H. Pylori - Resp Imunológ

Clinical Science (2006) 110, 305314 (Printed in Great Britain) doi:10.1042/CS20050232 305

R E V I E W

Immune responses to Helicobacter pyloricolonization: mechanisms and clinical outcomes

Cynthia PORTAL-CELHAY and Guillermo I. PEREZ-PEREZDepartment of Microbiology, NYU School of Medicine, VA Medical Center, 423 East 23rd Street, New York, N.Y. 10010U.S.A., and Departments of Medicine and Microbiology, NYU School of Medicine, VA Medical Center, 423 East 23rd Street,New York, N.Y. 10010 U.S.A.

A B S T R A C T

Helicobacter pylori colonizes the stomachs of half of the worlds population and usually persists inthe gastric mucosa of human hosts for decades or life. Although most H. pylori-positive peopleare asymptomatic, the presence of H. pylori is associated with increased risk for the developmentof peptic ulcer disease, gastric adenocarcinoma and gastric lymphoma. The development of a sus-tained gastric inflammatory and immune response to infection appears to be pivotal for the devel-opment of disease. During its long co-existence with humans, H. pylori has evolved complexstrategies to maintain a mild inflammation of the gastric epithelium while limiting the extent ofimmune effector activity. In this review, the nature of the host immune response to H. pyloriinfection and the mechanism employed by the bacterium to evade them is considered. Under-standing the mechanisms of colonization, persistence and virulence factors of the bacterium aswell as the innate and adaptive immune responses of the host are critically important for the devel-opment of new strategies to prevent the development of H. pylori-induced gastroduodenal disease.

INTRODUCTION

Helicobacter pylori is a Gram-negative microaerophilicspiral bacterium that colonizes the gastrointestinal mu-cosa of its host and, despite a strong persistent humoraland cellular immune response to H. pylori at the local andsystemic level, the organism persists for the lifetime ofits host. Virtually all people carrying H. pylori have co-existing gastric inflammation; however, only a small per-centage of colonized individuals develop clinically ap-parent sequelae.

Chronic gastritis induced by H. pylori increases therisk for a wide spectrum of clinical outcomes, rangingfrom peptic ulcer disease (gastric and duodenal ulcera-tion) to distal gastric adenocarcinoma and gastric mu-

cosal lymphoproliferative diseases, such as non-Hodgkins lymphoma [1]. Although the factors determ-ining the variable outcome of H. pylori infection are notwell understood, the development of a sustained gastricinflammatory and immune response to infection appearsto be pivotal for the development of disease. Enhancedrisk may be related to differences in the expression ofspecific bacterial products, to variations in the host in-flammatory response to the bacteria or to specificinteractions between host and microbe [2].

We are only now beginning to understand the bacterialand host factors that are involved in the hostpathogeninteraction during persistent colonization with H. pylori.The answers to these questions are likely to providenew and exciting directions for research in the fields of

Key words: colonization, gastric mucosa, Helicobacter pylori, immune response, inflammation.Abbreviations: APC, antigen-presenting cell; CI, confidence intervals; DC, dendritic cell; HcpA, Helicobacter cysteine-rich proteinA; IFN- , interferon- ; IL, interleukin; LPS, lipopolysaccharide; MAPK, mitogen-activated protein kinase; NF-B, nuclear factorB; NO, nitric oxide; PAI, pathogenicity island; sIgA, secretory IgA; TLR, Toll-like receptor; TNF-, tumour necrosis factor-;Treg, CD4+CD25+ regulatory T-cells.Correspondence: Dr Guillermo I. Perez-Perez (email [email protected]).

C 2006 The Biochemical Society

306 C. Portal-Celhay and G. I. Perez-Perez

Table 1 H. pylori proteins that may contribute to gastric colonization

Product Gene(s) Putative functions Targeted host defence

Urease ure operon Gastric acid neutralization Gastric acidFlagellae flaA , flaB, flg and flbA Motility Gastric peristalsisAdhesins babA and others Adherence to gastric epithelium Gastric peristalsisLewis antigens galT , futA , futB and futC Adherence to gastric epithelium Gastric peristalsis

? Molecular mimicry Humoral immune response

microbial pathogenesis and immunology. This reviewfocuses on the immune response to H. pylori colonizationand its effect on clinical outcomes.

EPIDEMIOLOGY

H. pylori is the most common chronic bacterial infectionin humans [3]. Infection with H. pylori occurs worldwide,but the prevalence varies greatly among countries andamong population groups within the same country.

H. pylori colonizes the stomachs of 50 % of the popul-ation in developed countries and approx. 80 % in the de-veloping world. The infection is acquired by oral inges-tion of the bacterium and is mainly transmitted withinfamilies. The main source of transmission is the motherwithin families [4]. Acquisition of H. pylori apparentlytakes place in early childhood and is rare or does not occurin adults. In industrialized countries, the rate of acquisi-tion of H. pylori has decreased substantially over recentdecades. Therefore the continuous increase in prevalenceof H. pylori with age is due mostly to a cohort effect, re-flecting more intense transmission at the time when mem-bers of earlier birth cohorts were children [5]. The organ-isms can be cultured from vomitus or diarrhoeal stools,suggesting the potential for transmission among familymembers during periods of illness [6].

The overall prevalence of H. pylori is strongly corre-lated with socioeconomic conditions. Factors such asdensity of housing, overcrowding, number of siblings,birth order, sharing a bed and lack of running waterhave all been linked to a higher acquisition of H. pyloriinfection [7].

PATHOGENESIS

Colonization of the gastric mucosaGastric acidity and peristalsis normally inhibit bacterialcolonization of the human stomach. However, naturalselection has provided H. pylori with several mechanismsto elude these primary defences and establish persistentinfection, such as the ability to withstand acidic gastricpH and motility (Table 1).

After being ingested, H. pylori has to evade thebactericidal activity of the gastric luminal contents and

establish intimate contact with the mucous layer. Theenzyme urease metabolizes urea to carbon dioxide andammonia to buffer the gastric acid. Flagella allow thebacterium to swim across the viscous gastric mucus andreach the more neutral pH below the mucus. KnockoutH. pylori mutants of urease or flagellar genes are defectivein gastric colonization in a gnotobiotic piglet model ofinfection [8,9].

Once below the mucus, H. pylori adheres tightly to theunderlying cells. The best characterized adhesin, BabA,is a 78 kDa outer-membrane protein that binds to thefucosylated Lewisb blood-group antigen.

