that the cag PAI is disrupted in the majority of isolates from around the world (142). The deletion of the cag PAI and of

30 particular genes such as virB9, cagE (virB4) and cagT (virB7) have been reported to reduce IL-8 production in vitro, and correlate with a less severe disease presentation (15, 18, 36, 48, 83, 142, 212).

1.9.6 Duodenal ulcer promoting gene

The duodenal ulcer promoting gene (dupA) was first described by Lu et al. in a study examining 14 vir gene homologues and their association with gastroduodenal disease (176). They found that the plasticity region genes jhp0917 and jhp0918 form one continuous ORF in clinical isolates due to a 1bp insertion. Their report sparked enormous interest for three reasons 1) dupA was disease specific being associated exclusively with DU and was protective against GC or GU, 2) the association was consistent between Asian and Western populations (Korea, Japan and Columbia) and 3) the association had biological plausibility with the gene associated with IL-8 induction in vitro. While later studies in patients from Brazil, India, China, USA, South Africa, Belgium, Iraq and Iran have confirmed that jhp0917 and jhp0198 are both present and form a continuous ORF, they have been unable to consistently replicate an association with gastroduodenal disease (7, 9, 67, 99, 127, 319). A recent meta-analysis by Hussein et al. has concluded that the effects of dupA may be population specific, predisposing to DU in some populations and GC and GU in others (128).

These differences may be in part attributable to the function of dupA. The function of dupA, a virB4 homologue, is still not fully understood. In Agrobacterium tumerfaciens virB4 encodes ATPase which is part of a TFSS involved in DNA uptake/DNA transfer and protein transfer. Furthermore, the association with IL-8 secretion in vitro (176), suggests that dupA like cagE, may act in combination with other vir homologues in the plasticity zone to form a functional TFSS similar to that encoded by the cag PAI (305). There is recent data suggesting that in only some strains dupA is able to induce IL-8 in a dupA-dependant manner and that there is an association between dupA and IL-12 production (305). It is possible that some strains encode a functional dupA TFSS, whilst others do not. This may account for the substantial variation in the association between dupA and gastroduodenal disease.

31 1.10 Aims

The underlying hypothesis of this thesis is that both host and bacterial factors play a role in disease development with differences in the presence of specific virulence determinants in circulating strains and host genetic make-up may play a role in susceptibility to H. pylori infection and severe gastroduodenal disease development. Thus, the overall aims of this thesis were two-fold: to investigate the host genetics, and the genomic variation and in vitro traits of H. pylori strains isolated from subjects from different populations and disease states.

Specific aims:

As variation between the predominant H. pylori strains circulating may contribute to the disparity in disease development between different ethnicities, Paper I aimed investigate the variability of the CagA EPIYA-motifs between different ethnic groups resident in Malaysia and Singpaore and assess whether the presence of specific classes, patterns or the number of EPIYA-motifs is associated with disease development.

In addition to CagA, several other H. pylori virulence factors have been proposed to be responsible for the differences in disease development. Paper II aimed to evaluate the reported association between the putative virulence factor dupA and disease development within different ethnic groups and populations, and to evaluate the diversity and functional relevance of this gene within these populations.

As GC development in the three major ethnic groups resident in Malaysia and Singapore is unlikely to be determined solely by one H. pylori virulence factor, Paper III aimed to investigate a novel, comprehensive combination of established and putative H. pylori virulence factors for a direct association with the disease state of the host and to assess the variation between ethnic groups which might underlie differences in disease development.

H. pylori-related disease development is determined by the complex interaction between the host and H. pylori, initiated by the immune system. Therefore, Paper IV aimed to examine the impact of H. pylori strains isolated from patients with different degrees of inflammtion on dendritic cell responses, a key immunological component of the response to H. pylori, and evaluate the importance of this interaction on the inflammatory response in persistent infection.

32 Differences in the genetic make up of the host, such as gene polymorphisms, may profoundly influence gene expression and thereby the response to H. pylori infection, including the predisposition to GC. Paper V aimed to investigate the association between a panel of host genetic polymorphisms and susceptibility to H. pylori infection and/or GC development in patients from different ethnic groups resident in Malaysia and Singapore.

33 1.11 References 1. 1989. Campylobacter pylori becomes Helicobacter pylori. Lancet 334:1019-1020. 2. Agudo, A., G. Perez-Perez, T. Alarcon, and M. Lopez-Brea. 2010. High prevalence of clarithromycin resistance Helicobacter pylori strains and risk factors associated with resistance in Madrid, Spain. J Clin Microbiol 0:JCM.00144-10v1 3. Akopyants, N. S., S. W. Clifton, D. Kersulyte, J. E. Crabtree, B. E. Youree, C. A. Reece, N. O. Bukanov, E. S. Drazek, B. A. Roe, and D. E. Berg. 1998. Analyses of the cag pathogenicity island of Helicobacter pylori. Mol Microbiol 28:37-53. 4. Alm, R. A., L. S. Ling, D. T. Moir, B. L. King, E. D. Brown, P. C. Doig, D. R. Smith, B. Noonan, B. C. Guild, B. L. deJonge, G. Carmel, P. J. Tummino, A. Caruso, M. Uria-Nickelsen, D. M. Mills, C. Ives, R. Gibson, D. Merberg, S. D. Mills, Q. Jiang, D. E. Taylor, G. F. Vovis, and T. J. Trust. 1999. Genomic- sequence comparison of two unrelated isolates of the human gastric pathogen Helicobacter pylori. Nature 397:176-180. 5. Amieva, M. R., R. Vogelman, A. Covacci, L. S. Tompkins, W. J. Nelson, and S. Falkow. 2003. Disruption of the epithelial apical-junctional complex by Helicobacter pylori CagA. Science 300:1430-1434. 6. Ando, T., R. M. J. Peek, Y. C. Lee, U. Krishna, K. Kusugami, and M. J. Blaser. 2002. Host cell responses to genotypically similar Helicobacter pylori isolates from United States and Japan. Clin Diagnostic Lab Immunol 9:167-175. 7. Arachchi, H. S., V. Kalra, B. Lal, V. Bhatia, C. S. Baba, S. Chakravarthy, S. Rohatgi, P. M. Sarma, V. Mishra, B. Das, and V. Ahuja. 2007. Prevalence of duodenal ulcer promoting gene (dupA) of Helicobacter pylori in patients with duodenal ulcer in north Indian population. Helicobacter 12:591-597. 8. Aras, R. A., W. Fischer, G. I. Perez-Perez, M. Crosatti, T. Ando, R. Haas, and M. J. Blaser. 2003. Plasticity of repetitive DNA sequences within a bacterial (type IV) secretion system component. J Exp Med 198:1349-1360. 9. Argent, R. H., A. Burette, V. Y. Miendje Deyi, and J. C. Atherton. 2007. The presence of dupA in Helicobacter pylori is not significantly associated with duodenal ulcer in Belgium, South Africa, China or North America. Clin Infect Dis 45:1204-1206. 10. Argent, R. H., J. L. Hale, E. M. El-Omar, and J. C. Atherton. 2008. Differences in Helicobacter pylori CagA tyrosine phosphorylation motif patterns between

34 western and East Asian strains, and influences on interleukin-8 secretion. J Med Microbiol 57:1062-1067. 11. Argent, R. H., M. Kidd, R. J. Owen, R. J. Thomas, M. C. Lim, and J. Atherton. 2004. Determinants and consequences of different levels of CagA phosphorylation for clinical isolates of Helicobacter pylori. Gastroenterology 127:514-523. 12. Arnold, I., J. Y. Lee, M. R. Amieva, A. Roers, R. A. Flavell, T. Sparwasser, and A. Müller. 2010. Tolerance rather than immunity protects from Helicobacter pylori-induced gastric preneoplasia. Gastroenterology. 13. Atherton, J. C., P. Cao, R. M. Peek, M. K. R. Tummuru, M. J. Blaser, and T. L. Cover. 1995. Mosaicism in vacuolating cytotoxin alleles of Helicobacter pylori. Association of specific vacA types with cytotoxin production and peptic ulceration. J Biol Chem 270:17771-17777. 14. Atherton, J. C., R. M. Peek, K. T. Tham, T. L. Cover, and M. J. Blaser. 1997. Clinical and pathological importance of heterogeneity in vacA, vacuolating cytotoxin gene in Helicobacter pylori. Gastroenterology 112:92-99. 15. Audibert, C., C. Burucoa, B. Janvier, and J. L. Fauchere. 2001. Implication of the structure of the Helicobacter pylori cag pathogenicity island in induction of interleukin-8 secretion. Infect Immun 69:1625-1629. 16. Azuma, T. 2004. Helicobacter pylori CagA protein variation associated with gastric cancer in Asia. J Gastroenterol 39:97-103. 17. Azuma, T., A. Yamakawa, S. Yamazaki, K. Fukuta, M. Ohtani, Y. Ito, M. Dojo, Y. Yamazaki, and M. Kuriyama. 2002. Correlation between variation of the 3' region of the cagA gene in Helicobacter pylori and disease outcome in Japan. J Infect Dis 186:1621-1630. 18. Azuma, T., A. Yamakawa, S. Yamazaki, M. Ohtani, Y. Ito, A. Muramatsu, H. Suto, Y. Yamazaki, Y. Keida, H. Higashi, and M. Hatakeyama. 2004. Distinct diversity of the cag pathogenicity island among Helicobacter pylori strains in Japan. J Clin Microbiol 42:2508-2517. 19. Azuma, T., S. Yamazaki, A. Yamakawa, M. Ohtani, A. Muramatsu, H. Suto, Y. Ito, M. Dojo, Y. Yamazaki, M. Kuriyama, Y. Keida, H. Higashi, and M. Hatakeyama. 2004. Association between diversity in the Src homology 2 domain-containing tyrosine phosphatase binding site of Helicobacter pylori CagA protein and gastric atrophy and cancer. J Infect Dis 189:820-827.

35 20. Azuma, T., S. Yamazaki, A. Yamakawa, M. Ohtani, A. Muramatsu, H. Suto, Y. Ito, M. Dojo, Y. Yamazaki, M. Kuriyama, Y. Keida, H. Higashi, and M. Hatakeyama. 2004. Association between diversity in the Src homology 2 domain-containing tyrosine phosphatase binding site of Helicobacter pylori CagA protein and gastric atrophy and cancer. J Infect Dis 189:820-827. 21. Backert, S., S. Moese, M. Selbach, V. Brinkmann, and T. F. Meyer. 2001. Phosphorylation of tyrosine 972 of the Helicobacter pylori CagA protein is essential for induction of a scattering phenotype in gastric epithelial cells. Mol Microbiol 42:631-44. 22. Backert, S., and M. Selbach. 2008. Role of type IV secretion in Helicobacter pylori pathogenesis. Cell Microbiol 10:1573-1581. 23. Backert, S., N. Tegtmeyer, and M. Selbach. 2010. The versatility of Helicobacter pylori CagA effector protein functions: the master key hypothesis. Helicobacter 15:163-176. 24. Backert, S., E. Ziska, V. Brinkmann, U. Zimny-Arndt, A. Fauconnier, P. R. Jungblut, M. Naumann, and T. F. Meyer. 2000. Translocation of the Helicobacter pylori CagA protein in gastric epithelial cells by a type IV secretion apparatus. Cell Microbiol 2:155-164. 25. Bäckhed, F., B. Rokbi, E. Torstensson, Y. Zhao, C. Nilsson, and D. Seguin. 2003. Gastric mucosal recognition of Helicobacter pylori is independent of Toll- like receptor 4. J Infect Dis 187:829-836. 26. Bäckstrom, A., C. Lundberg, D. Kersulyte, D. E. Berg, T. Boren, and A. Arnqvist. 2004. Metastability of Helicobacter pylori bab adhesin genes and dynamics of Lewis b antigen binding. Proc Natl Acad Sci USA 101. 27. Baker, G., R. J. Fraser, and G. Young. 2006. Subtypes of functional dyspepsia. World J Gastroenterol 12:2667-2671. 28. Basso, D., and M. Plebani. 2004. H. pylori infection: bacterial virulence factors and cytokine gene polymorphisms as determinants of infection outcome. Crit Rev Clin Lab Sci 41:313-337. 29. Basso, D., C.-F. Zambon, D. P. Letley, A. Stranges, A. Marchet, J. L. Rhead, S. Shiavon, G. Guariso, M. Ceroti, D. Nitti, M. Rugge, M. Plebani, and J. C. Atherton. 2008. Clinical relevance of Helicobacter pylori cagA and vacA gene polymorphisms. Gastroenterology 135:91-99.

36 30. Bazzoli, F., R. M. Zagari, and S. Fossi. 1993. Efficacy and tolerability of a short- term low-dose triple therapy for eradication of Helicobacter pylori. Gastroenterology 104:40A. 31. Ben Mansour, K., C. Fendri, M. Zribi, A. Masmoudi, M. Labbene, A. Fillali, N. Ben Mami, T. Najjar, A. Meherzi, T. Sfar, and C. Burucoa. 2010. Prevalence of Helicobacter pylori vacA, cagA, iceA and oipA genotypes in Tunisian patients. Ann Clin Microbiol Antimicrob 9:10. 32. Berkowicz, J., and A. Lee. 1987. Person to person transmission of Campylobacter pylori. Lancet 330:680-681. 33. Bimczok, D., R. H. Clements, K. B. Waites, L. Novak, D. E. Eckhoff, P. J. Mannon, P. D. Smith, and L. E. Smythies. 2010. Human primary gastric dendritic cells induce a Th1 response to H. pylori. Mucosal Immunol 3:260-269. 34. Björkholm, B., A. Lundin, A. Sillen, K. Guillemin, N. Salama, C. Rubio, J. I. Gordon, P. Falk, and L. Engstrand. 2001. Comparison of genetic divergence and fitness between two subclones of Helicobacter pylori. Infect Immun 69:7832- 7838. 35. Björkholm, B., M. Sjölund, P. G. Falk, O. G. Berg, L. Engstrand, and D. I. Andersson. 2001. Mutation frequency and biological cost of antibiotic resistance in Helicobacter pylori. Proc Natl Acad Sci U S A 98:14607-14612. 36. Björkholm, B. M., J. L. Guruge, J. D. Oh, A. J. Syder, N. Salama, K. Guillemin, S. Falkow, C. Nilsson, P. G. Falk, L. Engstrand, and J. I. Gordon. 2002. Colonization of germ-free transgenic mice with genotyped Helicobacter pylori strains from a case-control study of gastric cancer reveals a correlation between host responses and HsdS components of type I restriction-modification systems. J Biol Chem 277:34191-34197. 37. Blaser, M. J., and J. C. Atherton. 2004. Helicobacter pylori persistence: biology and disease. J Clin Invest 113:321-333. 38. Blaser, M. J., D. G. Newell, F. Fullerton, J. W. Yarnell, A. R. Stacey, N. Wald, and F. Sitas. 1991. Association between infection with Helicobacter pylori and risk of gastric cancer: evidence from a prospective investigation. Br Med J 302:1302-1305. 39. Blaser, M. J., G. I. Perez-Perez, H. Kleanthous, T. L. Cover, R. M. Peek, P. H. Chyou, G. N. Stemmermann, and A. Nomura. 1995. Infection with Helicobacter

37 pylori strains possessing cagA is associated with an increased risk of developing adenocarcinoma of the stomach. Cancer Res 55:2111-2115. 40. Blomstergren, A., A. Lundin, C. Nilsson, L. Engstrand, and J. Lundeburg. 2004. Comparative analysis of the complete cag pathogenicity island sequence in four Helicobacter pylori isolates. Gene 328:85-93. 41. Boren, T., P. Falk, K. A. Roth, G. Larson, and S. Normark. 1993. Attachment of Helicobacter pylori to human gastric epithelium mediated by blood group antigens. Science 262:1892-1895. 42. Brandt, S., T. Kwok, R. Hartig, W. König, and S. Backert. 2005. NF-κB activation and potentiation of proinflammatory responses by the Helicobacter pylori CagA protein. Proc Natl Acad Sci USA 102:9300-9305. 43. Brown, L. M. 2000. Helicobacter pylori: Epidemiology and routes of transmission. Epidemiol Rev 22:283-297. 44. Canedo, P., A. J. Castanheira-Vale, N. Lunet, F. Pereira, C. Figueiredo, L. Gioia-Patricola, F. Canzian, H. Moreira, G. Suriano, H. Barros, F. Carneiro, R. Seruca, and J. C. Machado. 2008. The interleukin-8-251*T/*A polymorphism is not associated with risk for gastric carcinoma development in a Portuguese population. Eur J Cancer Prev 17:28-32. 45. Canedo, P., G. Corso, F. Pereira, N. Lunet, G. Suriano, C. Figueiredo, C. Pedrazzani, H. Moreira, F. Carneiro, R. Seruca, F. Roviello, and J. C. Machado. 2008. The interferon gamma receptor 1 (IFNGR1) -56C/T gene polymorphism is associated with increased risk of early gastric adenocarcinoma. Gut 57:1504- 1508. 46. Canedo, P., C. Durães, F. Pereira, G. Regalo, N. Lunet, H. Barros, F. Carneiro, R. Seruca, J. Rocha, and J. C. Machado. 2008. Tumor necrosis factor alpha extended haplotypes and risk of gastric carcinoma. Cancer Epidemiol Biomarkers Prev 17:2416-2420. 47. Canedo, P., C. Figueiredo, and J. C. Machado. 2007. After Helicobacter pylori, genetic susceptibility to gastric carcinoma revisited. Helicobacter 12 Suppl 2:45- 49. 48. Censini, S., C. Lange, Z. Xiang, J. E. Crabtree, P. Ghiara, M. Borodovsky, R. Rappuoli, and A. Covacci. 1996. cag, a pathogenicity island of Helicobacter pylori, encodes type I-specific and disease associated virulence factors. Proc Natl Acad Sci U S A 93:14648-14653.

38 49. Christie, P. J. 2001. Type IV scretion: intracellular transfer of macromolecules by systems ancestrally related to conjugation machines. Mol Microbiol 40:294- 305. 50. Clyne, M., B. Dolan, and E. P. Reeves. 2007. Bacterial factors that mediate colonization of the stomach and virulence of Helicobacter pylori. FEMS Microbiol Lett 268:135-143. 51. Colbeck, J. C., L. M. Hansen, J. M. Fong, and J. V. Solnick. 2006. Genotypic profile of the outer membrane proteins BabA and BabB in clinical isolates of Helicobacter pylori. Infect Immun 74:4375-4378. 52. Correa, P. 1992. Human Gastric Carcinogenesis: A multistep and multifactorial process - first American cancer society award lecture on cancer epidemiology and prevention. Cancer Res 52:6735-6740. 53. Covacci, A., S. Censini, M. Bugnoli, R. Petracca, D. Burroni, G. Macchia, A. Massone, E. Papini, Z. Xiang, N. Figura, and R. Rappuoli. 1993. Molecular characterisation of the 128-kDa immunodominant antigen of Helicobacter pylori associated with cytotoxicity and duodenal ulcer. Proc Natl Acad Sci U S A 90:5791-5795. 54. Cover, T., and M. Blaser. 1992. Purification and characterization of the vacuolating toxin from Helicobacter pylori. J Biol Chem 267:10570-10575. 55. Cover, T., P. Cao, C. Lind, K. Tham, and M. Blaser. 1993. Correlation between vacuolating cytotoxin production by Helicobacter pylori isolates in vitro and in vivo. Infect Immun 61:5008-5012. 56. Cover, T. L., C. P. Dooley, and M. J. Blaser. 1990. Characterization and human serologic response to proteins in Helicobacter pylori broth culture supernatants with vacuolizing cytotoxin activity. Infect Immun 58:603-610. 57. Crabtree, J. E. 1996. Gastric mucosal inflammatory responses to Helicobacter pylori. Aliment Pharmacol Ther 10:29-37. 58. Crabtree, J. E., J. D. Taylor, J. I. Wyatt, R. V. Heatley, T. M. Shallcross, D. S. Tompkins, and B. J. Rathbone. 1991. Mucosal IgA-recognition of Helicobacter 120kDa-protein, peptic ulceration, and gastric pathology. Lancet 338:332-335. 59. Dabiri, H., P. Maleknejad, Y. Yamaoka, M. M. Feizabadi, F. Jafari, M. Rezadehbashi, F. A. Nakhjavani, A. Mirsalehian, and M. R. Zali. 2009. Distribution of Helicobacter pylori cagA, cagE, oipA and vacA in different major ethnic groups in Tehran, Iran. J Gastroenterol Hepatol 24:1380-1386.

39 60. Daugule, I., and M. Rowland. 2008. Helicobacter pylori infection in children. Helicobacter 13:41-46. 61. de Jonge, R., R. G. J. Pot, R. J. L. Loffeld, A. H. M. van Vliet, E. J. Kuipers, and J. G. Kusters. 2004. The functional status of the Helicobacter pylori sabB adhesin gene as a putative marker for disease outcome. Helicobacter 9:158-164. 62. Dhar, S. K., R. K. Soni, B. K. Das, and G. Mukhopadyay. 2003. Molecular mechanisms of action of major Helicobacter pylori virulence. Mol Cell Biochem 253:201-215. 63. Dickerson, L. M., and D. A. King. 2004. Evaluation and management of nonulcer dyspepsia. Am Fam Physician 70:107-114. 64. Doenges, J. L. 1938. Spirochetes in the gastric gland of Macaucus rhesus and humans without definite history of related disease. Proc Soc Exp Med Biol 38:536-538. 65. Dooley, C. P., H. Cohen, P. L. Fitzgibbons, M. Bauer, M. D. Appleman, G. I. Perez-Perez, and M. J. Blaser. 1989. Prevalence of Helicobacter pylori infection and histologic gastritis in asymptomatic persons. N Engl J Med 321:1562-1566. 66. Dossumbekova, A., C. Prinz, J. Mages, R. Lang, J. G. Kusters, A. H. M. Van Vliet, W. Reindl, S. Backert, D. Saur, R. M. Schmid, and R. Rad. 2006. Helicobacter pylori HopH (OipA) and bacterial pathogenicity: Genetic and functional genomic analysis of hopH gene polymorphisms. J Infect Dis 194:1346- 1355. 67. Douraghi, M., M. Mohammadi, A. Oghalaie, A. Abdirad, M. Mohagheghi, M. Hosseini, H. Zeraati, A. Ghasemi, M. Esmaieli, and N. Mohajerani. 2008. dupA as a risk determinant in Helicobacter pylori infection. J Med Microbiol 57:554-562. 68. Drumm, B., G. I. Perez-Perez, M. J. Blaser, and P. M. Sherman. 1990. Intrafamilial clustering of Helicobacter pylori infection. N Engl J Med 322:359-363. 69. Ekstrom, A. M., M. Held, L. E. Hansson, L. Engstrand, and O. Nyrén. 2001. Helicobacter pylori in gastric cancer established by CagA immunoblot as a marker of past infection. Gastroenterology 121:784-791. 70. El-Omar, E., C. Rabkin, M. Gammon, T. Vaughan, H. Risch, J. Schoenberg, J. Stanford, S. Mayne, J. Goedert, W. Blot, J. Fraumeni, and W. Chow. 2003. Increased risk of noncardia gastric cancer associated with proinflammatory cytokine gene polymorphisms. Gastroenterology 124:1193 - 201.

40 71. El-Omar, E. M. 2001. The importance of IL-1B in Helicobacter pylori associated disease. Gut 48:743-747. 72. El-Omar, E. M., M. Carrington, W. H. Chow, K. E. McColl, J. H. Bream, H. A. Young, J. Herrera, J. Lissowska, C. C. Yuan, N. Rothman, G. Lanyon, M. Martin, J. F. Fraumeni, Jr., and C. S. Rabkin. 2000. Interleukin-1 polymorphisms associated with increased risk of gastric cancer.[erratum appears in Nature 2001 Jul 5;412(6842):99]. Nature 404:398-402. 73. El-Omar, E. M., M. T. Ng, and G. L. Hold. 2008. Polymorphisms in Toll-like receptor genes and risk of cancer. Oncogene 27:244-252. 74. Eppinger, M., C. Baar, B. Linz, G. Raddatz, C. Lanz, H. Keller, G. Morelli, H. Gressmann, M. Achtman, and S. C. Schuster. 2006. Who ate whom? Adaptive Helicobacter genomic changes that accompanied a host jump from early humans to large felines. PLoS Genet 2:e120. 75. Epplein, M., A. M. Nomura, J. H. Hankin, M. J. Blaser, G. Perez-Perez, G. N. Stemmermann, and L. R. Wilkens. 2008. Associationof Helicobacter pylori infection and diet on the risk of gastric cancer: a case-control study. Cancer Causes Control 19:869-877. 76. Everhart, J. E., D. Kruszon-Moran, and G. Perez-Perez. 2002. Reliability of Helicobacter pylori and CagA serological assays. Clin Diagn Lab Immunol 9:412- 416. 77. Falk, P. G., A. J. Syder, J. L. Guruge, D. Kirschner, M. J. Blaser, and J. I. Gordon. 2000. Theoretical and experimental approaches for studying factors defining the Helicobacter pylori-host relationship. Trends Microbiol 8:321-329. 78. Falush, D., T. Wirth, B. Linz, J. K. Pritchard, M. Stephens, M. Kidd, M. J. Blaser, D. Y. Graham, S. Vacher, G. I. Perez-Perez, F. Mégraud, K. Otto, U. Reichard, E. Katzowitsch, X. Wang, M. Achtman, and S. Suerbaum. 2003. Traces of human migrations in Helicobacter pylori populations. Science 299:1582- 1585. 79. Farag, T. H., R. J. Stoltzfus, S. S. Khalfan, and J. M. Tielsch. 2007. Unexpected low prevalence of Helicobacter pylori infection among pregnant women on Pemba Island, Zanzibar. Trans R Soc Trop Med Hyg 101:915-922. 80. Ferlay, J., F. Bray, P. Pisani, and D. M. Parkin 2004, posting date. GLOBOCAN 2002: Cancer incidence, mortality and prevalence worldwide version 2.0. IARC CancerBase No. 5, Lyon, IARCpress, 2005. version 2.0. [Online.]

41 81. Figueiredo, C., J. Machado, P. Pharoah, R. Seruca, S. Sousa, R. Carvalho, A. Capelinha, W. Quint, C. Caldas, L. van Doorn, F. Carneiro, and M. Sobrinho- Simoes. 2002. Helicobacter pylori and interleukin 1 genotyping: an opportunity to identify high-risk individuals for gastric carcinoma. J Natl Cancer Inst 94:1680 - 1687. 82. Figura, N., C. Vindigni, A. Covacci, L. Presenti, D. Burroni, R. Vernillo, T. Banducci, F. Roviello, D. Marrelli, M. Biscontri, S. Kristodhullu, C. Gennari, and D. Vaira. 1998. cagA positive and negative Helicobacter pylori strains are simultaneously present in the stomach of most patients with non-ulcer dyspepsia: relevance to histological damage. Gut 42:772-778. 83. Fischer, W., J. Puls, R. Buhrdorf, B. Gebert, S. Odenbreit, and R. Haas. 2001. Systematic mutagenesis of the Helicobacter pylori cag pathogenicity island: essential genes for CagA translocation in host cells and induction of interleukin- 8.[erratum appears in Mol Microbiol. 2003 Mar;47(6):1759.]. Mol Microbiol 42:1337-1348. 84. Fock, K. M., P. Katelaris, K. Sugano, T. L. Ang, R. Hunt, N. J. Talley, S. K. Lam, S. D. Xiao, H. K. Tan, C.-Y. Wu, H. C. Jung, B. H. Hoang, U. Kachintorn, K. L. Goh, T. Chiba, and A. A. Rani. 2009. Second Asia-Pacific consensus guidelines for Helicobacter pylori infection. J Gastroenterol Hepatol 24:1587-1600. 85. Ford, A. C., B. C. Delaney, D. Forman, and P. Moayyedi. 2006. Eradication therapy for peptic ulcer disease in Helicobacter pylori positive patients. Cochrane Database Syst Rev. Apr 19:CD003840. 86. Franco, A. T., E. Johnston, U. Krishna, Y. Yamaoka, D. A. Israel, T. A. Nagy, L. E. Wroblewski, M. B. Piazuelo, P. Correa, and R. M. Peek. 2008. Regulation of gastric carcinogenesis by Helicobacter pylori virulence factors. Cancer Res 68:379- 387. 87. Freedburg, A. S., and L. E. Barron. 1940. The presence of spirochetes in human gastric mucosa. Am J Dig Dis 7:443-445. 88. Fujimoto, S., O. O. Ojo, A. Arnqvist, J. Y. Wu, S. Odenbreit, R. Haas, D. Y. Graham, and Y. Yamaoka. 2007. Helicobacter pylori BabA expression, gastric mucosal injury, and clinical outcome. Clin Gastroenterol Hepatol 5:49-58. 89. Gao, L., A. Nieters, and H. Brenner. 2009. Cell proliferation-related genetic polymorphisms and gastric cancer risk: systematic review and meta-analysis. Eur J Hum Genet 17:1658-1667.

42 90. Garza-Gonzalez, E., F. J. Bosques-Padilla, S. I. Mendoza-Ibarra, J. P. Flores- Gutierrez, H. J. Maldonado-Garza, and G. I. Perez-Perez. 2007. Assessment of the toll-like receptor 4 Asp299Gly, Thr399Ile and interleukin-8 -251 polymorphisms in the risk for the development of distal gastric cancer. BMC Cancer 7:70. 91. Gewirtz, A. T., Y. Yu, U. S. Krishna, D. A. Israel, S. L. Lyons, and R. M. Peek Jr. 2004. Helicobacter pylori flagellin evades toll-like receptor 5-mediated innate immunity. J Infect Dis 189:1914-1920. 92. Ghoshal, U. C. 2007. The Indian enigma of frequent H. pylori infection but infrequent gastric cancer: Is the magic key in Indian diet, host’s genetic make up, or friendly bug? Am J Gastroenterol 102:2113-2114. 93. Gianfagna, F., E. De Feo, C. M. van Duijn, G. Ricciardi, and S. Boccia. 2008. A systematic review of meta-analyses on gene polymorphisms and gastric cancer risk. Curr Genomics 9:361-374. 94. Glocker, E., C. Lange, A. Covacci, S. Bereswill, M. Kist, and H. L. Pahl. 1998. Proteins encoded by the cag pathogenicity island of Helicobacter pylori are required for NF-kappaB activation. Infect Immun 66:2346-2348. 95. Goh, K. L. 1997. Prevalence of and risk factors for Helicobacter pylori infection in a multi-racial dyspeptic Malaysian population undergoing endoscopy. J Gastroenterol Hepatol 12:S29-35. 96. Goh, K. L., P. L. Cheah, N. Md, M. Path, K. F. Quek, and N. Parasakthi. 2006. Ethnicity and H. pylori as risk factors for gastric cancer in Malaysia: a prospective case control study. Am J Gastroenterol 101:1-6. 97. Goh, K. L., and N. Parasakthi. 2001. The racial cohort phenomenon: seroepidemiology of Helicobacter pylori infection in a multiracial South-East Asian country. Eur J Gastroen Hepat 13:177-183. 98. Goll, R., F. Gruber, T. Olsen, G. Cui, G. Raschpichler, M. Buset, A. M. Asfeldt, A. Husebekk, and J. Florholmen. 2007. Helicobacter pylori stimulates a mixed adaptive immune response with a strong T-regulatory component in Human gastric mucosa. Helicobacter 12:185-192. 99. Gomes, L. I., G. A. Rocha, A. M. Rocha, T. F. Soares, C. A. Oliveira, P. F. Bittencourt, and D. M. Queiroz. 2008. Lack of association between Helicobacter pylori infection with dupA-positive strains and gastroduodenal diseases in Brazilian patients. Int J Med Microbiol 298:223-230.

43 100. Gorouhi, F., F. Islami, H. Bahrami, and F. Kamangar. 2008. Tumour-necrosis factor-A polymorphisms and gastric cancer risk: a meta-analysis. Br J Cancer 98:1443-1451. 101. Graham, D., H. Liu, and Y. Yamaoka. 2009. African, Asian or Indian enigma, the East Asian Helicobacter pyloriI: facts or medical myths. J Dig Dis 10:77-84. 102. Graham, D. Y. 2000. Helicobacter pylori infection is the primary cause of gastric cancer. J Gastroenterol 35 Suppl 12:90-97. 103. Graham, D. Y., P. D. Klein, D. J. Evans, D. G. Evans, L. C. Alpert, A. R. Opekun, and T. W. Boutton. 1987. Campylobacter pylori detected noninvasively by the 13C-Urea breath test Lancet 329:1174-1177. 104. Graham, D. Y., G. M. Lew, P. D. Klein, D. G. Evans, D. J. Evans Jr., Z. A. Saeed, and H. M. Malaty. 1992. Effect of treatment on the long-term recurrence of gastric or duodenal ulcer: a randomized, controlled study. Ann Intern Med 116:705-708. 105. Graham, D. Y., H. M. Malaty, D. G. Evans, D. J. Evans, P. D. Klein, and E. Adam. 1991. Epidemiology of Helicobacter pylori in an asymptomatic population in the United States. Gastroenterology 100:1495-1501. 106. Graham, D. Y., A. R. Opekun, M. S. Osato, H. M. T. El-Zimaity, C. K. Lee, Y. Yamaoka, W. A. Qureshi, M. Cadoz, and T. P. Monath. 2004. Challenge model for Helicobacter pylori infection in human volunteers. Gut 53:1235-1243. 107. Graham, D. Y., Y. Yamaoka, and H. M. Malaty. 2007. Thoughts about populations with unexpected low prevalences of Helicobacter pylori infection. Trans R Soc Trop Med Hyg 101:849-851. 108. Gregson, D. B., D. E. Low, M. M. Cohen, N. B. Cooter, J. J. Connon, S. L. Wolman, and A. E. Simor. 1989. The prevalence of Campylobacter pylori gastritis among asymptomatic adults. CMAJ 140:1449-1453. 109. Gu, H., L. Yang, Q. Sun, B. Zhou, N. Tang, R. Cong, Y. Zeng, and B. Wang. 2008. Gly82Ser polymorphism of the receptor for advanced glycation end products is associated with an increased risk of gastric cancer in a Chinese population. Clin Cancer Res 14:3627-3632. 110. Hacker, J., and J. B. Kaper. 2000. Pathogenicity islands and the evolution of microbes. Ann Rev Microbiol 54:641-679.

44 111. Hafsi, N., P. voland, S. Schwendy, R. Rad, W. Reindl, M. Gerhard, and C. Prinz. 2004. Human dendritic cells respond to Helicobacter pylori, promoting NK cell and Th1-effector responses in vitro. J Immunol 173:1249-1257. 112. Hamlet, A., and L. Olbe. 1996. The influence of Helicobacter pylori infection on postprandial duodenal acid load and duodenal bulb pH in humans. Gastroenterology 111:391-400. 113. Hamlet, A., A. C. Thoreson, O. Nilsson, A. M. Svennerholm, and L. Olbe. 1999. Duodenal Helicobacter pylori infection differs in cagA genotype between asymptomatic subjects and patients with duodenal ulcers. Gastroenterology 116:259-68. 114. Hansen, J. M., M. Wildner-Christensen, J. Hallas, and O. B. Schaffalitzky de Muckadell. 2008. Effect of a community screening for the Helicobacter pylori: a 5- yr follow-up study. Am J Gastroenterol 103:1106-1113. 115. Hansson, L. E., O. Nyrén, A. W. Hsing, R. Bergström, S. Josefsson, W. H. Chow, J. F. Fraumeni, and H. O. Adami. 1996. The risk of stomach cancer in patients with gastric or duodenal ulcer disease. N Engl J Med 335:242-249. 116. Hatakeyama, M. 2004. Oncogenic mechanisms of the Helicobacter pylori CagA protein. Nat Rev Cancer 4:688-694. 117. Heinzmann, A., I. Ahlert, T. Kurz, R. Berner, and K. A. Deichmann. 2004. Association study suggests opposite effects of polymorphisms within IL8 on bronchial asthma and respiratory syncytial virus bronchiolitis J Allergy Clin Immunol 114:671-676. 118. Helicobacter Cancer Collaborative G. 2001. Gastric cancer and Helicobacter pylori a combined analysis of 12 case control studies nested within prospective cohorts. Gut 49:347-353. 119. Hemmi, H., O. Takeuchi, T. Kawai, T. Kaisho, S. Sato, H. Sanjo, M. Matsumoto, K. Hoshino, H. Wagner, K. Takeda, and S. Akira. 2000. A Toll- like receptor recognizes bacterial DNA. Nature 408:740-745. 120. Higashi, H., A. Nakaya, R. Tsutsumi, K. Yokoyama, Y. Fujii, S. Ishikawa, M. Higuchi, A. Takahashi, Y. Kurashima, Y. Teishikata, S. Tanaka, T. Azuma, and M. Hatakeyama. 2004. Helicobacter pylori CagA induces Ras-independent morphogenetic response through SHP-2 recruitment and activation. J Biol Chem 279:17205-17216.

