Helicobacter pylori-induced epithelial cell signalling in gastric carcinogenesis

Transcription

1 Review TRENDS in Microbiology Vol.12 No.1 January Helicobacter pylori-induced epithelial cell signalling in gastric carcinogenesis Michael Naumann 1 and Jean E. Crabtree 2 1 Experimental Internal Medicine, Medical Faculty, Otto-von-Guericke-University, Magdeburg, Germany 2 Molecular Medicine Unit, St. James s University Hospital, Leeds, UK LS9 7TF Helicobacter pylori represents a highly successful human microbial pathogen that infects the stomach of more than half of the world s population. H. pylori induces gastric inflammation, and the diseases that can follow such infection include chronic gastritis, peptic ulcers and, more rarely, gastric cancer. The reasons why a minority of patients with H. pylori develops gastric cancer could be related to differences in host susceptibility, environmental factors and the genetic diversity of the organism. This review examines the features of H. pylori-induced epithelial cell signalling in gastric diseases. Clinical studies and animal models, and also evidence for H. pylori strain-related differences in gastric epithelial cell proliferation in vivo are discussed. In addition, the mechanisms by which H. pylori triggers hyperproliferative processes and takes direct command of epithelial cell signalling, including activation of tyrosine kinase receptors, cell cell interactions and cell motility are reviewed. The bacterium Helicobacter pylori has a major aetiological role in human gastric carcinogenesis and has been classified as a class I carcinogen. H. pylori is acquired primarily in childhood, and in the majority of cases infection and associated chronic gastritis is lifelong. There is a marked diversity in the clinical outcome of H. pylori infection with only a small percentage of infected subjects developing gastric adenocarcinoma or mucosaassociated lymphoid tissue (MALT) lymphoma [1]. The risk of gastric adenocarcinoma is greatest in infected subjects with non-ulcer dyspepsia or gastric ulceration who have developed severe gastric atrophy and intestinal metaplasia [2]. Bacterial virulence factors, such as the cag pathogenicity island (PAI) type IV secretory system [1] and genetic polymorphisms in proinflammatory and immunoregulatory cytokines [3], have been linked to an increased risk of developing gastric atrophy or intestinal-type gastric cancer. A key feature of the associated risk in developing gastric cancer with H. pylori infection is gastric epithelial hyperproliferation. The cellular and molecular signalling pathways used during H. pylori infection to promote epithelial hyperproliferation and transformation are being investigated. H. pylori is an excellent model system to use as a tool to investigate bacterial-induced epithelial cell signalling Corresponding author: J.E. Crabtree (j.crabtree@leeds.ac.uk). pathways of relevance to neoplasia. Bacterial virulence factors (Box 1) have been identified that mediate key phenotypic and genotypic changes in gastric epithelial cells. In this review the evidence for H. pylori-induced epithelial cell proliferation in human infection and in animal models is described, and the potential role of both bacterial virulence factors and inflammation in the hyperproliferative response is discussed. The molecular basis by which H. pylori triggers cell signalling cascades, and promotes inflammation and epithelial cell proliferative and motogenic responses is also described. Epithelial cell proliferation in H. pylori-induced gastric diseases Clinical studies There is increasing evidence that several chronic infections increase the risk of carcinogenesis. A key factor is probably infection-related alterations in epithelial cell homeostasis. H. pylori is associated with increased gastric epithelial cell proliferation in children [4] and adults [5,6]. Lifelong increased cell turnover is considered an important risk factor for increased mutational changes and the development of gastric cancer. Epithelial proliferation has been positively correlated with the degree of histological inflammation in the antrum and corpus mucosa of H. pylori-infected patients [5 7]. However, stepwise multiple regression analysis has indicated that the only independent predictor of epithelial cell proliferation in infected patients is the density of H. pylori colonization [7]. The enhanced epithelial proliferation observed with infection is probably promoted as a consequence of the inflammatory response and also by a route independent of inflammation. Eradication of H. pylori infection results in a significant decrease in gastric epithelial cell proliferation [8]. Box 1. H. pylori virulence factors associated with gastric cancer VacA: an 87 kd protein (140 kd precursor protein) that assembles into oligomeric structures. CagA: a 128 kd protein translocated by the type IV secretion system. It becomes phosphorylated within epithelial cells and contributes to H. pylori-induced cell scattering. FldA: a 19 kd putative flavodoxin protein as a potential marker for neoplasm. BabA2: a 78 kd adhesin that binds to LewisB on gastric epithelial cells X/$ - see front matter q 2003 Elsevier Ltd. All rights reserved. doi: /j.tim