Persistent colonization depends on the ability to res-pond to changing environmental conditions and circum-vent host defence mechanisms initiated during infection.Rearrangement of genomic DNA allows a variety ofpathogens to adapt expression of surface antigens andevade host immunity. H. pylori has the highest rate ofgenetic recombination of any known bacterial species,suggesting that this process confers a selective advantagein colonization. Loughlin et al. [10] have shown thatH. pylori mutants defective for homologous recom-bination were spontaneously cleared from the murinegastric mucosa, whereas the H. pylori wild-type strainestablished a persistent high level of colonization.

Virulence factorsThe major disease-associated genetic difference inH. pylori isolates is the presence or absence of the cag-PAI(pathogenicity island), a 40 kb genomic fragment con-taining ORFs (open reading frames) that represent 31genes. As in other bacterial pathogens, the PAI has beenacquired by horizontal transfer of a genetic cassette.

The cag-PAI is present in 6070 % of US clinicalstrains. A particular cluster of these genes encodes fora type IV bacterial secretion system which, in addition toplaying a role in the inflammatory response, has beenshown to deliver the immunodominant CagA proteinthat is encoded in the cag-PAI inside the host cells.Once inside the gastric epithelial cell, CagA is tyrosinephosphorylated by Src kinase activity, leading to agrowth-factor-like cellular response and cytokine pro-duction by the host cell [11]. Because of these functions,H. pylori cagA+ strains are associated with a significantlyincreased risk for severe gastritis, atrophic gastritis, peptic


Immune responses to Helicobacter pylori colonization 307

ulcer disease and distal gastric cancer compared withcagA strains [12,13].

A second locus of heterogenicity is the gene vacA,which encodes the vacuolating cytotoxin. VacA is asecreted protein toxin that is responsible for the gastricepithelial erosion observed in infected hosts. Despitethe fact that 100 % of H. pylori strains carry the vacAgene, approx. 50 % of H. pylori strains produce a func-tional VacA protein [14]. H. pylori strains that expressvacuolating activity are more common among patientswith peptic ulcer disease and distal gastric cancer thanamong infected patients with superficial gastritis alone[15]. Although vacA is conserved among all H. pyloristrains, significant polymorphism exists. vacA alleles pos-sess one of two types of signal region, s1 or s2, and one oftwo mid-regions, m1 or m2, occurring in all possible com-binations. H. pylori strains with different forms of vacAexhibit varied phenotypes and have particular associ-ations with gastroduodenal diseases. vacA s1/m1 strainsare most closely associated with gastric carcinoma [16].

HOST RESPONSE TO H. PYLORI

The immune response towards bacterial pathogens canbe divided into an innate and an adaptive response. Theinnate response towards bacterial infection is generallyan initial non-specific process, which reacts quickly withseveral bacterial molecules to signal infectious danger andwith the aim of killing the bacteria. By contrast, the adap-tive immune response is delayed, antigen-specific, leadsto the activation of T-, B- and memory cells and is shapedby the innate immune response.

Innate immunityRecognition of bacterial molecules by the innate immunesystem is mediated by TLRs (Toll-like receptors) expres-sed on APCs (antigen-presenting cells) such as monocytesand DCs (dendritic cells). Bacterial contact with mono-cytes and other APCs leads to the secretion of pro-inflammatory cytokines such as TNF- (tumour necrosisfactor-), IL (interleukin)-1 and IL-8. H. pylori infec-tion has been shown to be associated with increased levelsof these cytokines which, in turn, act as local chemo-attractants, inducing granulocytic infiltration [17].

Thus far, many of the studies on innate immune res-ponses to H. pylori in epithelial cells have focused onTLR4, the specific pathogen-recognition molecule(PRM) of Gram-negative LPS (lipopolysaccharide).However, gastric epithelial cell lines were non-res-ponsive to H. pylori LPS, even when relatively high con-centrations of this endotoxin were added to the cells[18]. Consistent with this, a TLR4-neutralizing antibodydid not block H. pylori-induced secretion of the pro-inflammatory cytokine IL-8 in AGS cells [19].

Two additional TLRs, TLR2 and TLR5, have beenshown to participate actively in innate immune responses

to Gram-negative bacteria through their recognition oflipoproteins and flagellin respectively. A role for TLR2and TLR5 signalling to H. pylori was suggested by astudy showing an increase in NF-B (nuclear factorB) luciferase reporter activity in H. pylori-stimulatedHEK-293 cells that had been co-transfected to expressTLR2, TLR4 or TLR5. Infection of the cultures withH. pylori induced NF-B activity in cells transfectedwith TLR2 and TLR5, but not TLR4 [20]. Other groups,however, demonstrated that flagellin-responsive epi-thelial cell lines do not detect native or recombinantH. pylori flagellin, suggesting that H. pylori flagellinevades TLR5 recognition [21].

Several studies have demonstrated that contact be-tween H. pylori and gastric epithelial cells results in arapid activation of NF-B, which is followed by increasedIL-8 expression [22]. The ability of H. pylori to activateNF-B in vitro has also been corroborated in vivo asactivated NF-B is present within gastric epithelial cellsof infected, but not uninfected, patients, which mirrorsthe location of increased IL-8 protein within colonizedmucosa [23].

In addition to NF-B, MAPKs (mitogen-activatedprotein kinases) have been implicated as mediators ofH. pylori-induced IL-8 expression. In vitro studies util-izing IL-8 reporter constructs have now revealed thatH.pylori-induced IL-8 gene expression is dependent uponactivation of both NF-B and AP-1 (via activation ofMAPKs), indicating that synergistic interactions betweenAP-1 and NF-B within the IL-8 promoter are requiredfor maximal H. pylori-induced IL-8 production [24].

Adaptive immunity

Cellular responseAdaptive immune responses towards H. pylori infectionhave also been identified. H. pylori causes continuousgastric inflammation in virtually all infected people. Thisinflammatory response initially consists of neutrophils,followed by lymphocytes (T- and B-cells), plasma cellsand macrophages, along with varying degrees of epithelialcell degeneration and injury [25].

Since H. pylori rarely, if ever, invades the gastricmucosa, the host response is triggered primarily by theattachment of bacteria to epithelial cells. The organismproduces a number of antigenic substances, includingHSP (heat-shock protein), urease and LPS, all of whichcan be taken up and processed by lamina propia macro-phages and activate T-cells [26]. Cellular disruption, es-peciallyadjacent toepithelial tight junctions,undoubtedlyenhances antigen presentation to the lamina propia andfacilitates immune stimulation. The net result is increasedproduction of inflammatory cytokines such as IL-1, IL-6,TNF- and, most notably, IL-8 [27,28].