45 121. Higashi, H., R. Tsutsumi, A. Fujita, S. Yamazaki, M. Asaka, T. Azuma, and M. Hatakeyama. 2002. Biological activity of the Helicobacter pylori virulence factor CagA is determined by variation in the tyrosine phosphorylation sites. Proc Natl Acad Sci U S A 99:14428-33. 122. Higashi, H., R. Tsutsumi, S. Muto, T. Sugiyama, T. Azuma, M. Asaka, and M. Hatakeyama. 2002. SHP-2 tyrosine phosphatase as an intracellular target of Helicobacter pylori CagA protein. Science 295:683-686. 123. Hofseth, L. J., S. i. Saito, S. P. Hussain, M. G. Espey, K. M. Miranda, Y. Araki, C. Jhappan, Y. Higashimoto, P. He, S. P. Linke, M. M. Quezado, I. Zurer, V. Rotter, D. A. Wink, E. Appella, and C. C. Harris. 2003. Nitric oxide-induced cellular stress and p53 activation in chronic inflammation. Proc Natl Acad Sci USA 100:143-148. 124. Holcombe, C., B. A. Omotara, J. Eldridge, and D. M. Jones. 1992. H. pylori, the most common bacterial infection in Africa: a random serological study. Am J Gastroenterol 87:28-30. 125. Hull, J., A. Thomson, and D. Kwiatkowski. 2000. Association of respiratory syncytial virus bronchiolitis with the interleukin 8 gene region in UK families. Thorax 55:1023-1027. 126. Hussain, S. P., L. J. Hofseth, and C. C. Harris. 2003. Radical causes of cancer. Nat Rev Cancer 3:276–285. 127. Hussein, N., M. Mohammadi, Y. Talebkhan, M. Doraghi, D. Letley, M. Muhammad, R. Argent, and J. Atherton. 2008. Differences in virulence markers between Helicobacter pylori strains from Iraq and those from Iran: potential importance of regional differences in H. pylori-associated disease. J Clin Microbiol 46:1774-9. 128. Hussein, N. R. 2010. The association of dupA and Helicobacter pylori-related gastroduodenal diseases. Eur J Clin Microbiol Infect Dis 29:817-821. 129. Hwang, H., J. Dwyer, and R. M. Russell. 1994. Diet, Helicobacter pylori infection, food preservation and gastric cancer risk: are there new roles for preventative factors? Nutr Rev 52:75-83. 130. IARC. 2006. Ingested nitrates and nitrites. IARC Monographs 44. 131. Ilver, D., A. Arnqvist, J. Ogren, I. M. Frick, D. Kersulyte, E. T. Incecik, D. E. Berg, A. Covacci, L. Engstrand, and T. Boren. 1998. Helicobacter pylori adhesin

46 binding fucosylated histo-blood group antigens revealed by retaging. Science 279:373-377. 132. Ishizak, K., M. Kato, M. Ikenaga, K. Honda, K. Ozawa, and J. Toguchida. 1990. Correlation of L-myc Genotypes to Metastasis of Gastric Cancer and Breast Cancer. J Natl Cancer Inst. 82:238-239. 133. Israel, D. A., N. Salama, U. Krishna, U. M. Rieger, J. C. Atherton, S. Falkow, and R. M. Peek, Jr. 2001. Helicobacter pylori genetic diversity within the gastric niche of a single human host. Proc Natl Acad Sci U S A 98:14625-30. 134. Ito, Y., T. Azuma, S. Ito, H. Miyaji, M. Hirai, Y. Yamazaki, F. Sato, T. Kato, Y. Kohli, and M. Kuriyama. 1997. Analysis and typing of the vacA gene from cagA-positive strains of Helicobacter pylori isolated in Japan. J Clin Microbiol 35:1710-1714. 135. Jang, S., K. R. Jones, C. H. Olsen, Y. M. Joo, Y. J. Yoo, I. S. Chung, J. H. Cha, and D. S. Merrell. 2009. An epidemiological link between gastric disease and polymorphisms in VacA and CagA. J Clin Microbiol 48:559-567. 136. Janzon, A., T. Bhuiyan, A. Lundgren, F. Qadri, A. M. Svennerholm, and A. Sjöling. 2009. Presence of high numbers of transciptionally active Helicobacter pylori in vomitus from Bangladeshi patients suffering from acute gastroenteritis. Helicobacter 14:237-247. 137. Ji, X., T. Fernandez, D. Burroni, C. Pagliaccia, J. C. Atherton, J.-M. Reyrat, R. Rappuoli, and J. L. Telford. 2000. Cell specificity of Helicobacter pylori cytotoxin is determined by a short region in the polymorphic midregion. Infect Immun 68:3754-3757. 138. Johnson, J. R. 1991. Virulence factors in Escherichia coli urinary tract infections. Clin Microbiol Rev 4:80-128. 139. Jones, K. R., Y. M. Joo, S. Jang, Y. J. Yoo, H. S. Lee, I. S. Chung, C. H. Olsen, J. M. Whitmire, D. S. Merrell, and J. H. Cha. 2009. Polymorphism in the CagA EPIYA motif impacts development of gastric cancer. J Clin Microbiol 47:959-968. 140. Kao, J. Y., M. Zhang, M. J. Miller, J. C. Mills, B. Wang, M. Liu, K. A. Eaton, W. Zou, B. E. Berndt, T. S. Cole, T. Takeuchi, S. Y. Owyang, and J. Luther. 2010. Helicobacter pylori immune escape is mediated by dendritic cell–induced Treg skewing and Th17 suppression in mice Gastroenterology 138:1046-1054. 141. Kato, S., N. Matsukura, K. Tsukada, N. Matsuda, T. Mizoshita, T. Tsukamoto, M. Tatematsu, Y. Sugisaki, Z. Naito, and T. Tajiri. 2007. Helicobacter pylori

47 infection-negative gastric cancer in Japanese hospital patients: incidence and pathological characteristics. Cancer Sci 98:790-794. 142. Kauser, F., A. A. Khan, M. A. Hussain, I. M. Carroll, N. Ahmad, S. Tiwari, Y. Shouche, B. Das, M. Alam, M. Ali, C. M. Habibullah, R. Sierra, F. Megraud, L. A. Sechi, and N. Ahmed. 2004. The cag pathogenicity island of Helicobacter pylori is disrupted in the majority of patient isolates from different human populations. J Clin Microbiol 42:5302-5308. 143. Kidd, M., and I. M. Modlin. 1998. A century of Helicobacter pylori: Paradigms lost - paradigms regained. Digestion 59:1-15. 144. Kido, M., J. Tanaka, N. Aoki, S. Iwamoto, H. Nishiura, T. Chiba, and N. Watanade. 2010. Helicobacter pylori promotes the production of thymic stromal lymphopoietin by gastric epithelial cells and induces dendritic cell-mediated inflammatory Th2 responses. Infect Immun 78:108-144. 145. Kim, C. C., E. A. Joyce, K. Chan, and S. Falkow. 2002. Improved analytical methods for microarray-based genome-composition analysis. Genome Biol 3:research 0065.1-0065.17. 146. Kim, S.-Y., Y.-C. Lee, H. K. Kim, and M. J. Blaser. 2006. Helicobacter CagA transfection of gastric epithelial cells induces interleukin-8. Cell Microbiol 8:97- 106. 147. Kim, S. F., D. A. Huri, and S. H. Snyder. 2005. Inducible nitric oxide synthase binds S-nitrosylates and activates cyclooxygenase-2. Science 310:1966–1970. 148. Kivi, M., Y. Tindberg, M. Sorberg, T. H. Casswall, R. Befrits, P. M. Hellstrom, C. Bengtsson, L. Engstrand, and M. Granstrom. 2003. Concordance of Helicobacter pylori strains within families. J Clin Microbiol 41:5604-5608. 149. Klusters, J. G., A. H. M. van Vliet, and E. J. Kuipers. 2006. Pathogenesis of Helicobacter pylori infection. Clin Microbiol Rev 19:449-490. 150. Kolb, H., and B. V. Kolb. 1992. Nitric oxide: a pathogenic factor in autoimmunity. Immunol Today 13:157-160. 151. Konno, M., S. Yokota, T. Suga, M. Takahashi, K. Sato, and N. Fujii. 2008. Predominance of mother-to-child transmission of Helicobacter pylori infection detected by random amplified polymorphic DNA fingerprinting analysis in Japanese families. Pediatr Infect Dis J 27:999-1003. 152. Kreig, A. M. 2002. CpG motifs in bacterial DNA and their immune effects. Annu Rev Immunol 20:709-760.

48 153. Kuipers, E. J., D. A. Israel, J. G. Kusters, M. M. Gerrits, J. Weel, A. van Der Ende, R. W. van Der Hulst, H. P. Wirth, J. Hook-Nikanne, S. A. Thompson, and M. J. Blaser. 2000. Quasispecies development of Helicobacter pylori observed in paired isolates obtained years apart from the same host. J Infect Dis 181:273- 282. 154. Kuipers, E. J., J. C. Thijs, and H. P. Festen. 1995. The prevalence of Helicobacter pylori in peptic ulcer disease. Aliment Pharmacol Ther 9:59-69. 155. Kuraoka, K., S. Matsumura, Y. Hamai, K. Nakachi, K. Imai, K. Matsusaki, N. Oue, R. Ito, H. Nakayama, and W. Yasui. 2003. A single nucleotide polymorphism in the transmembrane domain nucleotide polymorphism in the transmembrane domain coding region of HER-2 is associated with development and malignant phenotype of gastric cancer. Int J Cancer 107:593- 596. 156. Kwok, T., D. Zabler, S. Urman, M. Rohde, R. Hartig, S. Wessler, R. Misselwitz, J. Berger, N. Sewald, W. König, and S. Backert. 2007. Helicobacter exploits integrin for type IV secretion and kinase activation. Nature 449:862-867. 157. Lamouliatte, H., R. Cayla, and F. Regland. 1993. Amoxicillin-clarithromycin- omeprazole: the best therapy for Helicobacter pylori eradication. Acta Gastroenterol Belg 56:A139. 158. Lee, A., M. F. Dixon, S. J. Danon, E. J. Kuipers, F. Megraud, H. Larsson, and B. Mellgard. 1995. Local acid production and Helicobacter pylori: a unifying hypothesis of gastroduodenal disease. Eur J Gastroenterol Hepatol 7:461-465. 159. Lee, A., J. G. Fox, G. Otto, E. H. Dick, and S. Krakowka. 1991. Transmission of Helicobacter spp. A challenge to the dogma of faecal-oral spread. Epidemiol Infect 107:99-109. 160. Lee, J. U., and O. Kim. 2008. Natural maternal transmission of H. pylori in Mongolian gerbils. World J Gastroenterol 12:5663-5667. 161. Lejeune, P., P. Lagadec, N. Onier, D. Pinard, H. Ohshima, and J. F. Jeannin. 1994. Nitric oxide involvement in tumor-induced immunosuppression. J Immunol 152:5077-5083. 162. Leung, W. K., K. L. Siu, C. K. Kwok, S. Y. Chan, R. Sung, and J. J. Sung. 1999. Isolation of Helicobacter pylori from vomitus in children and its implication in gastro-oral transmission. Am J Gastroenterol 94:2881-2884.

49 163. Leunk, R. D., P. T. Johnson, B. C. David, W. G. Kraft, and D. R. Morgan. 1988. Cytotoxic activity in broth-culture filtrates of Campylobacter pylori. J Med Microbiol 26:93-99. 164. Levine, M. M. 1987. Escherichia coli that cause diarrhea: enterotoxigenic, enteropathogenic, enteroinvasive, enterohemorrhagic, and enteroadherent. J Infect Dis 155:377-389. 165. Li, S. D., D. Kersulyte, I. J. Lindley, B. Neelam, D. E. Berg, and J. E. Crabtree. 1999. Multiple genes in the left half of the cag pathogenicity island of Helicobacter pylori are required for tyrosine kinase-dependent transcription of interleukin-8 in gastric epithelial cells. Infect Immun 67:3893-3899. 166. Lim, G. C. C., H. Yahaya, and T. O. Lim. 2002. The first report of the national cancer registry - cancer incidence in Malaysia 2002. National cancer registry. 167. Linden, S., H. Nordman, J. Hedenbro, M. Hurtig, T. Boren, and I. Carlstedt. 2002. Strain- and blood group-dependent binding of Helicobacter pylori to human gastric MUC5AC glycoforms. Gastroenterology 123:1923-1930. 168. Lindholm, C., M. Quiding-Järbrink, H. Lönroth, A. Hamlet, and A. M. Svennerholm. 1998. Local cytokine response in H. pylori-infected subjects. Infect Immun 66:5964-5971. 169. Linz, B., F. Balloux, Y. Moodley, A. Manica, H. Liu, P. Roumagnac, D. Falush, C. Stamer, F. Prugnolle, S. W. van der Merwe, Y. Yamaoka, D. Y. Graham, E. Perez-Trallero, T. Wadstrom, S. Suerbaum, and M. Achtman. 2007. An African origin for the intimate association between humans and Helicobacter pylori. Nature 445:915-918. 170. Linz, B., and S. C. Schuster. 2007. Genomic diversity in Helicobacter and related organisms. Research Microbiol 158:737-744. 171. Liu, F., K. Pan, X. Zhang, Y. Zhang, L. Zhang, J. Ma, C. Dong, L. Shen, J. Li, D. Deng, D. Lin, and W. You. 2006. Genetic variants in Cyclooxygenase-2: expression and risk of gastric cancer and its precursors in a Chinese population. Gastroenterology 130:1975-1984. 172. Liu, H., D. S. Merrell, C. Semino-Mora, M. Goldman, A. Rahman, S. Mog, and A. Dubois. 2009. Diet synergistically affects Helicobacter pylori-induced gastric carcinogenesis in non-human primates. Gastroenterology 137:1367-1379.

50 173. Liu, L., W. Zhuang, C. Wang, Z. Chen, X. T. Wu, and Y. Zhou. 2010. Interleukin-8 -251 A/T gene polymorphism and gastric cancer susceptibility: a meta-analysis of epidemiological studies. Cytokine 50:328-334. 174. Lo, S., J. H. Chen, C. W. Wu, and W. Y. Lui. 2009. Functional polymorphism of NFKB1 promoter may correlate to susceptibility of gastric cancer in aged patients. Surgery 145:280-285. 175. Lochhead, P., and E. M. El-Omar. 2008. Gastric cancer. Br Med Bull 85:87-100. 176. Lu, H., P.-I. Hsu, D. Graham, and Y. Yamaoka. 2005. Duodenal ulcer promoting gene of Helicobacter pylori. Gastroenterology 128:833-848. 177. Lu, H., Y. Yamaoka, and D. Y. Graham. 2005. Helicobacter pylori virulence factors: facts and fantasies. Curr Opin Gastroenterol 21:653-659. 178. Lunet, N., C. Valbuena, A. L. Vieira, C. Lopes, C. Lopes, L. David, F. Carneiro, and H. Barros. 2007. Fruit and vegetable consumption and gastric cancer by location and histological type: case-control and meta-analysis. Eur J Cancer Prev 16:312-327. 179. Machado, J., C. Figueiredo, P. Canedo, P. Pharoah, R. Carvalho, S. Nabais, C. Alves, M. Campos, L. van Doorn, C. Caldas, R. Seruca, F. Carneiro, and M. Sobrinho-Simoes. 2003. A proinflammatory genetic profile increases the risk for chronic atrophic gastritis and gastric carcinoma. Gastroenterology 125:364- 371. 180. Machado, J. C., P. Pharoah, S. Sousa, R. Carvalho, C. Oliveira, C. Figueiredo, A. Amorim, R. Seruca, C. Caldas, F. Carneiro, and M. Sobrinho-Simoes. 2001. Interleukin 1B and interleukin 1RN polymorphisms are associated with increased risk of gastric carcinoma. Gastroenterology 121:823-829. 181. Maeda, S., F. Kanai, K. Ogura, H. Yoshida, T. Ikenoue, M. Takahashi, T. Kawabe, Y. Shiratori, and M. Omata. 1997. High seropositivity of anti-CagA antibody in Helicobacter pylori-infected patients irrelevant to peptic ulcers and normal mucosa in japan. Dig Dis Sci 42:1841-1847. 182. Maeda, S., K. Ogura, H. Yoshida, F. Kanai, T. Ikenoue, N. Kato, Y. Shiratori, and M. Omata. 1998. Major virulence factors, VacA and CagA, are commonly positive in Helicobacter pylori isolates in Japan. Gut 42:338-343. 183. Maeda, S., H. Yoshida, T. Ikenoue, K. Ogura, F. Kanai, N. Kato, Y. Shiratori, and M. Omata. 1999. Structure of cag pathogenicity island in Japanese Helicobacter pylori isolates. Gut 44:336-41.

51 184. Mahadeva, S., M. C. Raman, A. C. Ford, M. Follows, A. T. R. Axon, K. L. Goh, and P. Moayyedi. 2005. Gastro-oesophageal reflux is more prevalent in Western dyspeptics: a prospective comparison of British and South-East Asian patients with dyspepsia. Aliment Pharmacol Ther 21:1483-1490. 185. Mahdavi, J., B. Sonden, M. Hurtig, F. O. Olfat, L. Forsberg, N. Roche, J. Angstrom, T. Larsson, S. Teneberg, K.-A. Karlsson, S. Altraja, T. Wadstrom, D. Kersulyte, D. E. Berg, A. Dubois, C. Petersson, K.-E. Magnusson, T. Norberg, F. Lindh, B. B. Lundskog, A. Arnqvist, L. Hammarstrom, and T. Boren. 2002. Helicobacter pylori SabA adhesin in persistent infection and chronic inflammation. Science 297:573-578. 186. Mandell, L., A. P. Moran, A. Cocchiarella, J. Houghton, N. Taylor, and J. G. Fox. 2004. Intact Gram-negative Helicobacter pylori, Helicobacter felis, and Helicobacter hepaticus bacteria activate innate immunity via Toll-like receptor 2 but not Toll-like receptor 4. Infect Immun 72:6446-6454. 187. Marshall, B. J., J. A. Armstrong, D. B. McGrechie, and R. J. Glancy. 1985. Attempt to fulfill Koch's postulates for pyloric Campylobacter. Med J Aust 142:436-439. 188. Marshall, B. J., L. J. Barrett, C. Prakash, R. W. McCallum, and R. L. Guerrant. 1990. Urease protects Helicobacter (Campylobacter) pylori from the bactericidal effect of acid. Gastroenterology 99:697-702. 189. Marshall, B. J., D. B. McGrechie, P. A. Rogers, and R. J. Glancy. 1985. Pyloric Campylobacter infection and gastroduodenal disease. Med J Aust 142:439-444. 190. Marshall, B. J., and J. R. Warren. 1984. Unidentified curved bacilli in the stomach of patients with gastritis and peptic ulceration. Lancet 323:1311-1315. 191. McClain, M. S., P. Cao, H. Iwamoto, A. D. Vinion-Dubiel, G. Szabo, Z. Shao, and T. L. Cover. 2001. A 12-amino-acid segment, present in type s2 byt not type s1 Helicobacter pylori VacA proteins, abolishes cytotoxin activity and alters membrane channel formation. J Bacteriol 183:6499-6508. 192. Megraud, F. 2010. Antimicrobial resistance and approaches to treatment, p. 45- 68. In P. Sutton and H. M. Mitchell (ed.), Helicobacter pylori in the 21st century. CABI, Wallingford. 193. Megraud, F. 1995. Transmission of Helicobacter pylori: faecal-oral versus oral- oral route. Aliment Pharmacol Ther 9:85-91.

52 194. Miehlke, S., K. Kibler, J. C. Kim, N. Figura, S. M. Small, D. Y. Graham, and M. F. Go. 1996. Allelic Variation in the cagA Gene of Helicobacter pylori Obtained from Korea Compared to the United States. Am J Gastroenterol 91:1322-1325. 195. Milne, A. N., F. Carneiro, C. O'Morain, and G. J. A. Offerhaus. 2009. Nature meets nurture: molecular genetics of gastric cancer. Hum Genet 126:615-628. 196. Mimuro, H., T. Suzuki, J. Tanaka, M. Asahi, R. Haas, and C. Sasakawa. 2002. Grb2 is a key mediator of Helicobacter pylori CagA protein activities. Mol Cell 10:745-755. 197. Mirvish, S. S. 1994. Experimental evidence for inhibition of N-nitroso compound formation as a factor in the negative correlation between vitamin C consumption and the incidence of certain cancers. Cancer Res 54:1948s-1951s. 198. Mirvish, S. S. 1996. Inhibition by vitamins C and E of in vivo nitrosation and vitamin C occurrebce in the stomach. Eur J Cancer Prev 5:131-136. 199. Mirvish, S. S. 1995. Role of N-nitroso compounds (NOC) and N-nitrosation in etiology of gastric, esophageal, nasopharyngeal and bladder cancer and contribution to cancer of known exposures to NOC. Cancer Letters 93:17-48. 200. Mitchell, H. 2001. Epidemiology and infection, p. 7-18. In H. L. T. Mobley, G. L. Mendz, and S. L. Hazell (ed.), Helicobacter pylori: phyisology and genetics. ASM Press, Washington, D. C. 201. Mitchell, H., D. R. English, F. Elliott, M. Gengos, J. H. Barrett, G. G. Giles, and D. Forman. 2008. Immunoblotting using multiple antigens is essential to demontrate the true risk of Helicobacter pylori infection for gastric cancer. Aliment Pharmacol Ther 28:903-910. 202. Mitchell, H., and F. Megraud. 2002. Epidemiology and diagnosis of Helicobacter pylori infection. Helicobacter 7:8-16. 203. Mitchell, H. M., S. L. Hazell, Y. Y. Li, and P. J. Hu. 1996. Serological response to specific Helicobacter pylori antigens: antibody against CagA antigen is not predictive of gastric cancer in a developing country. Am J Gastroenterol 91:1785-1788. 204. Mitchell, J. D., H. M. Mitchell, and V. Tobias. 1992. Acute Helicobacter pylori infection in an infant, associated with gastric ulceration and serological evidence of intra-familial transmission. Am J Gastroenterol 87:382-386. 205. Miwa, H., M. F. Go, and N. Sato. 2002. H. pylori and gastric cancer: the Asian enigma. Am J Gastroenterol 97:1106-1112.

53 206. Moayyedi, P., S. Soo, J. Deeks, B. Delaney, A. Harris, M. Innes, R. Oakes, S. Wilson, A. Roalfe, C. Bennett, and D. Forman. 2006. Eradication of Helicobacter pylori for non-ulcer dyspepsia. Cochrane Database Syst Rev 19:CD002096. 207. Morris, A., and G. Nicholson. 1987. Ingestion of Campylobacter pyloridis causes gastritis and raised fasting gastric pH. Am J Gastroenterol 82:192-199. 208. Moss, S. F., and M. J. Blaser. 2005. Mechanisms of disease: inflammation and the origins of cancer. Nat Clin Pract Oncol 2:90-97. 209. Murata-Kamiya, N., Y. Kurashima, Y. Teishikata, Y. Yamahashi, Y. Saito, H. Higashi, H. Aburatani, T. Akiyama, R. M. Peek, T. Azuma, and M. Hatakeyama. 2007. Helicobacter pylori CagA interacts with E-cadherin and deregulates the beta-catenin signal that promotes intestinal transdifferentiation in gastric epithelial cells. Oncogene 26:4617-4626. 210. Nahar, S., K. M. K. Kibria, M. E. Hossain, J. Sultana, S. A. Sarker, L. Engstrand, P. K. Bardham, M. Rahman, and H. P. Endtz. 2009. Evidence of intra-familial transmission of Helicobacter pylori by PCR-based RAPD fingerprinting in Bangladesh. Eur J Clin Microbiol Infect Dis 28:767-773. 211. Naito, M., T. Yamazaki, R. Tsutsumi, H. Higashi, K. Onoe, S. Yamazaki, T. Azuma, and M. Hatakeyama. 2006. Influence of EPIYA-repeat polymorphism on the phosphorylation dependent biological activity of Helicobacter pylori CagA. Gastroenterology 130:1181-1190. 212. Nilsson, C., A. Sillen, L. Eriksson, M. Strand, H. Enroth, S. Normark, P. Falk, and L. Engstrand. 2003. Correlation between cag pathogenicity island composition and Helicobacter pylori-associated gastroduodenal disease. Infect Immun 71:6573-6581. 213. Nurgalieva, Z. Z., M. E. Conner, A. R. Opekun, C. Q. Zheng, S. N. Elliot, P. B. Ernst, M. Osato, M. K. Estes, and D. Y. Graham. 2005. B-cell and T-cell immune responses to experimental Helicobacter pylori infection in humans. Infect Immun 73:2999-3006. 214. O'Connor, H. J. 1994. The role of Helicobacter pylori in peptic ulcer disease. Scand J Gastroenterol Suppl 201:11-15. 215. Odenbreit, S., J. Puls, B. Sedlmaier, E. Gerland, W. Fischer, and R. Haas. 2000. Translocation of Helicobacter pylori CagA into gastric epithelial cells by type IV secretion. Science 287:1497-1500.

54 216. Ogiwara, H., D. Y. Graham, and Y. Yamaoka. 2008. vacA i-region subtyping. Gastroenterology 134:1267-1276. 217. Ogiwara, H., M. Sugimoto, T. Ohno, R. K. Vilaichone, V. Mahachi, D. Y. Graham, and Y. Yamaoka. 2009. Role of deletion located between the intermediate and middle regions of Helicobacter pylori vacA gene in cases of gastroduodenal diseases. J Clin Microbiol 47:3493-3500. 218. Oussalah, A. 2007. Uninvestigated dyspepsia or functional dyspepsia: that is the question. Am J Gastroenterol 102:2109-2110. 219. Owen, R. J. 1998. Helicobacter - species classification and identification. Br Med Bull 54:17-30. 220. Pan, Z., R. van der Hulst, M. Feller, S. Xiao, G. Tytgat, J. Dankert, and A. van der Ende. 1997. Equally high prevalences of infection with cagA-positive Helicobacter pylori in Chinese patients with peptic ulcer disease and those with chronic gastritis-associated dyspepsia. J Clin Microbiol 35:1344-1347. 221. Parkin, D. M., F. Bray, J. Ferlay, and P. Pisani. 2005. Global cancer statistics, 2002. CA Cancer J Clin 55:74-108. 222. Parsonnet, J. 1998. Helicobacter pylori: the size of the problem. Gut 43:6S-9. 223. Parsonnet, J., G. D. Friedman, N. Orentreich, and H. Vogelman. 1997. Risk for gastric cancer in people with CagA positive or CagA negative Helicobacter pylori infection. Gut 40:297-301. 224. Parsonnet, J., H. Shmuely, and T. Haggerty. 1999. Faecal-oral shedding of Helicobacter pylori from healthy infected adults. JAMA 282:2240-2245. 225. Pasare, C., and R. Medzhitov. 2005. Toll-like receptors: linking innate and adaptive immunity. Adv Exp Med Biol 560:11-18. 226. Pattis, I., E. Wiess, R. Laugks, R. Haas, and W. Fischer. 2007. The Helicobacter CagF protein is a type IV-secretion chaperone-like molecule that binds close to the C-terminal secretion signal of CagA. Microbiology 153:2896-2909. 227. Peek, R. M., and D. A. Israel. 2001. Review article: pathogenesis of Helicobacter pylori-induced gastric inflammation. Aliment Pharmacol Ther 15:1271-1290. 228. Peek, R. M., G. G. Miller, K. T. Tham, G. I. Perez-Perez, X. Zhao, J. C. Atherton, and M. J. Blaser. 1995. Heightened inflammatory response and cytokine expression in vivo to cagA+ Helicobacter pylori strains. Lab Invest 73:760-770.

55 229. Peek, R. M., S. F. Moss, K. T. Tham, G. I. Perez-Perez, S. Wang, G. G. Miller, J. Atherton, P. R. Holt, and M. J. Blaser. 1997. Helicobacter pylori cagA+ strains and dissociation of gastric cell proliferation from apoptosis. J Natl Cancer Inst 89:836-837. 230. Perez-Perez, G. I. 2004. Epidemiology of Helicobacter pylori infection. Helicobacter 9 Suppl 1:1-6. 231. Portal-Celhay, C., and G. I. Perez-Perez. 2006. Immune response to Helicobacter pylori colonisation: mechanisms and clinical outcomes. Clin Sci 110:305-314. 232. Pride, D. T., and M. J. Blaser. 2002. Concerted evolution between genetic elements in Helicobacter pylori. J Mol Biol 316:629-642. 233. Rad, R., W. Ballhorn, P. Voland, K. Eisenächer, J. Mages, L. Rad, R. Ferstl, R. Lang, H. Wagner, R. M. Schmid, S. Bauer, C. Prinz, C. J. Kirschning, and A. Krug. 2009. Extracellular and Intracellular Pattern Recognition Receptors Cooperate in the Recognition of Helicobacter pylori Gastroenterology 136:2247- 2257. 234. Renfrew, C. 2009. Where bacteria and languages concur. Science 323:467-468. 235. Reyes-Leon, A., J. C. Atherton, R. H. Argent, J. L. Puente, and J. Torres. 2007. Heterogeneity in the activity of Mexican Helicobacter pylori strains on gastric epithelial cells and its association with diversity in the cagA gene. Infect Immun 75:3445-3454. 236. Rhead, J. L., D. P. Letley, M. Mohammadi, N. Hussein, M. A. Mohagheghi, M. Hosseini, and J. C. Atherton. 2007. A new Helicobacter pylori vacuolating cytotoxin determinant, the intermediate region, is associated with gastric cancer. Gastroenterology 133:926-36. 237. Robert, A., A. S. Olafsson, C. Lancaster, and W. R. Zhang. 1991. Interleukin 1 is cytoprotective, antisecretory, stimulates PGE2 synthesis by the stomach, and retards gastric emptying. Life Sci 48:123-134. 238. Roberts-Thomson, I. C., and W. J. Butler. 2005. Polymorphism and gastric cancer. J Gastroenterol Hepatol 20:793-794. 239. Rokkas, T., S. Ladas, C. Liatsos, E. Petridou, G. Papatheodorou, S. Theocharis, A. Karameris, and S. Raptis. 1999. Relationship of Helicobacter pylori CagA status to gastric cell proliferation and apoptosis. Dig Dis Sci 44:487-493.

56 240. Romero-Gallo, J., E. J. Harris, U. Krishna, M. K. Washington, G. I. Perez-Perez, and R. M. Peek Jr. 2008. Effect of Helicobacter pylori eradication on gastric carcinogenesis. Lab Invest 88:328-336. 241. Romo-Gonzalez, C., N. R. Sala, J. Burgeno-Ferreira, V. Ponce-Castaneda, E. Lazcano-Ponce, M. Camorlinga-Ponce, and J. Torres. 2009. Differences in the genome content among Helicobacter pylori isolates from patients with gastritis, duodenal ulcer, or gastric cancer reveal novel disease-associated genes. Infect Immun 77:2201-2211. 242. Saadat, I., H. Higashi, C. Obuse, M. Umeda, N. Murata-Kamiya, Y. Saito, H. Lu, N. Ohnishi, T. Azuma, A. Suzuki, S. Ohno, and M. Hatakeyama. 2007. Helicobacter pylori CagA targets PAR1/MARK kinase to disrupt epithelial cell polarity. Nature 447:330-333. 243. Salama, N. R., K. Guillemin, T. K. McDaniel, G. Sherlock, L. Tompkins, and S. Falkow. 2000. A whole-genome microarray reveals genetic diversity among Helicobacter pylori strains. Proc Natl Acad Sci U S A 97:14668-14673. 244. Salih, B. A., B. K. Bolek, and S. Arikan. 2009. DNA sequence analysis of cagA 3' motifs of Helicobacter pylori strains from patients with peptic ulcer diseases. J Med Microbiol 59:144-148. 245. Schmidt, H.-M. A., K. L. Goh, K. M. Fock, I. Hilmi, S. Dhamodaran, D. Forman, and H. Mitchell. 2009. Distinct cagA EPIYA motifs are associated with ethnic diversity in Malaysia and Singapore. Helicobacter 14:256-263. 246. Schwarz, S., G. Morelli, B. Kusecek, A. Manica, F. Balloux, R. J. Owen, D. Y. Graham, S. Van der Merwe, M. Achtman, and S. Suerbaum. 2008. Horizontal versus familial transmission of Helicobacter pylori. PLOS Pathog 4:e1000180. 247. Segal, E. D., C. Lange, A. Covacci, L. S. Thompkins, and S. Falkow. 1997. Induction of host signal transduction pathways by Helicobacter pylori. Proc Natl Acad Sci U S A 94:7595-7599. 248. Selbach, M., S. Moese, C. R. Hauck, T. F. Meyer, and S. Backert. 2002. Src is the kinase of Helicobacter pylori CagA protein in vitro and in vivo. J Biol Chem 277:6775-6778. 249. Selbach, M., S. Moese, R. Hurwitz, C. R. Hauck, T. F. Meyer, and S. Backert. 2003. The Helicobacter pylori CagA protein induces cortactin dephosphorylation and actin rearrangement by c-Src inactivation. EMBO J 22:515-28.

57 250. Shanks, A.-M., and E. M. El-Omar. 2009. Helicobacter pylori infection, host genetics and gastric cancer. J Dig Dis 10:157-164. 251. Sharma, S. A., M. K. Tummuru, M. J. Blaser, and L. D. Kerr. 1998. Activation of IL-8 gene expression by Helicobacter pylori is regulated by transcription factor nuclear factor-kappa B in gastric epithelial cells. J Immunol 160:2401-2407. 252. Sheu, B. S., S. Odenbreit, K. H. Hung, C. P. Liu, S. M. Sheu, H. B. Yang, and J. J. Wu. 2006. Interaction between host gastric Sialyl-Lewis X and H. pylori SabA enhances H. pylori density in patients lacking gastric Lewis B antigen. Am J Gastroenterol 101:36-44. 253. Shin, A., H. R. Shin, D. Kang, S. K. Park, C.-S. Kim, and K.-Y. Yoo. 2005. A nested case-control study of the association of Helicobacter pylori infection with gastric adenocarcinoma in Korea. Br J Cancer 92:1273-1275. 254. Sierra, R., C. Une, V. Ramirez, W. Alpizar-Alpizar, M. I. González, J. A. Ramirez, A. de Mascarel, P. Cuenca, G. Perez-Perez, and F. Megraud. 2008. Relation of atrophic gastritis with Helicobacter pylori-CagA+ and interleukin-1 gene polymorphisms. World J Gastroenterol 14:6481-6487. 255. Sierra, R., C. Une, V. Ramirez, M. I. González, J. A. Ramirez, A. De Mascarel, R. Barahona, R. Salas-Aguilar, R. Paez, G. Avendano, A. Avalos, N. Broutet, and F. Megraud. 2006. Association of serum pepsinogen with atrophic body gastritis in Costa Rica. Clin Exp Med 6:72-78. 256. Singh, K., and U. C. Ghoshal. 2006. Causal role of Helicobacter pylori infection in gastric cancer: An Asian enigma. World J Gastroenterol 12:1346-1351. 257. Smith, M. F., A. Mitchell, G. Li, S. Ding, A. M. Fitzmaurice, K. Ryan, S. Crowe, and J. B. Goldberg. 2003. Toll-like receptor (TLR) 2 and TLR5, but not TLR4, are required for Helicobacter pylori-induced NF-κB activation and chemokine expression by epithelial cells. J Biol Chem 278:32552-32560. 258. Sobala, G. M., J. E. Crabtree, M. F. Dixon, C. J. Schorah, J. D. Taylor, B. J. Rathbone, R. V. Heatley, and A. T. Axon. 1991. Acute Helicobacter pylori infection: clinical features, local and systemic immune response, gastric mucosal histology, and gastric juice ascorbic acid concentration. Gut 32:1415- 1418. 259. Solnick, J. V., J. Fong, L. M. Hansen, K. Chang, D. R. Canfield, and J. Parsonnet. 2006. Acquisition of Helicobacter pylori infection in rhesus macaques is most consistent with oral-oral transmission. J Clin Microbiol 44:3799-3803.

58 260. Solnick, J. V., L. M. Hansen, N. R. Salama, J. K. Boonjakuakul, and M. Syvanen. 2004. Modification of Helicobacter pylori outermembrane protein expression during experimental infection of rhesus macaques. Proc Natl Acad Sci USA 101:2106-2111. 261. Solnick, J. V., and P. Vandamme. 2001. Taxonomy of the Helicobacter genus. In H. L. T. Mobley, G. L. Mendz, and S. L. Hazell (ed.), Helicobacter pylori: physiology and genetics. ASM Press, Washington. 262. Spach, D. H., F. E. Silverstein, and W. E. Stamm. 1993. Transmission of infection by gastrointestinal endoscopy and broncoscopy. Ann Int Med 118:117- 128. 263. Stein, M., F. Bagnoli, R. Halenbeck, R. Rappuoli, W. J. Fantl, and A. Covacci. 2002. c-Src/Lyn kinases activate Helicobacter pylori CagA through tyrosine phosphorylation of the EPIYA motifs. Mol Microbiol 43:971-80. 264. Stein, M., R. Rappouli, and A. Covacci. 2000. Tyrosine phosphorylation of the Helicobacter pylori CagA antigen after cag-driven host cell translocation. Proc Natl Acad Sci U S A 97:1263-1268. 265. Suerbaum, S., and C. Josenhans. 2007. Helicobacter pylori evolution and phenotypic diversification in a changing host. Nat Rev Microbiol 5:441-452. 266. Suerbaum, S., J. Maynard-Smith, K. Bapumia, G. Morelli, N. H. Smith, E. Kunstmann, I. Dyrek, and M. Achtman. 1998. Free recombination within Helicobacter pylori. Proc Natl Acad Sci USA 95:12619-12624. 267. Suerbaum, S., and P. Michetti. 2002. Helicobacter pylori infection. N Engl J Med 347:1175-1186. 268. Sugimoto, M., Y. Yamaoka, and T. Furuta. 2010. Influence of interleukin polymorphisms on development of gastric cancer and peptic ulcer. World J Gastroenterol 16:1188-1200. 269. Suzuki, H., T. Nishizawa, and T. Hibi. 2010. Helicobacter pylori eradication therapy. Future Microbiol 5:639-648. 270. Syder, A. J., J. L. Guruge, Q. Li, Y. Hu, C. M. Oleksiewicz, R. G. Lorenz, S. M. Karam, P. G. Falk, and J. I. Gordon. 1999. Helicobacter pylori attaches to NeuAcα2,3Galβ1,4 glycoconjugates produced in the stomach of transgenic mice lacking parietal cells Mol Cell 3:263-274.

59 271. Talley, N. J., and R. H. Hunt. 1997. What role does Helicobacter pylori play in dyspepsia and non-ulcer dyspepsia? Arguments for and against H. pylori being associated with dyspeptic symptoms. Gastroenterology 113:S67-S77. 272. Talley, N. J., A. R. Zinmeister, A. Weaver, E. P. DiMagno, H. A. Carpenter, G. I. Perez-Perez, and M. J. Blaser. 1991. Gastric adenocarcinoma and Helicobacter pylori infection. J Natl Cancer Inst 83:1734-1739. 273. Tammer, I., S. Brandt, R. Hartig, W. Konig, and S. Backert. 2007. Activation of Abl by Helicobacter pylori: A novel kinase for CagA and crucial mediator of host cell scattering. Gastroenterology 132:1309-1319. 274. Tay, A., H. Mitchell, Q. Dong, K. L. Goh, I. W. Dawes, and R. Lan. 2009. Population structure of Helicobacter pylori among ethnic groups in Malaysia: recent acquisition of the bacterium by the Malay population. BMC Microbiol 9:126. 275. Telford, J., P. Ghiara, M. Dell'Orco, M. Comanducci, D. Burroni, M. Bugnoli, M. Tecce, S. Censini, A. Covacci, and Z. Xiang. 1994. Gene structure of the Helicobacter pylori cytotoxin and evidence of its key role in gastric disease. J Exp Med 179:1653-1658. 276. Tindberg, Y., C. Bengtsson, F. Granath, M. Blennow, O. Nyren, and M. Granstrom. 2001. Helicobacter pylori infection in Swedish school children: lack of evidence of child-to-child transmission outside the family. Gastroenterology 121:310-316. 277. Tokudome, S., W. D. Samsuria, Soeripto, F. Triningsih, S. Suzuki, A. Hosono, T. Triono, I. Wijaya, Sarjadi, I. P. Miranti, R. Ghadimi, and M. A. Moore. 2005. Helicobacter pylori infection appears to be essential for stomach carcinogenesis: observations in Semarang, Indonesia. Cancer Sci 96:873-875. 278. Tomb, J.-F., O. White, A. R. Kerlavage, R. A. Clayton, G. G. Sutton, R. D. Fleischmann, K. A. Ketchum, H. P. Klenk, S. Gill, B. A. Dougherty, K. Nelson, J. Quackenbush, L. Zhou, E. F. Kirkness, S. Peterson, B. Loftus, D. Richardson, R. Dodson, H. G. Khalak, A. Glodek, K. McKenney, L. M. Fitzegerald, N. Lee, M. D. Adams, E. K. Hickey, D. E. Berg, J. D. Gocayne, T. R. Utterback, J. D. Peterson, J. M. Kelley, M. D. Cotton, J. M. Weidman, C. Fujii, C. Bowman, L. Watthey, E. Wallin, W. S. Hayes, M. Borodovsky, P. D. Karp, H. O. Smith, C. M. Fraser, and J. C. Venter. 1997. The complete genome sequence of the gastric pathogen Helicobacter pylori. Nature 388:539.