2 30 Review TRENDS in Microbiology Vol.12 No.1 January 2004 (a) (b) (c) (d) Figure 1. Epithelial cell proliferation is increased with gastric Helicobacter pylori infection. Proliferating epithelial cells in the Mongolian gerbil antral mucosa detected by immunohistochemistry using an anti-bromodeoxyuridine monoclonal antibody in (a) an uninfected gerbil and (b) a gerbil thirty-six weeks post-infection with H. pylori (SS1 strain). One hour before sacrifice gerbils were injected with bromodeoxyuridine, which is incorporated into proliferating cells. Haematoxylin and eosin stained sections of gastric antral mucosa of Mongolian gerbil, infected for 36 weeks with H. pylori SS1 strain. Gastric pathology in (c) H. pylori uninfected and (d) H. pylori-infected Mongolian gerbils. Scale bars represent 500 mm (a,b) and 100 mm (c,d). In addition, attenuation of epithelial cell proliferation has also been observed in those patients where H. pylori eradication has failed, probably because of decreased bacterial density or inflammation [9]. To date, only one study has failed to demonstrate a reduction in gastric epithelial cell proliferation after eradication of H. pylori infection [10]. Divergent results might be attributed to varying patient populations, labelling techniques or treatment with pharmacological agents, such as protonpump inhibitors. Animal models Animal models of gastric Helicobacter infection have been useful in delineating the potential role of the pathogen and associated inflammation in gastric epithelial cell proliferative responses. Experimental infection with H. pylori or related gastric Helicobacter species, such as H. felis, increases gastric epithelial cell proliferation in animals, including Mongolian gerbils [11 13] (Figure 1) and mice [14]. Gastric epithelial cell proliferation has been extensively examined in murine H. felis and H. pylori models. H. felis infection in mice causes severe inflammation and marked epithelial hyperplasia [14]. The extent of epithelial proliferation and apoptosis is dependent on the mouse strain [14]. In C57BL/6 mice H. felis induces extensive epithelial hyperproliferation in the corpus mucosa, with parietal and chief cells being replaced with mucous-secreting cells [14,15]. H. felis induces greater gastric epithelial cell proliferative responses in these mice than H. pylori [13], and interestingly, there are marked gender differences in proliferative responses [15]. Significant increases in epithelial proliferation and apoptosis are only observed in H. felis-infected C57BL/6 female mice, possibly reflecting sex differences in the immune response and cytokine production [15]. By contrast, in transgenic hypergastrinaemic (INS-GAS, insulin-gastrin) mice on a FVB/N background, H. pylori infection results in gastric cancer in males only [16]. Gastrin, which is elevated in H. pylori infection, is known to promote gastric epithelial hyperproliferation [17]. H. pylori-induced epithelial cell proliferation has been correlated with elevated serum gastrin in Mongolian gerbils [11]. In INS-GAS transgenic mice, H. felis and H. pylori infection accelerates the development of gastric adenocarcinoma [16,18]. However, the effects of H. felis or H. pylori infection on gastric epithelial cell proliferation in hypergastrinaemic mice remain to be examined. Experimental infection using gastric Helicobacter species in transgenic mice that have specific gene deletions has also been a useful approach to identify potential host factors that promote or regulate epithelial hyperproliferation. In immune-deficient mice, epithelial proliferative responses to infection with gastric Helicobacter species are reduced, emphasizing the importance of mucosal inflammation in the murine model. RAG 2/2 (recombinaseactivating gene-deficient) mice [19], those with severe combined immune deficiency disorder (SCID) [20] and also mice deficient in interferon regulatory factor [21] or gamma interferon [20], do not develop gastritis and associated epithelial hyperproliferative responses following gastric Helicobacter infection. By contrast, interleukin (IL)-10 2/2 mice [22] and those that lack functional transforming growth factor (TGF)-b type II receptor [23] develop severe hyperplastic gastritis during infection,