Chronic active gastritis is associated with an increasedCD4/CD8 T-cell ratio within the gastric mucosa, due



largely to the accumulation of CD4+ T-helper lympho-cytes in the lamina propia. H. pylori infection resultsin a Th1-predominant host immune response in thegastric mucosa, characterized by the induction of IFN- (interferon- ) and IFN- -related genes. A Th1-pre-dominant immune response is associated with elevatedlevels of the pro-inflammatory cytokines IL-12, IL-18and TNF- [11]. The severity of gastritis associatedwith H. pylori infection was correlated with mucosalexpression of the TNF- subunit CD68 and IFN- [29].A more robust mucosal Th1 response has also beenassociated with progression to atrophic gastritis and gast-ric cancer, as supported by animal models [30].

The host genetic background contributes to the im-mune and inflammatory response to H. pylori infection.The IL-1B gene encodes the expression of IL-1, a potentpro-inflammatory cytokine and powerful inhibitor ofgastric acid secretion that plays a major role in initiatingand amplifying the inflammatory response to H. pyloriinfection [31]. A polymorphic allele with T instead ofC at position 511 of the regulatory region of theIL-1B gene (IL-1B 511T) is associated with in-creased IL-1 production [32]. IL-1RN encodes theIL-1-receptor antagonist, an anti-inflammatory cytokinethat competitively binds to IL-1 receptors and therebymodulates the potentially damaging effects of IL-1 [33].The IL-1RN gene has a penta-allelic 86 bp variable num-ber tandem repeat in intron 2, of which allele 2 (IL-1RN2) is associated with a wide range of chronic inflam-matory and autoimmune conditions and enhancedIL-1 secretion. Hwang et al. [34] have shown that, inH. pylori-infected Japanese subjects, individuals whocarried an IL-1B gene polymorphism (IL-1B 511T/T)or those carrying the IL-1RN2 allele had higher mucosalIL-1 levels than non-carriers. These studies demonstratethat host genetic polymorphisms in inflammatoryresponse genes can influence the nature and extent ofH. pylori-mediated gastritis and thereby disease expres-sion.

Humoral responseIndividuals colonized with H. pylori elicit a strong speci-fic systemic and local antibody response to the infection.Tosi and Czinn [35] reported that binding of IgG toH. pylori promoted phagocytosis and killing in vitroby polymorphonuclear leucocytes. H. pylori strains aresusceptible to complement and activate it either via theclassical pathway, even in the absence of specific anti-bodies, or by the alternative pathway [36,37]. Despite thisvigorous immune response, H. pylori is not eradicatedunless an infected individual is treated with a com-bination of antibiotics, and lifelong chronic infectionusually develops. These observations suggest that gastricmucus may be a protective niche in which H. pyloriexist and are relatively inaccessible to specific antibodiesor their effector functions. The ineffective humoral res-

ponse generated towards H. pylori and its componentsmay actually contribute to pathogenesis. Some of themonoclonal antibodies directed against H. pylori cross-react with gastric epithelium of both mice and humansand delivery of these antibodies alone to mice can inducegastritis [2].

Clyne et al. [38] have shown that human serum fromboth infected and non-infected subjects exerted a bac-tericidal effect on H. pylori. Furthermore heat inactiv-ation abolished the killing effect of the serum samples onthe organism, strongly suggesting that it was complementmediated. Later it was reported that bovine normal serumand serum from immunized cows is highly bactericidalfor H. pylori. This bactericidal effect is destroyed byheating to 56 C for 30 min and restored by the additionof fetal calf serum as a source of complement, indicatingthat the bactericidal effect is probably dependent on anantibodycomplement system. However, the bactericidalactivity did not correlate with titres of specific antibodyor with IgG concentrations [39]. In the study by Clyneet al. [38], serum samples from infected subjects killed theorganism more effectively than serum from non-infectedindividuals, indicating that some of this effect is mediatedby the classical pathway.

Sampling of gastric secretions from H. pylori-infectedindividuals also reveals an active mucosal antibody res-ponse, primarily of the IgA isotype. This response isconsistent with the predominance of sIgA (secretory IgA)in the gastric secretions of healthy individuals. sIgA anti-(H. pylori) antibodies are also found in saliva and breastmilk [11].

Thomas et al. [40] have reported that children inGambia who were breastfed by mothers that had hightitres of specific anti-(H. pylori) sIgA in their milk wereprotected from infection for a longer period than childrenwhose mothers had lower anti-(H. pylori) antibody titres.It was postulated that specific antibodies in human milkpassed from mother to child during breast feeding areprotective against infection by H. pylori; however, mostchildren were infected by 12 months of age. Other studieshave questioned the role of breast feeding as protectivefactor in H. pylori acquisition. The relationship betweeninfection and childhood home environment has beenevaluated in German [41], Japanese [42] and Italian [43]populations. No association was seen between H. pyloriseropositivity and a history of breast feeding in any of thethree studies.

Studies have shown that sIgA can interfere with theability of some enteric pathogens to establish infection[44,45]. Others have shown that sIgA can inhibit bacterialadherence [46]. However, Clyne et al. [38] have shownthat the systemic antibody response and the sIgA res-ponse against H. pylori do not inhibit the organism fromadhering to gastric cells in vitro. This may explain whychronic infection develops in infected subjects despite avigorous immune response.



IMMUNOMODULATION BY H. PYLORI

To maintain prolonged colonization of the human gastricmucosa, H. pylori must avoid both innate and adaptiveimmune responses.

Following infection with H. pylori, DCs phagocytosebacterial proteins and express peptides on their surface,together with MHC and co-stimulatory molecules. Thispresentation of antigens effectively leads to the activationof CD4+ T-cells that react towards these antigens andtrigger the immune response. Naive CD4 T-cells candifferentiate upon activation into either Th1 or Th2 cells,which differ in the type of cytokines that they produceand therefore in their function. The factors that deter-mine whether a proliferating CD4 T-cell in vivo will dif-ferentiate into a Th1 or a Th2 cell are not fully understood.The cytokines elicited by infectious agents (principallyIFN- , IL-2 and IL-4), the co-stimulators used to drivethe response and the nature of the peptideMHC ligandall have an effect [47].