60 279. Torres, J., G. I. Perez-Perez, Y. Leal-Herrera, and O. Munoz. 1998. Infection with CagA+ Helicobacter pylori strains as a possible predictor of risk in the development of gastric adenocarcinoma in Mexico. Int J Cancer 78:298-300. 280. Tovey, F. 1979. Peptic ulcer in India and Bangladesh. Gut 20:329-347. 281. Tsugane, S., and S. Sasazuki. 2007. Dier and the risk of gastric cancer. Gastric Cancer 10:75-83. 282. Tsuji, S. 2006. The 'Costa Rican enigma' of Helicobacter pylori CagA and gastric cancer. J Gastroenterol 41:716-717. 283. Tsutsumi, R., H. Higashi, M. Higuchi, M. Okada, and M. Hatakeyama. 2003. Attenuation of Helicobacter pylori CagA x SHP-2 signaling by interaction between CagA and C-terminal Src kinase. J Biol Chem 278:3664 -3670. 284. Tu, H.-k., K.-f. Pan, Y. Zhang, W.-q. Li, L. Zhang, J.-l. Ma, J.-y. Li, and W.-c. You. 2010. Manganese superoxide dismutase polymorphism and risk of gastric lesions, and its effects on chemoprevention in a Chinese population. Cancer Epidemiol Biomarkers Prev 19:1089-1097. 285. Uemura, N., S. Okamoto, S. Yamamoto, N. Matsumura, S. Yamaguchi, M. Yamakido, K. Taniyama, N. Sasaki, and R. J. Schlemper. 2001. Helicobacter pylori infection and the development of gastric cancer. N Engl J Med 345:784- 789. 286. Unemo, M., A. Aspholm-Hurtig, D. Iver, J. Bergstrom, T. Boren, D. Danielsson, and S. Teneberg. 2005. The sialic acid binding SabA adhesin of Helicobacter pylori is essential for nonopsonic activation of human neutrophils. J Biol Chem 280:15390-15397. 287. Uyub, A. M., S. M. Raj, R. Visvanathan, M. Nazim, S. Aiyar, and M. Mansur. 1994. Helicobacter pylori infection in north-eastern peninsular Malaysia. Evidence for an unusually low prevalence. Scand J Gastroenterol 29:209-213. 288. van der Ende, A., E. A. Rauws, M. Feller, C. J. Mulder, G. N. Tytgat, and J. Dankert. 1996. Heterogeneous Helicobacter pylori isolates from members of a family with a history of peptic ulcer disease. Gastroenterology 111:638-647. 289. van Doorn, L. J., C. Figueiredo, R. Sanna, A. Plaisier, P. Schneeberger, W. de Boer, and W. Quint. 1998. Clinical relevance of the cagA, vacA, and iceA status of Helicobacter pylori. Gastroenterology 115:58-66. 290. Viala, J., C. Chaput, I. G. Boneca, A. Cardona, S. E. Girardin, A. P. Moran, R. Athman, S. Mémet, M. R. Huerre, A. J. Coyle, P. S. DiStefano, P. J. Sansonetti,

61 A. Labigne, J. Bertin, D. J. Philpott, and R. L. Ferrero. 2004. Nod1 responds to peptidoglycan delivered by the Helicobacter pylori cag pathogenicity island. Nat Immunol 5:1166-1174. 291. Vollmer, J. 2006. TLR9 in health and disease. Int Rev Immunol 25:155-181. 292. Wang, G., M. Z. Humayun, and D. E. Taylor. 1999. Mutation as an origin of genetic variability in Helicobacter pylori. Trends Microbiol 7:488-493. 293. Wang, J., T. G. Blanchard, and P. B. Ernst. 2001. Host inflammatory response to infection. In H. L. T. Mobley, G. L. Mendz, and S. L. Hazell (ed.), Helicobacter pylori: Physiology and genetics. ASM Press. 294. Wang, J. T., J. C. Sheu, J. T. Lin, T. H. Wang, and M. S. Wu. 1993. Direct DNA amplification and restriction pattern analysis of Helicobacter pylori in patients with duodenal ulcer and their families. J Infect Dis 168:1544-1548. 295. Wang, X. Q., P. D. Terry, and H. Yan. 2009. Review of salt consumption and stomach cancer risk: epidemiological and biological evidence. World J Gastroenterol 15:2204-2213. 296. Wang, Y.-H., J.-P. Gorvel, Y.-T. Chu, J.-J. Wu, and H.-Y. Lei. 2010. Helicobacter pylori impairs murine dendritic cell responses to infection. PLoS ONE 5:e10844. 297. Warren, J. R., and B. Marshall. 1983. Unidentified curved bacilli on gastric epithelium in active chronic gastritis. Lancet 321:1273-1275. 298. WCRF, and AICR. 2007. Food, nutrition, physical activity and the prevention of cancer: a global perspective. AICR, Washington, DC. 299. Weeks, D. L., S. Eskandari, D. R. Scott, and G. Sachs. 2000. A H+-gated urea channel: the link between Helicobacter pylori urease and gastric colonization. Science 287:482-485. 300. Wong, B. C. Y., W. H. Wang, D. E. Berg, F. M. Y. Fung, K. W. Wong, W. M. Wong, K. C. Lai, C. H. Cho, W. M. Hui, and S. K. Lam. 2001. High prevalence of mixed infections by Helicobacter pylori in Hong Kong: metronidazole sensitivity and overall genotype. Aliment Pharmacol Ther 15:493-503. 301. Xia, Y., Y. Yamaoka, Q. Zhu, I. Matha, and X. Gao. 2009. A comprehensive sequence and disease correlation analyses for the C-terminal region of CagA protein of Helicobacter pylori. PLoS One 4:e7736. 302. Xiang, Z., S. Censini, P. F. Bayedi, J. L. Telford, N. Figura, R. Rappuoli, and A. Covacci. 1995. Analysis of expression of CagA and VacA virulence factors in 43 strains of Helicobacter pylori reveals that clinical isolates can be divided into two

62 major types and that CagA is not necessary for the expression of the vacuolating cytotoxin. Infect Immun 63:94-98. 303. Xie, F., N. Luo, G. Blackhouse, R. Goeree, and H.-P. Lee. 2008. Cost- effectiveness analysis of Helicobacter pylori screening in prevention of gastric cancer in Chinese. Int J Technol Assess Health Care 24:87-95. 304. Yamaoka, Y. 2008. Roles of Helicobacter pylori BabA in gastroduodenal pathogenesis. World J Gastroenterol 14:4265-4272. 305. Yamaoka, Y. 2008. Roles of the plasticity regions of Helicobacter pylori in gastroduodenal pathogenesis. J Med Microbiol 57:545-553. 306. Yamaoka, Y., H. M. El-Zimaity, O. Gutierrez, N. Figura, J. G. Kim, T. Kodama, K. Kashima, D. Y. Graham, and J. K. Kim. 1999. Relationship between the cagA 3' repeat region of Helicobacter pylori, gastric histology, and susceptibility to low pH. Gastroenterology 117:342-349. 307. Yamaoka, Y., T. Kodama, K. Kashima, D. Graham, and A. Sepulveda. 1998. Variants of the 3' region of the cagA gene in Helicobacter pylori isolates from patients with different H. pylori-associated diseases. J Clin Microbiol 36:2258- 2263. 308. Yamaoka, Y., T. Kodama, M. Kita, J. Imanishi, K. Kashima, and D. Y. Graham. 1999. Relation between clinical presentation, Helicobacter pylori density, interleukin 1beta and 8 production, and cagA status. Gut 45:804-11.

309. Yamaoka, Y., D. H. Kwon, and D. Graham. 2000. A Mr 34,000 proinflammatory outermembrane protein (oipA) of Helicobacter pylori. Proc Natl Acad Sci U S A 97:7533-7538. 310. Yamaoka, Y., O. Ojo, S. Fujimoto, S. Odenbreit, R. Haas, O. Gutierrez, H. M. El-Zimaity, R. Reddy, A. Arnqvist, and D. Y. Graham. 2006. Helicobacter pylori outer membrane proteins and gastroduodenal disease. Gut 55:775-781. 311. Yamazaki, S., A. Yamakawa, Y. Ito, M. Ohtani, H. Higashi, M. Hatakeyama, and T. Azuma. 2003. The CagA protein of Helicobacter pylori is translocated into epithelial cells and binds to SHP-2 in human gastric mucosa. J Infect Dis 187:334-337. 312. Yang, H., S. V. Wu, S. Pichuantes, M. Song, J. Wang, D. Zhou, Z. Xu, S. Quan, A. Polito, and J. H. Walsh. 1999. High prevalence of cagA-positive strains in Helicobacter pylori-infected, healthy, young Chinese adults. J Gastroenterol Hepatol 14:476-480.

63 313. Yeh, J. M., K. M. Kuntz, M. Ezzati, and S. J. Goldie. 2008. Exploring the cost- effectiveness of Helicobacter pylori screening to prevent gastric cancer in China in anticipation of clinical trial results. Int J Cancer 124:157-166. 314. Yoshiyama, H., H. Nakamura, M. Kimoto, K. Okita, and T. Nakazawa. 1999. Chemotaxis and motility of Helicobacter pylori in a viscous environment. J Gastroenterol 34:18-23. 315. Young, K. A., Y. Akyon, D. S. Rampton, S. G. R. G. Barton, R. P. Allaker, J. M. Hardie, and R. A. Feldman. 2000. Quanitative culture of Helicobacter pylori from gastric juice: the potential for transmission. J Med Microbiol 49:343-347. 316. Zhang, J., C. Dou, Y. Song, C. Ji, S. Gu, Y. Xie, and Y. Mao. 2008. Polymorphisms of tumor necrosis factor-alpha are associated with increased susceptibility to gastric cancer: a meta-analysis. J Hum Genet 53:479-489. 317. Zhang, X. M., X. P. Miao, W. Tan, T. Sun, Y. L. Guo, D. Zhao, and D. X. Lin. 2006. [Genetic polymorphisms in the promoter region of cyclooxygenase-2 and their association with risk of gastric cancer] [Abstract only]. Zhongguo Yi Xue Ke Xue Yuan Xue Bao 28:119-123. 318. Zhang, Y., R. H. Argent, D. P. Letley, R. J. Thomas, and J. C. Atherton. 2005. Tyrosine Phosphorylation of CagA from Chinese Helicobacter pylori Isolates in AGS Gastric Epithelial Cells. J Clin Microbiol 43:786-790. 319. Zhang, Z., Q. Zheng, X. Chen, S. Xiao, W. Liu, and H. Lu. 2008. The Helicobacter pylori duodenal ulcer promoting gene, dupA in China. BMC Gastroenterol 8:49. 320. Zhou, W., S. Yamazaki, A. Yamakawa, M. Ohtani, Y. Ito, Y. Keida, H. Higashi, M. Hatakeyama, J. Si, and T. Azuma. 2004. The diversity of vacA and cagA genes of Helicobacter pylori in East Asia. FEMS Immunol Med Microbiol 40:81-87. 321. Zhu, Y. L., S. Zheng, Q. Du, K. D. Qian, and P. C. Fang. 2005. Characterisation of CagA variable region of Helicobacter pylori isolates from Chinese patients. World J Gastroenterol 11:880-884.

64 2 List of Publications

65 2.1 Paper I

Helicobacter ISSN 1523-5378

OriginalEPIYASchmidtBlackwellOxford,1083-43891523-5378©JournalXXX Distinct 2009 Motifs compilationTheUKHELHelicobacter et PublishingAr Authorsal. ticleare Associated © Ltd 2009 Blackwell with Distinct Publishing Ethnic Ltd, Groups Helicobacter XX: xxÐxxcagA EPIYA Motifs are Associated with Ethnic Diversity in Malaysia and Singapore Heather-Marie A. Schmidt,* Khean-Lee Goh,† Kwong Ming Fock,‡ Ida Hilmi,† Subbiah Dhamodaran,‡ David Forman§ and Hazel Mitchell*

*The School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, Australia, †Department of Medicine, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia, ‡Division of Gastroenterology, Department of Medicine, Changi General Hospital, Singapore, §Centre for Epidemiology and Biostatistics, Faculty of Medicine and Health, University of Leeds, UK

Keywords Abstract CagA, EPIYA motif, Helicobacter pylori, Malaysia, Background: In vitro studies have shown that the biologic activity of CagA is Singapore, gastric cancer. influenced by the number and class of EPIYA motifs present in its variable region Reprints request to: Hazel Mitchell, PhD, as these motifs correspond to the CagA phosphorylation sites. It has been Professor of Medical Microbiology, School of hypothesized that strains possessing specific combinations of these motifs may Biotechnology and Biomolecular Sciences, be responsible for gastric cancer development. This study investigated the University of New South Wales, Sydney, NSW prevalence of cagA and the EPIYA motifs with regard to number, class, and 2052, Australia. E-mail: [email protected] patterns in strains from the three major ethnic groups within the Malaysian and Singaporean populations in relation to disease development. Materials and methods: Helicobacter pylori isolates from 49 Chinese, 43 Indian, and 14 Malay patients with functional dyspepsia (FD) and 21 gastric cancer (GC) cases were analyzed using polymerase chain reaction for the presence of cagA and the number, type, and pattern of EPIYA motifs. Additionally, the EPIYA motifs of 47 isolates were sequenced. Results: All 126 isolates possessed cagA, with the majority encoding EPIYA-A (97.6%) and all encoding EPIYA-B. However, while the cagA of 93.0% of Indian FD isolates encoded EPIYA-C as the third motif, 91.8% of Chinese FD isolates and 81.7% of Chinese GC isolates encoded EPIYA-D (p < .001). Of Malay FD isolates, 61.5% and 38.5% possessed EPIYA-C and EPIYA-D, respectively. The majority of isolates possessed three EPIYA motifs; however, Indian isolates were significantly more likely to have four or more (p < .05). Conclusion: Although, H. pylori strains with distinct cagA-types are circulating within the primary ethnic groups resident in Malaysia and Singapore, these genotypes appear unassociated with the development of GC in the ethnic Chinese population. The phenomenon of distinct strains circulating within different ethnic groups, in combination with host and certain environmental factors, may help to explain the rates of GC development in Malaysia.

Helicobacter pylori infects the gastric mucosa of more than While in some populations cagA-positive H. pylori strains half of the world’s population [1,2]. Although all infected have been associated with the development of GC [5–8], individuals develop H. pylori-associated gastritis, a signiﬁcant the ﬁnding that in many East Asian populations the minority will progress to more severe gastroduodenal prevalence of cagA-positive strains is greater than 90% is diseases, such as peptic ulcer disease (PUD) and gastric likely to preclude a direct association between cagA per se cancer (GC). Morbidity and mortality from GC represents and the development of GC [9–12]. a major global burden with GC ranked as the fourth most It is now known that the cagA gene is located on a common cancer worldwide and the second most common pathogenicity island (the cagPAI), and that the CagA protein cause of cancer-related death [3,4]. Although studies is translocated into host gastric epithelial cells via a type IV investigating the etiology of GC have predominantly secretion system also encoded by the cagPAI. Once inside, focused on the role of bacterial factors, due to inconsistent CagA is tyrosine phosphorylated at the Glu-Pro-Ile-Tyr-Ala associations between studies and populations, the role and (EPIYA) motifs by the cellular kinases Src [13] and Abl relative importance of these factors remains controversial. [14], which are involved in early and sustained CagA

Schmidt et al. EPIYA Motifs are Associated with Distinct Ethnic Groups phosphorylation, respectively. CagA binds the phosphatase and Singapore, respectively [29,31,32] and a low rate of SHP-2 in a phosphorylation-dependent manner, resulting GC (ASR males: 2.6 and 6.3, and females: 1.3 and 4.0, in characteristic phenotypic changes such as cellular respectively) [33]. elongation, pseudopodia formation, and cytoskeletal Thus, Malaysia and Singapore represent an ideal arrangement in vitro, known as the hummingbird pheno- population in which to study the potential contribution of type [15–19]. These interactions are determined by the cagA to GC development by investigating the prevalence of presence, type, and number of the EPIYA tyrosine phos- cagA and the relationship between the number, class, and phorylation motifs, encoded in the 3’ variable region of the patterns of EPIYA motifs with respect to ethnicity and cagA gene [20]. EPIYA motifs have been divided into four disease state. classes, designated EPIYA-A to -D, determined by the 32– 40 amino-acid sequence flanking the core glutamine- proline-isoleucine-tyrosine-alanine motif [21] While, Materials and Methods EPIYA-A and -B are found in the majority of isolates, Patients EPIYA-C and -D have a distinct geographic association, with cagA possessing strains from the west most frequently As part of a larger study examining the role of bacterial, encoding a single EPIYA-A and -B motif, followed by one host, and environmental factors in GC development, to three EPIYA-C motifs [22]. In East Asia, the EPIYA-C gastric biopsies were obtained from consecutively enrolled motif has been shown to be replaced by a single EPIYA-D unrelated patients undergoing routine endoscopic motif in the majority of cagA possessing isolates [23,24]. examination of gastrointestinal symptoms at the University The EPIYA-C and -D motifs are the major sites of tyrosine Hospital, Kuala Lumpur, Malaysia, and the Changi Hospital, phosphorylation within CagA and are not only necessary Singapore (between mid-2004 and 2007). Based on endo- for the interaction between CagA and SHP-2, but also scopic and histologic examination, patients were diagnosed determine the strength of this interaction, the extent of with either distal GC or functional dyspepsia (FD). SHP-2 deregulation, and the extent of hummingbird phenotype induction in vitro [17,20,24]. H. pylori Culture and Identification In addition to the role of CagA in perturbing tyrosine phosphorylation-dependent pathways, a recent study in Gastric biopsies were stored at –70 °C in brain heart infusion vitro has shown that the EPIYA motif acts as a membrane (BHI) broth containing 20% glycerol and transported to targeting signal for CagA in a phosphorylation-independent Australia on dry ice. Gastric biopsies were smeared manner [25]. Thus, the heterologous nature of cagA and onto campylobacter selective medium, incubated, and resultant functional differences of these motifs within the identified using morphology, microscopy, and biochemical protein may play an important role in determining GC tests as previously described [35,36]. development as previously suggested [20,21,24,26]. To date primary cultures of H. pylori have been obtained Indeed, a recent study has demonstrated that the number from 126 patients (49 Chinese, 43 Indian, and 13 Malay of EPIYA-C motifs is an important determinant for cancer FD patients, and 21 Chinese GC patients). Clinical features risk in a western population [27]. However, this is yet to of the patients for which H. pylori isolates have been be confirmed in other populations. obtained (Table 1). Within Malaysia and Singapore a considerable disparity exists between the prevalence of H. pylori and the develop- Preparation of Genomic DNA ment of GC within the three primary ethnic groups [28]. Ethnic Chinese residing in these two countries have been Genomic DNA from one single colony isolate per patient reported to have an H. pylori seropositivity of between was extracted using the Gentra Puregene DNA extraction 26.7–58.6% in Malaysia [29–31] and 45.8–46.8% in kit (Gentra, Minneapolis, MN, USA). Singapore [32], and have a relatively high age standardized risk (ASR) for GC (males: 11.9 and 25.7, and females: 8.7 Analysis of cagA and the EPIYA Motifs by PCR and 12.6, respectively) [33,34]. Conversely, while the H. pylori seropositivity in Indians resident in Malaysia and Isolates were initially typed for cagA using two primer pairs Singapore (49.4–52.3% and 45.1–51.1%, respectively) is specific for the cagA constant region, cagAsbra F/R [37] and generally higher than ethnic Chinese [29–32], the rate of cagA F/R [38] as previously described [37]. GC is lower (ASR males: 12.9 and 8.4, and females: 6.3 and Polymerase chain reaction (PCR) for the EPIYA motifs 7.9, respectively) [33,34]. In contrast the ethnic Malays was performed in separate reactions according to the have both a low level of H. pylori infection with seroposi- method described by Argent et al. [21] using the common tivity rates of 11.9–29.2% and 26.1–30.1% in Malaysia forward primer cag2 with one of three reverse primers,

EPIYA Motifs are Associated with Distinct Ethnic Groups Schmidt et al. cagAP1C, cagAP2TA, and cagAP3E, for the amplification Western clinical isolates deposited in GenBank (accession of EPIYA-A, -B, and -C and -D, respectively. PCR was no. AB057038-47, AB057051-9, and AB057060-2), to performed in a 25 µL reaction containing 20–40 ng of enable specific amplification of EPIYA-C and -D, genomic DNA, 10 pmol of primer, 2 mmol/L deoxynucle- respectively, when used in combination with the forward oside triphosphates, 31.25 nmol magnesium chloride, 1× primer cag2. PCR was performed as described for the PCR reaction buffer, and 0.825 U of Taq polymerase EPIYA primer pairs above with an increased annealing (Fisher Biotech, Subiaco, WA, Australia). Samples were temperature of 62 °C. denatured for 5 minutes at 95 °C, and then cycled 35 times, 95 °C for 20 seconds, 56 °C for 20 seconds, and Validation by DNA Sequencing 72 °C for 40 seconds; followed by a final elongation at 72 °C for 7 minutes. All PCR products were visualized by To validate the specificity of CagAWest and CagAEast, agarose gel electrophoresis. sequencing was conducted for 47 isolates, including 10 Primer sequences are presented in Table 2. isolates which were ambiguous for CagAWest or CagAEast, using the cag2 forward and cag4 reverse [21] or the CagAv F/R primers [39]. Sequencing reactions were performed Primer Design and Detection of EPIYA-C and -D with the BigDye™ Terminator version 3.1 (Applied As cagAP3E does not discriminate between EPIYA-C and Biosystems, Foster City, CA, USA). Sequencing analysis -D we designed two additional reverse primers, cagAWest was performed on an ABI3730 Capillary DNA sequencer and cagAEast, based on sequences of East Asian and (GENterprise, Mainz, Germany).

Table 1 Clinical features of 126 patients for which H. pylori isolates were obtained

Total no. No. of female No. of male (mean age ± standard (mean age ± standard (mean age ± standard Ethnicity Diagnosis deviation) deviation) deviation)

Chinese Functional dyspepsia 49 (55.2 ± 14.3) 18 (56.6 ± 13.5) 31 (54.5 ± 14.9) Gastric cancer 21 (66.1 ± 8.2) 10 (65.5 ± 10.7) 11 (67.5 ± 5.5) Indian Functional dyspepsia 44 (47.3 ± 16.2) 25 (49.6 ± 15.6) 23 (44.6 ± 16.9) Malay Functional dyspepsia 14 (34.6 ± 11.6) 4 (41 ± 17.9) 10 (32 ± 7.9)

Table 2 PCR primers and conditions used in this study

Gene or region Primer name Primer sequence (5′ to 3′): Size (bp) cagA constant region CagAsbra (37) f: ATG ATG GCG TGA TGT TTG T 823 r: TTT TCA AGG TCG CTT TTT GC CagA (38) f: GAT AAC AGG CAA GCT TTT GAG G 349 r: CTG CAA AAG ATT GTT TGG CAG A cagA variable region CagA variable (39) f: GAA TGT CTG ATA AAC TTG AAA 1152 r: GCG TAT GTG GCT GTT AGT AGC G Cag2 (21) f: GGA ACC CTA GTC GGT AAT G 550–800 Cag4 (21) r: ATC TTT GAG CTT GTC TAT CG EPIYA-A CagAP1Cb (21) r: GTC CTG CTT TCT TTT TAT TAA CTT KAG Cc 264 EPIYA-B CagAP2TAb (21) r: TTT AGC AAC TTG AGC GTA AAT GGG 306 EPIYA-C, EPIYA-D CagAP3Eb (21) r: ATC AAT TGT AGC GTA AAT GGG 468 EPIYA-C CagAWestb,a r: TTT CAA AGG GAA AGG TCC GCC 501 EPIYA-D CagAEastb,a r: AGA GGG AAG CCT GCT TGA TT 495 aThis study. bUsed in combination with the forward primer cag2 F. f: forward primer; r: reverse primer. cK: G or T.

Schmidt et al. EPIYA Motifs are Associated with Distinct Ethnic Groups

Figure 1 Examples of PCR ampliﬁcation of all ﬁve EPIYA motifs; (A) shows a typical western strain with the EPIYA-ABC pattern and (B) a typical East Asian strain with the EPIYA-ABD pattern. Lane 1: molecular weight marker; Lane 2–6: results for PCR for EPIYA-A, -B, -C/D, -C, and -D, respectively.

Representative sequences have been deposited in observed in the presence of the cagA variable region GenBank under accession numbers EU369652–EU369656. between ethnic and disease groups, all isolates from Chinese, Indian, and Malay FD patients and from Chinese GC patients, respectively being shown to possess the cagA Statistical Analysis variable region (p > .05). Fischer’s exact test was used to calculate statistical significance, p < .05 was considered significant. Number, Class, and Pattern of EPIYA Motifs Between Ethnic Groups Ethical Approval A single forward and multiple reverse primers specific for This study was approved by the Human Research Ethics each motif class were used to differentially detect the Committee (HREC) of the University of New South Wales EPIYA motifs; enabling not only the class and number, but (HREC 08115 and HREC 02144). also the pattern in which they occur to be determined. An example of the PCR detection of the different EPIYA motifs is shown in Fig. 1. Isolates from Malaysia and Singapore Results were shown to possess between two and five EPIYA motifs. Although, most FD isolates (84 of 105, 80.0%) Prevalence of cagA possessed three EPIYA motifs, there was a significant One hundred and twenty-six clinical isolates from patients difference between Chinese and Indian FD isolates, with undergoing endoscopy in Malaysia and Singapore were Indian isolates possessing a greater number of motifs (> 3) analyzed by PCR using two primer pairs detecting the cagA significantly more frequently than Chinese isolates, the constant region. The majority of isolates from Chinese (48 prevalence being 30.2% and 6.1%, respectively (p < .01). of 49), Indian (42 of 43), and Malay (13 of 13) FD patients EPIYA-A and -B were found in the majority of isolates and all 21 Chinese GC isolates were shown to possess the from all groups (Table 3); however, the possession of cagA constant region with no significant difference between EPIYA-C and -D demonstrated a significant difference in ethnic or disease groups (p > .05). relation to ethnicity with 40 of 43 (93.0%) of Indian All isolates were evaluated for the presence of the cagA isolates possessing EPIYA-C, as compared with only four of variable region through the detection of the specific EPIYA 49 (8.2%) Chinese FD isolates (p < .0001). Conversely, motifs. From these analyses no significant difference was only three of 43 (7.0%) Indian isolates possessed EPIYA-D,

Table 3 Patterns, displaying classes, number, and repeats, for EPIYA motifs observed according to ethnicity and disease state

No. of isolates (%) EPIYA-BC EPIYA-BD EPIYA-ABC EPIYA-ABD EPIYA-ABBD EPIYA-ABCC EPIYA-ABCCC

Chinese FD 49 0 1 (2.04%) 2 (4.08%) 43 (87.76%) 1 (2.04%) 2 (4.08%) 0 Indian FD 43 1 (2.33%) 1 (2.33%) 26 (60.47%) 2 (4.65%) 0 12 (27.91%) 1 (2.33%) Malay FD 13 1 (7.69%) 0 6 (46.15%) 5 (38.46%) 0 1 (7.69%) 0 Chinese GC 21 0 0 0 18 (85.71%) 0 3 (14.29%) 0 Total 126 2 (1.59%) 2 (1.59%) 34 (26.98%) 68 (53.97%) 1 (0.79%) 18 (14.29%) 1 (0.79%)

FD, functional dyspepsia; GC, gastric cancer.

EPIYA Motifs are Associated with Distinct Ethnic Groups Schmidt et al. as compared with 45 of 49 (91.8%) Chinese FD isolates [12,41–45] and may be partially explained by the (p < .0001). difference in sequence and number of PCR primers The most common pattern observed in Indian FD used to detect cagA. The finding that two isolates were PCR isolates was EPIYA-ABC (26 of 43) with only two of 43 negative for the cagA constant region but were positive possessing ABD. Conversely, the most common pattern for the cagA variable region questions the adequacy of observed for the Chinese FD isolates was EPIYA-ABD (43 of utilizing a single primer set for the detection of cagA, 49) with two of 49 being ABC (p < .0001). Six of the 13 especially when strains are isolated from different Malay isolates possessed EPIYA-ABC and five of 13 populations [42,46,47]. The high prevalence of cagA possessed EPIYA-ABD, which was significantly different identified in the current study precluded any association from the frequency of EPIYA-ABD in Indian FD isolates between cagA per se and more severe disease development. (p < .01) and both EPIYA-ABC and EPIYA-ABD in Chinese In the current study we examined cagA subtypes in FD isolates (p < .001). terms of the biologically significant EPIYA motifs using the Repeats of the EPIYA motifs were observed most fre- methodology of Argent et al. [21], which employs single quently for EPIYA-C with 15 of 105 FD isolates (12 of 43 forward and multiple reverse primers specific for either Indian, two of 49 Chinese, one of 13 Malay) possessing EPIYA-A, EPIYA-B, or the consensus sequence of EPIYA- two copies of EPIYA-C and one of 105 FD isolates (one of C and EPIYA-D. However, this method is unable to differ- 43 Indian isolates) possessing three copies of EPIYA-C. entiate between the EPIYA-C and EPIYA-D motifs, an One isolate, from an ethnic Chinese FD patient, was important consideration in our study, and thus we shown to possess a repeated EPIYA-B motif. However, no designed two additional reverse primers, one specific for repeats were observed for EPIYA-A or EPIYA-D in the EPIYA-C and one specific for EPIYA-D. Although an isolates investigated. alternative method recently published which employs a single primer set which amplifies a product whose fragment length allows differentiation between three EPIYA Comparison of Number, Class, and Pattern of EPIYA motifs (ABC) or four EPIYA motifs (ABCC) [48], to our Motifs in Isolates from Ethnic Chinese FD and GC knowledge this method has not been tested on isolates As shown in Table 3 there was no significant difference with an East Asian-type CagA, specifically the EPIYA-D. observed between Chinese FD and GC isolates with respect Given that the authors acknowledged their method was to the number, class, or patterns observed (p > .05). All GC unable to differentiate between isolates with the same isolates possessed EPIYA-A and -B, while three of 24 number but different classes of the EPIYA motifs, we possessed EPIYA-C and 21 of 24 possessed EPIYA-D. The considered this method to be inappropriate for use in our most frequent EPIYA motif pattern detected was ABD (18 population which was likely to contain East Asian strains of 21). Eighteen of 21 Chinese GC isolates possessed three and thus the EPIYA-D motif. EPIYA motifs, while three of 21 possessed four motifs. In the current study we identified EPIYA-A in the majority Comparison of the Chinese FD and GC isolates, however, (97.6%) and EPIYA-B in all Malaysian and Singaporean revealed no significant difference in the prevalence of the isolates, a finding that is in agreement with previously larger number of repeats (p > .05). reported studies [1,48,49]. However, while 93.0% of isolates from ethnic Indian patients contained the typically Western EPIYA-C motif, and thus an EPIYA-ABC pattern Validation by DNA Sequencing (60.5%), the majority of isolates (91.8%) from ethnic Sequences obtained for the 35 isolates used for validation Chinese resident in Malaysia and Singapore contained the confirmed the results obtained by PCR analysis in all cases. typically east Asian EPIYA-D, resulting in an EPIYA-ABD EPIYA motifs for the nine isolates for which the PCR pattern (87.9%). These observations reflect reports that results had been ambiguous were obtained by sequencing isolates from mainland China and the Indian subcontinent and were included in the statistical analysis. have East Asian and Western-type cagA, respectively [9,50]. For example, in a study of isolates from mainland China, Zhu et al. [9] identified that the third EPIYA Discussion motif in Chinese isolates possessed a different amino acid We detected cagA in all of the H. pylori isolates investigated. sequence to that reported previously for Western isolates, Although this prevalence is higher than that reported including European strains. In contrast, in a study of previously for the Malaysian population [11,40], this H. pylori isolates from the Indian subcontinent, Devi et al. degree of variation between studies in an individual [50] reported 99% of isolates studied clustered with country is not appreciably different from other studies in hpEurope, rather than hpEastAsia, based on genotyping of east Asian populations, such as Korea, Thailand, and Japan the cagPAI including the cagA gene and EPIYA motifs.

Given this, the results of our study would suggest that urgently required to determine the contribution of such immigrants from China and India resident in Malaysia and factors and their role in combination with H. pylori virulence Singapore have retained over generations the distinct determinants in GC development. strains typical of their ethnic background. As the number of isolates we obtained from Indian GC Interestingly, ethnic Malays resident in Malaysia and patients was extremely small, we were unable to confirm Singapore displayed a more mixed H. pylori genotype than the claim by Basso et al. [27] that EPIYA-C repeats increase that observed in ethnic Chinese or Indians, with both East the GC risk among subjects infected with western strains. Asian (38.5%) and Western (61.5%) cagA-possessing However, it is possible to comment upon the frequency of strains being common. Although the number of Malay repeated EPIYA-C in strains from patients suffering FD. isolates available for analysis is small, power calculations The frequency of repeated EPIYA-C motifs reported in the have shown that a sample size of 13 is large enough to present study (13 of 43, 30.2%) is not significantly differ- detect statistical significance at a level of 0.05 with 80% ent from that reported in comparable studies from Greece power in a two-sided test (results not shown). Interestingly, (12 of 58 adults, 20.7%) and Italy (12 of 42, 28.6%) a recent study in the Thai population observed a similar [27,48]. When comparing the GC rates (ASR) of these finding, 26.8% of 41 patients having a western-type cagA populations, Italy has a slightly higher rate (9.7 and 18.8 and 53.7% an eastern-type cagA [43]. for females and males, respectively), than Greece (5.9 and Our finding that the Malay population had both East 12.0 for females and males, respectively) or Indians resi- Asian and Western EPIYA motifs strongly supports the dent in Malaysia and Singapore (6.3 and 7.9 for females, hypothesis of Graham et al. that ethnic Malays only and 12.9 and 8.4 for males, respectively) [33,34,56]. Any recently acquired H. pylori from Indian and Chinese immi- apparent differences may be explained by the contribution grants to Malaysia and Singapore [51]. This hypothesis of host genetics and environmental factors, which are provides an explanation for the comparatively low well known to vary between populations. Indeed, studies H. pylori infection and related disease development rates in of the Indian population have emphasized the role of this population [51]. Further studies on a larger number of diet in explaining this enigma [30,57], although this is isolates are, however, required to confirm this hypothesis. controversial [58]. The finding that strains with distinct EPIYA patterns In conclusion, the present study has demonstrated for were associated with specific ethnic groups within a single the first time that distinct H. pylori strains, with regard to country not only reflects previous observations in Malay- the EPIYA motifs, are circulating within the three primary sia regarding vacA but also studies of vacA and iceA in dif- ethnic groups, Chinese, Indian, and Malay, resident in ferent ethnic groups resident in the USA [40,52]. Malaysia and Singapore. However, in Chinese subjects Previous studies have suggested that repeats of EPIYA-C the possession of EPIYA motifs reported to be associated or the presence of the EPIYA-D motif are associated with with distinct morphologic changes appear not to be tighter SHP-2 binding and more potent deregulation of associated with the development of GC in this population. cellular signaling [20,24,53], thus would be expected to be Functional studies of the strains isolated from different associated with the development of more severe H. pylori- ethnic groups and disease outcomes in this study may related disease. However, in the current study there was help elucidate any association between the EPIYA motifs no significant difference between Chinese FD and GC and GC development. isolates with regards to the number, class, or pattern (p > 0.05) of the EPIYA motifs. The lack of association between specific EPIYA motifs or the number of EPIYA Acknowledgements and Disclosures motifs, and GC development is similar to the findings of We would like to thank Dr Quanjiang Dong for his assistance with two recent study in East Asia, specifically, Korea and primer design and Dr Ruiting Lan for his assistance with the China, where no association between EPIYA motifs and statistical analysis. We would also like to thank Prof Lars Engstrand for the use of his laboratory and facilities for a part of the development of gastroduodenal disease was found this study. [9,54]. It is now well accepted that GC development is a Declaration of Interests: This study was funded by Cancer multifactorial process with host genetic and dietary factors, Council NSW, Australia (REF 66/04). also playing a role in GC development. Our finding in the Chinese population that no association exists between the References EPIYA motifs and GC development would suggest that other factors, including host predispositions such as IL-1 1 Azuma T. Helicobacter pylori CagA protein variation associated with gastric cancer in Asia. J Gastroenterol 2004;39:97–103. gene polymorphisms or environmental stimuli, such as 2 Dooley CP, Cohen H, Fitzgibbons PL, et al. Prevalence of high dietary salt intake, may play a more determining role Helicobacter pylori infection and histologic gastritis in asymptomatic in GC development (reviewed in [55]). Further studies are persons. N Engl J Med 1989;321:1562–6.

3 Parkin DM, Bray FI, Devesa SS. Cancer burden in the year 2000. 22 Dong Q, O’Sullivan M, Hall W, et al. Identification of a new The global picture. Eur J Cancer 2001;37(Suppl. 8):S4–S66. segment involved in cagA 3′ region variation of Helicobacter pylori. 4 Parkin DM, Bray F, Ferlay J, Pisani P. Estimating the world cancer FEMS Immunol Med Microbiol 2002;33:51–5. burden: Globocan 2000. Int J Cancer 2001;94:153–6. 23 Hatakeyama M, Higashi H. Helicobacter pylori CagA: a 5 Covacci A, Censini S, Bugnoli M, et al. Molecular characterisation new paradigm for bacterial carcinogenesis. Cancer Sci of the 128-kDa immunodominant antigen of Helicobacter pylori 2005;96:835–43. associated with cytotoxicity and duodenal ulcer. Proc Natl Acad Sci 24 Higashi H, Tsutsumi R, Fujita A, et al. Biological activity of the USA 1993;90:5791–5. Helicobacter pylori virulence factor CagA is determined by variation 6 Blaser MJ, Perez-Perez GI, Kleanthous H, et al. Infection with in the tyrosine phosphorylation sites. Proc Natl Acad Sci USA Helicobacter pylori strains possessing cagA is associated with an 2002;99:14428–33. increased risk of developing adenocarcinoma of the stomach. 25 Higashi H, Yokoyama K, Fujii Y, et al. EPIYA motif is a membrane Cancer Res 1995;55:2111–5. targeting signal of Helicobacter pylori virulence factor CagA in 7 Parsonnet J, Friedman GD, Orentreich N, Vogelman H. Risk for mammalian cells. J Bacteriol 2005;280:23130–7. gastric cancer in people with CagA positive or CagA negative 26 Hatakeyama M. Oncogenic mechanisms of the Helicobacter pylori Helicobacter pylori infection. Gut 1997;40:297–301. CagA protein. Nat Rev Cancer 2004;4:688–94. 8 Huang JQ, Zheng GF, Sumanac K, Irvine EJ, Hunt RH. Meta- 27 Basso D, Zambon C-F, Letley DP, et al. Clinical relevance of analysis of the relationship between cagA seropositivity and gastric Helicobacter pylori cagA and vacA gene polymorphisms. cancer. Gastroenterology 2003;125:1636–44. Gastroenterology 2008;135:91–9. 9 Zhu YL, Zheng S, Du Q, Qian KD, Fang PC. Characterisation of 28 Kang JY, Yeoh KG, Ho KY, et al. Racial differences in CagA variable region of Helicobacter pylori isolates from Chinese Helicobacter pylori seroprevalence in Singapore: correlation patients. World J Gastroenterol 2005;11:880–4. with differences in peptic ulcer frequency. J Gastroenterol Hepatol 10 Yu J, Leung WK, Go MYY, et al. Relationship between Helicobacter 1997;12:655–9. pylori babA2 status with gastric epithelial cell turnover and 29 Goh KL. Helicobacter pylori infection in Malaysia. In: Faculty of premalignant gastric lesions. Gut 2002;51:480–4. Medicine (vol. Doctor of Medicine). Kuala Lumpur: University of 11 Ramelah M, Aminuddin A, Alfizah H, et al. cagA gene variants in Malaya, 1997. Malaysian Helicobacter pylori strains isolated from patients of 30 Goh KL, Cheah PL, Md N, Path M, Quek KF, Parasakthi different ethnic groups. FEMS Immunol Med Microbiol 2005;44:239– N. Ethnicity and H. pylori as risk factors for gastric cancer in 42. Malaysia: a prospective case control study. Am J Gastroenterol 12 Zhou W, Yamazaki S, Yamakawa A, et al. The diversity of vacA and 2007;102:40–5. cagA genes of Helicobacter pylori in East Asia. FEMS Immunol Med 31 Goh KL, Parasakthi N. The racial cohort phenomenon: Microbiol 2004;40:81–7. seroepidemiology of Helicobacter pylori infection in a multiracial 13 Selbach M, Moese S, Hauck CR, Meyer TF, Backert S. Src is the South-East Asian country. Eur J Gastroen Hepat 2001;13:177–83. kinase of Helicobacter pylori CagA protein in vitro and in vivo. J Biol 32 Ang TL, Fock KM, Dhamodaran S, Teo EK, Tan J. Racial differences Chem 2002;277:6775–8. in Helicobacter pylori, serum pepsinogen and gastric cancer 14 Tammer. Activation of Abl by Helicobacter pylori: a novel kinase for incidence in an urban Asian population. J Gastroenterol Hepatol CagA and crucial mediator of host cell scattering. Gastroenterology 2005;20:1603–9. 2007;132:1309–19. 33 Look M, Gao F, Low CH, Nambiar R. Gastric cancer in Singapore. 15 Bagnoli F, Buti L, Tompkins L, Covacci A, Amieva MR. Helicobacter Gastric Cancer 2001;4:219–22. pylori CagA induces a transition from polarized to invasive 34 Lim GCC, Halimah Y. Second report of the national registery. phenotypes in MDCK cells. Proc Natl Acad Sci USA Cancer incidence in Malaysia 2003. Kuala Lumpur: National 2005;102:16339–44. Cancer Registry; 2004. 16 Segal ED, Cha J, Lo J, Falkow S, Thompkins LS. Altered states: 35 Lee A, O’Rourke J, Ungria MCD, Robertson B, Daskalopoulos G, involvement of phosphorylated CagA in the induction of host Dixon MF. Erratum: a standardized mouse model of Helicobacter cellular growth changes by Helicobacter pylori. Proc Natl Acad Sci USA pylori infection – introducing the Sydney strain. Gastroenterology 1999;96:14559–64. 1997;113:732. 17 Higashi H, Tsutsumi R, Muto S, et al. SHP-2 tyrosine phosphatase 36 Lee A, O’Rourke J, Ungria MCD, Robertson B, Daskalopoulos G, as an intracellular target of Helicobacter pylori CagA protein. Science Dixon MF. A standardized mouse model of Helicobacter pylori 2002;295:683–6. infection – introducing the Sydney strain. Gastroenterology 18 Higashi H, Nakaya A, Tsutsumi R, et al. Helicobacter pylori CagA 1997;112:1386–97. induces Ras-independent morphogenetic response through SHP-2 37 Nilsson C, Sillen A, Eriksson L, et al. Correlation between cag recruitment and activation. J Biol Chem 2004;279:17205–16. pathogenicity island composition and Helicobacter pylori-associated 19 Backert S, Moese S, Selbach M, Brinkmann V, Meyer TF. gastroduodenal disease. Infect Immun 2003;71:6573–81. Phosphorylation of tyrosine 972 of the Helicobacter pylori CagA 38 Yamaoka Y, El-Zimaity HM, Gutierrez O, et al. Relationship protein is essential for induction of a scattering phenotype in between the cagA 3′ repeat region of Helicobacter pylori, gastric gastric epithelial cells. Mol Microbiol 2001;42:631–44. histology, and susceptibility to low pH. Gastroenterology 20 Naito M, Yamazaki T, Tsutsumi R, et al. Influence of EPIYA-repeat 1999;117:342–9. polymorphism on the phosphorylation dependent biological 39 Azuma T, Yamakawa A, Yamazaki S, et al. Correlation between activity of Helicobacter pylori CagA. Gastroenterology 2006;130:1181– variation of the 3′ region of the cagA gene in Helicobacter pylori and 90. disease outcome in Japan. J Infect Dis 2002;186:1621–30. 21 Argent RH, Zhang Y, Atherton JC. Simple method for 40 Tan HJ, Rizal AM, Rosmadi MY, Goh KL. Distribution of determination of the number of Helicobacter pylori CagA variable Helicobacter pylori cagA, cagE and vacA in different ethnic region EPIYA tyrosine phosphorylation motifs by PCR. J Clin groups in Kuala Lumpur, Malaysia. J Gastroenterol Hepatol Microbiol 2005;43:791–5. 2005;20:589–94.