The consequences of inducing Th1 versus Th2 cellsare profound: the selective production of Th1 cells leadsto cell-mediated immunity and the production of opso-nizing antibody classes (predominantly IgG), whereas theproduction of predominantly Th2 cells provides humoralimmunity, especially IgM, IgA and IgE.

Most intracellular bacteria induce Th1 responses,whereas extracellular pathogens stimulate Th2-type res-ponses. Based on the fact that H. pylori is non-invasiveand that infection is accompanied by an exuberant humo-ral response, one might predict that a Th2 responsewould be predominant within H. pylori-colonized gastricmucosa. Paradoxically, the majority of H. pylori antigen-specific T-cell clones isolated from infected gastricmucosa produce higher levels of IFN- than IL-4, whichis reflective of a Th1-type response [48]. H. pylori alsostimulates production of IL-12 in vitro, a cytokinethat promotes Th1 differentiation. These findings raisethe hypothesis that an aberrant host response (Th1)to an organism predicted to induce secretory immuneresponses (Th2) may influence and perpetuate gastricinflammation. Animal models of H. pylori-induced gast-ritis have supported these conclusions [49]. Furthermore,a bacterial factor contributing to the initiation of Th1polarization of H. pylori-specific immune response wasrecently characterized. Deml et al. [50] identified HcpA(Helicobacter cysteine-rich protein A) as a novel pro-inflammatory and Th1-promoting protein. Using spleniccells of H. pylori-negative naive mice, the authors foundthat HcpA induces a substantial secretion of pro-inflam-matory (IL-6 and TNF-) and Th1-promoting cytokines(IL-12 and IFN- ), but no significant release of IL-2 andthe Th2-promoting cytokines IL-4 and IL-5.

A major mechanism for self/non-self discrimination bythe immune system and establishment of self-toleranceis the clonal deletion of self-reactive T- and B-cells ex-

posed to self-antigens during development in the thy-mus [51]. The deletion mechanism is not complete, how-ever, and potentially hazardous self-reactive lymphocytesare present in the periphery of normal individuals. It hasbecome increasingly evident that active suppression ofself-reactive T-cells by regulatory T-cells takes place inthe periphery of normal individuals avoiding the onsetof harmful autoimmunity. Three phenotypically distinctsubsets of suppressor-regulatory T-cell have been de-scribed based on one or more surface-marker antigensand/or cytokine-production profiles: the naturalCD4+CD25+ regulatory T-cells (Treg) [52], the IL-10-secreting Tr1 cells [53] and the TGF- (transforminggrowth factor-)-secreting Th3 cells [54], which func-tionally both in vitro and in vivo have been shown tosuppress the proliferation and cytokine secretion of effec-tor T-cells.

The recognition of an important role for regulat-ory T-cells in the suppression of pathogen-induced in-flammatory responses has just started to emerge. Recentevidence suggests that the activation of regulatoryT-cells, including both Treg and Tr1 cells, might resultin decreased pathological responses and prolonged per-sistence of infection as a mechanism for the maintenanceof pathogen-specific immunological memory [55].

Treg seem to play a role in modulating inflammation inH. pylori infection. A recent study [56] has demonstratedthat infection of athymic nu/nu mice with H. pylori re-sulted in considerably lower colonization in mice recon-stituted previously with lymph node cells depleted ofTreg compared with those reconstituted with controlcells. In another study, Lundgren et al. [57] observed thatmemory T-cell responses were increased upon specificantigen stimulation of peripheral blood lymphocytestaken from H. pylori-infected subjects when the cellpopulation was depleted of Treg. These studies indicatethat Treg may reduce immunopathology in H. pylorigastritis, possibly by reducing activation of IFN- -producing CD4+ T-cells, but at the expense of a higherH. pylori bacterial load.

EVASION OF IMMUNE RESPONSE BYH. PYLORI (Figure 1)

Evasion of innate responseNO (nitric oxide) is a key component of the innateimmune system and an effective antimicrobial agent [58].Gobert et al. [59] have shown that H. pylori preventsNO production by host cells by producing the enzymearginase. This enzyme is encoded by the gene rocF, andcompetes with NOS2 (NO synthase 2) for the substratel-arginine and converts it into urea and l-ornithine,rather than NO. Mutation of rocF results in efficient kill-ing of the bacteria in an NO-dependent manner, whereaswild-type bacteria survive under these conditions.



Figure 1 H. pylori pathogenesis and the inflammatory responseH. pylori resides in the gastric lumen and colonizes the gastric epithelium using urease. Binding of H. pylori to epithelial cells and injection of CagA results in theproduction of IL-8 and other chemokines, and activation of the innate and adaptive immune systems. To evade the immune response, H. pylori has evolved mechanismsto reduce recognition by immune sensors, down-regulate activation of immune cells and escape immune effectors. PMN, polymorphonuclear cells.

Furthermore, the rocF mutant is mildly attenuated in itsability to colonize mice.

It has been shown that, during H. pylori infection,bacteria can survive intracellularly within macrophagesby interfering with lysosomal proteins, similar to Myco-bacterium tuberculosis [60]. Therefore, although thereis an innate response to the bacteria, it is not effectiveenough to eliminate the infection. In one study, Allenet al. [61] have shown a delayed uptake of bacteria intomacrophages followed by the formation of megasomes asa result of phagosome fusion. These megasomes protectintracellular bacteria from efficient killing. Furthermore,using human blood monocytes and polymorphonuclearcells, Ramarao et al. [62] have shown that H. pylori canactively block its own uptake, as well as the uptake of co-cultured bacteria of other species and latex beads. Both ofthese phenotypes depended on the presence of Cag-PAI.

Bacterial LPSs are classic mediators of inflammation,because of their activation of phagocytic cells, endothelialand epithelial cells and lymphocytes [63]. However, de-spite a general conservation of LPS structure, large differ-ences in their pro-inflammatory activity have been noted.When compared with LPS from Enterobacteriaceae,H. pylori LPS is 1000-fold less active and only weaklyactivates macrophages [64]. Therefore it has been sug-

gested that there is selective pressure in H. pylori cellsto minimize pro-inflammatory activities to permit long-term colonization, since enhanced inflammation, leadingto atrophic gastritis, would lead to a loss of niche.

Finally, it has been recently demonstrated [65] thatmembers of the and Proteobacteria, includingH. pylori, possess flagellin molecules that cannot berecognized by TLR5. Their unique flagellin sequencescontain amino acid differences in the TLR5 recognitionsite that permit TLR5 evasion, as well as compensatorymutations that preserve bacterial motility. These resultssuggest that TLR5 evasion is critical for the survival ofH. pylori at mucosal sites.