41 Shin A, Shin HR, Kang D, Park SK, Kim C-S, Yoo K-Y. A nested Heterogeneity in the activity of Mexican Helicobacter pylori strains case–control study of the association of Helicobacter pylori infection on gastric epithelial cells and its association with diversity in the with gastric adenocarcinoma in Korea. Br J Cancer 2005;92:1273– cagA gene. Infect Immun 2007;75:3445–54. 5. 50 Devi SM, Ahmed I, Francalacci P, et al. Ancestral European 42 Miehlke S, Kibler K, Kim JC, et al. Allelic Variation in the cagA roots of Helicobacter pylori in India. BMC Genomics 2007;8:184. gene of Helicobacter pylori obtained from Korea compared to the 51 Graham DY, Yamaoka Y, Malaty HM. Thoughts about populations United States. Am J Gastroenterol 1996;91:1322–5. with unexpected low prevalences of Helicobacter pylori infection. 43 Yamazaki S, Kato S, Matsukura N, et al. Identiﬁcation of Trans R Soc Trop Med Hyg 2007;101:849–51. Helicobacter pylori and the cagA genotype in gastric biopsies using 52 Yamaoka Y, Malaty HM, Osato MS, Graham DY. Conservation of highly sensitive real-time PCR as a new diagnostic tool. FEMS Helicobacter pylori genotypes in different ethnic groups in Houston, Immunol Med Microbiol 2005;44:261–8. Texas. J Infect Dis 2000;181:2083–6. 44 Vilaichone R, Mahachai V, Tumwasorn S, Wu J, Graham DY, 53 Fu HY, Asahi K, Hayashi Y, et al. East Asian-type Helicobacter pylori Yamaoka Y. Molecular epidemiology and outcome of Helicobacter cytotoxin-associated gene A has a more signiﬁcant effect on pylori infection in Thailand: a cultural cross roads. Helicobacter growth of rat gastric mucosal cells than the Western type. J 2004;9:453–9. Gastroenterol Hepatol 2006;22:355–62. 45 Ito Y, Azuma T, Ito S, et al. Analysis and typing of the vacA gene 54 Choi KD, Kim N, Lee DH, et al. Analysis of the 3′ variable region from cagA-positive strains of Helicobacter pylori isolated in Japan. of the cagA gene of Helicobacter pylori in isolated Koreans. Dig Dis Sci J Clin Microbiol 1997;35:1710–4. 2007;52:960–6. 46 Pan Z, van der Hulst R, Feller M, et al. Equally high prevalences of 55 Miwa H, Go MF, Sato N. H. pylori and gastric cancer: the Asian infection with cagA-positive Helicobacter pylori in Chinese patients enigma. Am J Gastroenterol 2002;97:1106–12. with peptic ulcer disease and those with chronic gastritis- 56 Ferlay J, Bray F, Pisani P, Parkin DM. GLOBOCAN 2002: cancer associated dyspepsia. J Clin Microbiol 1997;35:1344–7. incidence, mortality and prevalence worldwide version 2.0. (vol. 47 Zhou J, Zhang J, Xu C, He L. cagA genotype and variants in Chinese 2005). Version 2.0 ed. Lyon: IARC Cancer Base No. 5, Lyon, IARC Helicobacter pylori strains and relationship to gastroduodenal press, 2005.; 2004. diseases. J Med Microbiol 2004;53:231–5. 57 Singh K, Ghoshal UC. Causal role of Helicobacter pylori infection in 48 Panayotopoulou EG, Sgouras DN, Papadakos K, et al. Strategy gastric cancer: an Asian enigma. World J Gastroenterol to characterize the number and type of repeating EPIYA 2006;12:1346–51. phosphorylation motifs in the carboxyl terminus of CagA 58 Ghoshal UC. The Indian enigma of frequent H. pylori infection but protein in Helicobacter pylori clinical isolates. J Clin Microbiol infrequent gastric cancer: Is the magic key in Indian diet, host’s 2007;45:488–95. genetic make up, or friendly bug? Am J Gastroenterol 49 Reyes-Leon A, Atherton JC, Argent RH, Puente JL, Torres J. 2007;102:2113–4.

74 Gut Pathogens BioMed Central

Research Open Access The prevalence of the duodenal ulcer promoting gene (dupA) in Helicobacter pylori isolates varies by ethnic group and is not universally associated with disease development: a case-control study Heather-Marie A Schmidt1, Sönke Andres2,3, Nadeem O Kaakoush1, Lars Engstrand2,3, Lena Eriksson3, Khean-Lee Goh4, Kwong Ming Fock5, Ida Hilmi4, Subbiah Dhamodaran5, David Forman6 and Hazel Mitchell*1

Address: 1School of Biotechnology and Biomolecular Sciences, University of New South Wales, NSW, Sydney, Australia, 2Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, Stockholm, Sweden, 3Department of Bacteriology, Swedish Institute for Infectious Disease Control, Solna, Sweden, 4Department of Medicine, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia, 5Division of Gastroenterology, Department of Medicine, Changi General Hospital, Singapore and 6Centre for Epidemiology and Biostatistics, Faculty of Medicine and Health, Leeds University, Leeds, UK Email: Heather-Marie A Schmidt - [email protected]; Sönke Andres - [email protected]; Nadeem O Kaakoush - [email protected]; Lars Engstrand - [email protected]; Lena Eriksson - [email protected]; Khean- Lee Goh - [email protected]; Kwong Ming Fock - [email protected]; Ida Hilmi - [email protected]; Subbiah Dhamodaran - [email protected]; David Forman - [email protected]; Hazel Mitchell* - [email protected] * Corresponding author

Published: 11 March 2009 Received: 20 January 2009 Accepted: 11 March 2009 Gut Pathogens 2009, 1:5 doi:10.1186/1757-4749-1-5 This article is available from: http://www.gutpathogens.com/content/1/1/5 © 2009 Schmidt et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

Abstract Background: The putative H. pylori pathogenicity-associated factor dupA has been associated with IL-8 induction in vitro, and duodenal ulcer (DU) and gastric cancer (GC) development in certain populations, but this association is inconsistent between studies. We aimed to investigate dupA prevalence in clinical isolates from Sweden, Australia and from ethnic Chinese, Indians and Malays resident in Malaysia and Singapore and to examine the association with DU and GC. In addition we investigated the sequence diversity between isolates from these diverse groups and compared the level of IL-8 secretion in isolates possessing and lacking dupA. Methods: PCR primers were designed to amplify over the C/T insertion denoting a continuous dupA. PCR products from 29 clinical isolates were sequenced and compared with sequences from three additional strains obtained from GenBank. Clinical isolates from 21 Malaysian patients (8 dupA-positive, 14 dupA-negative) were assessed for their ability to induce IL-8 in AGS cells in vitro. Statistical analysis was performed using Fisher's exact test. Results: The prevalence of dupA in isolates from Swedish functional dyspepsia (FD) control patients (65%, 13/20) was higher and in isolates from Indian FD patients (7.1%, 3/42) was lower as compared with isolates from Chinese (28.9%, 13/49, P = 0.005, P = 0.025), Malay (35.7%, 5/14, P = 0.16, P = 0.018) and Australian (37.8%, 17/45, P = 0.060, P < 0.001) FD patients. dupA was associated with DU and GC development in Chinese with 62.5% (10/16) and 54.6% (12/22) of isolates possessing dupA respectively as compared with FD controls (28.9%) (P = 0.015, P = 0.032). No significant difference in prevalence of dupA between FD controls, DU (63.6%, 7/11) and GC

Page 1 of 8 (page number not for citation purposes) Gut Pathogens 2009, 1:5 http://www.gutpathogens.com/content/1/1/5

(61.9%, 13/21) cases (P = 1.000) was observed in the Swedish population. Sequence analysis revealed a pairwise variation of 1.9% and all isolates possessed the C/T insertion. The average IL- 8 induction was 1330 pg/mL for dupA-positive isolates and 1378 pg/mL for dupA-negative isolates. Conclusion: Although dupA is highly conserved when present, we identified no consistent association between dupA and DU or GC development across the ethnic groups investigated, with the dupA prevalence in control groups varying significantly. Our results would suggest that in the clinical isolates investigated dupA is not associated with IL-8 induction in vitro.

Introduction gesting that dupA, and the TFSS it is associated with, may Although Helicobacter pylori infection invariably results in be a major determinant in DU-development. In contrast, gastritis, a significant minority of those infected will two recent studies, one in Brazil [14] and a second in progress to more severe gastroduodenal pathologies, which subjects from Belgium, China, South Africa and the including Duodenal Ulcer (DU) and Gastric Cancer (GC). USA were examined [15] did not find an association The factors resulting in progression to severe H. pylori- between dupA and DU development. Given that H. pylori related disease are poorly understood, however there is pathogenicity-associated factors in isolates from different compelling evidence to suggest a co-contribution of bac- populations have been shown to possess distinct geno- terial virulence factors, host genetics and environmental types [16-18], it is important that the association between stimuli [1]. a newly identified putative pathogenicity-associated factor, such as dupA and specific disease outcomes, be To date a number of H. pylori pathogenicity-associated fac- assessed in clinical isolates from a range of different tors, including flagella, adhesins, urease, the vacuolating regions and ethnic groups [19]. cytotoxin and the cag pathogenicity island (cag PAI) have been associated with the development of more serious H In the present study we determined the prevalence of dupA pylori-related disease outcomes. The cag PAI encodes a across a range of countries and ethnic groups including type IV secretion system (TFSS), homologous to the well- Australia, Sweden and the three primary ethnic groups res- studied virB/D4 TFSS of the plant pathogen Agrobacterium ident within Malaysia, namely Chinese, Indians and tumefaciens, as well as the effector protein CagA [2,3]. Malays and examined the relationship between dupA and Translocation of CagA by the TFSS into gastric epithelial DU and GC in the Swedish and ethnic Chinese popula- cells has been shown to drive deregulation of intracellular tions. We also sequenced dupA PCR products from isolates signaling pathways resulting in a myriad of cellular effects from each ethnic and disease group and determined that such as increased motility and elongation [4]. In a process they possessed the C/T insertion required for a continuous independent of CagA, the TFSS also induces the produc- dupA gene, and additionally that there was a only a small tion of the chemo-attractant and pro-inflammatory degree of sequence diversity between isolates. Finally, we cytokine IL-8 by facilitating the entry of peptidoglycan determined that the induction of IL-8 secretion by AGS into the host epithelial cell [5]. In vitro studies have dem- cells was not associated with the presence of dupA in the onstrated that the virB4 homologous ATPase, CagE clinical isolates investigated in vitro. (HP0544, JHP0492), is essential for IL-8 induction [6-10]. In addition to cagE, multiple virB4 homologues exist Methods within the H. pylori genome, with both sequenced strains Patients, H. pylori strains and extraction of genomic DNA J99 and 26695 possessing three additional virB4 homo- Malaysian & Singaporean logues [11,12]. Recently Lu et al. demonstrated that the As part of a larger study examining the role of bacterial, partial virB4 homologues JHP0917 and JHP0918, which host and environmental factors in GC development, gas- are present in the plasticity zone of J99 but not 26695, tric biopsies were obtained from consecutively enrolled form one continuous gene in clinical isolates the duode- unrelated patients undergoing routine endoscopic exami- nal ulcer promoting gene (dupA). and that this gene was a nation of gastrointestinal symptoms at the University functional virB4 homologous ATPase [13]. dupA, like Hospital, Kuala Lumpur, Malaysia and the Changi Hospi- cagE, is associated with IL-8 induction [13] and thus it is tal, Singapore (between mid-2004 and 2007). Based on possible that this may be part of an as yet unidentified endoscopic and histological examination, patients were TFSS. In the original study examining the role of dupA in diagnosed with either distal GC or functional dyspepsia more severe H. pylori related disease, dupA possessing iso- (FD). Gastric biopsies were stored at -70°C in brain heart lates were found to be associated with the development of infusion (BHI) broth containing 20% glycerol and trans- DU in patients from Columbia, Japan and Korea [13]. In ported to Australia on dry ice. Gastric biopsies were contrast no association was found with gastric cancer sug- smeared onto campylobacter selective medium (CSA),

Page 2 of 8 (page number not for citation purposes) Gut Pathogens 2009, 1:5 http://www.gutpathogens.com/content/1/1/5

incubated and identified using morphology, microscopy ation for 5 minutes at 95°C, followed by 35 cycles of and biochemical tests as previously described [20,21]. 95°C for 20 seconds, 55°C for 20 seconds and 72°C for Genomic DNA from one single colony isolate per patient 40 seconds, and a final at 72°C for 7 minutes. PCR prod- was extracted using the Gentra Puregene DNA extraction ucts were visualized by agarose gel electrophoresis. kit (Gentra, Minneapolis, Minnesota, USA). In the present study, 142 single colony isolates from 52 Chinese, 42 Validation of primers and investigation of dupA by DNA Indian and 14 Malay FD patients, 22 Chinese GC patients, sequencing and 16 Chinese DU patients were included. 29 single colony isolates (4 Chinese, 2 Indian, 4 Malay, 3 Australian and 3 Swedish FD isolates, 4 Chinese and 3 Swedish Swedish DU isolates, and 3 Chinese and 3 Swedish GC In a previous case-control study which recruited patients isolates) were sequenced to validate the specificity of the from eight hospitals in Sweden biopsies were collected, primers dupA F and dupA R and to ascertain whether cultured for H. pylori and bacterial DNA extracted as pre- JHP0917-0918 formed a single continuous gene in these viously described [22]. In the present study 52 single col- isolates. Sequencing was performed using the BigDye™ ony isolates from Swedish patients diagnosed with FD Terminator version 3.1 (Applied Biosystems, Foster City, (20), DU (11) and GC (21) were included. California, USA). Sequencing analysis was performed on an ABI3730 Capillary DNA sequencer (GENterprise, Australian Mainz, Germany). Sequences were verified using Blast In a previous study which recruited patients from Sydney, [24] available from Biomanager by ANGIS http:// Australia, biopsies were collected, cultured for H. pylori www.angis.org.au. Sequence alignments of the 29 strains, and bacterial DNA extracted as previously described [23]. with the original dupA gene sequence from Gene Bank In the present study, 45 single colony isolates from Cau- (accession no. AB196363) [13] and two additional dupA- casian Australian patients with FD were included. like sequences (accession no. EF076755 and EF076756) available in the GenBank database http:// Primer design and detection of dupA in single colony www.ncbi.nlm.nih.gov/Genbank/index.html were cre- isolates ated using ClustalW accurate [25] also available from Primer pair dupA1274F and dupA1674R (table 1) were BioManager. designed to amplify a 399–400 bp fragment corresponding to nucleotides 1016315–1016714 in J99 (accession Development of a dupA multiplex PCR no. AE001439). The PCR primers amplified over the To confirm that isolates negative for the dupA PCR truly region of dupA containing the 1 bp insertion/deletion lacked the dupA gene, rather than being the result of detected by Lu et al. 2005 [13], that is the forward primer sequence diversity over the primer annealing site, a multi- bound to a sequence from JHP0917 and the reverse plex PCR was designed involving the use of 2 forward primer bound to a sequence from JHP0918, suggesting (dupA113F and dupA1274F) and 3 reverse (dupA1083R, that PCR positive isolates have both JHP0917 and dupA1674R and dupA1830R) primers which produce 5 JHP0918. PCR was performed in a 25 μL reaction contain- differently sized PCR products if all primers bind (Figure ing 0.825 U Taq DNA polymerase (Fisher Biotech, Subi- 1). Primer sequences are presented in table 1. aco, Australia), 45 nmol magnesium-chloride, 10 pmol of each primer and 10 ng of genomic DNA. All PCR runs The PCR was performed in a 25 uL reaction as described included a negative control (no DNA) and a positive con- above, with an initial denaturation for 5 minutes at 95°C, trol (J99). PCR was performed as follows: initial denatur- followed by 35 cycles of 95°C for 30 seconds, 50°C for 30

Table 1: PCR primers used in this study.

Primer name Nucleotide sequence (5'-3') Reference

dupA113F GAC GAT TGA GCG ATG GGA ATA T [15]

dupA1083R CTG AGA AGC CTT ATT ATC TTG TTG G [15]

dupA1274F GCG TGA TCA ATA TGG ATG CTT TTG C This study

dupA1674R TTG TCT GGC TCT CAT GTC CGT GTT G This study

dupA1830R CTT CCT TAT AAG TTT CTT GGT TTG C [15]

Page 3 of 8 (page number not for citation purposes) Gut Pathogens 2009, 1:5 http://www.gutpathogens.com/content/1/1/5

A M J99 26695 FD527

2686 1583 1116 859 692 501/489 404

B 113F 1274F

dupA

1083R 1674R 1830R 400bp 556bp 970bp 1561bp 1717bp

FigurePCR amplification 1 using the dupA multiplex PCR PCR amplification using the dupA multiplex PCR. (A)Genomic DNA from H. pylori strains J99 (positive control), 26695 (negative control) and FD527 (clinical isolate) were used to PCR amplify dupA using the forward primers dupA113F and dupA1274F, and the reverse primers dupA1083R, dupA1674R and dupA1830R. M, size markers (in base pairs). (B) Schematic representation of dupA showing the annealing positions of the forward primers dupA113F and dupA1274F, and the reverse primers dupA1083R, dupA1674R and dupA1830R, and the expected sizes of the amplified PCR products. seconds and 65°C for 2 minutes, and a final at 65°C for 7 sent the average for these experiments. The concentration minutes. of IL-8 in the supernatant was analysed by ELISA, using the BD OptEIA: Human IL-8 ELISA set (BD, San Diego, Infection assay and IL-8 induction in vitro CA, USA) according to the manufacturers instructions. AGS cells (ATCC CRL-1739) were routinely maintained in RMPI1640 supplemented with 10% FCS and glutamine Statistical Analysis (All from Gibco Invitrogen, Paisley, UK), and were cul- Fisher's exact test was used to calculate statistical signifi- tured at 37°C in 5%CO2/95% air. Cells were seeded in 24- cance; two-sided P values < 0.05 were considered signifi- well tissue culture plates and infected with approximately cant. 3 × 107 bacterial cells. 22 H. pylori isolates from Malaysian patients (8 dupA-positive and 14 dupA-negative) were Sequences have been deposited in GenBank under the fol- available for in vitro analysis. Two H. pylori control strains lowing accession numbers: EU253504–EU253532. (67:20 (cag PAI negative) and 67:21 (cag PAI positive) [26] were also available for in vitro analysis. H. pylori iso- Results lates were cultured overnight in brucella broth (Sigma) Prevalence of dupA supplemented with 5% FCS (Gibco Invitrogen). Bacteria Following PCR analysis, dupA was identified in 15/52 were added to AGS cells in duplicate, and were co-incu- (28.9%) Chinese, 5/14 (35.7%) Malay, 3/42 (7.1%) bated for 6 hours before cell culture supernatant was col- Indian, 17/45 (37.8%) Australian and 13/20 (65.0%) lected and stored at -20°C prior to use. At least three Swedish FD clinical isolates (table 2). The prevalence of independent experiments were performed; results repre- dupA in Swedish FD isolates was higher (P = 0.005, P =

Page 4 of 8 (page number not for citation purposes) Gut Pathogens 2009, 1:5 http://www.gutpathogens.com/content/1/1/5

Table 2: Comparison of dupA prevalence between ethnic groups and countries.

Ethnicity/ Australian Chinese Indian Malay Swedish Country

Diagnosis FD FD DU GC FD FD FD DU GC

Ethnicity Diagnosis Prevalence 37.4 28.9 62.5 54.6 7.1 37.7 65.0 63.6 61.9 (%)

Australian FD (n = 45) 37.4 P = 1.000 P = 0.237 P < 0.001 P = 1.000 P = 0.060

Chinese FD (n = 52) 28.9 P = 0.237 P = 1.000 P = 0.015 P = 0.032 P = 0.025 P = 0.052 P = 0.005 DU (n = 16) 62.5 P = 0.015 P = 1.000 P = 0.744 P = 1.000 GC (n = 22) 54.6 P = 0.032 P = 0.744 P = 1.000 P = 0.760

Indian FD (n = 42) 7.1 P < 0.001 P = 0.025 P = 1.000 P = 0.018 P < 0.001

Malay FD (n = 14) 37.7 P = 1.000 P = 0.052 P = 0.018 P = 1.000 P = 0.163

Swedish FD (n = 20) 65.0 P = 0.060 P = 0.005 P < 0.001 P = 0.163 P = 1.000 P = 1.000 P = 1.000 DU (n = 11) 63.6 P = 1.000 P = 1.000 P = 1.000 P = 1.000 GC (n = 21) 61.9 P = 0.760 P = 1.000 P = 1.000 P = 1.000

Comparison of the prevalence of dupA-positive clinical isolates as determined by PCR between FD (functional dyspepsia), DU (duodenal ulcer) and GC (gastric cancer) patients from Australia, Sweden and the 3 primary ethnic groups of Malaysia and Singapore (Chinese, Indian and Malay). The statistical significance (P-values) of each valid comparison is displayed.

0.162, P = 0.025) and the prevalence in Indian FD isolates with GenBank accession no. EF076755. Of the 4 Chinese, was lower (P = 0.025, P = 0.0517, P = 0.094) as compared 2 Indian, 4 Malay FD, 3 Swedish FD and 3 Australian FD with the Chinese, Malay and Australian FD isolates inves- isolates, 4 Chinese and 3 Swedish DU isolates, and 3 Chi- tigated. There was no significant difference in the preva- nese and 3 Swedish GC isolates, all isolates were observed lence of dupA between isolates from Chinese, Malay and to possess a 1 bp C or T insertion after position 1385 rel- Australian FD patients (P = 0.052 to P = 1.000). Among ative to gene JHP0917 in J99 suggesting the presence of 1 isolates from Chinese or Swedish DU patients there was continuous open-reading frame. no significant difference between the two ethnicities, with 10/16 (62.5%) and 7/11 (63.6%) of the isolates possess- Sequence alignments from the 29 clinical isolates with 3 ing dupA respectively (P = 1.000). Similarly, there was no comparable sequences deposited in GenBank (AB196363, significant difference between isolates from Chinese and EF076755, EF076756) revealed that dupA possessed a Swedish GC patients with 12/22 (54.6%) and 13/21 pair-wise sequence variation of only 1.90/100 bp (maxi- (61.9%) isolates possessing dupA respectively (P = 0.759). mum variation: 14.71/100 bp, minimum variation 0.46/ Among isolates from Chinese patients, the prevalence of 100 bp, median variation 0.92/100 bp, standard devia- isolates possessing dupA was significantly higher in iso- tion 3.16/100 bp). Two isolates examined (the Chinese lates from DU and GC patients than in isolates from FD FD isolate FD553 (GenBank sequence: EU253520) and control patients (P = 0.015, P = 0.032), suggesting an asso- Iranian isolate EF076756), demonstrated a sequence vari- ciation with the development of disease in this popula- ations of 14.71/100 bp and 12.64/100 bp respectively, tion. There was no significant difference in dupA being more than 2 standard variations above the mean. In prevalence between the isolates from Chinese DU and the case of FD553 the high degree of sequence variation Chinese GC patients (P = 0.744). Conversely, there was was the direct result of a 60 bp insertion at base pair 1465 no significant difference in the prevalence of dupA relative to AB196363. The remaining isolates were within between isolates from Swedish FD, DU and GC patients 1 standard deviation of the mean. (P = 1.000). Multiplex PCR Sequence Analysis The use of the multiplex PCR revealed no additional iso- Blast analysis of sequences from 29 clinical isolates lates positive for dupA, when compared to the use of revealed significant sequence similarity to the dupA gene dupA1274F and dupA1674R alone, suggesting that the use sequence deposited in GenBank, accession no. of the single primer pair is adequate to determine the AB196363[13] and the sequence of a dupA-like protein prevalence of dupA in both western and east Asian popu-

Page 5 of 8 (page number not for citation purposes) Gut Pathogens 2009, 1:5 http://www.gutpathogens.com/content/1/1/5

lations. An example of the multiplex PCR is presented in dupA (65.0%) than isolates from Australia (26.3%) or eth- figure 1; all five bands are clearly distinguishable in clini- nic Chinese (28.9%), Indians (7.1%) or Malays (35.7%) cal isolate FD527. resident in Malaysia or Singapore. Interestingly isolates from ethnic Indian FD patients were significantly less In vitro IL-8 production likely to possess dupA (7.1%) than isolates from any of the The uninoculated control secreted 13 pg/ml, with the neg- other groups. This difference in dupA prevalence between ative control strain (67:20) secreting 80 pg/ml and the countries and ethnic groups is supported by several stud- positive control strain (67:21) secreting 1368 pg/ml (fig- ies [15,27] including the initial Lu et al. 2005 [13] which ure 2A). Individual IL-8 secretion levels varied between identified that dupA was more prevalent in isolates from 126 pg/ml and 1855 pg/ml for dupA-positive isolates, and Columbian gastritis patients (39%) than in isolates from 123 pg/ml and 1633 pg/ml for dupA-negative isolates. Japanese (14%) and Koreans (7%) gastritis patients. Fur- However, no significant difference (p > 0.5) was observed thermore, there is no significant difference in dupA preva- in the average level of IL-8 secreted by AGS cells between lence reported previously for isolates from Chinese clinical isolates possessing dupA (1330 pg/mL) and those patients (25% and 35.3% respectively) [15,28] and for lacking dupA (1378 pg/mL) (figure 2B). isolates from ethnic Chinese Malaysians and Singapore- ans reported in the present study (28.9%), despite differ- Discussion ent primer pairs being used, suggesting that the primer The current investigation of the prevalence of dupA in clin- pair designed in this study is appropriate. Conversely, the ical isolates from Australia and Sweden as well as from the prevalence reported in a north Indian population was three major ethnic groups (Chinese, Indians and Malays) considerably higher than that found in the ethnic Indian resident in Malaysia and Singapore clearly shows that Malaysians in the present study (16/70 as compared with there is significant variability in the prevalence of dupA 3/42) [29]. One possible explanation is that the Indian between not only geographical locations, but also Malaysians are likely to be predominantly ethnic Tamil between ethnic groups resident in the same country. In [29], originating from Southern India, rather than the the present study the prevalence of dupA in H. pylori iso- north Indian Indeed, two studies in Brazil reported signif- lates collected from FD patients varied significantly icantly different prevalences for dupA (92.3% vs. 62%) between nationalities/ethnicities, with isolates from [14,30]. Swedish patients being significantly more likely to possess

A 2500 2000 1500 1000

IL-8 (pg/mL) 500 0 67:20 67:21 GC79 FD613 FD484 FD657 FD535 FD544 FD514 FD599 FD517 FD616 FD675 FD661 FD527 FD537 FD457 FD621 FD437 FD550 FD509 FD516 FD443 FD607 Uninocculated B Controls dupA + dupA -

2000 1000 0 IL-8 (pg/mL) IL-8 dupA + (n=8) dupA - (n=14)

IL-8Figure induction 2 in AGS cells by clinical isolates in vitro IL-8 induction in AGS cells by clinical isolates in vitro. (A) Induction of IL-8 secretion in AGS cells incubated with indicated strains for 6 hours. The error bars show the standard deviations of each independent experiment. (B) Average induction of IL-8 secretion in AGS cells incubated with dupA positive (blue) or negative (red) strains. The error bars show the standard deviations for each group of strains.

Page 6 of 8 (page number not for citation purposes) Gut Pathogens 2009, 1:5 http://www.gutpathogens.com/content/1/1/5

We also demonstrated that the association between dupA tion. An alternate explanation, since dupA is located in the and severe gastroduodenal disease was inconsistent plasticity zone, is that the acquisition was a relatively between the Chinese and Swedish populations. In isolates recent phenomenon. Although there is, as yet, no specific from ethnic Chinese patients the prevalence of dupA was evidence for either scenario, we believe the former expla- significantly higher in patients diagnosed with DU nation is more likely. Whole gene sequence analysis of (62.5%) or GC (54.6%) as compared with those diag- dupA over an increased sample size is necessary to ade- nosed with FD. This is in contrast to the observations of quately investigate these points. Lu et al. 2005 [13] and Zhang et al. 2008 [28] who reported a negative association with GC, but supports the Although the initial study by Lu et al reported that dupA observations by Argent et al. [15]. In contrast, in the Swed- was associated with IL-8 induction in vitro in both knock- ish population there was no significant difference in the out studies and in clinical isolates [13], in the present prevalence of dupA in isolates from patients diagnosed study we observed no significant difference in the level of with DU, GC or FD, similar to findings reported from Bra- IL-8 induction between strains with or without dupA. It is zil [14,30]. Our observations are consistent with the possible however that the strains used in the present study reported variation in other H. pylori virulence factors such lacked the other essential components of the dupA type IV as the cagA and vacA genes [31], and further emphasise the secretion system, however, as none of these components fact that before a new virulence factor is associated with a have yet been identified it is impossible to assess whether specific disease outcome, studies must be undertaken in a this is responsible for the lack of association with IL-8 range of geographic locations as well in different ethnic induction in the present study. More comprehensive stud- groups. ies with strains lacking genes known to contribute to IL-8 induction such as cag PAI components and with oipA off Lu et al. have reported that, with the exception of J99, would be beneficial in elucidating the contribution of JHP0917–JHP0918 forms a continuous open reading dupA to IL-8 induction. frame, dupA, due to the presence of a 1 bp C/T insertion in clinical isolates [13]. In the present study, all clinical iso- Conclusion lates sequenced possessed a continuous dupA gene, signi- Although based on our findings and those of others, it is fied by the presence of the C/T insertion, a finding that is unlikely that dupA is itself a determinant or indicator of a consistent with previous reports in other populations specific clinical outcome across all populations; it may be [13,14,32]. Our results also indicate that dupA is highly shown to be an important virulence factor if a role as part conserved over the region sequenced with an average par- of a novel TFSS is identified. tial sequence variation of only 1.9%. This indicates a high degree of conservation comparable to housekeeping Competing interests genes such as atpA, ureI, efp, and ppa [33]. We must The authors declare that they have no competing interests. acknowledge that as only 29 clinical isolates were sequenced and the sequences used for comparison repre- Authors' contributions sent less than 400 bp, the pairwise sequence variation for HMAS designed the study, prepared the Malaysian iso- the entire gene and in a larger population may be higher lates for analysis, designed and carried out the molecular than that described here. Interestingly in a recent study genetic studies, carried out the sequence alignment and Douraghi et al. compared 3 regions of the dupA gene – analysis, performed the statistical analysis, helped design JHP0917 (289 bp), JHP0918 (259 bp) and the junction and participated in the in vitro studies and drafted the region over JHP0917 and JHP0918 i.e. 'dupA' (216 bp) in manuscript. SA helped design and participated in the in 6 Iranian strains with that of 10 Brazilian and 3 Indian vitro studies. NK participated in the molecular genetic strains whose sequences had been deposited in GenBank studies. LEr prepared the Swedish isolates for analysis. [32]. The reported sequence similarities in that study HM was awarded grant to support the study, participated ranged from 86.1%–100% for JHP0917, 88–98.8% for in the design of the study and helped to draft the manu- JHP0918 and 93.4–99.5% for 'dupA, a finding that would script. All authors read and approved the final manu- support the high degree of conservation of this gene script. observed in our study. Surprisingly in the Douraghi et al. study none of the strains were compared for all 3 regions. Acknowledgements This work was supported by The Cancer Council of New South Wales, The high degree of sequence conservation reported and Australia (REF 66/04). Sönke Andres was supported by a grant from the the finding that dupA is present in strains from different Marie Curie Early Stage Research Training Fellowship of the European continents and in populations with different genetic back- Community's Sixth Framework Program called IMO-train. grounds, suggests that dupA may confer a fitness advan- We would like to thank Alfred Tay for providing the calculations on the tage of some sort, which has resulted in the conservation pairwise sequence variation for the Falush et al. [33] study. We would also of the gene and the sequence, throughout human migra- like to thank Quanjiang Dong for his help with primer design.

Page 7 of 8 (page number not for citation purposes) Gut Pathogens 2009, 1:5 http://www.gutpathogens.com/content/1/1/5

References pylori infection – introducing the Sydney strain. Gastroenterol 1. Pritchard DM, Crabtree JE: Helicobacter pylori and gastric can- 1997, 112:1386-1397. cer. Curr Opin Gastroenterol 2006, 22:620-625. 22. Enroth H, Kraaz W, Engstrand L, Nyrén O, Rohan T: Helicobacter 2. Odenbreit S, Puls J, Sedlmaier B, Gerland E, Fischer W, Haas R: pylori strain types and risk of gastric cancer: a case-control Translocation of Helicobacter pylori CagA into gastric epithe- study. Cancer Epidemiol Biomarkers Prev 2000, 9:981-985. lial cells by type IV secretion. Science 2000, 287:1497-1500. 23. Borody T, Clancy R, Warren EF, Surace R, Brusentsev S, Mitchell H: 3. Backert S, Ziska E, Brinkmann V, Zimny-Arndt U, Fauconnier A, Jung- Antibiotic sensitivities of Helicobacter pylori vary at differ- blut PR, Naumann M, Meyer TF: Translocation of the Helico- ent gastric mucosal sites. In Helicobacter pylori: Basic mechanisms bacter pylori CagA protein in gastric epithelial cells by a type to clinical cure 2002 Edited by: Hunt R, Tytgat G. Dordrecht: Kluwer IV secretion apparatus. Cell Microbiol 2000, 2:155-164. Academic Publishers; 2003:373-382. 4. Backert S, Selbach M: Role of type IV secretion in Helicobacter 24. Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, Lip- pylori pathogenesis. Cell Microbiol 2008, 10:1573-1581. man DJ: Gapped BLAST and PSI-BLAST: a new generation of 5. Viala J, Chaput C, Boneca IG, Cardona A, Girardin SE, Moran AP, Ath- protein database search programs. Nucl Acids Res 1997, man R, Mémet S, Huerre MR, Coyle AJ, et al.: Nod1 responds to 25:3389-3402. peptidoglycan delivered by the Helicobacter pylori cag patho- 25. Thompson JD, Higgins DG, Gibson TJ: CLUSTAL W: improving genicity island. Nat Immunol 2004, 5:1166-1174. the sensitivity of progressive multiple sequence alignment 6. Maeda S, Ogura K, Yoshida H, Kanai F, Ikenoue T, Kato N, Shiratori through sequence weighting, position-specific gap penalties Y, Omata M: Major virulence factors, VacA and CagA, are and weight matrix choice. Nucl Acids Res 1994, 22:4673-4680. commonly positive in Helicobacter pylori isolates in Japan. 26. Bjorkholm B, Lundin A, Sillen A, Guillemin K, Salama N, Rubio C, Gut 1998, 42:338-343. Gordon JI, Falk P, Engstrand L: Comparison of genetic diver- 7. Owen RJ, Sharp S, Lawson AJ, Durrani Z, Rijpkema S, Kidd M: Inves- gence and fitness between two subclones of Helicobacter tigation of the biological relevance of Helicbacter pylori cagE pylori. Infect Immun 2001, 69:7832-7838. locus diversity, presence of CagA motifs and vacuolating 27. Hussein N, Mohammadi M, Talebkhan Y, Doraghi M, Letley D, cytotoxin genotype on IL-8 induction in gastric epithelial Muhammad M, Argent R, Atherton J: Differences in virulence cells. FEMS Immunol Med Microbiol 2003, 36:135-140. markers between Helicobacter pylori strains from Iraq and 8. Tummuru MK, Sharma SA, Blaser MJ: Helicobacter pylori picB, a those from Iran: potential importance of regional differences homologue of the Bordetella pertussis toxin secretion pro- in H. pylori-associated disease. J Clin Microbiol 2008, tein, is required for induction of IL-8 in gastric epithelial 46:1774-1779. cells. Mol Microbiol 1995, 18:867-876. 28. Zhang Z, Zheng Q, Chen X, Xiao S, Liu W, Lu H: The Helicobacter 9. Day AS, Jones NL, Lynett JT, Jennings HA, Fallone CA, Beech R, Sher- pylori duodenal ulcer promoting gene, dupA, in China. BMC man PM: cagE is a virulence factor associated with Helico- Gastroenterol 2008, 8:49. bacter pylori-induced duodenal ulceration in children. J Infect 29. Arachchi HS, Kalra V, Lal B, Bhatia V, Baba CS, Chakravarthy S, Dis 2000, 181:1370-1375. Rohatgi S, Sarma PM, Mishra V, Das B, Ahuja V: Prevalence of duo- 10. Censini S, Lange C, Xiang Z, Crabtree JE, Ghiara P, Borodovsky M, denal ulcer promoting gene (dupA) of Helicobacter pylori in Rappuoli R, Covacci A: cag, a pathogenicity island of Helico- patients with duodenal ulcer in north Indian population. Heli- bacter pylori, encodes type I-specific and disease associated cobacter 2007, 12:591-597. virulence factors. Proc Natl Acad Sci USA 1996, 93:14648-14653. 30. Pacheco AR, Proença-Módena JL, Sales AI, Fukuhara Y, Silveira WDd, 11. Alm RA, Ling LS, Moir DT, King BL, Brown ED, Doig PC, Smith DR, Pimenta-Módena JL, Oliveira RBd, Brocchi M: Involvement of the Noonan B, Guild BC, deJonge BL, et al.: Genomic-sequence com- Helicobacter pylori plasticity region and cag pathogenicity parison of two unrelated isolates of the human gastric path- island genes in the development of gastroduodenal diseases. ogen Helicobacter pylori. Nature 1999, 397:176-180. Eur J Clin Microbiol Infect Dis 2008, 27:1053-1059. 12. Tomb J-F, White O, Kerlavage AR, Clayton RA, Sutton GG, Fleis- 31. Ahmed N, Sechi LA: Helicobacter pylori and gastroduodenal chmann RD, Ketchum KA, Klenk HP, Gill S, Dougherty BA, et al.: The pathology: new threats from an old friend. Ann Clin Microbiol complete genome sequence of the gastric pathogen Helico- Antimicrob 2005, 4:1. bacter pylori. Nature 1997, 388:539. 32. Douraghi M, Mohammadi M, Oghalaie A, Abdirad A, Mohagheghi M, 13. Lu H, Hsu P-I, Graham D, Yamaoka Y: Duodenal ulcer promoting Hosseini M, Zeraati H, Ghasemi A, Esmaieli M, Mohajerani N: dupA gene of Helicobacter pylori. Gastroenterol 2005, 128:833-848. as a risk determinant in Helicobacter pylori infection. J Med 14. Gomes LI, Rocha GA, Rocha AM, Soares TF, Oliveira CA, Bittencourt Microbiol 2008, 57:554-562. PF, Queiroz DM: Lack of association between Helicobacter 33. Falush D, Wirth T, Linz B, Pritchard JK, Stephens M, Kidd M, Blaser pylori infection with dupA-positive strains and gastroduode- MJ, Graham DY, Vacher S, Perez-Perez GI, et al.: Traces of human nal diseases in Brazilian patients. Int J Med Microbiol 2008, migrations in Helicobacter pylori populations. Science 2003, 298:223-230. 299:1582-1585. 15. Argent RH, Burette A, Miendje Deyi VY, Atherton JC: The presence of dupA in Helicobacter pylori is not significantly associated with duodenal ulcer in Belgium, South Africa, China or North America. Clin Infect Dis 2007, 45:1204-1206. 16. Miehlke S, Kibler K, Kim JC, Figura N, Small SM, Graham DY, Go MF: Allelic variation in the cagA gene of Helicobacter pylori obtained from Korea compared to the United States. Am J Gastroenterol 1996, 91:1322-1325. 17. Cao P, Lee KJ, Blaser MJ, Cover TL: Analysis of hopQ alleles in Publish with BioMed Central and every East Asian and Western strains of Helicobacter pylori. FEMS scientist can read your work free of charge Microbiol Lett 2005, 251:37-43. 18. Tan HJ, Rizal AM, Rosmadi MY, Goh KL: Distribution of Helico- "BioMed Central will be the most significant development for bacter pylori cagA, cagE and vacA in different ethnic groups in disseminating the results of biomedical research in our lifetime." Kuala Lumpur, Malaysia. J Gastroenterol Hepatol 2005, Sir Paul Nurse, Cancer Research UK 20:589-594. 19. Lu H, Yamaoka Y, Graham DY: Helicobacter pylori virulence fac- Your research papers will be: tors: facts and fantasies. Curr Opin Gastroenterol 2005, 21:653-659. available free of charge to the entire biomedical community 20. Lee A, O'Rourke J, Ungria MCD, Robertson B, Daskalopoulos G, Dixon MF: Erratum: a standardized mouse model of Helico- peer reviewed and published immediately upon acceptance bacter pylori infection – introducing the Sydney strain. Gas- cited in PubMed and archived on PubMed Central troenterol 1997, 113:732. 21. Lee A, O'Rourke J, Ungria MCD, Robertson B, Daskalopoulos G, yours — you keep the copyright Dixon MF: A standardized mouse model of Helicobacter Submit your manuscript here: BioMedcentral http://www.biomedcentral.com/info/publishing_adv.asp