Evasion of adaptive responseH. pylori has evolved to subvert not only the innate, butalso the adaptive immune response. It has been shownthat H. pylori specifically can block antigen-dependentproliferation of T-cells. This effect is mediated by thevirulence factor VacA, which acts as an immunomodu-lator by interfering with the IL-2 signalling pathway inT-cells by blocking Ca2+ mobilization and the activity ofthe Ca2+/calmodulin-dependent phosphatase calcineurin[66]. Another possible function of VacA in subvertingthe adaptive immune response is its ability to interfere



Figure 2 H. pylori-induced diseases in humansH. pylori is acquired in early childhood, and it may take decades to lead to symptomatic disease. The clinical course is highly variable and depends on bacterial andhost factors. High acid outputs predispose patients to duodenal ulcers, whereas lower acid outputs are associated with gastric atrophy and cancer.

with antigen presentation mediated by MHC class II[67]. A recent study [68] strongly supports the possibilitythat VacA is immunosuppressive, but the mechanism de-scribed involves a direct action on T-cells rather thanAPCs. The authors [68] show that the toxin inhibits theactivation and proliferation of T-cells.

Several findings indicate that a strong inflammatoryreaction is necessary for the elimination of H. pylori and,at least in animal models, it seems to be actively repressedby the bacterium: an increased inflammatory reaction, asseen in IL-10 knockout mice, is associated with clearanceof H. pylori from the stomach within 8 days of theinfection [69]. Similarly, increased inflammation due todeletion of the gene encoding PHOX (NADPH oxidase)results in a marked reduction in bacterial numbers [70].

Genomic DNA recombination has been shown to becritical for mediating persistence of H. pylori throughthe induction of ineffective immune responses. Robinsonet al. [71] found that the H. pylori wild-type strain andthe ruvC mutant (which encodes a Holliday junctionresolvase) elicited oppositely polarized Th2 cell responsesthat correlated with persistence and clearance respect-ively. Temporary colonization by the ruvC mutant con-ferred significant protection against subsequent challengewith the wild-type strain.

CLINICAL OUTCOMES

Everyone who carries H. pylori in their stomach developsa cellular infiltrate in their gastric mucosa, termed chronic

gastritis (Figure 2). In most patients (80 %), H. pyloridoes not cause clinical symptoms and the infection canpersist for a lifetime without further problems. A pro-portion (1020 %) of infected patients will develop gastrichyperacidity and peptic ulcers, but can be cured by anti-biotic treatment. However a small percentage (0.1 4 %)of infected patients will develop distal gastric adeno-carcinoma, depending on the circumstances of the in-fection and the individual immune response towards thebacterium.

Although the factors determining this variable out-come are not well understood, the development of asustained gastric inflammatory and immune response toinfection appears to be pivotal for the development ofdisease [72]. Mucosal T-cells in infected individuals arepolarized towards the production of IFN- , rather thanIL-4 or IL-5, indicating a strong bias towards a Th1-typeresponse [48].

Bacterial determinants of virulence are considered cri-tical for initiating close interactions with host epithelialcells and inducing mucosal inflammation. H. pyloristrains containing the cag-PAI are associated with moresevere antral inflammation, higher mucosal levels of IL-1 and IL-8 [73], peptic ulceration [74] and increased acidsecretion [75].

Other virulence-independent factors, such as the dura-tion and density of infection, may also influence the out-come, but have been studied less well. Surveys of the earlyphases of infection (in children) are few, but indicate that apro-inflammatory response and Th1-type cytokines were



detected in response to infection [76,77]. In contrast withadults, in whom neutrophilic infiltrates predominate,children develop a predominantly mononuclear infiltrateand have a higher degree of lymphoid follicular hyper-plasia. The mechanisms underlying the differences inhistopathology and disease pattern between children andadults and the early and late stages of infection are stillnot well understood.

Recent studies have begun to shed light on specifichost genetic determinants of risk for H. pylori-associateddisease outcomes. Polymorphisms that increase the IL-1response to H. pylori are associated with an increased riskof developing hypochlorhydria, gastric atrophy and ad-enocarcinoma [78,79]. In a large population-based case-control study of gastrointestinal carriers, El Omaret al. [80] recently demonstrated that pro-inflammatorygenotypes of TNF- and IL-10 were each associated withmore than doubling the risk of noncardia gastric cancerand that carriage of multiple pro-inflammatory polymor-phisms conferred even greater risks.

When the influence of CagA and VacA status on histo-logical changes was analysed in subjects stratified accord-ing to IL-1B and IL-RN polymorphisms, it appeared thatmore severe gastric abnormalities correlated both withhigh-risk polymorphisms (IL-1B 511T and IL-RN2alleles) and virulence factors (cagA+ /vacAs1+).In vacAs1/IL-1B 511T patients and cagA-positive/IL1RN2/2, the odds ratios for gastric carcinoma were87 [95 % CI (confidence intervals), 11679] and 23 res-pectively (95 % CI, 7.072) [81].

CONCLUSIONS

H. pylori continues to be one of the most common bac-terial infections in humans, colonizing the stomach andpersisting for the hosts lifetime. Although the humanhost mounts a vigorous innate and adaptive immune re-sponse against the bacterium, H. pylori escapes and evadeshost responses by a variety of mechanisms, leading topersistent colonization and chronic active inflammation.An intriguing characteristic of H. pylori-induced gastritisis its capacity to persist for decades without causingserious damage in most cases. In its long association withhumans, H. pylori has established a fine balance betweenestablishing a comfortable niche and avoiding the immuneconsequences of its colonization. Clinical complicationsof H. pylori colonization, such as peptic ulcer disease andgastric cancer, are therefore likely to represent imbalancesin gastric homoeostasis that are disadvantageous for bothmicrobe and host, particularly if death of the host ensues.Considerable efforts have focused on delineating theprecise mechanisms by which H. pylori colonizationleads to gastric inflammation. It is likely that thepolarized Th1 responses and the inflammatory cytokinesinduced by H. pylori gene products play major roles

in the gastric epithelial inflammation associated withH. pylori disease. Genetic or environmentally influencedvariability in the propensity to mount these types of in-flammatory responses may, at least in part, help to explainthe different outcomes of H. pylori-induced disease indifferent individuals.