Page 8 of 8 (page number not for citation purposes) 2.3 Paper III

82 Eur J Clin Microbiol Infect Dis (2010) 29:439–451 DOI 10.1007/s10096-010-0881-7

ARTICLE

The cag PAI is intact and functional but HP0521 varies significantly in Helicobacter pylori isolates from Malaysia and Singapore

H.-M. A. Schmidt & S. Andres & C. Nilsson & Z. Kovach & N. O. Kaakoush & L. Engstrand & K.-L. Goh & K. M. Fock & D. Forman & H. Mitchell

Received: 15 October 2009 /Accepted: 5 January 2010 /Published online: 16 February 2010 # Springer-Verlag 2010

Abstract Helicobacter pylori-related disease is at least Singapore, and evaluated whether the cag PAI was intact partially attributable to the genotype of the infecting strain, and functional in vitro. Polymerase chain reaction (PCR) particularly the presence of specific virulence factors. We was used to detect dupA, cagA, cagE, cagT, cagL and investigated the prevalence of a novel combination of H. babA, and to type vacA, the EPIYA motifs, HP0521 alleles pylori virulence factors, including the cag pathogenicity and oipA ON status in 159 H. pylori clinical isolates. island (PAI), and their association with severe disease in Twenty-two strains were investigated for IL-8 induction isolates from the three major ethnicities in Malaysia and and CagA translocation in vitro. The prevalence of cagA, cagE, cagL, cagT, babA, oipA ON and vacA s1 and i1 was >85%, irrespective of the disease state or ethnicity. The : : : : H.-M. A. Schmidt C. Nilsson Z. Kovach N. O. Kaakoush prevalence of dupA and the predominant HP0521 allele and H. Mitchell (*) EPIYA motif varied significantly with ethnicity (p<0.05). School of Biotechnology and Biomolecular Sciences, A high prevalence of an intact cag PAI was found in all University of New South Wales, Sydney, NSW 2052, Australia ethnic groups; however, no association was observed e-mail: [email protected] between any virulence factor and disease state. The novel : : association between the HP0521 alleles, EPIYA motifs and S. Andres C. Nilsson L. Engstrand host ethnicity indicates that further studies to determine the Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, function of this gene are important. Stockholm, Sweden : : S. Andres C. Nilsson L. Engstrand Introduction Department of Bacteriology, Swedish Institute for Infectious Disease Control, Solna, Sweden Helicobacter pylori infects the gastric mucosa of more than half of the world’s population, making it the world’s most K.-L. Goh common infection. H. pylori infection invariably results in Department of Medicine, Faculty of Medicine, University of Malaya, superficial gastritis, and is the primary risk factor for peptic Kuala Lumpur, Malaysia ulcer disease (PUD), gastric cancer (GC) and mucosa- associated lymphoid tissue (MALT) lymphoma development. K. M. Fock Although the mechanisms resulting in severe disease devel- Division of Gastroenterology, Department of Medicine, Changi General Hospital, opment are poorly understood, a major factor is likely to be H. Singapore, Singapore pylori-induced gastric injury and inflammation, including the induction of the chemoattractant and proinflammatory D. Forman marker IL-8. Furthermore, the rates of severe disease Centre for Epidemiology and Biostatistics, Faculty of Medicine and Health, Leeds University, development vary between human populations, and it has Leeds, UK been postulated that differences in H. pylori genotypes, 440 Eur J Clin Microbiol Infect Dis (2010) 29:439–451 particularly in terms of the virulence factors of the circulating it has been controversially associated with the induction of strains, may partially explain these differences [1]. While a IL-8 [20, 21] and with duodenal ulcer (DU) and GC [20– number of virulence factors have been identified in H. pylori, 22]. This data indicates that dupA may be part of an as yet including the cag pathogenicity island (PAI), the vacuolating unidentified TFSS and, as such, may play a significant role cytotoxin (VacA), the duodenal ulcer promoting gene A in inducing inflammation and gastric injury. (dupA) and the adhesins BabA and OipA, the association The vacuolating cytotoxin (encoded by vacA) generates between individual factors or combinations of these factors a cascade of cellular effects, including inducing cell and disease remains controversial [2]. vacuolation and inducing IL-8 secretion [23, 24]. VacA The importance of the cytotoxin associated gene A production and activity is determined by polymorphic (cagA)inH. pylori virulence is underlined by the signal and mid-region sequences encoded in the vacA gene, pronounced cellular damage and effects associated with with s1m1 strains being the most virulent [25]. The recently CagA in vitro and in vivo [3]. The CagA protein is discovered polymorphism in the intermediate region has translocated into gastric epithelial cells via a type IV been proposed to be the best marker for VacA-associated secretion system (TFSS), at which point it may be tyrosine pathogenicity [26]. Indeed, strains possessing s1 and i1 phosphorylated by the Src family kinases and Abl kinase at vacA types have been previously associated with GC in the EPIYA motifs, which are encoded in the cagA 3′ some populations [11, 26, 27]. variable region [4, 5]. CagA is then able to interact with The adherence of H. pylori to gastric epithelial cells is SHP-2, causing aberrant activity of the cellular tyrosine partially mediated by the outer membrane inflammatory phosphatase, such as the induction of the hummingbird protein A (OipA, also called HopH) and the blood group phenotype, which is characterised by cellular elongation antigen binding adhesin (BabA), which are also important and cytoskeletal arrangement in vitro [6, 7]. The interaction in H. pylori pathogenesis. An additional role for OipA between CagA and SHP-2 is dependent upon the presence in directly inducing IL-8 when in frame (ON) [25]has of EPIYA-C or EPIYA-D motifs, which are typically also been proposed, although this role is controversial present in CagA from Western and East Asian isolates, [26, 27] respectively. The EPIYA-D motif exhibits greater binding Studies have shown considerable disparity in the to SHP-2 than single or repeated EPIYA-C motifs, resulting association between these virulence factors and severe in more severe deregulation of cellular pathways in vitro disease development [2, 20]. This has been particularly [7–9], Furthermore, the EPIYA-D and repeats of the evident in East Asia, where studies have often failed to EPIYA-C motif have been associated with an increased replicate associations observed in the West, often due to the risk of GC in a small number of studies [10, 11], illustrating high prevalence of these virulence factors within many East the importance of identifying these motifs when genotyping Asian populations [2, 23, 28–32]. Given these discrep- cagA. ancies, it may be more likely that it is a combination of The 40-kb cag PAI encodes the cagA gene and up to 30 virulence factors that determines disease development in other genes, including the 18 genes of the TFSS required susceptible individuals. for CagA translocation and cag PAI-associated IL-8 induc- As isolates from different populations possess distinct tion. In addition, there are a several genes with as yet genotypes [23, 29 , 33], it is important to investigate a unassigned functions, such as the allelic HP0521 [12]. Due range of different regions and ethnic groups [2]and,to to the prohibitively large number of genes on the cag PAI, date, few studies exist which fulfill these requirements. investigators have attempted to use genes such as cagA and Malaysia and Singapore represent ideal populations to cagE as markers of an intact cag PAI [13, 14]. However, conduct such studies, due to the presence of three primary this approach has been shown to be inadequate [15]. A ethnic groups, Chinese, Indian and Malay, which we have more comprehensive analysis of the cag PAI using multiple previously reported to correspond to an East Asian, genes and oligonucleotide primer pairs has revealed that the Western and a ‘mixed’ population, respectively [34]. cag PAI was not intact in the majority of H. pylori isolates Thus, this study aimed to examine the prevalence and from several countries and continents [15–19]. However, association with disease of a unique combination of given that considerable variation in nucleotide sequences established and putative H. pylori virulence determinants, between strains from different geographic regions is known including five cag PAI genes cagA, cagE, cagL, cagT to occur, which may result in primers failing to anneal, and and HP0521, vacA subtypes, babA and oipA in the that the verification of cag PAI intactness in vitro has not three primary ethnic groups resident in Malaysia and been performed in these studies, there is a need for further Singapore. In addition, we investigated a subset of strains validation. with either intact, intact 521 ES or partial cag PAIs for The dupA is a recently proposed virulence factor for H. their ability to translocate CagA and induce IL-8 secretion pylori. While little is known about its role in pathogenesis, in vitro. Eur J Clin Microbiol Infect Dis (2010) 29:439–451 441

Materials and methods Determination of oipA ON/OFF status

Patients, H. pylori strains and extraction of genomic DNA PCR products were purified using the Qiagen PCR purification kit, and sequencing was performed using the As part of a larger study, biopsies from unrelated BigDye™ Terminator version 3.1 (Applied Biosystems) consecutively enrolled symptomatic adult patients present- and performed on an ABI3730 Capillary DNA sequencer ing with gastrointestinal symptoms for routine endoscopy (GENterprise, Mainz, Germany). were collected, cultured for H. pylori and bacterial DNA All oipA sequences were aligned using the CLC Free extracted as previously described [20]. In the present study, Workbench (version 4.5.1, available from http://www. 159 single-colony isolates from 52 Chinese, 49 Indian and clcbio.com) and translated using translator software from 20 Malay functional dyspepsia (FD) patients, 22 Chinese Life Science Tools (available from http://www.fr33.net). GC patients and 16 Chinese DU patients were included (Table 1). Co-culture of H. pylori and AGS gastric epithelial cells

Determination of virulence genotypes AGS cells (ATCC CRL-1739) were routinely maintained in RPMI 1640 supplemented with glutamine, HEPES and Polymerase chain reaction (PCR) was used to detect the 10% foetal bovine serum (FBS; Gibco Invitrogen, Paisley, babA, cagA, cagE, cagL, cagT, dupA,HP0521,oipA and UK), and were cultured at 37°C in 5% CO2. Cells were vacA genes and the cagA EPIYA motifs. Primer pairs for seeded into 24-well (IL-8 induction) or 12-well (CagA babA, cagA, the EPIYA motifs, cagE, cagT, dupA, translocation) tissue culture plates. HP0521, oipA and vacA were selected from the previously Twenty-two clinical isolates and two control strains published literature and are presented in Table 2.Where (67:20 and 67:21), which are cag PAI-negative and more than one primer pair was used, isolates positive for -positive, respectively, [37] were used for in vitro analysis. one primer pair were considered to be positive for that H. pylori isolates were cultured overnight in Brucella broth gene/locus. (Sigma, St. Louis, MO, USA) supplemented with 5% FBS Allelotyping of HP0521 was performed using two (Gibco Invitrogen) and IsoVitaleX (Becton Dickinson, primer pairs selected from the previously published Cockeysville, MD, USA) under microaerobic conditions. literature (HP0521A and HP0521B) and the HP0521 empty The optical density at 600 nm adjusted to 0.1 (IL-8) or 0.05 site (ES) primer pair kindly provided by Dr. Christina (CagA translocation) corresponded to a multiplicity of Nilsson. The HP0521 ES primer pair was used to detect a infection of approximately 100 or 25, respectively. Follow- restricted deletion of HP0521 and amplifies differentially ing the addition of H. pylori strains and cell culture media, sized fragments, reflecting the allele present. the plates were centrifuged at 800 rpm for 5 min to Additional primer pairs were designed for babA and synchronise infection, and were incubated for 6 h at 37°C cagL using Primer-BLAST from the NCBI (http://www. in a 10% CO2 incubator. ncbi.nlm.nih.gov/tools/primer-blast/index.cgi?LINK_ LOC=BlastHome)[35] and evaluated against six completely CagA translocation and phosphorylation in vitro sequenced strains using in silico PCR amplification, available from http://insilico.ehu.es/ [36]. Following co-incubation, protein was extracted using RIPA PCR amplification was performed in a 25–30-µl standard buffer containing Protease Inhibitor Cocktail and Phosphatase reaction with 0.825U Taq polymerase (Fisher Biotech, Inhibitor Cocktail 2 (Sigma) and snap frozen. Samples were Subiaco, Australia), 20 ng of template DNA and 10 nmol of clarified by centrifugation, then combined with 2× SDS- each primer using an Applied Biosystems Thermocycler PAGE buffer and denatured for 5 min. Proteins were separated (Foster City, CA, USA) (Table 2). by SDS-PAGE on precast 7.5% polyacrylamide gels

Table 1 Details of the Malaysian and Singaporean Ethnicity Diagnosis Total number (female, male) Mean age (range) patients enrolled in this study Chinese FD 52 (21, 31) 54 years (24–76) DU 16 (7, 9) 49 years (31–74) GC 22 (10, 12) 66 years (37–76) FD functional dyspepsia; DU Indian FD 49 (26, 23) 47 years (18–77) duodenal ulcer; GC gastric Malay FD 20 (8, 12) 39 years (21–63) cancer 442 Eur J Clin Microbiol Infect Dis (2010) 29:439–451

Table 2 Polymerase chain reaction (PCR) primer pairs and PCR conditions used for genotyping Helicobacter pylori isolates

Genes Primer sequence (5′→3′) PCR product PCR conditions References (bp) vacA s1/s2 f: ATG GAA ATA CAA CAA ACA CAC 259/286a 95°C, 5 min; 35× (95°C, 20 s; 52°C, 20 s; [33] r: CTG CTT GAA TGC GCC AAA C 72°C, 40 s); 72°C, 7 min m1/m2 f: CAA TCT GTC CAA TCA AGC GAG 570/645a 95°C, 5 min; 35× (95°C, 20 s; 52°C, 20 s; [33] r: GCG TCT AAA TAA TTC CAA GG 72°C, 40 s); 72°C, 7 min i1 f: GTT GGG ATT GGG GGA ATG CCG 426 95°C, 5 min; 35× (95°C, 20 s; 55°C, 20 s; [26] r: TTA ATT TAA CGC TGT TTG AAG 72°C, 40 s); 72°C, 7 min i2 f: GTT GGG ATT GGG GGA ATG CCG 432 95°C, 5 min; 35× (95°C, 20 s; 55°C, 20 s; [26] r: GAT CAA CGC TCT GAT TTG A 72°C, 40 s); 72°C, 7 min cagA cagAsbra f: ATG ATG GCG TGA TGT TTG T 823 95°C, 5 min; 30× (95°C, 30 s; 50°C, 30 s; [18] r: TTT TCA AGG TCG CTT TTT GC 72°C, 1 min); 72°C, 7 min cagA f: GAT AAC AGG CAA GCT TTT 349 95°C, 5 min; 30× (95°C, 30 s; 50°C, 30 s; [68] GAG G 72°C, 1 min); 72°C, 7 min r: CTG CAA AAG ATT GTT TGG CAG A cagAv f: GAA TGT CTG ATA AAC TTG AAA 1,152 95°C, 5 min; 35× (95°C, 30 s; 58.5°C, 30 s; [69] r: GCG TAT GTG GCT GTT AGT AGC G 72°C, 90 s); 72°C, 7 min cagAvZ f: CAAAAGATAACGGATAAAGT 818–1,052 95°C, 5 min; 35× (95°C, 30 s; 58.5°C, 30 s; [70] r: CTGTTAGTAGCGTAATTGTC 72°C, 90 s); 72°C, 7 min cag2 f: GGA ACC CTA GTC GGT AAT G 550–800 95°C, 5 min; 35× (95°C, 30 s; 58.5°C, 30 s; [71] cag4 r: ATC TTT GAG CTT GTC TAT CG 72°C, 90 s); 72°C, 7 min EPIYA motifs cag2 f: GGA ACC CTA GTC GGT AAT G 495 95°C, 5 min; 35× (95°C, 20 s; 62°C, 20 s; [34] cagAWest r: TTT CAA AGG GAA AGG TCC GCC 72°C, 40 s); 72°C, 7 min cag2 f: GGA ACC CTA GTC GGT AAT G 501 95°C, 5 min; 35× (95°C, 20 s; 62°C, 20 s; [34] cagAEast r: AGA GGG AAG CCT GCT TGA TT 72°C, 40 s); 72°C, 7 min cagE picB f: TGT TTG GTT TCC CTG 1,335 95°C, 5 min; 35× (95°C, 30 min; 53°C, 30 s; [33] r: ACG CAT TCC TTA ACG 72°C, 1 min); 72°C, 7 min cagE f: TTG AAA ACT TCA AGG ATA GGA 508 95°C, 5 min; 35× (95°C, 30 min; 53°C, 30 s; [72] TAG AGC 72°C, 1 min); 72°C, 7 min r: GCC TAG CGT AAT ATC ACC ATT ACC C cagL cagL f: AAG CGT CTG TGA AGC AGT GA 433 95°C, 5 min; 35× (95°C, 20 min; 55°C, 20 s; This study r: GAC CAA CCA ACA AGT GCT CA 72°C, 40 s); 72°C, 7 min cagT cagT f: CCA TGT TTA TAC GCC TGT GT 301 95°C, 5 min; 35× (95°C, 20 min; 55°C, 20 s; [14] r: CAT CAC CAC ACC CTT TTG AT 72°C, 40 s); 72°C, 7 min HP0521 HP0521A f: AAT AAA TGG CTC AAT GAA 179/381b 95°C, 5 min; 35× (95°C, 20 min; 55°C, 20 s; [55] AAT GG 72°C, 40 s); 72°C, 7 min r: TTT TTG AAA TTC TTT TTT CTT TTC AT HP0521B f: CTA AAG ACT TCT AAT ATC 526 95°C, 5 min; 30× (95°C, 30 s; 50°C, 30 s; [55] ATA GG 72°C, 1 min); 72°C, 7 min r: CTT CAT CAG CCA CAT TCT TG HP0521 ES f: TGT GTC TTG AGC GGT GCT AT 860/1,100/ 95°C, 5 min; 30× (95°C, 30 s; 50°C, 30 s; This study r: TCC CCA AAG CTT TTG TCA TT 1,110/690c 72°C, 1 min); 72°C, 7 min babA babA2 f: AAT CCA AAA AGG AGA AAA AGT 607 95°C, 5 min; 35× (95°C, 20 min; 55°C, 20 s; [45] ATG AAA 72°C, 40 s); 72°C, 7 min babA607 r: CTT TGA GCG CGG GTA AGC [46] Eur J Clin Microbiol Infect Dis (2010) 29:439–451 443

Table 2 (continued)

Genes Primer sequence (5′→3′) PCR product PCR conditions References (bp)

babAZ f: GCT GAA GAC GAC GGC TTT TA 392 95°C, 5 min; 5× (95°C, 30 s; 61°C, 30 s; This study r: GATCGATACCCTGGCTCGT 72°C, 1 min), 40× (95°C, 30 s; 55°C, 30 s; 72°C, 1 min); 72°C, 7 min oipA oipA f: CAA GCG CTT AAC AGA TAG GC 450 95°C, 5 min; 35× (95°C, 20 min; 55°C, [73] r: AAG GCG TTT TCT GCT GAA GC 20 s; 72°C, 40 s); 72°C, 7 min dupA dupA1274F f: GCG TGA TCA ATA TGG ATG CTT 400 95°C, 20 s; 55°C, 20 s; 72°C, 40 s [20] TTG C (35 cycles) dupA1674R r: TTG TCT GGC TCT CAT GTC CGT GTT G a Allele 1/allele 2 b Size in 26695 genome/size in J99 genome c Approximate size for HP0521A/HP0521A(large)/HP0521B/HP0521 ES

(Bio-Rad, Hercules, CA, USA) and electro-transferred onto preferentially using the Chi-square test in order to allow PVDF membranes (Bio-Rad). Membranes were probed with for the simultaneous evaluation of multiple groups. Fisher’s one of the following primary antibodies: cagA (A10) (Santa exact test was used when the groups were too small or Cruz Biotechnology, Santa Cruz, CA, USA), anti-H. pylori when the Chi-square p-value<0.2. Cag antigen IgG fraction (polyclonal) (Austral Biologicals, To examine the association between virulence factors San Ramon, CA, USA) or p-Tyr (PY99) (Santa Cruz themselves, the cross-tabs function and Chi-square test, Biotechnology) to detect translocated CagA or tyrosine available in PASW Statistics 17.0 (Release 17.0.2, March phosphorylated CagA. Sheep anti-mouse conjugated HRP 11, 2009; SPSS Inc., Chicago, IL, USA) were used. A two- and goat anti-rabbit conjugated HRP secondary antibodies tailed p-value<0.05 was considered to be significant. (Amersham International, Amersham, UK) were used. Western blots were developed with ECL western blotting detection reagents (Amersham) according to the manufac- Results turer’s instructions. Results were obtained from at least two independent experiments. H. pylori virulence genotypes in clinical isolates

IL-8 induction by AGS cells The prevalence of 19 putative and established virulence loci was assessed in 159 clinical H. pylori isolates from IL-8 induction by AGS cells was measured as described Malaysia and Singapore. A summary of the prevalence of previously [20]. Briefly, following incubation, cell culture individual virulence factors is presented in Table 3. supernatants were collected and IL-8 levels measured by The cag PAI genes cagE, cagL and cagT were highly enzyme-linked immunosorbent assay (ELISA), using the prevalent in all of the groups analysed (ranging between 86 BD OptEIA Human IL-8 ELISA Set (BD, San Diego, CA, and 100%) (FE p>0.5). USA) according to the manufacturer’s instructions. At least The HP0521 alleles varied significantly between the three independent experiments were performed; the results three ethnic groups. The small HP0521A allele was present represent the average of these experiments. in the majority of isolates from Indian and Malay FD patients (73.5 and 55%, respectively), in contrast to the Statistical analysis Chinese FD patient group, where HP0521 ES (restricted deletion) was dominant (88.5%) (χ2 p<0.0001). The The Chi-square test (χ2) and the more sensitive Fisher’s HP0521 ES was predominant in the isolates from Chinese exact (FE) test available in GraphPad InStat version 3.06 patients (86.3%–100%), irrespective of disease state (FE for Windows (GraphPad Software, San Diego, CA, USA; p>0.05). http://www.graphpad.com) were used as appropriate for the All Chinese, Indian and Malay FD isolates were found to comparison of individual virulence factors between ethnic possess babA. The prevalence of babA was also universally groups and disease states. Analyses were performed high in Chinese GC and DU isolates (FE p>0.05). 444 Eur J Clin Microbiol Infect Dis (2010) 29:439–451

Table 3 Summary of the prevalence of 16 loci, and Ethnicity Chinese Indian Malay resulting cag pathogenicity island (PAI) and vacA geno- Diagnosis FD GC DU FD FD types, in 159 clinical H. pylori Total number 52 22 16 49 20 isolates from Malaysia and Singapore, with respect to cagA 100.0% 100.0% 100.0% 100.0% 100.0% disease state and ethnicity EPIYA-C 9.6% 13.6% 31.3% 89.8% 70.0% of the host EPIYA-D 90.4% 86.4% 68.8% 10.2% 30.0% cagE 92.3% 95.5% 93.8% 93.9% 100.0% cagL 86.5% 100.0% 100.0% 98.0% 100.0% cagT 96.2% 86.4% 100.0% 98.0% 100.0% HP0521 ES 88.5% 86.4% 100.0% 20.4% 25.0% HP0521 A 7.7% 13.6% 0.0% 73.5% 55.0% HP0521 A+ 0.0% 0.0% 0.0% 4.1% 10.0% HP0521 B 3.8% 0.0% 0.0% 2.0% 10.0% cag PAI intact 7.7% 13.6% 0.0% 73.5% 75.0% cag PAI intact 521 ES 73.1% 72.7% 81.3% 16.3% 25.0% cag PAI partial 19.2% 13.6% 18.8% 10.2% 0.0% vacA s1 100.0% 100.0% 100.0% 91.8% 100.0% vacA m1 51.9% 50.0% 75.0% 65.3% 60.0% vacA i1 100.0% 100.0% 100.0% 89.8% 100.0% vacA s1i1m1 51.9% 50.0% 75.0% 61.2% 60.0% vacA s1i1m2 48.1% 50.0% 25.0% 26.5% 40.0% vacA s2m2i2 0.0% 0.0% 0.0% 8.2% 0.0% babA 100.0% 95.5% 100.0% 100.0% 100.0% OipA ON 86.5% 86.4% 87.5% 91.8% 95.0% dupA 32.7% 59.1% 62.5% 14.3% 30.0%

The oipA gene was detected in all but five isolates (all in classified as high virulence (s1i1m1/s1i1m2) and low the Chinese subpopulation). Based on sequencing of the virulence (non-s1i1m1/s1i1m2) between the Chinese and PCR product, oipA was determined to be ON in the majority Indian FD groups (FE p=0.011). A meaningful statistical of isolates (90–95%), with no significant difference between comparison of the Malay and Indian FD groups could not the groups (FE p>0.05). be performed due to the low numbers of isolates and low For vacA, the signal sequence and intermediate region frequency of low-virulence strains. types were uniform, with all isolates from the Chinese FD, The prevalence of dupA, cagA and the EPIYA motifs GC and DU patients being positive for s1 and i1. In have been reported in this population previously, and addition, 91.8% isolates from Indian FD patients and all although we have added isolates to the Malay FD group, isolates from Malay FD patients were also positive for s1 the trends have not altered [21, 35]. The prevalence of the and i1, leaving only 8.2% isolates from Indian FD patients EPIYA motifs in the Chinese DU group has not previously positive for s2 and i2 (FE p>0.05). Although the been reported. We found the EPIYA-C motif to be prevalence of the mid-region m1 differed in isolates from significantly more prevalent in isolates from DU than FD Chinese patients with 51.9% FD, 50% GC and 75% DU patients (FE p=0.0474). However, no significant difference positivity, this did not reach statistical significance (χ2 was found between the Chinese FD, DU and GC groups p>0.2). No significant difference was found in the when analysed in a 3 × 2 Chi-square test (χ2 p=0.0978), prevalence of m1 in isolates from Chinese FD patients as nor when we compared the Chinese DU and GC groups or compared with Indian and Malay FD patients (χ2 p>0.2). the Chinese FD and GC groups (FE p=0.2431 and When all three vacA regions were combined, the high- p=0.6880, respectively). virulence s1i1m1 and s1i1m2 genotypes were found in 100% of isolates from Chinese FD, GC and DU patients Prevalence of an intact cag PAI and 87.8% and 100% of Indian and Malay FD isolates, respectively. Other genotypes, classified as low virulence, In the present study, the cag PAI was classified as either were observed in a minority of isolates. There was a intact, intact 521 ES or partial, according to the presence of significant difference between the proportion of isolates cagA, cagE, cagL, cagT and HP0521. An intact cag PAI Eur J Clin Microbiol Infect Dis (2010) 29:439–451 445 was defined as PCR-positive for all cag PAI genes example of the immunoblot used to determine CagA investigated. An isolate with a restricted deletion over translocation and tyrosine phosphorylation is shown in HP0521 (HP0521 ES) but which was PCR-positive for all Fig. 1. CagA was translocated and phosphorylated by 18 of of the other cag PAI genes investigated, including HP0520 the 22 clinical isolates investigated. All strains found to and HP0522 (HP0521 ES PCR), was classified as intact translocate CagA possessed either an intact cag PAI (10/18) 521 ES. Isolates found to lack one or more of the cag PAI or an intact 521 ES cag PAI (8/18). Of the four clinical genes investigated, with the exception of HP0521, were isolates that did not translocate CagA, two possessed a classified as possessing a partial cag PAI. partial cag PAI (one was cagE−, one cagE− cagL −), one We found that none of the isolates investigated com- possessed an intact cag PAI and one possessed an intact pletely lacked the cag PAI, and that isolates with partial cag 521 ES cag PAI. PAI were present at a low frequency (0–19%), as shown in Table 3. Isolates with an intact cag PAI were detected at a IL-8 induction by AGS cells, and CagA translocation high frequency within the Indian and Malay FD populations and tyrosine phosphorylation in vitro (73.5 and 75%, respectively), whereas an intact 521 ES cag PAI was predominant in isolates from Chinese FD patients To assess the whether the cag PAI TFSS was likely to be (73.1%) (χ2 p<0.0001). The intact 521 ES cag PAI was also functional, isolates were grouped according to the presence predominant in isolates from Chinese GC or DU patients of cagE and dupA, cag PAI status and the ability to (72.7 and 81.3%, respectively) (FE p>0.05). translocate CagA, as shown in Fig. 2a–c. The presence of both virB4 homologues (cagE and dupA) did not result in Association between virulence loci higher IL-8 induction, with no significant difference detected between the cagE+ dupA+ group (n=7, 1,330 pg/ Whether there was a significant tendency for the various ml) and the cagE+ dupA− group (n=13, 1,396 pg/ml). As loci examined to occur together in this population was also we reported previously [20], there was no correlation examined (Table 4). As true associations between virulence between dupA and IL-8 induction. However, isolates factors should be a feature of the strains and should not possessing cagE appeared to induce higher IL-8 than those relate to the ethnicity and disease state of the host, we lacking cagE (Fig. 2a). There was no significant difference analysed all isolates together. We observed a significant between the average levels of IL-8 induced by isolates association between the HP0521 alleles and EPIYA motifs, possessing an intact cag PAI (n=12, 1,393 pg/ml) and an with EPIYA-C tending to be present with HP0521A intact 521 ES cag PAI (n=8, 1,344 pg/ml) (Fig. 2b). and EPIYA-D tending to be present with HP0521 ES Although the number was too small for statistical analysis, (p<0.0001). This association did disappear when the results these groups differed considerably from the partial cag PAI were stratified by disease state. babA and cagA and high- group (n=2, 268 pg/ml) (Fig. 2b). The average IL- virulence vacA genotypes co-occurred with a high frequency. 8 secretion of the isolates which did not translocate CagA Significant associations were observed between oipA ON/ (622 pg/ml) was considerably lower than the average of OFF and vacA m1/m2 (p=0.001) and between the EPIYA those which did translocate CagA (1,416 pg/ml) (Fig. 2c). motifs and dupA (p=0.039); however, when stratified by the In addition, we grouped strains according to oipA ON disease state, the latter association disappeared (data not status and vacA subtypes. We found no considerable shown). difference in the levels of IL-8 induction between oipA In addition, we assessed whether any single marker was ON (n=19, 1,313 pg/ml) and OFF ( n=3, 1,018 pg/ml) associated with an intact cag PAI by combining both the (Fig. 2d), nor between vacA s1m1i1 (n=11, 1,344 pg/ml) intact and intact 521 ES cag PAI groups. We found that and vacA s1m2i1 strains (n=10, 1,309 pg/ml) (Fig. 2e). The cagE, cagL and cagT were all significantly associated with single strain with vacA s2m2i2 genotype induced a low an intact cag PAI (p<0.0001), but no locus investigated level of IL-8 (123 pg/ml). absolutely predicted an intact cag PAI. The HP0521 alleles were not associated significantly with cag PAI status (p=0.084). Discussion

CagA translocation and tyrosine phosphorylation in vitro The genome of H. pylori exhibits enormous variability with regard to both sequence divergence within conserved genes CagA translocation and phosphorylation was used, in and gene content between strains [38]. Furthermore, strains combination with IL-8 induction, to assess whether the can differ remarkably between ethnic groups, even within cag PAI-encoded TFSS was functional, and, therefore, the same country, reinforcing the necessity to investigate whether the cag PAI was likely to be intact. A typical potential virulence factors and disease determinants in 446 Eur J Clin Microbiol Infect Dis (2010) 29:439–451

Table 4 Comparison of the tendency for selected virulence Virulence factors Number (total n=159) p-value loci to occur together within a single strain, showing the cag PAI, vacA number of isolates positive for cag PAI intact/521 ES, vacA s1i1m1/m2 132 each combination of virulence cag PAI intact/521 ES, vacA s2i2m2 6 loci, as well as the probability of such occurrences cag PAI partial, vacA s1i1m1/m2 21 1 cag PAI, oipAa cag PAI intact/521 ES, oipA ON 124 cag PAI intact/521 ES, oipA OFF 12 cag PAI partial, oipA ON 15 cag PAI partial, oipA OFF 3 0.387 cag PAI, babA cag PAI intact/521 ES, babA+ 138 cag PAI partial, babA+20 cag PAI partial, babA− 1 0.132 cag PAI, dupA cag PAI intact/521 ES, dupA+44 cag PAI intact/521 ES, dupA− 94 cag PAI partial, dupA+9 cag PAI partial, dupA− 12 0.330 EPIYA, HP0521 EPIYA-C, HP0521A 43 EPIYA-C, HP0521A+4 EPIYA-C, HP0521B 2 EPIYA-C, HP0521 ES 15 EPIYA-D, HP0521A 11 EPIYA-D, HP0521A+0 EPIYA-D, HP0521B 3 EPIYA-D, HP0521 ES 81 <0.0001 cagA, babA, vacA cagA+, babA+, vacA s1i1m1/m2 152 cagA+, babA−, vacA s1i1m1/m2 1 cagA+, babA+, vacA s2i2m2 6 1.000 babA, dupA babA+, dupA+52 babA+, dupA− 106 babA−, dupA+ 1 0.333 oipA, babAa oipA ON, babA+ 139 oipA OFF, babA+15b oipA, vacAa oipA ON, vacA s1i1m1/m2 134 oipA ON, vacA s2i2m2 5 0.465 oipA OFF, vacA s1i1m1/m2 14 oipA OFF, vacA s2i2m2 1 a As some isolates were not able oipA, dupAa to be typed for oipA, for comparisons involving oipA ON, dupA+46 oipA,n = 154 oipA ON, dupA− 93 b Could not be computed oipA OFF, dupA+5 because one virulence locus oipA OFF, dupA− 10 1.000 is constant Eur J Clin Microbiol Infect Dis (2010) 29:439–451 447