ACKNOWLEDGMENTS

This study was supported by RO3 CA099512-01A1 fromthe National Cancer Institute (National Institutes ofHealth), by the NYU School of Medicine Institute forUrban and Global Health, and by grant 02816-9 from theThrasher Research Fund.

REFERENCES

1 Parsonnet, J., Friedman, G. D., Vandersteen, D. P. et al.(1994) Helicobacter pylori infection and gastric lymphoma.N. Engl. J. Med. 330, 12671271

2 Israel, D. A. and Peek, R. M. (2001) Pathogenesis ofHelicobacter pylori-induced gastric inflammation.Aliment. Pharmacol. Ther. 15, 12711290

3 Cave, D. R. (1996) Transmission and epidemiology ofHelicobacter pylori. Am. J. Med. 100, 12S17S

4 Perez-Perez, G. I., Rothenbacher, D. and Brenner, H.(2004) Epidemiology of Helicobacter pylori infection.Helicobacter 9, 1S6S

5 Banatvala, N., Mayo, K., Megraud, F., Jennings, R., Deeks,J. J. and Feldman, R. A. (1993) The cohort effect andHelicobacter pylori. J. Infect. Dis. 168, 219221

6 Parsonnet, J., Shmuely, H. and Haggerty, T. (1999) Fecaland oral shedding of Helicobacter pylori from healthyinfected adults. JAMA, J. Am. Med. Assoc. 282,22402245

7 Webb, P. M., Knight, T., Greaves, S. et al. (1994) Relationbetween infection with Helicobacter pylori and livingconditions in childhood: evidence for person to persontransmission in early life. Br. Med. J. 308, 750753

8 Eaton, K. A. and Krakowka, S. (1994) Effect of gastric pHon urease-dependent colonization of gnotobiotic piglets byHelicobacter pylori. Infect. Immun. 62, 36043607

9 Eaton, K. A., Suerbaum, S., Josenhans, C. andKrakowka, S. (1996) Colonization of gnotobiotic pigletsby Helicobacter pylori deficient in two flagellingenes. Infect. Immun. 64, 24452448

10 Loughlin, M. F., Barnard, F. M., Jenkins, D., Sharples, G. J.and Jenks, P. J. (2003) Helicobacter pylori mutants defectivein RuvC Holliday junction resolvase display reducedmacrophage survival and spontaneous clearance from themurine gastric mucosa. Infect. Immun. 71, 20222031

11 Tummala, S., Keates, S. and Kelly, C. P. (2004) Update onthe immunologic basis of Helicobacter pylori gastritis.Curr. Opin. Gastroenterol. 20, 592597

12 Kuipers, E. J., Perez-Perez, G. I., Meuwissen, S. G. et al.(1995) Helicobacter pylori and atrophic gastritis:importance of the cagA status. J. Natl. Cancer Inst. 87,17771780

13 Blaser, M. J., Perez-Perez, G. I., Kleanthous, H. et al.(1995) Infection with Helicobacter pylori strains possessingcagA is associated with an increased risk of developingadenocarcinoma of the stomach. Cancer Res. 55, 21112115

14 Cover, T. L. and Blaser, M. J. (1992) Purification andcharacterization of the vacuolating toxin from Helicobacterpylori. J. Biol. Chem. 267, 1057010575

15 Atherton, J. C., Cao, P., Peek, R. M. et al. (1995) Mosaicismin vacuolating cytotoxin alleles of Helicobacter pylori.Association of specific vacA types with cytotoxinproduction and peptic ulceration. J. Biol. Chem. 270,1777117777



16 Miehlke, S., Kirsch, C., Agha-Amiri, K. et al. (2000) TheHelicobacter pylori vacA s1 m1 genotype and cagA isassociated with gastric carcinoma in Germany.Int. J. Cancer 87, 322327

17 Crabtree, J. E. (1996) Immune and inflammatory responsesto Helicobacter pylori infection. Scand. J. Gastroenterol.215, 3S10S

18 Backhed, F., Rokbi, B., Torstensson, E. et al. (2003) Gastricmucosal recognition of Helicobacter pylori is independentof Toll-like receptor 4. J. Infect. Dis. 187, 829836

19 Su, B., Ceponis, P. J., Lebel, S., Huynh, H. and Sherman,P. M. (2003) Helicobacter pylori activates Toll-likereceptor 4 expression in gastrointestinal epithelial cells.Infect. Immun. 71, 34963502

20 Smith, Jr, M. F., Mitchell, A., Li, G. et al. (2003) Toll-likereceptor (TLR) 2 and TLR5, but not TLR4 are required forHelicobacter pylori-induced NF-B activation andchemokine expression by epithelial cells. J. Biol. Chem.278, 3255232560

21 Gewirtz, A. T., Yu, Y., Krishna, U. S., Israel, D. A., Lyons,S. L. and Peek, R. M. (2004) Helicobacter pylori flagellinevades toll-like receptor 5-mediated innate immunity.J. Infect. Dis. 189, 19141920

22 Sharma, S. A., Tummuru, M. K., Blaser, M. J. et al. (1998)Activation of IL-8 gene expression by Helicobacter pyloriis regulated by transcription factor nuclear factor-B ingastric epithelial cells. J. Immunol. 160, 24012407

23 Keates, S., Hitti, Y. S., Upton, M. et al. (1997) Helicobacterpylori infection activates NF-B in gastric epithelial cells.Gastroenterology 113, 10991109

24 Aihara, M., Tsuchimoto, D., Takizawa, H. et al. (1997)Mechanisms involved in Helicobacter pylori-inducedinterleukin-8 production by a gastric cancer cell line,MKN45. Infect. Immunol. 65, 32183224

25 Goodwin, C. S., Armstrong, J. A. and Marshall, B. J. (1986)Campylobacter pyloridis, gastritis, and peptic ulceration.J. Clin. Pathol. 39, 353365

26 Di Tommaso, A., Xiang, Z., Bugnoli, M. et al. (1995)Helicobacter pylori-specific CD4+ T-cell clones fromperipheral blood and gastric biopsies. Infect. Immun. 63,11021106

27 Crabtree, J. E. (1996) Gastric mucosal inflammatoryresponses to Helicobacter pylori. Aliment. Pharmacol.Ther. 10, 1S10S

28 Fan, X. G., Chua, A., Fan, X. J. and Keeling, P. W. (1995)Increased gastric production of interleukin-8 and tumornecrosis factor in patients with Helicobacter pyloriinfection. J. Clin. Pathol. 48, 133136

29 Lehmann, F. S, Terraiano, L., Carena, I. et al. (2002) In situcorrelation of cytokine secretion and apoptosis inHelicobacter pylori-associated gastritis. Am. J. Physiol.Gastrointest. Liver Physiol. 283, G481G488

30 Houghton, J., Fox, J. G. and Wang, T. C. (2002)Gastric cancer: laboratory bench to clinic.J. Gastroenterol. Hepatol. 17, 495502

31 Noach, L. A., Bosma, N. B., Jansen, J., Hoek, F. J.,van Deventer, S. J. and Tygat, G. N. (1994) Mucosal tumornecrosis factor-, interleukin-1, and interleukin-8production in patients with Helicobacter pylori infection.Scand. J. Gastroenterol. 29, 425429

32 El-Omar, E. M., Carrington, M., Chow, W. H. et al. (2000)Interleukin-1 polymorphisms are associated with increasedrisk of gastric cancer. Nature (London) 404, 398402

33 Basso, D., Scrigner, M., Toma, A. et al. (1996) Helicobacterpylori infection enhances mucosal interleukin-1,interleukin-6, and the soluble receptor of interleukin-2.Int. J. Clin. Lab. Res. 26, 207210

34 Hwang, I. R., Kodama, T., Kikuchi, S. et al. (2002) Effectof interleukin 1 polymorphisms on gastric mucosalinterleukin 1beta production in Helicobacter pyloriinfection. Gastroenterology 123, 17931803

35 Tosi, M. F. and Czinn, S. J. (1990) Opsonic activity ofspecific human IgG against Helicobacter pylori. J. Infect.Dis. 162, 156162

36 Gonzalez-Valencia, G., Perez-Perez, G. I., Washburn,R. G. and Blaser, M. J. (1996) Susceptibility of Helicobacterpylori to the bactericidal activity of human serum.Helicobacter 1, 2833

37 Berstad, A. E, Holbjorn, K., Bukholm, G., Moran, A. P.and Brandtzaeg, P. (2001) Complement activation directlyinduced by Helicobacter pylori. Gastroenterology 120,11081116

38 Clyne, M., Thomas, J. E., Weaver, L. and Drumm, B.(1997) In vitro evaluation of the role of antibodies againstHelicobacter pylori in inhibiting adherence of the organismto gastric cells. Gut 40, 731738

39 Korhonen, H., Syvaoja, E. L., Ahola-Luttila, H. et al.(1995) Bactericidal effect of bovine normal and immuneserum, colostrums and milk against Helicobacter pylori.J. Appl. Bacteriol. 78, 655662

40 Thomas, J. E., Austin, S., Dale, A. et al. (1993) Protectionby human milk IgA against Helicobacter pylori infection ininfancy. Lancet 342, 121

41 Rothenbacher, D., Bode, G. and Brenner, H. (2002)History of breastfeeding and Helicobacter pylori infectionin pre-school children: results of a population-based studyfrom Germany. Int. J. Epidemiol. 31, 632637

42 Ueda, M., Kikuchi, S., Kasugai, T., Shunichi, T. andMiyake, C. (2003) Helicobacter pylori risk associated withchildhood home environment. Cancer Sci. 94, 914918

43 Dore, M. P., Malaty, H. M., Graham, D. Y., Faniciulli, G.,Delitala, G. and Realdi, G. (2002) Risk factors associatedwith Helicobacter pylori infection among children in adefined geographic area. Clin. Infect. Dis. 35, 240245

44 Michetti, P., Mahan, M. J., Slauch, J. M., Mekalanos, J. J.and Neutra, M. R. (1992) Monoclonal secretoryimmunoglobulin A protects mice against oral challengewith the invasive pathogen Salmonella typhimurium.Infect. Immun. 60, 17681792

45 Ruiz-Palacios, G. M., Calva, J. J., Pickering, L. K. et al.(1990) Protection of breast-fed infant againstCampylobacter diarrhea by antibodies in human milk.J. Pediatr. 116, 707713

46 Cravioto, A., Tello, A, Villafan, H., Ruiz, J., del Vedovo, S.and Neeser, J. R. (1991) Inhibition of localized adhesionsof enteropathogenic Escherichia coli to Hep-2 cells byimmunoglobulin and oligosaccharide fractions of humancolostrums and breast milk. J. Infect. Dis. 163, 12471255

47 Janeway, C. A., Travers, P., Walport, M. and Shlomchik, M.(2005) Immunobiology: the Immune System in Health andDisease, 6th edn, Garlan Science Publishing, New York

48 Bamford, K. B, Fan, X., Crowe, S. E. et al. (1998)Lymphocytes in the human gastric mucosa duringHelicobacter pylori have a T helper cell 1 phenotype.Gastroenterology 114, 482492

49 Sawai, N., Kita, M., Kodama, T. et al. (1999) Role of interferon in Helicobacter pylori induced gastricinflammatory responses in a mouse model. Infect. Immun.67, 279285

50 Deml, L., Aigner, M., Decker, J. et al. (2005)Characterization of the Helicobacter pylori cysteine-richprotein A as a T-helper cell type 1 polarizing agent.Infect. Immun. 73, 47324742

51 Medzhitov, R. and Janeway, Jr, C. A. (2002) Decoding thepatterns of self and nonself by the innate immune system.Science 296, 298300

52 Sakaguchi, S., Sakaguchi, N., Asano, M., Itoh, M. andToda, M. (1995) Immunologic self-tolerance maintained byactivated T cells expressing IL-2 receptor -chains CD25.Breakdown of a single mechanism of self-tolerance causesvarious autoimmune diseases. J. Immunol. 155, 11511164

53 Groux, H., OGarra, A., Bigler, M. et al. (1997) A CD4+T-cell subset inhibits antigen-specific T-cell responses andprevents colitis. Nature (London) 389, 737742

54 Weiner, H. L. (2001) Induction and mechanism of action oftransforming growth factor--secreting Th3 regulatorycells. Immunol. Rev. 182, 207214

55 Raghavan, S. and Holmgren, J. (2005) CD4+ CD25+suppressor T cells regulate pathogen induced inflammationand disease. FEMS Immunol. Medical Microbiol. 44,121127

56 Raghavan, S., Fredriksson, M., Svennerholm, A. M. et al.(2003) Absence of CD4+ CD25+ regulatory T cells isassociated with a loss of regulation leading to increasedpathology in Helicobacter pylori-infected mice.Clin. Exp. Immunol. 132, 393400