1 2 3 4 lations. This is the largest number of putative H. pylori A virulence loci to be typed in a multi-ethnic study. Both oipA switched ON and babA were found to be highly prevalent in isolates from Malaysia and Singapore, B which is consistent with previous studies in East Asia [40– 44]. Primer pairs for babA have, to date, been designed to differentiate between the non-functional babA1 and func- Fig. 1 The ability of Helicobacter pylori clinical isolates to – translocate CagA (a) and tyrosine phosphorylation of translocated tional babA2 alleles [42, 45 48]. However, these primers CagA (b) by AGS cells in vitro was determined by immunoblot. were not able to detect babA in all babA2-positive Lanes 1–3 show three clinical isolates (FD657, FD661 and FD675) for completely sequenced strains [36]. The primer pair which CagA was translocated (a) and tyrosine phosphorylated (b), designed in the present study is capable of amplifying whilst lane 4 shows a clinical isolate (FD613) for which CagA was not translocated (a) nor tyrosine phosphorylated (b). Variations in the size babA in all six completely sequenced strains, but cannot of CagA as detected by specific antibodies for CagA and tyrosine differentiate between the babA alleles. We argue that this phosphorylated (PY-99) CagA reflect differences in the size of CagA approach is valid, given that, despite a large number of as determined by EPIYA motifs and other gene variations studies, only one H. pylori isolate has been reported to possess babA1 [49–51]. Thus, we are unlikely to have over- multi-ethnic studies [2, 34, 39]. This study was designed to estimated babA2 prevalence through the use of this primer investigate a novel combination of 19 loci, most of which pair. are known putative virulence factors, in 159 clinical H. The high-virulence vacA s1i1m1 and s1i1m2 genotypes pylori isolates from three major ethnic groups in Singapore were predominant in all three ethnic groups, irrespective of and Malaysia, including GC and DU cases in the ethnic disease state. This strongly suggests that this factor is Chinese subpopulation and FD controls from all subpopu- unlikely to be a predictor of disease within this population,

pg/ml pg/ml pg/ml 2000 2000 2000

1500 1500 1500

1000 1000 1000

500 500 500

A 0 B 0 C 0 cagE+ cagE+ cagE- cagE- cagPAI cagPAI cagPAI Tranlocation + Tranlocation - dupA- dupA- dupA+ dupA- intact intact ES partial

pg/ml pg/ml 2000 2000

1500 1500

1000 1000

500 500

D 0 E 0 oipA oipA vacA vacA vacA ON OFF s1i1m1 s1i1m2 s2i2m2

Fig. 2 Induction of IL-8 secretion by AGS cells in vitro in response The results correspond to the average of at least three independent to incubation with 22 clinical H. pylori isolates for 6 h. Each panel experiments, with each sample performed in duplicate. Each isolate is represents the distribution of IL-8 secretion when the results for represented by a point and the average IL-8 secreted (pg/ml) is individual isolates were grouped by: (a) the presence of cagE and represented by a bar within the figure, except where only one sample dupA,(b) cag PAI status, (c) ability to translocate cagA,(d) oipA ON is present within a group and OFF status, or (e) the vacA subtypes s1m1i1, s1m2i1 and s2m2i2. 448 Eur J Clin Microbiol Infect Dis (2010) 29:439–451 as has been suggested for other populations [11, 26, 27]. interaction with other cag PAI genes require investigation, These findings are consistent with previous studies in East and we are currently undertaking additional studies on this Asia [44, 52–54], although we are the first to conduct a intriguing gene (Nilsson et al., in preparation). study in Malaysia and Singapore. Our approach to determining whether the cag PAI was The most significant findings of this study relate to the likely to be intact differs to previous studies due to our cag PAI, where we typed a unique combination of loci, study population, as well as the use of a novel combination namely, cagA, EPIYA-C, EPIYA-D, cagE, cagL, cagT, of loci and primer pairs. An intact cag PAI was commonly HP0521A, HP0521B and the HP0521 ES by PCR, to assess present in the ethnic Indian and ethnic Malay subpopula- whether the cag PAI was likely to be intact. In the initial tions (73.5 and 75%, respectively), whilst in the ethnic study by Blomstergren et al., HP0521 was reported to be Chinese population, an intact 521ES cag PAI was more allelic, with 54% of Swedish isolates possessing the novel common (72.7–81.3%) due to the high prevalence of HP0521B allele [55]. In contrast, we found HP0521B to be HP0521 ES in this population. Our results are consistent comparatively rare in all three ethnicities (2–10%), with the with the few studies conducted in East Asia [30–32], but smaller HP0521A allele present in the majority of isolates contrast with reports from predominantly western populations from ethnic Indian and Malay FD patients. This is [15, 17, 18, 60, 61]. Despite previous reports of an surprising, as isolates from ethnic Indians would typically association between the cag PAI and severe H. pylori-related cluster with western strains [34, 56] and, so, would be disease development [18, 62, 63], the overall high prevalence expected to be similar to Swedish strains. However, as of an intact cag PAI in our study, as in other East Asian these are the only two ‘western’ populations to investigate populations [30–32], has precluded an association with HP0521 alleles, further investigation is necessary. Within disease development. However, the high prevalence of an the ethnic Chinese subpopulation, the HP0521 ES was intact cag PAI in the Indian subpopulation was unexpected, highly prevalent, irrespective of disease status. This is given two previous studies which reported the prevalence of similar to the only other study to examine HP0521 in Asia, an intact cag PAI to be 8.3 and 12% [15, 60] using the same where Azuma et al. reported that 8/11 Japanese isolates primer pairs. The methodology used, where isolates negative investigated lacked HP0521 [57]. Our observation of a for any one primer pair were not considered to have an intact significant association between the HP0521 alleles and the cag PAI, is likely to have underestimated the true prevalence EPIYA motifs, namely, between HP0521 ES and EPIYA-D, of an intact cag PAI. Indeed, using the same primer pairs, and HP0521A and EPIYA-C, is intriguing, considering that Ikenoue et al. acknowledged that sequence variability within there is no known functional relationship between HP0521 the cagA promoter was a problem when designing primers and cagA, with in vitro mutagenesis studies demonstrating and using that region as a marker [14]. In addition, other no involvement of HP0521 in either CagA translocation or studies have indicated that it is necessary to use multiple IL-8 induction [12]. While the function is currently primer pairs for individual cag PAI genes to avoid false- unknown, both HP0521A and HP0521B contain domains negatives [63, 64]. with similarity to DNA topoisomerase I [18, 57]. In A shortcoming of many previous studies is the lack of addition to their role in DNA replication, DNA top- functional validation when classifying the cag PAI as intact oisomerases have been implicated in the pathogenesis of or partial, particularly when only one or two genes were bacteria such as Shigella flexneri and Agrobacterium examined. To validate our approach, we investigated the tumefaciens. Expression of a DNA Topoisomerase I in S. ability of a subset of strains to translocate CagA and induce flexneri has been shown to up-regulate virulence gene IL-8 secretion by AGS cells in vitro to determine whether transcription [58], and in A. tumefaciens, the TFSS utilises the cag PAI was fully functional, suggestive of an intact the topoisomerase activity of VirD1 to translocate DNA to cag PAI. Ninety percent of strains classified as intact or target plant cells [59]. This poses several questions when intact 521ES possessed a fully functional cag PAI, with the taken together with our data. For example, whether HP0521 ability to both translocate CagA and induce IL-8, sugges- could be involved in a new function for the cag PAI and/or tive of a truly intact cag PAI. The two strains possessing a the cag PAI-encoded TFSS. Furthermore, why is this gene partial cag PAI were unable to induce IL-8 and translocate selectively deleted in high-virulence and disease-associated CagA, as would be expected given that the cag PAI genes strains? A further question is whether there is a novel investigated have important roles in these cag PAI functional relationship between the HP0521 alleles and the functions. A further two strains which possessed either an EPIYA motifs, which might explain their association. It is intact or intact 521 ES cag PAI, that is, possessed all of the also possible that the correlation between the HP0521 assayed genes (except HP0521), did not have fully alleles and EPIYA motifs is not functional per se, but, functional cag PAIs, with both of them being unable to rather, are independently associated with the ethnic groups translocate CagA. Whilst one of these strains (an intact cag investigated. Clearly, the HP0521 alleles, their role and PAI) was not compromised in its ability to induce IL-8, the Eur J Clin Microbiol Infect Dis (2010) 29:439–451 449 other (an intact 521 ES cag PAI) only induced a low level between ethnic groups in terms of HP0521 is of consider- of IL-8 secretion in vitro. This may suggest that there are able interest, given the fact that the functions of the HP0521 one or more low-frequency disruptions elsewhere in the cag alleles are currently unknown. Our finding that the major H. PAI not detected by our analysis. Indeed, while HP0524 is pylori virulence factors, vacA s1i1m1 and s1i1m2, oipA and essential for CagA translocation, it is not required for IL- babA, in addition to an intact cag PAI, were highly 8 induction [12] and IL-8 induction has even been shown to prevalent in Malaysian and Singaporean isolates, irrespec- be enhanced through the inactivation of HP0524 (virD4) tive of the ethnicity or disease state of the host, shows that [65, 66], which may explain why one strain which was these virulence factors are not reliable predictors of disease unable to translocate CagA was not compromised in its in this population. However, given the disparity in severe ability to induce IL-8. A disruption of any one or more of disease development between these ethnic groups, some or the cag PAI genes involved in both translocation and IL- all of these factors may play an integral role when other, as 8 induction may be responsible for the non-functional cag yet unidentified, predisposing host and environmental PAI TFSS in the strain unable to translocate CagA or factors are also present. induce IL-8 induction. However, due to the prohibitively large size of the cag PAI, and as single-nucleotide Acknowledgements AGS cells were kindly donated by Dr. Laura polymorphisms may also inactivate the cag PAI [12], it is Plant, Swedish Institute for Infectious Disease Control (Smittskyddsin- difficult to assess the precise nature of the disruption. stitutet), and the Microbiology and Tumour Biology Centre at Karolinska Interestingly, IL-8 induction by cag PAI-negative strains Institutet, Solna, Sweden. This work was supported by the Cancer Council of New South Wales, Australia (REF 66/04). Sönke Andres was has also been previously reported [40, 67], and it is, supported by a grant from the Marie Curie Early Stage Research Training therefore, possible that the IL-8 induction in these two Fellowship of the European Community’s 6th Framework Program, isolates may instead be cag PAI-independent. under contract number MEST-CT-2004-8475. Given that strains with HP0521 ES were not compromised in their ability to translocate CagA or induce IL-8 in vitro, it is likely that they possess a fully functional cag References PAI. It is, therefore, appropriate to classify such strains as ‘intact 521 ES’, rather than partial. It is possible, however, 1. Yamaoka Y, Kato M, Asaka M (2008) Geographic differences in that our proposed classification system may require gastric cancer incidence can be explained by differences between Helicobacter pylori strains. Intern Med 47:1077–1083 modification when the function of HP0521 is finally 2. Lu H, Yamaoka Y, Graham DY (2005) Helicobacter pylori elucidated. virulence factors: facts and fantasies. Curr Opin Gastroenterol 21 We found that no single marker absolutely corresponded (6):653–659 to an intact cag PAI, confirming the findings of Jenks et al. 3. Hatakeyama M (2008) SagA of CagA in Helicobacter pylori pathogenesis. Curr Opin Gastroenterol 11(1):30–37 and Ko and Seo [63, 64]. Given the statistical association 4. Tammer I, Brandt S, Hartig R et al (2007) Activation of Abl by with cag PAI status, we propose that the combined use of Helicobacter pylori: a novel kinase for CagA and crucial mediator cagA, cagE, cagL and cagT, but not HP0521, is likely to of host cell scattering. Gastroenterology 132(4):1309–1319 reliably determine whether the cag PAI is functional and, 5. Selbach M, Moese S, Hauck CR et al (2002) Src is the kinase of the Helicobacter pylori CagA protein in vitro and in vivo. J Biol therefore, intact in an East Asian population. We recom- Chem 277(9):6775–6778 mend that future studies employing new marker-based 6. Segal ED, Cha J, Lo J et al (1999) Altered states: involvement of methods to determine the likelihood of an intact cag PAI phosphorylated CagA in the induction of host cellular growth should conduct functional studies to validate their approach changes by Helicobacter pylori. Proc Natl Acad Sci USA 96:14559–14564 and results. 7. Higashi H, Tsutsumi R, Muto S et al (2002) SHP-2 tyrosine In conclusion, although significant differences in the phosphatase as an intracellular target of Helicobacter pylori CagA prevalence of dupA and the EPIYA motif and HP0521 types protein. Science 295(5555):683–686 were observed among the three ethnicities resident in 8. Higashi H, Tsutsumi R, Fujita A et al (2002) Biological activity of the Helicobacter pylori virulence factor CagA is determined by Malaysia and Singapore, no association was observed variation in the tyrosine phosphorylation sites. Proc Natl Acad Sci between any single virulence factor or combination of USA 99(22):14428–14433 virulence factors and either DU or GC in the Chinese 9. Naito M, Yamazaki T, Tsutsumi R et al (2006) Influence of population. However, in this study, we have established that EPIYA-repeat polymorphism on the phosphorylation-dependent biological activity of Helicobacter pylori CagA. Gastroenterology a combination of four cag PAI genes (cagA, cagE, cagL 130:1181–1190 and cagT) could be used to estimate whether the cag PAI 10. Jones KR, Joo YM, Jang S et al (2009) Polymorphism in the was intact and likely to be functional in vitro. Using this CagA EPIYA motif impacts development of gastric cancer. J Clin – novel approach, the majority of isolates from Malaysia and Microbiol 47(4):959 968 11. Basso D, Zambon C-F, Letley DP et al (2008) Clinical relevance Singapore were found to have an intact and functional cag of Helicobacter pylori cagA and vacA gene polymorphisms. PAI. The novel finding that significant variation exists Gastroenterology 135:91–99 450 Eur J Clin Microbiol Infect Dis (2010) 29:439–451

12. Fischer W, Püls J, Buhrdorf R et al (2001) Systematic mutagenesis 30. Sheu SM, Sheu BS, Yang HB et al (2002) Presence of iceA1 but of the Helicobacter pylori cag pathogenicity island: essential not cagA, cagC, cagE, cagF, cagN, cagT,ororf13 genes of genes for CagA translocation in host cells and induction of Helicobacter pylori is associated with more severe gastric interleukin-8. Mol Microbiol 42(5):1337–1348; erratum in Mol inflammation in Taiwanese. J Formos Med Assoc 101(1):18–23 Microbiol. 2003 Mar;47(6):1759 31. Hsu P-I, Hwang I-R, Cittelly D et al (2002) Clinical presentation 13. Censini S, Lange C, Xiang Z et al (1996) cag, a pathogenicity in relation to diversity within the Helicobacter pylori cag island of Helicobacter pylori, encodes type I-specific and disease- pathogenicity island. Am J Gastroenterol 97:2231–2238 associated virulence factors. Proc Natl Acad Sci USA 93 32. Maeda S, Yoshida H, Ikenoue T et al (1999) Structure of cag (25):14648–14653 pathogenicity island in Japanese Helicobacter pylori isolates. Gut 14. Ikenoue T, Maeda S, Ogura K et al (2001) Determination of 44(3):336–341 Helicobacter pylori virulence by simple gene analysis of the cag 33. Tan HJ, Rizal AM, Rosmadi MY et al (2005) Distribution of pathogenicity island. Clin Diagn Lab Immunol 8(1):181–186 Helicobacter pylori cagA, cagE and vacA in different ethnic groups 15. Kauser F, Khan AA, Hussain MA et al (2004) The cag in Kuala Lumpur, Malaysia. J Gastroenterol Hepatol 20(4):589–594 pathogenicity island of Helicobacter pylori is disrupted in the 34. Schmidt H-MA, Goh KL, Fock KM et al (2009) Distinct cagA majority of patient isolates from different human populations. J EPIYA motifs are associated with ethnic diversity in Malaysia and Clin Microbiol 42(11):5302–5308 Singapore. Helicobacter 14:256–263 16. Ali M, Khan AA, Tiwari SK et al (2005) Association between 35. Rozen S, Skaletsky HJ (2000) Primer3 on the WWW for general cag-pathogenicity island in Helicobacter pylori isolates from users and for biologist programmers. In: Krawetz S, Misener S peptic ulcer, gastric carcinoma, and non-ulcer dyspepsia subjects (eds) Bioinformatics methods and protocols: methods in molecular with histological changes. World J Gastroenterol 11(43):6815–6822 biology. Humana Press, Totowa, pp 365–386 17. Louw JA, Kidd MS, Kummer AF et al (2001) The relationship 36. Bikandi J, San Millán R, Rementeria A et al (2004) In silico between Helicobacter pylori infection, the virulence genotypes of analysis of complete bacterial genomes: PCR, AFLP-PCR and the infecting strain and gastric cancer in the African setting. endonuclease restriction. Bioinformatics 20(5):798–799 Helicobacter 6(4):268–273 37. Björkholm B, Lundin A, Sillén A et al (2001) Comparison of 18. Nilsson C, Sillén A, Eriksson L et al (2003) Correlation between cag genetic divergence and fitness between two subclones of pathogenicity island composition and Helicobacter pylori-associated Helicobacter pylori. Infect Immun 69(12):7832–7838 gastroduodenal disease. Infect Immun 71(11):6573–6581 38. Alm RA, Ling LS, Moir DT et al (1999) Genomic-sequence 19. Mattar R, dos Santos AF, Eisig JN et al (2005) No Correlation of comparison of two unrelated isolates of the human gastric babA2 with vacA and cagA genotypes of Helicobacter pylori and pathogen Helicobacter pylori. Nature 397(6715):176–180 grading of gastritis from peptic ulcer disease patients in Brazil. 39. Yamaoka Y, Malaty HM, Osato MS et al (2000) Conservation of Helicobacter 10(6):601–608 Helicobacter pylori genotypes in different ethnic groups in 20. Schmidt H-MA, Andres S, Kaakoush NO et al (2009) The Houston, Texas. J Infect Dis 181:2083–2086 prevalence of the duodenal ulcer promoting gene (dupA)in 40. Yamaoka Y, Kwon DH, Graham DY (2000) A M(r) 34,000 Helicobacter pylori isolates varies by ethnic group and is not proinflammatory outer membrane protein (oipA) of Helicobacter universally associated with disease development: a case–control pylori. Proc Natl Acad Sci USA 97(13):7533–7538 study. Gut Pathog 1(5):5 41. Kim YS, Kim N, Kim JM et al (2009) Helicobacter pylori 21. Lu H, Hsu P-I, Graham DY et al (2005) Duodenal ulcer promoting genotyping findings from multiple cultured isolates and mucosal gene of Helicobacter pylori. Gastroenterology 128(4):833–848 biopsy specimens: strain diversities of Helicobacter pylori isolates 22. Argent RH, Burette A, Miendje Deyi VY et al (2007) The in individual hosts. Eur J Gastroenterol Hepatol 21(5):522–528 presence of dupA in Helicobacter pylori is not significantly 42. Mizushima T, Sugiyama T, Komatsu Y et al (2001) Clinical associated with duodenal ulceration in Belgium, South Africa, relevance of the babA2 genotype of Helicobacter pylori in China, or North America. Clin Infect Dis 45(9):1204–1206 Japanese clinical isolates. J Clin Microbiol 39(7):2463–2465 23. Cao P, Lee KJ, Blaser MJ et al (2005) Analysis of hopQ alleles in 43. Lai CH, Kuo CH, Chen YC et al (2002) High prevalence of cagA- East Asian and Western strains of Helicobacter pylori. FEMS and babA2-positive Helicobacter pylori clinical isolates in Microbiol Lett 251(1):37–43 Taiwan. J Clin Microbiol 40(10):3860–3862 24. Hisatsune J, Nakayama M, Isomoto H et al (2008) Molecular 44. Chomvarin C, Namwat W, Chaicumpar K et al (2008) Prevalence characterization of Helicobacter pylori VacA induction of IL-8 in of Helicobacter pylori vacA, cagA, cagE, iceA and babA2 U937 cells reveals a prominent role for p38MAPK in activating genotypes in Thai dyspeptic patients. Int J Infect Dis 12(1):30–36 transcription factor-2, cAMP response element binding protein, 45. Gerhard M, Lehn N, Neumayer N et al (1999) Clinical relevance and NF-kappaB activation. J Immunol 180:5017–5027 of the Helicobacter pylori gene for blood-group antigen-binding 25. Dunn BE, Cohen H, Blaser MJ (1997) Helicobacter pylori. Clin adhesin. Proc Natl Acad Sci USA 96(22):12778–12783 Microbiol Rev 10(4):720–741 46. Zambon C-F, Navaglia F, Basso D et al (2003) Helicobacter 26. Rhead JL, Letley DP, Mohammadi M et al (2007) A new pylori babA2, cagA, and s1 vacA genes work synergistically in Helicobacter pylori vacuolating cytotoxin determinant, the inter- causing intestinal metaplasia. J Clin Pathol 56:287–291 mediate region, is associated with gastric cancer. Gastroenterology 47. Zhang Z, Zheng Q, Chen X et al (2008) The Helicobacter pylori 133(3):926–936 duodenal ulcer promoting gene, dupA in China. BMC Gastroenterol 27. Sugimoto M, Zali MR, Yamaoka Y (2009) The association of 8:49 vacA genotypes and Helicobacter pylori-related gastroduodenal 48. Han YH, Liu WZ, Zhu HY et al (2004) Clinical relevance of iceA diseases in the Middle East. Eur J Clin Microbiol Infect Dis 28 and babA2 genotypes of Helicobacter pylori in a Shanghai (10):1227–1236 population. Chin J Dig Dis 5(4):181–185 28. Huang JQ, Zheng GF, Sumanac K et al (2003) Meta-analysis of 49. Yamaoka Y (2008) Roles of Helicobacter pylori BabA in the relationship between cagA seropositivity and gastric cancer. gastroduodenal pathogenesis. World J Gastroenterol 14(27):4265– Gastroenterology 125(6):1636–1644 4272 29. Miehlke S, Kibler K, Kim JG et al (1996) Allelic variation in the 50. Ilver D, Arnqvist A, Ogren J et al (1998) Helicobacter pylori cagA gene of Helicobacter pylori obtained from Korea compared adhesin binding fucosylated histo-blood group antigens revealed to the United States. Am J Gastroenterol 91(7):1322–1325 by retagging. Science 279:373–377 Eur J Clin Microbiol Infect Dis (2010) 29:439–451 451

51. Fujimoto S, Olaniyi Ojo O, Arnqvist A et al (2007) Helicobacter gastroduodenal disease in South African Helicobacter pylori pylori BabA expression, gastric mucosal injury, and clinical isolates. Gut 49:11–17 outcome. Clin Gastroenterol Hepatol 5:49–58 63. Jenks PJ, Mégraud F, Labigne A (1998) Clinical outcome after 52. Sheu SM, Hung K-H, Sheu BS et al (2009) Association of infection with Helicobacter pylori does not appear to be reliably nonsynonymous substitutions in the intermediate region of the predicted by the presence of any of the genes of the cag vacA gene of Helicobacter pylori with gastric diseases in Taiwan. pathogenicity island. Gut 43:752–758 J Clin Microbiol 47(1):249–251 64. Ko JS, Seo JK (2002) cag pathogenicity island of Helicobacter 53. Ogiwara H, Graham DY, Yamaoka Y (2008) vacA i-region pylori in Korean children. Helicobacter 7(4):232–236 subtyping. Gastroenterology 134:1267–1278 65. Crabtree JE, Farmery SM, Lindley IJD et al (1994) CagA/ 54. Yamaoka Y, Kodama T, Gutierrez O et al (1999) Relationship cytotoxic strains of Helicobacter pylori and interleukin-8 in between Helicobacter pylori iceA, cagA, and vacA status and gastric epithelial cell lines. J Clin Pathol 47:945–950 clinical outcome: studies in four different countries. J Clin 66. Eaton KA, Kersulyte D, Mefford M et al (2001) Role of Microbiol 37(7):2274–2279 Helicobacter pylori cag region genes in colonization and gastritis 55. Blomstergren A, Lundin A, Nilsson C et al (2004) Comparative in two animal models. Infect Immun 69(5):2902–2908 analysis of the complete cag pathogenicity island sequence in four 67. Audibert C, Burucoa C, Janvier B et al (2001) Implication of the Helicobacter pylori isolates. Gene 328:85–93 structure of the Helicobacter pylori cag pathogenicity island in 56. Devi SM, Ahmed I, Francalacci P et al (2007) Ancestral European induction of interleukin-8 secretion. Infect Immun 69(3):1625– roots of Helicobacter pylori in India. BMC Genomics 8:184–192 1629 57. Azuma T, Yamakawa A, Yamazaki S et al (2004) Distinct 68. Yamaoka Y, El-Zimaity HM, Gutierrez O et al (1999) Relationship diversity of the cag pathogenicity island among Helicobacter between the cagA 3′ repeat region of Helicobacter pylori,gastric pylori strains in Japan. J Clin Microbiol 42(6):2508–2517 histology, and susceptibility to low pH. Gastroenterology 117 58. McNairn E, Ni Bhriain N, Dorman CJ (1995) Overexpression of (2):342–349 the Shigella flexneri genes coding for DNA topoisomerase IV 69. Azuma T, Yamakawa A, Yamazaki S et al (2002) Correlation compensates for loss of DNA topoisomerase I: effect on virulence between variation of the 3′ region of the cagA gene in gene expression. Mol Microbiol 15(3):507–517 Helicobacter pylori and disease outcome in Japan. J Infect Dis 59. Atmakuri K, Cascales E, Burton OT et al (2007) Agrobacterium 186(11):1621–1630 ParA/MinD-like VirC1 spatially coordinates early conjugative 70. Zhou J, Zhang J, Xu C et al (2004) cagA genotype and variants in DNA transfer reactions. EMBO J 26(10):2540–2551 Chinese Helicobacter pylori strains and relationship to gastroduo- 60. Tiwari SK, Khan AA, Ahmed KS et al (2005) Polymerase chain denal diseases. J Med Microbiol 53:231–235 reaction based analysis of the cytotoxin associated gene pathogenic- 71. Argent RH, Zhang Y, Atherton JC (2005) Simple method for ity island of Helicobacter pylori from saliva: an approach for rapid determination of the number of Helicobacter pylori CagA molecular genotyping in relation to disease status. J Gastroenterol variable-region EPIYA tyrosine phosphorylation motifs by PCR. Hepatol 20(10):1560–1566 J Clin Microbiol 43(2):791–795 61. Mattar R, Marques SB, Monteiro Mdo S et al (2007) Helicobacter 72. Ando T, Peek RM Jr, Lee Y-C et al (2002) Host cell responses to pylori cag pathogenicity island genes: clinical relevance for peptic genotypically similar Helicobacter pylori isolates from United ulcer disease development in Brazil. J Med Microbiol 56(Pt 1):9– States and Japan. Clin Diagn Lab Immunol 9(1):167–175 14 73. Franco AT, Johnston E, Krishna U et al (2008) Regulation of 62. Kidd M, Lastovica AJ, Atherton JC et al (2001) Conservation of gastric carcinogenesis by Helicobacter pylori virulence factors. the cag pathogenicity island is associated with vacA alleles and Cancer Res 68(2):379–387 2.4 Paper IV

95 RESEARCH ARTICLE Helicobacter pylori de¢nes local immune response through interaction with dendritic cells Sonke¨ Andres1,2, Heather-Marie A. Schmidt3, Hazel Mitchell3, Mikael Rhen2, Markus Maeurer1,2 & Lars Engstrand1,2

1Department of Microbiology, Tumor and Cell Biology, Karolinska Institutet, Stockholm, Sweden; 2Swedish Institute for Infectious Disease Control, Solna, Sweden; and 3School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, NSW, Australia

Correspondence: Lars Engstrand, Swedish Abstract Institute for Infectious Disease Control, Solna 171 82, Sweden. Tel.: 146 8 457 24 15; The bacterial pathogen Helicobacter pylori colonizes the human gastric and fax: 146 8 32 83 30; e-mail: duodenal mucosa, evades clearance by the host response and is associated with [email protected] peptic ulcer disease and an increased risk of gastric adenocarcinoma. Dendritic cells (DCs) are initiators of the immune response to H. pylori. The aim of the Received 17 June 2010; revised 7 October current study was to investigate the interaction between H. pylori with DCs. To 2010; accepted 26 October 2010. determine the impact of H. pylori on the maturation and the activation of monocyte-derived DCs, the effect of 20 clinical H. pylori strains with different DOI:10.1111/j.1574-695X.2010.00761.x inflammatory backgrounds on adenocarcinoma gastric epithelial cells was investigated. The inflammatory background was defined according to the degree of Editor: Nicholas Carbonetti lymphocyte and granulocyte infiltration and the bacterial density at the site of infection. DC maturation and activation varied after exposure to the different Keywords Helicobacter pylori; dendritic cell; epithelial cell; strains. While maturation appeared to be independent of any virulence factor interleukin-12; virulence factor. tested, a significant increase in the average level of cytokine production was observed for the proinflammatory cytokines interleukin-12 (IL-12), tumour necrosis factor-a, IL-6 and IL-1b when comparing strains with low inflammatory backgrounds with those of the medium or high backgrounds. In conclusion, the DC response towards different strains in vitro was associated with the clinical outcome of the individual host, suggesting a major role of this cell type in modulating strain-specific H. pylori infection.

including the 18 genes that encode a type IV secretion Introduction system. Among these, the ATPase CagE has been shown to Helicobacter pylori persistently colonizes the gastric mucosa be involved in the process of IL-8 induction (Tummuru of about one half of the world’s population causing chronic et al., 1995; Censini et al., 1996). The blood group antigen- gastric inflammation, which may progress to peptic ulcer binding adhesin (BabA) is a mediator of H. pylori binding to disease or gastric adenocarcinoma (Ernst & Gold, 2000). epithelial cells is by interacting with Lewis B-like structures The inflammatory response towards this Gram-negative on host cells (Ilver et al., 1998). Epithelial and infiltrating bacterium is characterized by the recruitment of macro- inflammatory cells release an array of cytokines including phages, dendritic cells (DCs), B and T lymphocytes, and in IL-1b, IL-2, IL-6, IL-8, IL-12, tumour necrosis factor-a particular, neutrophils to the site of infection (Crabtree, (TNF-a), interferon-g and transforming growth factor-b 1996). Interleukin-8 (IL-8) is the main chemoattractor (Crabtree, 1996; Lindholm et al., 1998; Suerbaum & for neutrophils and is released by gastric epithelial cells Michetti, 2002). Despite the vigorous immune response,

IMMUNOLOGY & MEDICAL MICROBIOLOGY (Zarrilli et al., 1999; Backhed et al., 2003). IL-8 release is H. pylori evades clearance by the downregulation of induced by the delivery of peptidoglycan fragments into the immune effectors, inhibition of innate immunity or anti- host cell by the cytotoxin associate gene (cag) pathogenicity genic variation, among other mechanisms (Blaser & Ather- island (PAI). The cag PAI comprises of up to 30 genes, ton, 2004). The H. pylori vacuolating cytotoxin VacA, for

FEMS Immunol Med Microbiol ]] (2010) 1–11 c 2010 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved 2 S. Andres et al. example, has been shown to suppress T-cell proliferation Materials and methods through downregulation of IL-2 transcription (Gebert et al., 2003). Bacterial cultures and strain material DCs act as important mediators of the innate and the Clinical H. pylori isolates were obtained from corpus biop- adaptive immune response through their ability to capture sies of 20 individuals (ages 52–75) from two communities in and transfer antigens and regulate the T-cell-mediated Northern Sweden, who were included in a population-based immune response (Banchereau et al., 2000). Pathogen- upper-endoscopy study, the Kalixanda study (Aro et al., associated molecular patterns, cytokines and other mole- 2004, 2006). This study was approved by the Ethics Com- cules determine the differentiation of immature DCs mittee at Umea˚ University, Sweden. Histological criteria (iDCs) and thereby immunity or immunological tolerance used for scoring the patient’s gastric biopsies have been (Sabatte et al., 2007). This differentiation is characterized by described previously by Storskrubb et al. (2005). Briefly, two the upregulation of costimulatory molecules and chemokine biopsies for H. pylori culture were collected from the gastric receptors on the cell surface that play important roles in cell corpus and antrum, respectively. A Warthin–Starry silver migration and T-cell activation (Sozzani et al., 1998; Guer- stain was performed to detect H. pylori infection. Histologi- monprez et al., 2002). Furthermore, DCs can penetrate the cal features of the biopsies were evaluated by two experi- intestinal barrier to sample bacteria (Niess et al., 2005) enced pathologists. One hundred randomly chosen subjects as has recently been shown for transepithelial DCs in were re-evaluated by a third pathologist. For this study, we the human H. pylori-infected gastric epithelium (Necchi selected 20 H. pylori strains isolated from corpus biopsies, et al., 2009). In addition, H. pylori can disrupt epithelial based on the grade of lymphocyte and granulocyte infiltra- apical–junction complexes (Amieva et al., 2003). The tion, as well as H. pylori density at the site of infection. These Toll-like receptors (TLRs) 2, 4 and 9 are probably involved parameters had been determined previously according to in DC activation but little is known regarding bacterial the updated Sydney System score definitions (Dixon et al., factors (Rad et al., 2007, 2009). Whole H. pylori cells and 1996). Strains from biopsies with high levels of inflamma- individual components such as lipopolysaccharides strongly tory cell infiltration and a low density of H. pylori were induce the activation and maturation of human iDCs. This designated as ‘high inflammatory potential’ and strains effect is comparable with that induced by Escherichia coli from biopsies with low levels of inflammatory cell infiltra- lipopolysaccharides, a recognized DC stimulating agent tion but a high density of H. pylori were designated as ‘low (Kranzer et al., 2004, 2005). Lewis blood group structures inflammatory potential’. The two control strains (67:20 and on the O antigen of H. pylori lipopolysaccharides bind with 67:21) have been described previously (Enroth et al., 2000). a high affinity to the dendritic cell-specific intercellular The bacterial strains were passaged a maximum of three adhesion molecule-3 (ICAM-3)-grabbing nonintegrin (DC- times. All strains used in the study are listed in Table 1. SIGN) receptor (Appelmelk et al., 2003). The bacteria are able The strains were incubated under microaerophilic condi- to escape this binding by switching off the Lewis antigen tions (5% O , 10% CO and 85% N ) for 24–72 h at 37 1C expression through phase variation and may thereby suppress 2 2 2 on Columbia II agar base BBL plates (Becton-Dickinson, the TLR-mediated immune response through DC-SIGN Cockeysville, MD) containing 8.5% horse blood and 10% (Bergman et al., 2004). horse serum. For liquid culture, bacteria were grown to In this study, we investigated the DC response to different exponential phase in Brucella broth (Becton-Dickinson, H. pylori strains by evaluating their capacity to impact Cockeysville, MD) supplemented with 5% foetal bovine upon maturation and cytokine release from the cells. The serum (Sigma, St. Louis, MO) and 1% IsoVitaleX (Becton- analysis was based on a collection of H. pylori strains Dickinson, Cockeysville, MD) pH 7.0. sampled from the Swedish adult general population, as part of the Kalixanda study (Aro et al., 2004, 2006). These strains come from patients with different levels of inflammation. To PCR analyses measure the IL-8 response and to detect translocation of CagA, the strains were exposed to human epithelial adeno- DNA for PCR analyses was prepared using the Dneasys carcinoma cells. Based on the comparison of H. pylori tissue kit (Qiagen, Hilden, Germany). Amplification of H. strains from biopsies with low and high lymphocyte/granu- pylori genes was performed for cagA, vacA [s (signal) and m locyte infiltration, no association was found with DC (middle) regions] and babA. An empty site PCR, using maturation or the induction of IL-8 release by adenocarci- primers targeting the flanking regions of the cag PAI, was noma cells. In contrast, the inflammatory state of the used to distinguish an intact or partial cag PAI from an biopsies from which the H. pylori strain was isolated absent cag PAI. All reactions have been described previously correlated with the cytokines released following DC infec- (Atherton et al., 1999; Gerhard et al., 1999; Nilsson et al., tion with these strains. 2003; Lehours et al., 2004).

Table 1. Characteristics of the Helicobacter pylori strains used in this study Inﬁltration of Cag AGS

Strains Age/sex Lc Gc Hp A PAI Translocation IL-8 vacA babA Low HPKX_208 64/M 1 1 3 11 1 1 s1m2 1 HPKX_422 58/F 1 1 3 À 1 ÀÀs1m1 1 HPKX_1213 63/F 1 1 3 11 1 1 s1m2 1 HPKX_1222 54/M 1 1 3 11 1 1 s1m1 À HPKX_1226 51/F 1 1 3 ÀÀ À À s1m2 À HPKX_1311 71/M 1 1 3 11 1 1 s1m2 À HPKX_1343 46/F 1 1 3 11 1 1 s1m2 1 Medium HPKX_271 57/F 2 1 1 11 ÀÀs1m1 À HPKX_476 69/F 2 1 1 11 1 1 s1m1 1 HPKX_627 56/F 2 1 1 11 1 1 s1m1 1 HPKX_1200 61/F 2 2 1 11 1 1 s1m1 1 HPKX_1215 61/F 2 2 1 ÀÀ À À s2m2 À HPKX_1363 54/F 2 1 1 11 1 1 s1m2 À HPKX_1377 62/M 2 1 1 11 1 1 s1m1 1 High HPKX_439 52/F 2 3 1 11 ÀÀs1m2 1 HPKX_519 57/M 3 3 1 11 1 1 s1m1 1 HPKX_1008 75/F 2 3 1 11 1 1 s1m2 1 HPKX_1116 73/F 3 3 1 11 1 1 s1m1 À HPKX_1165 64/F 2 3 1 11 1 1 s1m1 À HPKX_1191 52/M 2 3 0 11 1 1 s1m1 1 C HP_67:20 90/F ÀÀ À À s1m2 1 HP_67:21 90/F 11 1 1 s1m2 1

The following strains were selected based on the grade of inﬁltration by lymphoplasmatic (Lc) and granulocytic cells (Gc) and the density of H. pylori (Hp) at the side of infection (0 = none, 1 = slight, 2 = moderate, 3 = high, according to the recommendations of the updated Sydney System). HP67:20 and HP67:21 were used as negative and positive controls, respectively. Strains were characterized according to the presence of the virulence factors cagA, cag PAI, vacA (s and m alleles) and babA, and their ability to induce IL-8 secretion (HP67:21 = IL-8 4 100 pg mLÀ1; HP67:20 = IL-8 o 100 pg mLÀ1)and translocate CagA in vitro. AGS, adenocarcinoma.

In vitro IL-8 induction and CagA translocation fluoride membranes (Bio-Rad). Membranes were probed with anti-CagA (A10) (Santa Cruz Biotechnology, Santa Adenocarcinoma cells (ATCC CRL-1739) were routinely Cruz) or anti-H. pylori Cag antigen immunoglobulin G maintained in RMPI 1640 supplemented with glutamine, fraction (polyclonal) (Austral Biologicals, San Ramon). HEPES and 10% foetal calf serum (FCS) (Gibco, Invitrogen, Sheep anti-mouse-conjugated horseradish peroxidase Paisley, UK), and cultured at 37 1Cin5%CO. Cells seeded 2 (HRP) and goat anti-rabbit-conjugated HRP immunoglo- on to 24-well tissue culture plates for IL-8 induction and 12- bulins (Amersham International, Amersham, UK) were well tissue culture plates for CagA translocation assays were used as secondary antibodies. Detection of the antibodies cultured to 80–90% confluence. The OD of overnight was performed using the ECLTM Eastern Blotting Detection H. pylori liquid cultures was measured at 600 nm (OD 8 Reagents Kit (Amersham International) according to the 0.5 = 1 Â 10 bacteria) and adjusted to a multiplicity of manufacturer’s instructions. infection (MOI) of approximately for 25 for IL-8 induction Cytokine concentrations were analysed by enzyme-linked and CagA translocation, respectively. After addition of H. immunosorbent assay (ELISA) using BD OptEIA: Human pylori strains in cell culture media, the plates were centri- IL-8 ELISA set (Becton-Dickinson, San Diego, CA). fuged at 112 g for 5 min to synchronize infection and then co-incubated for 6 h at 37 1C in a 10% CO incubator. 2 Infection of immature DCs and analysis of DC To detect translocated CagA within the adenocarcinoma maturation and activation cells following co-incubation, proteins were extracted using RIPA buffer containing a Protease Inhibitor Cocktail and a Monocyte isolation, generation and culturing of iDCs were Phosphatase Inhibitor Cocktail 2 (Sigma). Samples were performed as described by Littmann et al. (2009). Briefly, separated by sodium dodecyl sulphate-polyacrylamide gel peripheral blood was obtained as standard buffy coat from electrophoresis on precast 7.5% polyacrylamide gels (Bio- anonymous healthy donors at the Karolinska University Rad, Hercules) and electrotransferred onto polyvinylidene Hospital. Monocytes were purified using Ficoll–Hypaque

FEMS Immunol Med Microbiol ]] (2010) 1–11 c 2010 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved 4 S. Andres et al. density separation using the RosetteSepTM human mono- strains possesed s1m1-, 40% (8/20) s1m2- and 5% (1/20) cyte enrichment cocktail (Stemcell Technologies). DCs were s2m2-vacA genotypes, respectively. Sixty percent (12/20) of derived in granulocyte-macrophage colony-stimulating fac- strains were positive for babA. The presence of virulence tor (GM-CSF) and IL-4-enriched medium, and their phe- factors were distributed between the low-, medium- notype was assessed by CD1a and CD11c expression. Cells at and high-level inflammatory potential groups as shown in a concentration of 1 Â 106 mLÀ1 were incubated with H. Table 1. pylori at an MOI of 10 for 24 h in RPMI 1640 supplemented with 10% heat-inactivated FCS and 2 mM L-glutamine (Invitrogen). DCs were fixed in 2% paraformaldehyde IL-8 production after in vitro adenocarcinoma before analysis on a FACSCalibur (Becton-Dickinson, San stimulation Diego, CA) using FLOW JO (Tree Star Inc.) and CELLQUEST (BD Biosciences). The activation of infected cells was determined The ability of the clinical isolates to induce IL-8 secretion by by examination of the mean fluorescence intensity (MFI) of adenocarcinoma cells in vitro was used as a surrogate marker surface maturation makers using fluorescein isothiocyanate- for the strain’s inflammatory potential. The presence of a conjugated anti-human CD80 and PE-Cy5-conjugated anti- functional cag PAI was determined through the detection of human CD86 (both BD Pharmingen). The cytokines IL-8, translocated CagA into the adenocarcinoma cells in an IL-1b, IL-6, IL-10, TNF-a and IL-12p70 in the supernatant immunoblot assay. Of the 17 strains that were positive for of the 24-h DC/H. pylori co-incubation were measured cagA and possessed a complete or partial cag PAI, 15 were simultaneously by Cytometric Bead Array (CBA) using the able to translocate CagA into the epithelial cells. For the Human Inflammation Kit (according to the manufacturer’s other two strains and the three cagA-negative strains, no instructions; BD Biosciences). CagA could be detected in the immunoblot analysis (Sup- porting Information, Fig. S1b). The average IL-8 released after 6 h of co-incubation was Statistical analyses 474 pg mLÀ1 with levels ranging from 21 to 619 pg mLÀ1 To determine the level of IL-8 secretion in the cell culture (Fig. S1a). A significant increase in IL-8 secretion was supernatant, at least three independent experiments were observed when comparing adenocarcinoma cells infected performed in duplicate. Results from the translocation assay with strains with and without a functional cag PAI were obtained from at least two independent experiments. (P o 0.001). The secretion of IL-8 induced by the CagA The data shown for DC activation and maturation were translocation-negative H. pylori differed marginally from obtained from four to five independent experiments using uninfected adenocarcinoma cells (P = 0.115). The presence DCs from different donors, with at least two infections each. of babA or different vacA variants was not associated with The Mann–Whitney and the Kruskal–Wallis tests were used differences in the level of IL-8 secretion (data not shown). for statistical analysis. The analysis was performed using Furthermore, when strains were grouped according to PASW for Windows v17 (IPSS Inc., Chicago, IL) graphs were inflammatory potential, no difference was detected in the s s made using ORIGIN 8 (OriginLab , Northampton, MA). level of IL-8 secretion by adenocarcinoma cells (Fig. 1). For the IL-8 induction experiments, the identical supernatant of positive and negative controls HP67:20 and HP67:21 were used. The data were normalized to Induction of maturation of iDCs by H. pylori À1 HP67:21 = 650 pg mL . iDCs were co-incubated with 22 clinical H. pylori strains at an MOI of 10 for 24 h. A significant increase in MFI or the maturation markers CD80 and CD86 was detected after Results stimulation with all H. pylori strains when compared with cells in culture media without stimuli. The average level of Characterization of the 20 clinical H. pylori maturation was comparable with that induced by E. coli strains lipopolysaccharides, ranging from 77% to 97% positive for CD80 and 85% to 102% positive for CD86, when the data Virulence factors were normalized to lipopolysaccharides stimulation, which PCR was used to characterize the H. pylori strains according was set at 100% (Fig. S2). No significant differences between to the presence of the major virulence factors cagA, the cag the strains were identified with respect to the presence of the PAI, vacA (s and m region) and babA. Ninety percent (18/ major H. pylori virulence factors (data not shown) or 20) of the strains possessed either a complete or a partial cag lymphocyte/granulocyte infiltration in the biopsies from PAI as indicated by a positive empty side PCR. Of these, 17 which the bacteria were isolated (the inflammatory potential strains were cagA positive. Fifty-five percent (11/20) of the of the bacteria) (Fig. 2).