57 Lundgren, A., Suri-Payer, E., Enarsson, K. et al. (2003)Helicobacter pylori-specific CD4+CD25 high regulatoryT cells suppress memory T cell responses to H. pylori ininfected individuals. Infect. Immun. 71, 17551762

58 Nathan, C. and Shiloh, M. U. (2000) Reactive oxygen andnitrogen intermediates in the relationship betweenmammalian hosts and microbial pathogens. Proc. Natl.Acad. Sci. U.S.A. 97, 88418848

59 Gobert, A. P., McGee, D. J., Akhtar, M. et al. (2001)Helicobacter pylori arginase inhibits nitric oxideproduction by eukaryotic cells: a strategy for bacterialsurvival. Proc. Natl. Acad. Sci. U.S.A. 98, 1384413849

60 Ramarao, N. and Meyer, T. F. (2001) Helicobacter pyloriresists phagocytosis by macrophages: quantitativeassessment by confocal microscopy and fluorescence-activated cell sorting. Infect. Immun. 69, 26042611

61 Allen, L. A., Schlesinger, L. S. and Kang, B. (2000) Virulentstrains of Helicobacter pylori demonstrate delayedphagocytosis and stimulate homotypic phagosome fusionin macrophages. J. Exp. Med. 191, 115128

62 Ramarao, N., Gray-Owen, S. D., Backert, S. and Meyer,T. F. (2000) Helicobacter pylori inhibits phagocytosis byprofessional phagocytes involving type IV secretioncomponents. Mol. Microbiol. 37, 13891404

63 Rietschel, E. T., Brade, O., Brandenburg, K. et al. (1987)Chemical structure and biologic activity of bacterial andsynthetic lipid A. Rev. Infect. Dis. 9 (Suppl.), 527536

64 Perez-Perez, G. I., Shepherd, V. L., Morrow, J. D. andBlaser, M. J. (1995) Activation of human THP-1 cellsand rat bone marrow-derived macrophages byHelicobacter pylori lipopolysaccharide. Infect. Immun.63, 11831187

65 Andersen-Nissen, E., Smith, K. D., Strobe, K. L. et al.(2005) Evasion of Toll-like receptor 5 by flagellatedbacteria. Proc. Natl. Acad. Sci. U.S.A. 102, 92479252

66 Boncristiano, M., Paccani, S. R., Barone, S. et al. (2003) TheHelicobacter pylori vacuolating toxin inhibits T cellactivation by two independent mechanisms. J. Exp. Med.198, 18871897

67 Molinari, M., Salio, M., Galli, C. et al. (1998) Selectiveinhibition of Ii-dependent antigen presentation byHelicobacter pylori toxin VacA. J. Exp. Med. 187, 135140

68 Gebert, B., Fischer, W., Weiss, R., Hoffmann, R. andHaas, R. (2003) Helicobacter pylori vacuolating cytotoxininhibits T-lymphocyte activation. Science 301, 10991102

69 Ismail, H. F., Zhang, J., Lynch, R. G., Wang, Y. and Berg,D. J. (2003) Role for complement in development ofHelicobacter-induced gastritis in interleukin-10-deficientmice. Infect. Immun. 71, 71407148

70 Blanchard, T. G., Yu, F., Hsieh, C. L. and Redline, R. W.(2003) Severe inflammation and reduced bacteria load inmurine helicobacter infection caused by lack of phagocyteoxidase activity. J. Infect. Dis. 187, 16091615

71 Robinson, K., Louglin, M. F., Potter, R. and Jenks, P. J.(2005) Host adaptation and immune modulation aremediated by homologous recombination in Helicobacterpylori. J. Infect. Dis. 191, 579587

72 Cover, T. L., Berg, D. E., Blaser, M. J. and Mobley, H. L. T.(2001) Helicobacter pylori pathogenesis, In Principles ofBacterial Pathogenesis (Groisman, E. A., ed.), pp. 509558,Academic Press, New York

73 Peek, R. M., Miller, G. G., Tham, K. T. et al. (1995)Heightened inflammatory response and cytokineexpression in vivo to cagA+ Helicobacter pylori strains.Lab. Invest. 73, 760770

74 Van Doorn, L. J., Figueiredo, C., Sanna, R. et al. (1998)Clinical relevant of the cagA, vacA, and iceA status ofHelicobacter pylori. Gastroenterology 115, 5866

75 El-Omar, E. M., Penman, I. D., Ardill, J. E., Chittajallu,R. S., Howie, C. and McColl, K. E. (1995) Helicobacterpylori infection and abnormalities of acid secretion inpatients with duodenal ulcer disease. Gastroenterology109, 681691

76 Guiraldes, E., Duarte, I., Pena, A. et al. (2001)Proinflammatory cytokine expression in gastric tissuefrom children with Helicobacter pylori-associated gastritis.J. Pediatr. Gastroenterol. Nutr. 33, 127132

77 Luzza, F., Parrello, T., Sebkova, L. et al. (2001) Expressionof proinflammatory and Th1 but not Th2 cytokines isenhanced in gastric mucosa of Helicobacter pylori infectedchildren. Dig. Liver Dis. 33, 1420

78 Furuta, T., El-Omar, E. M., Xiao, F., Shirai, N.,Takashima, M. and Sugimura, H. (2002) Interleukin 1polymorphisms increase risk of hypochlorhydria andatrophic gastritis and reduce risk of duodenal ulcerrecurrent in Japan. Gastroenterology 123, 92105

79 Machado, J. C., Figueiredo, C., Canedo, P. et al. (2003)A proinflammatory genetic profile increases the risk forchronic atrophic gastritis and gastric carcinoma.Gastroenterology 125, 364371

80 El-Omar, E. M., Rabkin, C. S, Gammon, M. D. et al. (2003)Increased risk of noncardia gastric cancer associated withproinflammatory cytokine gene polymorphisms.Gastroenterology 124, 11931201

81 Figueiredo, C., Machado, J. C., Pharoah, P. et al. (2002)Helicobacter pylori and interleukin 1 genotyping: anopportunity to identify high-risk individuals for gastriccarcinoma. J. Natl. Cancer Inst. 94, 16801687

Received 28 July 2005/15 September 2005; accepted 10 October 2005Published on the Internet 10 February 2006, doi:10.1042/CS20050232


Documents

H. Pylori - Resp Imunológ