No significant difference was observed in cytokine release when comparing strains with and without cagA or the cag PAI. Neither was a significant difference seen when comparing vacA s1m1 and s1m2 strains, although the two groups tended to differ with regard to the induction of IL-1b (P o 0.1). A trend was also observed for IL-12 when comparing strains with a functional and a nonfunctional or no cag PAI and for TNF-a when comparing babA-positive and -negative strains. Significant differences in the release of IL-12, TNF-a, IL-6 (each P o 0.05) and IL-1b (P o 0.01) were detected when strains from biopsies with a low inflammatory background and those with higher levels were compared. Furthermore, the level of cytokines secreted by strains with a low inflammatory potential corresponding to Fig. 1. IL-8 release by adenocarcinoma cells after Helicobacter pylori a high bacterial density at the site from which the biopsy was co-infection is dependent on cagA translocation but does not correlate taken and those with a low bacterial density differed with the histology background of the strains. IL-8 levels were measured significantly [IL-1b (P o 0.05)] or on a descriptive level after 6 h of adenocarcinoma–H. pylori co-incubation. Four independent [IL-12, TNF-a, IL-10, IL-6 (P o 0.1)] (Fig. 3a). experiments were performed, each in duplicate. To allow comparison between experiments, data were normalized to the positive control HP67:21 a cag PAI-positive strain; HP67:20 is the cag PAI-negative variant of the strain (solid black columns). Results represent the average Discussion for each group of strains. Striped columns show the comparison of strains with low, medium and high inflammatory potential (Low, Med- In the current study, we evaluated 20 H. pylori strains with ium and High, respectively) and solid white columns show the compar- different inflammatory backgrounds for their effect on the ison of strains that did or did not translocate CagA (TL1 and TL À , Ã ÃÃ induction of inflammation by epithelial cells (adenocarci- respectively). The error bars indicate SEs. P o 0.05, P o 0.01, ÃÃÃ noma) as well as activation and maturation of monocyte- P o 0.001 as compared between groups or between noninfected and H. pylori-infected adenocarcinoma cells. The results of IL-8 secretion derived DCs. The majority of strains investigated (18/20) and immunoblot analysis for each strain are depicted in Fig. S1. were shown to be of the vacA s1 genotype and were cag PAI positive, i.e. were high virulent strains, consistent with previous studies (Yamaoka et al., 1999; Olfat et al., 2005). The adenocarcinoma cell response, which was determined Stimulation of cytokine production matches by IL-8 induction after co-infection with H. pylori, could be strain origin clearly linked to the capacity of the strains to translocate the To assess the impact of the strains on DC activation, the virulence factor cagA into epithelial cells. Therefore, we supernatants were collected after co-incubation with H. pylori concluded that the cag PAI was functional in 15 strains as strains, and the concentrations of the proinflammatory they were both translocation positive and associated with a cytokines IL-12p70, TNF-a, IL-6 and IL-1b, the anti-inflam- high level of IL-8 induction. Four additional strains were cag matory cytokine IL-10 and the chemokine IL-8 were mea- PAI positive by PCR but did not possess a functional cag PAI sured by CBA. The level of cytokines produced by the DCs by our in vitro analysis. Although IL-8 induction by cag PAI- varied greatly between the donors from which they were negative strains has been reported previously (Yamaoka derived (Table S1). The mean stimulation by the different H. et al., 2000; Audibert et al., 2001), we did not detect this pylori strains was comparable with E. coli lipopolysaccharides effect in the current study; however, our small sample size for TNF-a, IL-10, IL-6, IL-1b and IL-8 but not for IL-12, may have had an influence on our ability to detect this trait. where no or small amounts of cytokine were detected after The histological background of the strains tested could not stimulation with E. coli lipopolysaccharides (Table S1). The be linked to a functional cag PAI or any of the examined strains were found to induce an altered release of cytokines virulence factors. with commonly donor-specific and cytokine-specific differ- To study H. pylori–DC interaction, DCs were generated ences being observed. Using DCs from four unrelated donors, from monocytes in culture medium containing GM-CSF we failed to find a combination of cytokines, virulence factors and IL-4. The phenotype of gastric mucosal DCs is affected or inflammatory background that consistently yielded a by the natural components of the mucosa. It would have significant correlation (Fig. 3b). Our data did suggest, how- been interesting to study the effects of H. pylori on primary ever, that certain H. pylori strains were more effective in gastric DCs purified from the stomach. This could not, activating DC’s than other strains (data not shown). however, be accomplished in this study due to the laborious

Fig. 2. Helicobacter pylori strains have a consistently high impact on the maturation of iDCs. (a) iDCs were generated with GM-CSF and IL-4. The phenotype was assessed before stimulation by examination of CD11c and CD1a. The figure depicts the stimulation of DC with media, Escherichia coli lipopolysaccharides (LPS) and two strains that differed in their impact on maturation. (b) CD80 and CD86 expression of matured DCs is shown as percent positive after normalization to E. coli lipopolysaccharides (LPS) (40 ng mLÀ1, 100%). The data depict the results from five to six independent experiments where iDCs generated from different donors for 24 h after co-infection with H. pylori at a MOI of 10. ‘Low, Medium and High’ represent the averaged maturation induced by strains grouped as having a low, medium and high inflammatory potential (see Table 1). The combined results for Ã ÃÃ ÃÃÃ all stimulations are shown in Fig. S2. P o 0.05, P o 0.01, P o 0.001, for unstimulated DCs vs. stimulated DCs.

isolation protocols required and the amount of cells re- studies, we chose 20 H. pylori strains from tissues histologi- quired for the study. cally defined as having high-, medium- or low-grade in- Recently, Necchi et al. (2009) described a direct contact flammation. The number of strains investigated is higher between DCs and H. pylori in inflamed but not normal than in earlier studies in which at most eight strains were mucosa. The interaction of monocyte-derived DCs and H. included based on either disease outcome or bacterial pylori was characterized here in vitro. In contrast to previous genotype (Kranzer et al., 2005; Hansson et al., 2006). A

Fig. 3. DC cytokine profiles correlate with the histology background of Helicobacter pylori strains. (a) The average cytokine secretion in response to the inflammatory potential of H. pylori, the density of H. pylori in the biopsy taken, CagA translocation by H. pylori into adenocarcinoma cells and vacA genotype. Cytokines released into the supernatant by DCs were measured by CBA after 24 h of coculture. Each data point reflects the average DC cytokine expression from three independent experiments in response to one H. pylori strain. Differences between IL-12, TNF-a, IL-10, IL-6 and IL-1b secretion by DCs in response to H. pylori strains with differing inflammatory potentials, IL-12 secretion towards CagA translocating and not translocating strains and IL-1b secretion towards H. pylori vacA s1m1 and s1m2 strains are depicted. No difference could be detected in IL-8 secretion or the levels of TNF-a, IL-10, IL-6 and IL-1b with regard to CagA translocation, or IL-12, TNF-a, IL-10 and IL-6 with regard to the vacA genotype (data not shown). Ã ÃÃ #P o 0.1, P o 0.05, P o 0.01. (b) Differences in cytokine release were detected between the four donors. The table depicts differences in the DC response towards H. pylori (Hp) with different inflammatory potentials, different capacities to translocate CagA and different vacA genotypes (1 =1of 4, 11 = 2 of 4, 111 = 3 of 4).

c 2010 Federation of European Microbiological Societies FEMS Immunol Med Microbiol ]] (2010) 1–11 Published by Blackwell Publishing Ltd. All rights reserved DC role in H. pylori infection 7 potential limitation of these types of studies is that although 2001; Sipponen, 2001). Of the strains chosen for this study, the specificity of histology is high, the sensitivity can vary the seven strains with high inflammatory cell infiltration due to the uneven distribution of H. pylori (Leodolter et al., and low bacterial density and the six strains with low

(a)

(b) IL-12TNF-α IL-10 IL-6 IL-1β IL-8 Inﬂammatory potential + ++ ++ +++ ++ + Hp density + ++ + + + cag PAI translocation +++ + + + ++ vacA +

FEMS Immunol Med Microbiol ]] (2010) 1–11 c 2010 Federation of European Microbiological Societies Published by Blackwell Publishing Ltd. All rights reserved 8 S. Andres et al. inflammatory cell infiltration and high bacterial density (Th1) phenotype (Wang et al., 2001). An IL-12-dependent were the most extreme strains, with regard to these histolo- Th1 response is required for vaccine-induced protection gical properties of the 336 available from the Kalixanda against the bacterium (Akhiani et al., 2002). We report here Study. We are, therefore, confident that the inflammatory that strains with a functional cag PAI induce a lower IL-12 potential of these strains is different between the different response in DCs but have no appreciable effect on the groups. secretion of other cytokines. This effect on IL-12 secretion It has previously been shown that the presence of H. pylori seems biologically plausible given that the cag PAI prevents stimulates human DC maturation (Hafsi et al., 2004; efficient uptake and killing by professional phagocytes Kranzer et al., 2004). This induction is independent (Allen et al., 2000; Ramarao et al., 2000) and that maximal of the vacA genotype and the presence of the cag PAI production of IL-12 towards H. pylori requires phagocytosis (Kranzer et al., 2005), as we have confirmed here. Further- of intact bacterial cells (Hafsi et al., 2004; Kranzer et al., more, we did not find that the babA status of the 2005). The production of cytokines, including IL-12, by strains played a role in the process. Although the strains monocytic cells has been shown previously to be cag PAI and had different potentials for stimulating DC maturation, VacA independent (de Jonge et al., 2001; Kranzer et al., this did not correlate with the inflammatory status of 2005). This is in contrast to the results published by the host. Conversely, a correlation could be detected for Hansson and colleagues who in a study of only four strains the average activation profiles of the stimulated cells. A described that a vacA s2m2, cag PAI-negative strain to significant increase in the production of the proinflamma- induce less DC migration, less cytokine response and less tory cytokines IL-12, TNF-a, IL-6 and IL-1b was shown T-cell migration than vacA s1, cag PAI-positive strains. It is between the groups of strains from biopsies with low, likely that factors other than VacA, cag PAI or BabA play a medium or high inflammatory cell infiltration. In parallel role in DC activation. Indeed, Kao et al. (2006) have H. pylori strains isolated from biopsies with a high bacterial reported that the level of IL-12 released by DCs was density induced a lower secretion of the same cytokines significantly lower when stimulated by H. pylori as com- as well as IL-10. It is tempting to speculate that these pared with another gastritis-causing pathogen Acinetobacter strains avoid a high production of proinflammatory lwoffii and proposed that an as yet undescribed secreted cytokines to allow them to flourish in a less hostile environ- factor caused inhibition of the response rather than H. pylori ment. It is important to point out, however, that the being weakly immunogenic. Voland et al. (2003) described detected differences were not consistent for all tested do- recognition of the outer membrane proteins HpaA and nors. It is possible that strains with high inflammatory Omp18 by human DCs followed by maturation, as well as potential require a susceptible host for this potential to be antigen presentation. Another recently described mechan- fully realized. Thus, the contribution of host genetic predis- ism of IL-12 and IL-6 downregulation in parallel with positions is likely to play an important role in H. pylori- enhanced IL-10 expression has been reported to be due to induced diseases. Indeed, host genetic polymorphisms in the the active control of the DC-SIGN signalosome by fucose- acid inhibitor IL-1b have been linked to an increased risk of expressing pathogens like H. pylori (Gringhuis et al., 2009). gastric cancer and its precursors (El-Omar et al., 2000). This While the mechanism behind the DC response to H. pylori association could not, however, be confirmed in a recent infection is still debated, H. pylori lipopolysaccharides seem study, where patients from a Swedish case-control study to play a crucial role, despite low biological activity as were analysed (Persson et al., 2009). Other polymorphisms compared with lipopolysaccharides from other bacteria such as those in TNF-a and IL-10, have also been identified such as E. coli or Salmonella spp. and the reported antag- as risk factors for noncardia gastric cancer (El-Omar et al., onizing effect of the lipopolysaccharides of certain H. pylori 2003). strains towards TLR4 signalling (Birkholz et al., 1993; The proinflammatory cytokines TNF-a, IL-6, IL-1b and Lepper et al., 2005). IL-8 are secreted in the gastric mucosa during H. pylori A generally accepted model of gastric carcinogenesis infection (Lindholm et al., 1998; Moss et al., 1994). The reports a cascade of events: chronic active gastritis, multi- observation made here, based on the cytokines secreted by focal atrophy, intestinal metaplasia, dysplasia and invasive DCs, support the proposition that monocytes play an carcinoma (Correa, 1992). Chronic gastritis as a first step is important role in the determination of cytokine levels in characterized by the infiltration of leucocytes to the site of the gastric mucosa. This outcome could be triggered by infection in response to an accumulation of proinflamma- characteristics of the infecting H. pylori; however, it is tory cytokines. Polymorphonuclear infiltration weakens unlikely that they are the putative virulence factors analysed within weeks and the lymphocytic infiltration within a in the present study. few months of successful eradication of H. pylori (Dixon Evaluation of the H. pylori-specific immune response in et al., 1996). In the present study, we have shown that the gastric mucosa has been shown to be of a T helper cell 1 strains isolated from regions with high lymphocyte and

c 2010 Federation of European Microbiological Societies FEMS Immunol Med Microbiol ]] (2010) 1–11 Published by Blackwell Publishing Ltd. All rights reserved DC role in H. pylori infection 9 granulocytic infiltration have a higher impact on monocyte- pathogenicity island in induction of interleukin-8 secretion. derived DC activation in vitro. We propose that H. pylori Infect Immun 69: 1625–1629. defines the local immune response in susceptible hosts by Backhed F, Torstensson E, Seguin D, Richter-Dahlfors A & Rokbi interacting with DCs in a strain-specific manner. This B (2003) Helicobacter pylori infection induces interleukin-8 interaction can lead to either mild infection or to disease receptor expression in the human gastric epithelium. Infect development. Immun 71: 3357–3360. Banchereau J, Briere F, Caux C, Davoust J, Lebecque S, Liu YJ, Pulendran B & Palucka K (2000) Immunobiology of dendritic Acknowledgements cells. Annu Rev Immunol 18: 767–811. Bergman MP, Engering A, Smits HH, van Vliet SJ, van This research project has been supported by a Marie Curie Bodegraven AA, Wirth HP, Kapsenberg ML, Vandenbroucke- Early Stage Research Training Fellowship of the European Grauls CM, van Kooyk Y & Appelmelk BJ (2004) Helicobacter Community’s Sixth Framework Programme under contract pylori modulates the T helper cell 1/T helper cell 2 balance number MEST-CT-2004-8475 (S.A.), the Swedish Research through phase-variable interaction between Council and the Swedish Cancer Society. We thank Lars lipopolysaccharide and DC-SIGN. J Exp Med 200: 979–990. Agreus, Tom Storskrubb, Jukka Ronkainen, Pertti Aro and Birkholz S, Knipp U, Nietzki C, Adamek RJ & Opferkuch W Nicholas J. Talley for providing the Kalixanda strains, Lena (1993) Immunological activity of lipopolysaccharide of Eriksson and Kristina Schonmeyr¨ for technical assistance Helicobacter pylori on human peripheral mononuclear blood and Laura Plant for the introduction to in vitro techniques. cells in comparison to lipopolysaccharides of other intestinal bacteria. FEMS Immunol Med Mic 6: 317–324. Blaser MJ & Atherton JC (2004) Helicobacter pylori persistence: References biology and disease. J Clin Invest 113: 321–333. Censini S, Lange C, Xiang Z, Crabtree JE, Ghiara P, Borodovsky Akhiani AA, Pappo J, Kabok Z, Schon K, Gao W, Franzen LE & M, Rappuoli R & Covacci A (1996) cag, a pathogenicity island Lycke N (2002) Protection against Helicobacter pylori infection of Helicobacter pylori, encodes type I-specific and disease- following immunization is IL-12-dependent and mediated by associated virulence factors. P Natl Acad Sci USA 93: Th1 cells. J Immunol 169: 6977–6984. 14648–14653. Allen LA, Schlesinger LS & Kang B (2000) Virulent strains of Correa P (1992) Human gastric carcinogenesis: a multistep and Helicobacter pylori demonstrate delayed phagocytosis and stimulate homotypic phagosome fusion in macrophages. J Exp multifactorial process – First American Cancer Society Award Med 191: 115–128. Lecture on Cancer Epidemiology and Prevention. Cancer Res Amieva MR, Vogelmann R, Covacci A, Tompkins LS, Nelson WJ 52: 6735–6740. & Falkow S (2003) Disruption of the epithelial Crabtree JE (1996) Gastric mucosal inflammatory responses to apical–junctional complex by Helicobacter pylori CagA. Science Helicobacter pylori. Aliment Pharm Therap 10 (suppl 1): 29–37. 300: 1430–1434. de Jonge R, Kusters JG, Timmer MS et al. (2001) The role of Appelmelk BJ, van Die I, van Vliet SJ, Vandenbroucke-Grauls Helicobacter pylori virulence factors in interleukin production CM, Geijtenbeek TB & van Kooyk Y (2003) Cutting edge: by monocytic cells. FEMS Microbiol Lett 196: 235–238. carbohydrate profiling identifies new pathogens that interact Dixon MF, Genta RM, Yardley JH & Correa P (1996) with dendritic cell-specific ICAM-3-grabbing nonintegrin on Classification and grading of gastritis. The updated Sydney dendritic cells. J Immunol 170: 1635–1639. System. International Workshop on the Histopathology of Aro P, Ronkainen J, Storskrubb T et al. (2004) Valid symptom Gastritis, Houston 1994. Am J Surg Pathol 20: 1161–1181. reporting at upper endoscopy in a random sample of the El-Omar EM, Carrington M, Chow WH et al. (2000) Interleukin- Swedish adult general population: the Kalixanda study. Scand J 1 polymorphisms associated with increased risk of gastric Gastroentero 39: 1280–1288. cancer. Nature 404: 398–402. Aro P, Storskrubb T, Ronkainen J, Bolling-Sternevald E, El-Omar EM, Rabkin CS, Gammon MD et al. (2003) Increased Engstrand L, Vieth M, Stolte M, Talley NJ & Agreus L (2006) risk of noncardia gastric cancer associated with Peptic ulcer disease in a general adult population: the proinflammatory cytokine gene polymorphisms. Kalixanda study: a random population-based study. Am J Epidemiol 163: 1025–1034. Gastroenterology 124: 1193–1201. Atherton JC, Cover TL, Twells RJ, Morales MR, Hawkey CJ & Enroth H, Kraaz W, Engstrand L, Nyren O & Rohan T (2000) Blaser MJ (1999) Simple and accurate PCR-based system for Helicobacter pylori strain types and risk of gastric cancer: a typing vacuolating cytotoxin alleles of Helicobacter pylori. case-control study. Cancer Epidem Biomar 9: 981–985. J Clin Microbiol 37: 2979–2982. Ernst PB & Gold BD (2000) The disease spectrum of Helicobacter Audibert C, Burucoa C, Janvier B & Fauchere JL (2001) pylori: the immunopathogenesis of gastroduodenal ulcer and Implication of the structure of the Helicobacter pylori cag gastric cancer. Annu Rev Microbiol 54: 615–640.

Gebert B, Fischer W, Weiss E, Hoffmann R & Haas R (2003) Lindholm C, Quiding-Jarbrink M, Lonroth H, Hamlet A & Helicobacter pylori vacuolating cytotoxin inhibits T Svennerholm AM (1998) Local cytokine response in lymphocyte activation. Science 301: 1099–1102. Helicobacter pylori-infected subjects. Infect Immun 66: Gerhard M, Lehn N, Neumayer N, Boren T, Rad R, Schepp W, 5964–5971. Miehlke S, Classen M & Prinz C (1999) Clinical relevance of Littmann M, Albiger B, Frentzen A, Normark S, Henriques- the Helicobacter pylori gene for blood-group antigen-binding Normark B & Plant L (2009) Streptococcus pneumoniae evades adhesin. P Natl Acad Sci USA 96: 12778–12783. human dendritic cell surveillance by pneumolysin expression. Gringhuis SI, den Dunnen J, Litjens M, van der Vlist M & EMBO Mol Med 1: 211–222. Geijtenbeek TB (2009) Carbohydrate-specific signaling Moss SF, Legon S, Davies J & Calam J (1994) Cytokine gene through the DC-SIGN signalosome tailors immunity to expression in Helicobacter pylori associated antral gastritis. Gut Mycobacterium tuberculosis, HIV-1 and Helicobacter pylori. 35: 1567–1570. Nat Immunol 10: 1081–1088. Necchi V, Manca R, Ricci V & Solcia E (2009) Evidence for Guermonprez P, Valladeau J, Zitvogel L, Thery C & Amigorena S transepithelial dendritic cells in human H. pylori active (2002) Antigen presentation and T cell stimulation by gastritis. Helicobacter 14: 208–222. dendritic cells. Annu Rev Immunol 20: 621–667. Niess JH, Brand S, Gu X et al. (2005) CX3CR1-mediated Hafsi N, Voland P, Schwendy S, Rad R, Reindl W, Gerhard M & dendritic cell access to the intestinal lumen and bacterial Prinz C (2004) Human dendritic cells respond to Helicobacter clearance. Science 307: 254–258. pylori, promoting NK cell and Th1-effector responses in vitro. Nilsson C, Sillen A, Eriksson L, Strand ML, Enroth H, Normark S, J Immunol 173: 1249–1257. Falk P & Engstrand L (2003) Correlation between cag Hansson M, Lundgren A, Elgbratt K, Quiding-Jarbrink M, pathogenicity island composition and Helicobacter pylori- Svennerholm AM & Johansson EL (2006) Dendritic cells associated gastroduodenal disease. Infect Immun 71: express CCR7 and migrate in response to CCL19 (MIP-3beta) 6573–6581. after exposure to Helicobacter pylori. Microbes Infect 8: Olfat FO, Zheng Q, Oleastro M, Voland P, Boren T, Karttunen R, 841–850. Engstrand L, Rad R, Prinz C & Gerhard M (2005) Correlation Ilver D, Arnqvist A, Ogren J, Frick IM, Kersulyte D, Incecik ET, of the Helicobacter pylori adherence factor BabA with duodenal Berg DE, Covacci A, Engstrand L & Boren T (1998) ulcer disease in four European countries. FEMS Immunol Med Helicobacter pylori adhesin binding fucosylated histo-blood Mic 44: 151–156. group antigens revealed by retagging. Science 279: 373–377. Persson C, Engstrand L, Nyren O, Hansson LE, Enroth H, Kao JY, Rathinavelu S, Eaton KA, Bai L, Zavros Y, Takami M, Ekstrom AM & Ye W (2009) Interleukin 1-beta gene Pierzchala A & Merchant JL (2006) Helicobacter pylori-secreted polymorphisms and risk of gastric cancer in Sweden. Scand J factors inhibit dendritic cell IL-12 secretion: a mechanism of Gastroentero 44: 339–345. ineffective host defense. Am J Physiol 291: G73–G81. Rad R, Brenner L, Krug A et al. (2007) Toll-like receptor- Kranzer K, Eckhardt A, Aigner M, Knoll G, Deml L, Speth C, dependent activation of antigen-presenting cells affects Lehn N, Rehli M & Schneider-Brachert W (2004) Induction of adaptive immunity to Helicobacter pylori. Gastroenterology maturation and cytokine release of human dendritic cells by 133: 150–163.e3. Helicobacter pylori. Infect Immun 72: 4416–4423. Rad R, Ballhorn W, Voland P et al. (2009) Extracellular and Kranzer K, Sollner L, Aigner M, Lehn N, Deml L, Rehli M & intracellular pattern recognition receptors cooperate in the Schneider-Brachert W (2005) Impact of Helicobacter pylori recognition of Helicobacter pylori. Gastroenterology 136: virulence factors and compounds on activation and 2247–2257. maturation of human dendritic cells. Infect Immun 73: Ramarao N, Gray-Owen SD, Backert S & Meyer TF (2000) 4180–4189. Helicobacter pylori inhibits phagocytosis by professional Lehours P, Menard A, Dupouy S, Bergey B, Richy F, Zerbib F, phagocytes involving type IV secretion components. Mol Ruskone-Fourmestraux A, Delchier JC & Megraud F (2004) Microbiol 37: 1389–1404. Evaluation of the association of nine Helicobacter pylori Sabatte J, Maggini J, Nahmod K, Amaral MM, Martinez D, virulence factors with strains involved in low-grade gastric Salamone G, Ceballos A, Giordano M, Vermeulen M & Geffner mucosa-associated lymphoid tissue lymphoma. Infect Immun J (2007) Interplay of pathogens, cytokines and other stress 72: 880–888. signals in the regulation of dendritic cell function. Cytokine Leodolter A, Wolle K & Malfertheiner P (2001) Current standards Growth F R 18: 5–17. in the diagnosis of Helicobacter pylori infection. Digest Dis 19: Sipponen P (2001) Update on the pathologic approach to the 116–122. diagnosis of gastritis, gastric atrophy, and Helicobacter pylori Lepper PM, Triantafilou M, Schumann C, Schneider EM & and its sequelae. J Clin Gastroenterol 32: 196–202. Triantafilou K (2005) Lipopolysaccharides from Helicobacter Sozzani S, Allavena P, D’Amico G, Luini W, Bianchi G, Kataura pylori can act as antagonists for Toll-like receptor 4. Cell M, Imai T, Yoshie O, Bonecchi R & Mantovani A (1998) Microbiol 7: 519–528. Differential regulation of chemokine receptors during

dendritic cell maturation: a model for their trafficking Yamaoka Y, Kwon DH & Graham DY (2000) A M(r) 34,000 properties. J Immunol 161: 1083–1086. proinflammatory outer membrane protein (oipA) of Storskrubb T, Aro P, Ronkainen J et al. (2005) A negative Helicobacter pylori. P Natl Acad Sci USA 97: 7533–7538. Helicobacter pylori serology test is more reliable for exclusion Zarrilli R, Ricci V & Romano M (1999) Molecular response of of premalignant gastric conditions than a negative test for gastric epithelial cells to Helicobacter pylori-induced cell current H. pylori infection: a report on histology and H. pylori damage. Cell Microbiol 1: 93–99. detection in the general adult population. Scand J Gastroentero 40: 302–311. Suerbaum S & Michetti P (2002) Helicobacter pylori infection. New Engl J Med 347: 1175–1186. Supporting Information Tummuru MK, Sharma SA & Blaser MJ (1995) Helicobacter pylori Additional Supporting Information may be found in the picB, a homologue of the Bordetella pertussis toxin secretion online version of this article: protein, is required for induction of IL-8 in gastric epithelial cells. Mol Microbiol 18: 867–876. Fig. S1. Adenocarcinoma IL-8 induction and CagA translo- Voland P, Hafsi N, Zeitner M, Laforsch S, Wagner H & Prinz C cation after Helicobacter pylori infection. (2003) Antigenic properties of HpaA and Omp18, two outer Fig. S2. Helicobacter pylori strain-specific impact on matura- membrane proteins of Helicobacter pylori. Infect Immun 71: tion of iDC. 3837–3843. Wang J, Brooks EG, Bamford KB, Denning TL, Pappo J & Ernst Table S1. Cytokine release of monocyte-derived DCs after PB (2001) Negative selection of T cells by Helicobacter pylori as Helicobacter pylori co-infection. a model for bacterial strain selection by immune evasion. J Immunol 167: 926–934. Please note: Wiley-Blackwell is not responsible for the Yamaoka Y, Kodama T, Gutierrez O, Kim JG, Kashima K & content or functionality of any supporting materials sup- Graham DY (1999) Relationship between Helicobacter pylori plied by the authors. Any queries (other than missing iceA, cagA, and vacA status and clinical outcome: studies in material) should be directed to the corresponding author four different countries. J Clin Microbiol 37: 2274–2279. for the article.

107 SCHMIDT ET AL. SNPs in H. pylori-related disease

Variation in human genetic polymorphisms, their association with H pylori acquisition and gastric cancer in a multi-ethnic country

Heather-Marie A Schmidt a, Dung Mai Ha a, Faye Taylor b, Zsuzsanna Kovach a, Khean-Lee Goh c, Kwong Ming Fock d, Jennifer H. Barret b, David Forman b & Hazel Mitchell a*

a School of Biotechnology and Biomolecular Sciences, University of New South Wales, Sydney, Australia. b Centre for Epidemiology and Biostatistics, Faculty of Medicine and Health, Leeds University, Leeds, United Kingdom. c Department of Medicine, Faculty of Medicine, University of Malaya, Kuala Lumpur, Malaysia. d Division of Gastroenterology, Department of Medicine, Changi General Hospital, Singapore. To whom correspondence should be addressed: [email protected]

Background and Aim: The contribution of human genetic polymorphisms to H. pylori infection and gastric cancer (GC) development remains unclear due to geographic variation in the association between specific host genetic polymorphisms and GC. In the current study we investigated the association between polymorphisms related to immune and cancer-related pathways and H. pylori infection among the major ethnicities, Chinese, Malay and Indian, resident in Singapore and Malaysia as well as the association between these polymorphisms and GC development in ethnic Chinese patients. Methods: Thirty-four polymorphisms in 26 genes were typed by mass spectrometry in 422 patients undergoing endoscopy (162 Chinese, 113 Indian and 87 Malay controls and 60 Chinese GC cases). Patients were assessed for evidence of H. pylori infection. Odds Ratios and Confidence intervals were obtained using logistic regression models. Results: The prevalence of 16 polymorphisms varied significantly among the ethnicities. In the Chinese subgroup, significant associations were shown between 1) EBBR2+1963G (rs1801200) and H. pylori infection (OR: 0.48, 95%CI 0.23,0.98, p=0.04), 2) PTGS2-1195G (rs689466) and an increased risk of GC on adjusting for H. pylori status (OR: 1.53, 95% CI 0.99,2.37, p=0.05), and 3) IL1B-1473C (rs1143623) and a decreased risk of GC (OR: 0.64, 95%CI 0.41,0.99, p=0.05). Borderline significant associations were seen between IL2-330G (rs2069762) (OR 1.45, 95%CI 0.95,2.15, p=0.06) and IL13-1111T (rs1800925) (OR 0.65, 95%CI 0.42,1.01, p=0.06) and H. pylori infection. Conclusion: These findings contribute to the understanding of the genetic variation between ethnicities which may influence H. pylori susceptibility and the outcome of infection.

Gastric cancer (GC) is a significant global facilitating H. pylori induced mucosal damage burden, being the fourth most common and GC development 2. A subsequent study cancer and the second most common cause by the same group demonstrated a greater of cancer-related death world wide 1. While than 20-fold increased risk of GC associated H. pylori infection is a recognized as a major with the presence of three or more pro- risk factor for GC development, less than 1% inflammatory polymorphisms in the genes of of individuals infected with H. pylori will cytokines IL-1β, IL-1RN, IL-10 and tumor develop GC, suggesting that inter-individual necrosis factor (TNF)-α 3. variation in the form of genetic Genetic polymorphisms may not only predict polymorphisms that alter gene expression GC development in individuals, but and functionality may play a significant role. variations in their frequency may explain In 2001 El-Omar et al. reported based on a differences in GC incidence between ethnic study of patients from Scotland and Poland groups, such as the three ethnicities resident that carriers of the IL1B-31 C allele and in Malaysia, where ethnic Chinese have a high homozygous carriers of the IL1RN*2 allele rate of GC (Age standardized rate of had an increased risk of GC 2. The IL-1β-31 11.3/100,000 for males), while in ethnic C allele has been reported to enhance IL-1β Indians (5.5/100,000) and ethnic Malays production resulting in increased (2.1/100,000) the rate is much lower4. inflammation and decreased gastric acid Recently, our group demonstrated significant secretion, the latter allowing spread of H. variation in the frequencies of IL1B-511, pylori from the antrum to the corpus, thus IL1RN, IL10-1082 and IL10-592 among the

1 SCHMIDT ET AL. SNPs in H. pylori-related disease

three ethnic groups resident in Malaysia and Immunodeficiency Virus were excluded. Singapore 5. Such findings necessitate the Two 5 ml vials of peripheral blood were validation of putative associations with collected at the time of endoscopy and serum human genetic polymorphisms in multiple was immediately separated from one vial. ethnic groups. Informed consent was obtained from each Ethnic variation in the frequency of genetic subject and a brief socio-demographic polymorphisms may also explain the questionnaire completed, including considerable difference in H. pylori prevalence information pertaining to gender, age and between ethnic Indians (50-60%), ethnic ethnicity. This study was approved by the Chinese (35-55%) and ethnic Malays (10- Human Research Ethics Committee (HREC) 25%) 6, As yet few studies have investigated of the University of New South Wales whether any such associations exist 7. H. (HREC 08115 and HREC 02144). pylori infection initiates an early innate immune response, which is quickly replaced Serological evaluation of past/current H. by a vigorous adaptive response which is pylori infection: unable to eliminate H. pylori infection 8. The An in-house ELISA to detect antibodies to human genetic factors driving the ineffective H. pylori, with a sensitivity of 100% and response to H. pylori infection and specificity of 94%, was performed on the sera carcinogenesis are mostly unknown, although of all 422 patients as described by Mitchell et a variety of interconnected pathways have al. 10. Briefly, H. pylori strains NTCC16369 been implicated 9. To date, few studies have and UNSWP10 were used for the solid phase been conducted which not only investigate a and sera were absorbed with C. jejuni strains panel of polymorphisms from these Cj100 and Cj23. Antibodies to H. pylori were contributory pathways in relation to GC, but visualized using a mouse-antihuman IgG also examine the relationship with H. pylori alkaline phosphatase conjugated secondary infection independent of GC status, and the antibody (Sigma Aldrich, St Louis, USA) variation in prevalence between ethnic (diluted 1:10,000) and the substrate disodium groups. p-nitrophenylphosphate (d-pNPP) Therefore, this study aimed to describe in the (Invitrogen, Carlsbad, USA), prepared three ethnic groups resident in Malaysia and according to the manufacturer’s instructions. Singapore, the frequency of a panel of 34 The absorbance of duplicate serum samples polymorphisms in 26 genes which may was read at 405nm. provide a basis for the well characterized In addition to ELISA, serum samples from differences in H. pylori prevalence and GC 12 GC patients shown to be H. pylori negative incidence. We further sought to specifically by ELISA were investigated further by assess the association between these Immunoblot (MPD Helico Blot 2.1, MP polymorphisms and H. pylori infection status Biomedicals, Australia) according to the in all three ethnicities, as well as GC in the manufacturer’s instructions. ethnic Chinese population. Selection of polymorphisms for analysis: Methods: Thirty-four polymorphisms over 26 genes Study population and Samples: were selected for analysis from the available Between January 2004 and April 2007, 422 literature based on known or potential unrelated patients undergoing endoscopy for functionality, and/or previous associations routine gastrointestinal symptoms were with cancer development, inflammatory enrolled at the University Hospital of diseases or H. pylori infection. All selected Malaysia, Kuala Lumpur, Malaysia and polymorphisms were reported to have a Changi General Hospital, Singapore. minor allele frequency >1% in the National Patients included in this study were Center for Biotechnology Information diagnosed with functional dyspepsia (FD) or (NCBI) dbSNP 11 or in published literature. GC based on endoscopy and histological Polymorphism sequences were identified assessment as appropriate. Patients taking using dbSNP. anti-microbial agents, non-steroidal anti- DNA extraction and analysis of inflammatory drugs or acid suppressants in polymorphisms by mass spectrometry: the two months prior to recruitment and DNA was prepared from each subject’s patients known to be infected with Human whole blood using the DNA Blood Mini kit

2 SCHMIDT ET AL. SNPs in H. pylori-related disease

Table 1: Oligonucleotide primer pairs and restriction enzymes used for PCR-RFLP based confirmation of five polymorphisms. Annealing dbSNP temperatur Restriction reference SNP Primer Sequence e Enzyme Reference GAA ACT AAC ATG TTG AAC AG IL1B - TTG AAT GGG TGA ATG GGA RS1143623 1473 AT 56°C StyI 5 AACACTACTAAGGCT IL10 - TCTTGAGGA RS1800896 1082 AGCACATAGAATGAAACCTTGG 56°C BseRI 5 TGG CAT TGA TCT GGT TCA TC RS16944 IL1B -511 GTT TAG GAA TCT TCC CAC TT 56°C AvaI 32 AGG CAA TAG GTT TTG AGG TNFA - GCC AT RS1800629 308 TCC TCC CTG CTC CGA TTC CG 60°C NcoI 33 TLR5 GGT AGC CTA CAT TGA TTT GC RS5744168 +1174 GAT TCT CTG AAG GGG TTT G 57°C DdeI 5

(Qiagen, Australia) as per the manufacturer’s Statistical Analysis: instructions, with the following modification: To assess differences in patient demographics DNA was rehydrated in sterile water. and H. pylori prevalence between control Samples, normalized to 10ng/uL, were sent groups, and controls and cases, the one-way to Australian Genome Research Facility Ltd analysis of variance (ANOVA) and unpaired (AGRF), St Lucia, University of Queensland, t-test available in GraphPad InStat version Australia, for SNP analysis by mass 3.06 statistical analysis package for Windows spectroscopy. (GraphPad Software, San Diego, USA, www.graphpad.com) were used. PCR-RFLP validation of mass Hardy-Weinberg equilibrium (HWE) analysis spectroscopy results for polymorphism for each polymorphism and pair wise linkage typing: disequilibrium (LD) measures for Polymerase chain reaction (PCR)-restriction polymorphisms on the same chromosome fragment length polymorphism (RFLP) was were performed using the FD group for each conducted to validate the results obtained by ethnicity. These tests were available in Mass Spectroscopy in a subgroup of subjects STATA version 10 (STATA Corporation, (n=373) for 5 polymorphisms over 4 genes: College Station, USA) using the genhwi and IL1B-511 (RS16944), IL1B-1473 pwld commands respectively. (RS1143623), IL10-1082 (RS1800896), Allele frequencies in the controls (FD) were TNFA-308 (RS1800629) and TLR5+1174 compared for each polymorphism between (RS5744168). PCR primers (Sigma Aldrich), the 3 ethnic groups using 2x3 chi-square tests annealing temperatures and restriction in Microsoft Excel 2007 (Microsoft enzymes (New England Biolabs, Australia) Corporation, Redmond, USA) to determine used to detect polymorphisms are presented variation between ethnic groups. Logistic in Table 1. Briefly, PCR was performed in a regression analyses were undertaken in 25uL reaction which included 4mM MgCl2 STATA version 10 and were used to obtain (Qiagen), 200μM dNTP (Qiagen), 1 unit Hot odds ratios (OR) and 95% confidence Star Taq polymerase (Qiagen) and 1μg DNA intervals (95% CI) as measures of association template. PCR cycling consisted of an initial and precision between polymorphism denaturation at 95°C for 15 minutes, 35 genotypes, H. pylori infection and GC. The cycles of 94°C for 30 seconds, 56-60°C for polymorphisms were measured as a 30 seconds and 72°C for 50 seconds, and a continuous variable, where if the frequency final extension step at 72°C for 5 minutes. of allele x was greater than allele y, then RFLP was conducted on PCR products xx=0, xy=1 and yy=2. according to the manufacturer’s recommendations.

3 SCHMIDT ET AL. SNPs in H. pylori-related disease

Table 2: Details of patients included in the analysis – percentage (%) of males, mean age and standard deviation (SD), and percentage of patients positive (+ve) for H. pylori. Ethnicity Diagnosis Total No. % Males Mean Age SD %HP +ve FD 162 46.3 54.14 14.03 77.8 Chinese GC 60 61.7 64.51 13.43 86.7 Indian FD 113 40.7 48.67 14.49 83.2 Malay FD 87 40.2 42.28 14.4 54.0

The results presented represent the ORs per Twenty-eight patients with call-rates below increment of the minor allele. Associations 80% were also excluded from analysis. Four were considered statistically significant if P ≤ polymorphisms (RS1800566, RS1800871, 0.05 or if the 95% CI excluded 1.0 in the RS4073 and RS3136674) found to depart adjusted models. significantly from HWE within Chinese controls were also excluded from analysis Results: (P<0.05). Patient demographics and H. pylori By linkage disequilibrium analysis, RS16944 status: (IL1B -511) and RS1143623 (IL1B -1473) on Patient demographics including H. pylori chromosome 2 were linked in the Chinese status of the 162 Chinese, 113 Indian and 87 and Malay populations (r2=0.76 and r2=0.58 Malay FD controls and 60 Chinese GC cases respectively), but not in the Indian are presented in Table 2. population (r2=0.29). RS2227307 (IL8 +396) There was a significant difference in the and rs2227306 (IL8 +781), on chromosome mean age between the three control groups 4, were linked in all three ethnic groups (r2 (P<0.001) and between Chinese cases and =0.55-0.68). All other polymorphism controls, despite matching for gender and age combinations produced r2 ≤ 0.5 (data not (±10 years, mean difference 10 years) shown). (P<0.001). The prevalence of H. pylori was significantly Variation in polymorphisms between higher in Chinese and Indian controls as ethnic groups: compared with Malay controls (P<0.0001), Sixteen of the remaining 26 polymorphisms but there was no significant difference (62%) were found to vary significantly between the case (GC) and control (FD) (P≤0.05) between the three ethnic groups groups in the Chinese sub-population. (Table 3). Further analysis by logistic regression was restricted to ethnicity-adjusted Confirmation of Mass Spectrometry analysis for polymorphisms that varied results using PCR-RFLP: between the ethnic groups. PCR-RFLP analysis showed a high concordance with and confirmed the findings Polymorphisms in relation to H. pylori of mass spectrometry; the results for status: RS1800896, RS1143623, RS16944, A borderline non-significant increase (OR RS1800629 and RS5744168 agreeing in 1.45, 95% CI 0.98, 2.15) in the odds of being 98.93%, 96.78%, 94.91%, 99.2% and 97.84% H. pylori positive with carriage of the minor G of cases respectively. Given this, all allele was observed for RS2069762 (IL2 -330) polymorphisms were analysed according to when adjusting for ethnicity, although this the results of mass spectrometry. association became non-significant in the stratified analysis of the Chinese. In the Quality Control, Call Rates, HWE and stratified analysis of Chinese controls, a LD Data Analysis: significantly decreased risk of being H. pylori Four polymorphisms (RS28720239, positive with carriage of the EBBR2 +1963G RS1800371, RS2069709 and RS5743708) allele (RS1801200) (OR 0.48, 95% CI 0.23, were found to be exceptionally rare (<1%) or 0.98) was observed. No other significant absent and were excluded from further associations were observed. The results of analysis. The polymorphism call rate was this analysis are presented in Table 4. found to be acceptable (0-6.45% failure rate).

4 SCHMIDT ET AL. SNPs in H. pylori-related disease

Polymorphisms in relation to GC in Malaysia and Singapore, exhibited a greater ethnic Chinese: similarity to Caucasians than to East Asians Carriage of the minor C allele of IL1B -1473 with respect to the IL1B polymorphisms, a (RS1143623) was associated with a finding that is consistent with the fact that significant decrease in the odds ratio for GC Malaysian and Singaporean Indians are Tamil (OR 0.64, 95% CI 0.41,0.99, P=0.05). migrants from Southern India, a population Carriage of the minor G allele for PTGS2 - reported to be genetically half way between 1195 (RS689466) resulted in a borderline Caucasians and East Asians 17. significant increase in the odds ratio for GC The IL2-330G allele has been postulated to (OR 1.53, 95% CI 0.99, 2.37, P=0.05). No positively influence the development of a other significant associations were observed. Th1-type immune response through The results of this analysis are presented in increased production of IFN- 18, which is Table 4. ineffective at eradicating the initial H. pylori infection. Although of borderline Discussion: significance, this allele was shown to be The number of host genetic polymorphisms associated with an increased risk of H. pylori proposed to modify the response to H. pylori infection (OR 1.45, 95% CI 0.98,2.15, infection and carcinogenesis has increased p=0.06), contrasting with a recent study by exponentially over the last decade, however Queiroz et al. who found IL2-330G to be none has yet provided a sufficiently reliable associated with a decreased risk of H. pylori marker across all populations. The present infection in Brazilian adults and with study investigated a panel of 34 increased serum concentrations of IL-2 18. polymorphisms from 26 genes, for variation Studies have shown that the IL-13 -1111 T between ethnicities, for the association with allele results in increased binding of nuclear H. pylori seropositivity in these ethnic groups proteins and expression of IL-13 from Th2 and for the association with GC development polarised CD4+ T cells, thus suggesting an in the ethnic Chinese population. Sixteen of enhanced predisposition to Th2 responses 19- the 34 (47%) genetic polymorphisms studied 20 which may protect against GC varied significantly between the three ethnic development and H. pylori infection. A groups investigated. borderline significant decrease in the odds of The frequencies of the IL1B-511A and H. pylori seropositivity associated with the T +3954T alleles were 63.5% and 8.5% in allele was observed in the unadjusted analysis Malays, 52.5% and 2% in Chinese, and 32.5% (OR 0.65, 95%CI 0.42,1.01, p=0.06) which is and 10.5% in the ethnic Indians. A meta- consistent with a Japanese study, which analysis by Camargo et al. similarly reported reported decreased odds of H. pylori considerable variation in the frequency of the seropositivity associated with carriers of the IL1B-511T (corresponding to the A allele in T allele 19. our analysis) and +3954T alleles between The IL10-1082 promoter polymorphism was Caucasian (33% and 23% respectively) and found to vary significantly between the ethnic Asian populations (50% and 5% respectively) groups (p<0.01), with Indians having a 12. Our finding that there was no association comparatively lower frequency of the A allele with either of these polymorphisms and GC (79%), as compared with Chinese and Malays development in the Chinese population is (98% and 89% respectively). Interestingly, consistent with meta-analyses that suggest we did not find a significant difference in the that such associations are largely confined to prevalence of IL10-592 between our ethnic Caucasian populations 12-15. We did, groups as anticipated based on a recent meta- however, find a significantly decreased analysis by Loh et al. which demonstrated association between the IL1B-1473 C allele considerable heterogeneity between and GC development in ethnic Chinese (OR associations in Caucasian and east Asian 0.41, 95% CI 0.64,0.99, p=0.05), an populations 14. The three IL10 association previously reported in a Korean polymorphisms investigated in our study population 16. Significantly the presence of were not associated with either H. pylori the G allele for this polymorphism has been seropositivity or GC, in agreement with shown to result in decreased promoter recent studies in China and Ireland which activity 16. likewise failed to substantiate an association Interestingly the ethnic Indians resident in between polymorphisms in the IL10 gene

5 SCHMIDT ET AL. SNPs in H. pylori-related disease

and GC 21-22. It is however, at variance with certainly be intriguing to evaluate in a several previous studies conducted in Korea, subgroup analysis. Japan and China which previously In conclusion, this study has examined a large demonstrated associations between one or range of polymorphisms in genes specifically more of these polymorphisms and GC 23-25. associated with the innate and acquired Oxidative stress is believed to play a key role immune responses, oxidative stress, cell cycle in carcinogenesis 26. PTGS2 encodes the regulation and cell signaling. Importantly we inducible enzyme cyclooxygenase-2, which is have included the association between these involved in the synthesis of prostaglandins polymorphisms and not only GC, but also H. and is thus pro-inflammatory, giving the pylori seropositivity. This is an important polymorphism biological plausibility for a consideration as the strong correlation role in GC development. The PTGS2-1195A between H. pylori infection and GC may polymorphism is reported to create a C-MYC mean an association between a binding site and result in higher transcription polymorphism and H. pylori is mistaken for 27-28. In the current study carriage of the weak association with GC. Importantly, we PTGS2 -1195 G allele was associated with identified significant differences in the increased odds of GC in the Chinese (OR prevalence of 16 polymorphisms across the 1.53, 95% CI 0.99, 2.37, p=0.05). In ethnic Chinese, Indian and Malay groups, a contrast, two previous studies in China finding likely to explain the failure to reported a significantly increased risk of GC consistently demonstrate significant associated with the A allele 28-29. However, associations between specific polymorphisms Liu et al. acknowledged that selection bias in and GC in disparate populations. In the their study may have occurred as GC cases ethnic Chinese, significant associations were were selected from a different population to identified between EBBR2+1963 and H. controls 28 . pylori infection, and PTGS2-1195 and IL1B- The proto-oncogene ERBB2, encoding 1473 and GC. Clearly further studies are HER-2/neu, has been associated with the required to investigate the importance of development and poor prognosis in a range these findings in other populations. cancers 30. The ERBB2+1963 (Ile655Val) polymorphism has previously been linked to Acknowledgements: GC development and progression in Japan 31. This work was supported by The Cancer Intriguingly, while we did not find an Council of New South Wales, Sydney, association with GC development, the minor Australia (Grant no. 66/04). G allele (valine), was associated with a significant decrease in the odds of H. pylori References: infection in the Chinese subgroup (OR 0.48, [1] Parkin DM, Bray F, Ferlay J, Pisani 95% CI 0.23, 0.99, p=0.04). The mechanism P. Estimating the world cancer burden: through which this gene and polymorphism Globocan 2000. Int J Cancer. 2001; 94: 153-6. may influence H. pylori infection is unclear. [2] El-Omar EM, Carrington M, Chow Our study is typical of many other published WH, et al. Interleukin-1 polymorphisms genetic association studies, in that it is associated with increased risk of gastric underpowered to detect associations of the cancer.[erratum appears in Nature 2001 Jul magnitude that might be expected, especially 5;412(6842):99]. Nature. 2000; 404: 398-402. given the observed heterogeneity between the [3] El-Omar EM, Rabkin CS, Gammon ethnic groups. Despite this, our study adds MD, et al. increased risk of noncardia gastric considerably to the available body of cancer associated with proinflammatory knowledge, although the low power and cytokine polymorphisms. Gastroenterology. multiple testing issues mean that our findings 2003; 124: 1193-201. may potentially include false positives and [4] Lim GCC, Yahaya H, Lim TO. The negatives, necessitating validation of these first report of the national cancer registry - results in further studies. Due to the low cancer incidence in Malaysia 2002. Kuala incidence of GC in ethnic Indian and Malays, Lumpur: National cancer registry 2002:56. as well as the low prevalence of H. pylori in [5] Ha M-D. The role of specific host the Malays, we were unable to recruit factors, specific dietary factors and sufficient numbers of these groups for Helicobacter pylori infection on the risk of detailed analysis. These populations would gastric cancer. School of biotechnology and

6 SCHMIDT ET AL. SNPs in H. pylori-related disease

biomolecular sciences. Sydney: University of New caste populations of southwest India and South Wales 2007:280. their affinity with related Indian and global [6] Goh KL, Parasakthi N. The racial groups. BMC Genetics. 2004; 5: 23. cohort phenomenon: seroepidemiology of [18] Queiroz DM, Saraiva IE, Rocha GA, Helicobacter pylori infection in a multiracial et al. IL2-330G polymorphic allele is South-East Asian country. Eur J Gastroen associated with decreased risk of Helicobacter Hepat. 2001; 13: 177-83. pylori infection in adulthood. Microbes Infect. [7] Hishida A, Matsuo K, Goto Y, et al. 2009; 11: 980-7. No association between AICDA 7888 C/T [19] Togawa S, Joh T, Itoh M, et al. polymorphism, Helicobacter pylori Interleukin-2 gene polymorphisms associated seropositivity, and the risk of atrophic with increase risk of gastric atrophy from gastritis and gastric cancer in Japanaese. Helicobacter pylori infection. Helicobacter. 2005; Gastric Cancer. 2010; 13: 43-9. 10: 172-8. [8] Stoicov C, Li H, Cerny J, Houghton [20] Hamajima N, Naito M, Kondo T, JM. How the study of Helicobacter infection Goto Y. Genetic factors involved in the can contribute to the understanding of development of Helicobacter pylori-related carcinoma development. Clin Microbiol Infect. gastric cancer. Cancer Sci. 2006; 97: 1129-38. 2009; 15: 813-22. [21] Hussain SK, Mu LN, Cai L, et al. [9] Milne AN, Carneiro F, O'Morain C, Genetic variation in immune regulation and Offerhaus GJA. Nature meets nurture: DNA repair pathways and stomach cancer in molecular genetics of gastric cancer. Hum China. Cancer Epidemiol Biomarkers Prev. 2009; Genet. 2009; 126: 615-28. 18: 2304-9. [10] Mitchell HM, Lee A, Berkowicz J, [22] Murphy G, Thornton J, McManus R, Borody T. The use of serology to diagnose et al. Association of gastric disease with active Campylobacter pylori infection. Med J polymorphisms in the inflammatory related Aust. 1988; 149: 604-9. genes IL-1B, IL-1RN, IL-10, TNF and [11] Sherry ST, Ward MH, Kholodov M, TLR4. Eur J Gastroenterol Hepatol. 2009; 21: et al. dbSNP: the NCBI database of genetic 630-5. variation. Nucl Acids Res. 2001; 29: 308-11. [23] Sugimoto M, Furuta T, Shirai N, et [12] Camargo MC, Mera R, Correa P, et al. Effects of interleukin-10 gene al. Interleukin-1beta and interleukin-1 polymorphism on the development of gastric receptor antagonist gene polymorphisms and cancer and peptic ulcer in Japanese subjects. J gastric cancer: a meta-analysis. Cancer Gastroenterol Hepatol. 2007; 22: 1443-9. Epidemiol Biomarkers Prev. 2006; 15: 1674-87. [24] Kang JM, Kim N, Lee DH, et al. The [13] Vincenzi B, Patti G, Galluzzo S, et al. effects of genetic polymorphisms of IL-6, IL- Interleukin 1beta-511T gene (IL1beta) 8, and IL-10 on Helicobacter pylori-induced polymorphism is correlated with gastric gastroduodenal diseases in Korea. J Clin cancer in the Caucasian population: results Gastroenterol. 2009; 43: 420-8. from a meta-analysis. Oncol Rep. 2008; 20: [25] Lu W, Pan K, Zhang L, Lin D, Miao 1213-20. X, You W. Genetic polymorphisms of [14] Loh M, Koh KX, Yeo BH, et al. interleukin (IL)-1B, IL-1RN, IL-8, IL-10 and Meta-analysis of genetic polymorphisms and tumour necrosis factor alpha and risk of gastric cancer risk: Variability in associations gastric cancer in a Chinese population. according to race. European Journal of Cancer. Carcinogenesis. 2005; 26: 631-6. 2009; 45: 2562-8. [26] Cerutti PA. Prooxidant states and [15] Gianfagna F, De Feo E, van Duijn tumor promotion. Science. 1985; 227: 375-81. CM, Ricciardi G, Boccia S. A systematic [27] Zhang X, Miao X, Tan W, et al. review of meta-analyses on gene Identification of functional genetic variants in polymorphisms and gastric cancer risk. Curr cyclooxygenase-2 and their association with Genomics. 2008; 9: 361-74. risk of esophageal cancer. Gastroenterology. [16] Lee KA, Ki CS, Kim HJ, et al. Novel 2005; 129: 565 - 76. interleukin 1beta polymorphism increased the [28] Liu F, Pan K, Zhang X, et al. Genetic risk of gastric cancer in a Korean population. variants in Cyclooxygenase-2: expression and J Gastroenterol. 2004; 39: 429-33. risk of gastric cancer and its precursors in a [17] Rajkumar R, Kashyap V. Genetic Chinese population. Gastroenterology. 2006; structure of four socio-culturally diversified 130: 1975-84.

7 SCHMIDT ET AL. SNPs in H. pylori-related disease

[29] Zhang XM, Miao XP, Tan W, et al. [Genetic polymorphisms in the promoter region of cyclooxygenase-2 and their association with risk of gastric cancer] [Abstract only]. Zhongguo Yi Xue Ke Xue Yuan Xue Bao. 2006; 28: 119-23. [30] Gravalos C, Jimeno A. HER2 in gastric cancer: a new prognostic factor and a novel therapeutic target. Ann Oncol. 2008; 19. [31] Kuraoka K, Matsumura S, Hamai Y, et al. A single nucleotide polymorphism in the transmembrane domain nucleotide polymorphism in the transmembrane domain coding region of HER-2 is associated with development and malignant phenotype of gastric cancer. Int J Cancer. 2003; 107: 593-6. [32] di Giovine FS, Takhsh E, Blakemore AI, Duff GW. Single base polymorphism at - 511 in the human interleukin-1 beta gene (IL1 beta). Hum Mol Genet. 1992; 1: 450. [33] Wilson AG, Di Giovine FS, Blakemore AI, Duff GW. Single base polymorphism in the human tumour necrosis factor alpha (TNF alpha) gene detectable by NcoI restriction of PCR product. Hum Mol Genet. 1992; 1: 353.

8 SCHMIDT ET AL. SNPs in H. pylori-related disease

1 2 Table 3: Details of the 34 polymorphisms (33 SNPs and 1 INDEL) and their prevalence in the three major ethnic groups resident in Malaysia and Singapore. Variation between Polymorphism Details Polymorphism Prevalence ethnic SNP or Chinese GC Chinese FD Indian FD Malay FD groups dbSNP Polymorphism INDEL Chromosomal accession Gene Location (x/y) Location xx xy yy xx xy yy HWE xx xy yy HWE xx xy yy HWE P rs689466 PTGS2 -1195 a/g 1q25.2-25.3 0.19 0.51 0.31 0.31 0.46 0.23 0.51 0.68 0.27 0.05 0.74 0.42 0.43 0.15 0.80 <0.01 rs1800896 IL10 -1082 a/g 1q31-32 0.95 0.05 0.00 0.92 0.07 0.01 0.23 0.64 0.30 0.06 0.24 0.81 0.16 0.03 0.22 <0.01 rs1800871 IL10 -819 t/c 1q31-32 0.41 0.41 0.19 0.52 0.32 0.16 <0.01 0.19 0.39 0.41 0.06 0.44 0.35 0.21 0.09 † rs1800872 IL10 -592 a/c 1q31-32 0.33 0.53 0.14 0.45 0.40 0.16 0.07 0.44 0.41 0.15 0.30 0.28 0.54 0.18 0.64 0.27 rs5744168 TLR5 +1174 c/t 1q41-42 0.93 0.07 0.00 0.94 0.06 0.00 1.00 0.78 0.22 0.00 0.61 0.79 0.21 0.00 1.00 <0.01 rs1143623 IL1B -1473 g/c 2q14 0.35 0.57 0.08 0.30 0.46 0.25 0.32 0.43 0.39 0.18 0.15 0.36 0.43 0.21 0.33 0.07 rs16944 IL1B -511 a/g 2q14 0.32 0.52 0.17 0.30 0.45 0.25 0.41 0.07 0.51 0.42 0.26 0.39 0.49 0.12 0.62 <0.01 rs1143634 IL1B +3954 c/t 2q14 0.98 0.02 0.00 0.96 0.04 0.00 1.00 0.80 0.19 0.01 1.00 0.83 0.17 0.00 1.00 <0.01 rs5743836 TLR9 -1237 t/c 3p21.3 1.00 0.00 0.00 0.89 0.09 0.02 0.27 0.79 0.19 0.03 0.19 0.90 0.07 0.02 1.00 <0.01 rs4073 IL8 -251 t/a 4q13-21 0.27 0.55 0.18 0.38 0.39 0.23 0.02 0.44 0.43 0.13 0.51 0.28 0.52 0.20 0.64 † rs2227307 IL8 +396 t/g 4q13-21 0.18 0.47 0.35 0.37 0.45 0.18 0.60 0.19 0.39 0.42 0.14 0.11 0.48 0.41 1.00 0.53 rs2227306 IL8 +781 c/t 4q13-21 0.39 0.48 0.13 0.46 0.43 0.11 1.00 0.54 0.34 0.12 0.09 0.55 0.40 0.05 1.00 0.26 rs28720239 NFkB1 -94 attg/- 4q24 1.00 0.00 0.97 0.03 1.00 0.00 1.00 0.00 † rs2069762 IL2 -330 t/g 4q26-27 0.37 0.53 0.10 0.48 0.39 0.13 0.20 0.27 0.41 0.32 0.11 0.25 0.54 0.21 0.64 <0.01 rs5743708 TLR2 +2408 g/a 4q32 1.00 0.00 0.00 0.99 0.00 0.01 1.00 1.00 0.00 0.00 1.00 0.98 0.02 0.00 1.00 0.18 rs1800925 IL13 -1111 c/t 5q31 0.70 0.30 0.00 0.73 0.26 0.01 0.53 0.64 0.31 0.06 0.39 0.61 0.35 0.04 0.75 0.08 rs2070874 IL4 -33 t/c 5q31.1 0.54 0.39 0.07 0.68 0.27 0.05 0.19 0.06 0.43 0.51 0.62 0.30 0.40 0.30 0.14 <0.01 rs1800629 TNFA -308 g/a 6p21.3 0.82 0.17 0.02 0.85 0.15 0.00 0.60 0.90 0.10 0.00 1.00 0.91 0.90 0.00 1.00 0.26 rs2234711 IFNGR1 -56 c/t 6q23-24 0.20 0.64 0.16 0.37 0.44 0.19 0.31 0.57 0.35 0.08 0.61 0.49 0.42 0.09 1.00 <0.01 rs4880 SOD2 +47 t/c 6q25.3 0.72 0.26 0.02 0.76 0.22 0.03 0.29 0.29 0.49 0.22 0.85 0.51 0.42 0.07 1.00 <0.01 rs16906079 TLR4 +690 a/g 9q32-33 0.98 0.00 0.02 0.99 0.01 0.00 1.00 0.95 0.04 0.01 0.05 1.00 0.00 0.00 1.00 0.02 rs4986790 TLR4 +896 a/g 9q32-33 0.98 0.02 0.00 0.91 0.08 0.01 1.00 0.71 0.26 0.03 0.72 0.89 0.09 0.03 1.00 <0.01 rs2069709 IFNG -179 g/t 12q14 1.00 0.00 0.00 1.00 0.00 0.00 1.00 0.00 0.00 1.00 0.00 0.00 † rs2301756 PTPN11 intron 3 +86 c/t 12q24 0.83 0.15 0.02 0.74 0.25 0.01 0.74 0.10 0.47 0.44 0.82 0.44 0.44 0.13 0.80 <0.01 3' UNTR rs696 NFkBIA (STOP+2) g/a 14q13 0.32 0.55 0.13 0.27 0.54 0.19 0.33 0.25 0.58 0.17 0.08 0.31 0.49 0.20 1.00 0.95 rs2070600 AGER +242 g/a 14q32 0.55 0.40 0.05 0.63 0.31 0.06 0.46 0.79 0.19 0.02 0.62 0.83 0.15 0.02 0.43 <0.01 rs1801275 IL4R +1902 a/g 16q11.2-12.1 0.69 0.28 0.03 0.56 0.36 0.08 0.83 0.60 0.38 0.02 0.23 0.43 0.43 0.13 1.00 0.05

9 SCHMIDT ET AL. SNPs in H. pylori-related disease

rs1800566 NQO1 +609 c/t 16q22.1 0.26 0.55 0.19 0.32 0.39 0.29 0.01 0.45 0.49 0.06 0.10 0.42 0.42 0.16 0.46 † rs1800371 TP53 +139 c/t 17p13.1 1.00 0.00 0.00 1.00 0.00 0.00 1.00 0.00 0.00 1.00 0.00 0.00 † rs1042522 TP53 +215 g/c 17p13.1 0.35 0.47 0.18 0.27 0.51 0.23 1.00 0.21 0.54 0.25 0.43 0.24 0.56 0.20 0.82 0.57 rs3136674 CCL13 -887 t/c 17q11.2 0.88 0.12 0.00 0.92 0.06 0.03 <0.01 0.95 0.04 0.01 0.07 0.98 0.02 0.00 1.00 † rs2297518 NOS2 +1823 g/a 17q11.2-12 0.69 0.25 0.05 0.73 0.24 0.03 0.36 0.71 0.27 0.02 1.00 0.74 0.25 0.01 1.00 0.78 rs1801200 ERBB2 +1963 a/g 17q21.1 0.81 0.19 0.00 0.77 0.20 0.03 0.27 0.68 0.28 0.04 0.73 0.62 0.37 0.01 0.44 0.17 rs1059293 IFNGR2 +2083 t/c 21q22.11 0.67 0.33 0.00 0.63 0.33 0.04 0.44 0.48 0.45 0.07 0.33 0.47 0.38 0.16 0.58 0.01 1 FD: Functional dyspepsia, GC: Gastric cancer, HP: H. pylori, HWE: Hardy-Weinberg Equilibrium, INDEL: insertion-deletion polymorphism, x: major allele, y: minor allele. † 2 Polymorphism is not in HWE or is monoallelic; therefore variation between ethnic groups cannot be assessed.

10 SCHMIDT ET AL. SNPs in H. pylori-related disease

1 Table 4: The association of each polymorphism included in statistical analysis was assessed with 1) H. pylori infection in the FD control group and 2) GC in the ethnic 2 Chinese subgroup. The ORs presented here represent the OR per increment of the minor allele. Polymorphism Association between polymorphism and HP Association between polymorphism and GC All ethnicities (n=334) Chinese subgroup analysis Chinese only (n=213) dbSNP Gene Polymorphism SNP or Unadjusted Adjusting for ethnicity (n=153) Unadjusted Adjusting for HP accession Location INDEL (x/y) OR 95% CI P OR 95% CI P OR 95% CI P OR 95% CI P OR 95% CI P rs689466 PTGS2 -1195 a/g 0.82 0.55,1.21 0.31 0.76 0.44,1.32 0.33 1.48 0.96,2.26 0.07 1.53 0.99,2.37 0.05 rs1800896 IL10 -1082 a/g 1.02 0.57,1.82 0.95 2.93 0.39,22.03 0.30 0.60 0.17,2.06 0.41 0.56 0.16,1.95 0.37 rs1800872 IL10 -592 a/c 0.89 0.63,1.27 0.52 0.96 0.66,1.39 0.81 1.48 0.83,2.67 0.19 1.18 0.77,1.81 0.46 1.17 0.76,1.80 0.48 rs5744168 TLR5 +1174 c/t 1.14 0.53,2.45 0.73 0.57 0.14,2.34 0.43 1.02 0.31,3.39 0.97 1.08 0.32,3.60 0.91 rs1143623 IL1B -1473 g/c 1.01 0.72,1.43 0.95 1.02 0.71,1.47 0.90 0.90 0.52,1.55 0.70 0.64 0.41,0.99 0.04 0.64 0.41,0.99 0.05 rs16944 IL1B -511 a/g 1.16 0.77,1.73 0.49 0.95 0.53,1.69 0.85 1.30 0.85,1.99 0.22 1.30 0.85,1.99 0.22 rs1143634 IL1B +3954 c/t 1.20 0.52,2.79 0.67 1.32 0.15,11.69 0.81 0.41 0.05,3.45 0.41 0.40 0.05,3.38 0.40 rs5743836 TLR9 -1237 t/c 0.85 0.41,1.77 0.67 0.93 0.27,3.23 0.91 rs2227307 IL8 +396 t/g 1.05 0.73,1.50 0.81 0.99 0.68,1.45 0.96 1.11 0.62,1.98 0.73 1.07 0.69,1.66 0.75 1.07 0.69,1.65 0.77 rs2227306 IL8 +781 c/t 1.29 0.86,1.93 0.22 1.23 0.81,1.88 0.34 1.31 0.69,2.49 0.40 1.24 0.78,1.98 0.36 1.23 0.77,1.97 0.38 rs2069762 IL2 -330 t/g 1.45 0.98,2.15 0.06 1.62 0.86,3.06 0.13 1.17 0.76,1.81 0.48 1.15 0.74,1.78 0.53 rs1800925 IL13 -1111 c/t 0.65 0.42,1.01 0.06 0.67 0.42,1.07 0.09 0.76 0.35,1.65 0.49 1.00 0.53,1.86 0.99 1.03 0.55,1.93 0.93 rs2070874 IL4 -33 t/c 0.80 0.53,1.21 0.29 1.10 0.54,2.25 0.80 1.47 0.90,2.39 0.13 1.47 0.90,2.41 0.12 rs1800629 TNFA -308 g/a 1.43 0.63,3.23 0.39 1.35 0.58,3.14 0.49 1.34 0.42,4.24 0.62 1.31 0.62,2.75 0.48 1.32 0.63,2.77 0.47 rs2234711 IFNGR1 -56 c/t 1.02 0.69,1.51 0.92 0.73 0.42,1.26 0.26 1.31 0.84,2.04 0.23 1.33 0.85,2.08 0.21 rs4880 SOD2 +47 t/c 0.96 0.62,1.48 0.85 1.47 0.60,3.58 0.40 1.14 0.62,2.07 0.68 1.11 0.61,2.03 0.73 rs16906079 TLR4 +690 a/g 0.46 0.10,2.07 0.31 2.84 0.40,20.23 0.30 2.65 0.38,18.74 0.33 rs4986790 TLR4 +896 a/g 0.56 0.27,1.16 0.12 0.42 0.12,1.55 0.19 0.21 0.03,1.70 0.14 0.23 0.03,1.81 0.16 rs2301756 PTPN11 intron 3 +86 c/t 1.08 0.69,1.68 0.74 0.87 0.38,1.99 0.75 0.64 0.32,1.30 0.22 0.64 0.31,1.30 0.21 3' UNTR rs696 NFkBIA (STOP+2) g/a 0.89 0.61,1.30 0.55 0.87 0.59,1.28 0.49 1.00 0.56,1.79 0.99 0.77 0.49,1.20 0.25 0.77 0.49,1.21 0.26 rs2070600 AGER +242 g/a 1.28 0.75,2.20 0.37 0.98 0.50,1.91 0.96 1.27 0.77,2.08 0.34 1.27 0.77,2.08 0.35 rs1801275 IL4R +1902 a/g 0.85 0.55,1.31 0.46 0.95 0.49,1.81 0.87 0.63 0.37,1.09 0.10 0.64 0.37,1.10 0.10 rs1042522 TP53 +215 g/c 1.11 0.77,1.59 0.58 1.07 0.73,1.56 0.73 1.17 0.67,2.05 0.57 0.76 0.49,1.16 0.21 0.76 0.50,1.17 0.21 rs2297518 NOS2 +1823 g/a 1.10 0.67,1.81 0.72 1.05 0.62,1.77 0.85 1.10 0.51,2.36 0.81 1.16 0.67,1.99 0.59 1.15 0.67,1.97 0.62 rs1801200 ERBB2 +1963 a/g 0.66 0.42,1.04 0.07 0.66 0.41,1.07 0.09 0.48 0.23,0.98 0.04 0.73 0.36,1.46 0.37 0.75 0.37,1.52 0.43 rs1059293 IFNGR2 +2083 t/c 1.19 0.75,1.87 0.46 0.83 0.40,1.72 0.61 0.75 0.41,1.37 0.36 0.77 0.42,1.41 0.40 3 CI: Confidence interval, FD: Functional dyspepsia, GC: Gastric cancer, HP: H. pylori, OR: Odds ratio. 4 Missing values: polymorphisms not analysed as varied significantly between the ethnic groups.

11 3 Concluding Remarks and Recommendations

H. pylori is the world’s most common bacterial infection, however, the strong positive correlation between socio-economic status and H. pylori infection means that the burden of infection is predominantly borne by poorer countries. Infection with H. pylori is an essential initiating factor in both PUD and GC, severe gastroduodenal diseases which may develop many years after the initial infection. The rate of severe disease development varies widely between populations and this variation is not always reconcilable with differences in H. pylori prevalence. Furthermore, not all infected individuals will progress to severe disease, the majority remaining asymptomatic. The eventual outcome of H. pylori infection is likely determined by the complex interaction of bacterial, host and environmental factors, and these factors should remain consistent across populations. This thesis examined host genetic polymorphisms and bacterial virulence traits that may contribute to susceptibility to H. pylori infection and to severe disease development in patients from different ethnic groups and disease states.

Investigation of the major virulence factors of H. pylori cagA, vacA s1i1, babA, oipA ON and an intact cag PAI in Paper III showed that all were in general highly prevalent in the three primary ethnic groups resident in Malaysia and Singapore. No association was observed between these virulence factors, singly or in combination, and disease development in ethnic Chinese, suggesting that in this population strains associated with high virulence in Western populations are common in this ethnic group. Given this it is unlikely that these virulence factors are the sole determining factors in progression to GC or PUD in this population. Interestingly distinct variations were shown to exist in the predmoninant circulating strains of H. pylori in the Chinese, Indian and Malay ethnic groups, despite the interaction between these ethnic groups. In particular variation in the prevalence of the cag PAI gene HP0521, the CagA EPIYA- motifs and the putative virulence determinant dupA was detected in strains isolated from the three ethnic groups. Interestingly the predominant CagA EPIYA motif varied significantly between the three ethnic groups, with Indians representing a western population (EPIYA-C) and Chinese representing a typically east Asian population (EPIYA-D) (Papers I and III). Strains with EPIYA-C and –D were isolated at similar frequencies from Malay patients, lends support to the contention that ethnic Malays were an H. pylori-free population who have acquired H. pylori from other populations

119 only recently. Paper III demonstrated that in Malaysia and Singapore at least, the predominant HP0521 alleles in the western-type strains differed from that in east-asian type strains, revealing an association between the HP0521 allele and the EPIYA-motif. This novel association suggests a functional correlation between these two loci. Since the function of the HP0521 alleles has not yet been elucidated, this association is intriguing, and warrants further investigation. The prevalence of dupA was also shown to differ significantly between isolates from different geographic regions and among different ethnicities (Papers II and III). While an association with the development of both DU and GC was observed in ethnic Chinese, a novel finding at the time of investigation, it was not replicated in the Swedish population, arguing against a consistent role in disease development. In addition, the previously proposed association with IL-8 induction could not be replicated. The allelic nature of dupA was first described after this study was conducted, and it is possible that the demonstrated lack of variation in dupA in the strains investigated in this study may play a role. As dupA is a virB4 homologous ATPase, it is almost certainly a component of an as-yet unidentified TFSS, a fact that is likely to account for the variable association with the induction of cytokines. Identification of the proteins with which DupA is associated is likely to hold the key.

In Paper IV, H. pylori strains associated with different levels of mucosal inflammation in vivo were not found to have significantly different effects on the induction of IL-8 by gastric epithelial cells in vitro, a typical marker of the inflammatory potential of H. pylori. However, these strains did have a differential impact on DC maturation and cytokine release by DCs. Since DCs are important in determining the Th-type response to H. pylori, these results suggest that H. pylori strains may be able to define the local immune response through their interaction with DCs. However, the interaction between DCs and H. pylori, as well as the precise role of DCs in H. pylori-associated disease remains poorly defined. While the virulence factors cagA and vacA did not appear to play a role in this interaction, it is apparent that there is an intrinsic factor in these strains that does. Recently, dupA has been associated with the induction of IL-12, suggesting a role for this gene in the maturation of DCs. Further characterisation of these strains and the relationship between H. pylori virulence factors and DCs is required. It would also be of significant interest to compare the well-characterised H. pylori strains from the three ethnic groups in Singapore and Malaysia with each other and the Swedish strains for their effects on DC maturation and cytokine release. In

120 addition, future studies should determine whether the mechanisms behind induction of cytokine production by H. pylori in vitro, are the same as those that define the infiltration of immune cells in H. pylori infection, leading to disease development in the susceptible host.

Genetic polymorphisms, which influence gene expression and function, alter the normal response to H. pylori infection, predisposing to persistent infection and disease development. Paper V identified significant variation in a large number of polymorphisms between three ethnic groups (Chinese, Indian and Malay) resident in Malaysia and Singapore, several of which had not been previously assessed in relation to H. pylori and/or GC. Such genetic variation has the potential to explain the differences in H. pylori infection and GC development rates between these ethnic groups. Further work is required to determine whether the associations between the polymorphisms investigated and H. pylori seropositivity or disease status, as identified in the Chinese, a population with a high GC risk, are consistent in Indians, a population with a low GC and high DU risk, and Malays, a population with an unusually low H. pylori prevalence and low GC risk. Due to the low rates of H. pylori in Malays and GC in both Indians and Malays, insufficient numbers of patients were recruited to permit such an analysis in this study. While EBBR2+1963G has been linked to GC previously, we identified a novel association with H. pylori seropositivity. This is certainly worthy of validation and subsequent characterisation, as the biological foundations for such an association are unclear. We also identified associations between the polymorphisms PTGS2-1195G and IL1B-1473C, and GC. The number of patients recruited to this study, combined with the low frequency of many of the polymorphisms investigated, meant that the power of the study was too low to permit the analysis of haplotypes. This is certainly a factor that should be considered in future studies. In addition, recent studies have identified many functional or putatively polymorphisms in genes with a biologically plausible role infection, inflammation and other immune response and carcinogenesis, which would be worth including in future studies, such as polymorphisms in the Nod genes. As microarrays and next generation sequencing technologies become cheaper the potential to assess all polymorphisms in the genome means that these approcahes are likely to replace more traditional methods of polymorphism-typing such PCR-RFLP and mass-spectrometry. This will clearly enable investigation of many new and novel polymorphisms.

121 The next logical step is the simultaneous investigation of host genetic polymorphisms in light of H. pylori virulence factors. Unfortunately the variability in these factors across populations necessitates the recruitment of a large number of patients. To date, we have been unable to recruit sufficient numbers of patients and matched strains to permit such an analysis. Furthermore, the extremely high prevalence of virulence factors in strains from Malaysia and Singapore, irrespective of disease state, means it will be difficult to detect associations relying on the combined effect of bacterial virulence factors and host genetic polymorphisms. Clearly much larger studies are needed.

In conclusion, the work presented in this thesis not only adds substantially to the body of research on H. pylori, but provides novel insights into H. pylori virulence and host genetic factors, particularly in relation to gastroduodenal disease development, revealing several avenues for further investigation.

122 4 Appendix 4.1 Author contributions to publications included in this thesis

Paper I:

HMS: designed the study, prepared the Malaysian isolates for analysis, designed and carried out the molecular genetic studies, performed all data analysis and prepared the manuscript. KMF and KLG: provided clinical specimens. IH and SD: provided patient clinical details. All authors approved the manuscript. DF: conceived of the study. HM: conceived of the study, participated in the design the study, supervised the study and helped prepare the manuscript.

Paper II:

HMS: designed the study, prepared the Malaysian isolates for analysis, designed and carried out the molecular genetic studies, carried out the sequence alignment and analysis, performed the statistical analysis, designed and participated in the in vitro studies, conducted the data analysis and prepared the manuscript. SA: helped design and participated in the in vitro studies. NK: participated in the molecular genetic studies. LEn: supervised the study. LEr: prepared the Swedish isolates for analysis. KMF and KLG: provided clinical specimens. IH and SD: provided patient details. DF: conceived of the study. HM: conceived of the study, participated in the design the study, supervised the study and helped prepare the manuscript. All authors approved the manuscript.

Paper III:

HMS: designed the study, prepared the Malaysian isolates for analysis, designed and carried out the molecular genetic studies, designed and participated in the in vitro studies, conducted the data analysis and prepared the manuscript. SA: helped design and participated in the in vitro studies. NK: participated in the molecular genetic studies. ZK: designed one primer set. LE: supervised the study. KMF and KLG: provided clinical specimens. IH and SD: provided patient details. DF: conceived of the study. HM: conceived of the study, participated in the design the study, supervised the study and helped prepare the manuscript.

123 Paper IV:

SA: designed the study, performed the molecular genetic analysis, designed and conducted the DC-related in vitro studies, participated in and helped design the CagA translocation and AGS cell activation in vitro studies, performed the statistical analysis and prepared the manuscript, HMS: helped design the study, participated in and helped design the CagA translocation and AGS cell activation in vitro studies, assisted with the manuscript preparation. LE, MR, MM and HM: assisted in the design of the study and manuscript preparation. All authors approved the manuscript.

Paper V:

HMS: Designed the study, prepared the DNA, performed the serological studies, performed and assisted with data analysis including statistical analysis and prepared the manuscript. DMH: Performed the PCR-RFLP studies and participated in DNA extraction, FT: Performed the statistical analysis, ZK: Assisted with the serological studies. KLG and KMF: provided patient samples, IH: provided patient demographic data, JKB: Assisted with the statistical analysis, DF: conceived of the study and oversaw the statistical analysis, HM: Conceived of the study, participated in the design of the study, supervised the study, assisted with the serological studies and assisted with manuscript preparation. All authors approved the manuscript.

124