3 Review TRENDS in Microbiology Vol.12 No.1 January indicating the importance of anti-inflammatory and immune-regulatory cytokines in bacterial-induced epithelial hyperproliferation. Influence of H. pylori strains on epithelial proliferation H. pylori is a genomically diverse pathogen and several bacterial virulence factors, including the cag PAI, VacA and adhesins such as BabA2, are considered to have a key role in disease pathogenesis [1,24] (Box 1). Only strains containing the cag PAI trigger signalling cascades in gastric epithelial cells, resulting in nuclear factor kappa B (NF-kB) activation [25] and multiple associated changes in epithelial gene expression [26] (Table 1). In human gastric epithelial cells, the upregulation of IL-8 and other neutrophil chemotactic C-X-C chemokines, such as GRO-a (growthregulated oncogene protein-alpha), is considered crucial in the association between infection with cag PAI-positive strains, neutrophilic responses and more severe gastroduodenal disease [27]. The profile of gene expression in gastric epithelial cells, as determined by cdna array analysis, is markedly different after culture with wild-type cag PAI-positive and negative strains [26], with many genes involved in cell cycle control and apoptosis being differentially expressed [26]. By contrast, a recent study demonstrated that isogenic cag PAI-negative mutants failed to elicit significant changes in epithelial gene expression [28]. This implies that in cag PAI-positive strains, the cag PAI is the major mediator of changes in gene expression but that wild-type cag PAI-negative strains induce changes in epithelial gene expression by other pathways. It is currently unclear whether gastric epithelial cell proliferation differs in patients that are infected with cag PAI-positive or negative strains. In one study, gastric epithelial cell proliferation was greater in patients infected with caga-positive strains than caga-negative strains [5], whereas, another study in patients with nonulcer dyspepsia did not confirm these observations [6]. The presence or absence of bacterial adhesins could account for earlier discrepant results. A recent study based in China, where 98% of patients were infected with caga-positive H. pylori strains, found increased epithelial cell proliferation in patients infected with strains expressing the blood group antigen-binding adhesin baba2 [29]. There have also been divergent clinical studies regarding the effects of the cag PAI on apoptosis in human gastric epithelial cells [5,6]. In addition to the varying patient populations studied, other factors, such as the use of non-steroidal anti-inflammatory agents that decrease apoptosis in the gastric epithelium [8], might account for divergent results. Infection of animal models with genetically defined H. pylori strains is an alternative approach to use when investigating the importance of the cag PAI on gastric epithelial proliferation and apoptosis. In the gerbil, cag PAI strains of H. pylori induce gastritis with greater severity [12,30] and increased epithelial proliferation and apoptosis [12], compared with strains that lack a functional cag PAI. In the early stages of infection, the epithelial hyperproliferative responses in the gerbil are confined to the antrum, with subsequent progression to the corpus [12,13], therefore mirroring the pathology and proliferative responses observed during human infection [6]. As a result, the gerbil is a useful model to use when analysing the role of H. pylori virulence factors on gastric epithelial proliferation and apoptosis. Further studies in the gerbil model using isogenic mutants of other virulence factors will be important to delineate their importance in the epithelial hyperproliferative response. H. pylori affects apoptosis and cell cycle control Apoptosis and cell cycle control are processes required for the regulation of cellular homeostasis. The disturbed equilibrium of apoptosis and cell proliferation associated with H. pylori infection could lead to an overall increase in cellular turnover and persistence of mutated cells, which will favour the development of neoplasia. Apoptosis is often difficult to observe in vivo because the dying cells are rapidly phagocytosed by tissue macrophages. This phagocytosis is different from that seen in inflammation, when activated macrophages are recruited from outside the immediate area of cell death. From a biological viewpoint, the chronic imbalance between apoptosis and cell proliferation is the first step of gastric carcinogenesis, as in all tumours. The cell cycle, the programme for cell growth and division (proliferation), consists of four phases that are known as G 1 and G 0,S,G 2 and M. The important protein families used during this cycle include the cyclins, the cyclindependent kinases (Cdks), the Cdk inhibitors and the tumour-supressor genes (in particular, Rb and p53) [31]. The H. pylori toxin VacA induces gastric epithelial cell apoptosis, suggesting that differences in levels of gastric mucosal apoptosis among infected persons might result from strain-dependent variations in VacA structure [32]. In addition, Shibayama et al. [33] have shown that g-glutamyl transpeptidase from H. pylori membrane preparations induces apoptosis. At the molecular level, Table 1. Type IV secretion system dependent and independent signalling a,b Type IV secretion system-dependent Type IV secretion system-independent Kinase IKK, JNK, p38 MEK, ERK GTPases Rac1, Cdc42 Ras, Rap1 Transcription factors NF-kB, AP-1 BP1 and 2, USF1 and 2, CREB Genes Proinflammatory cytokines and chemokines Histidine decarboxylase, Cyclooxygenase 2 a Type IV secretion system-dependent signalling leads to the induction of proinflammatory cytokine or chemokine genes, activation of immediate early response transcription factors (NF-kB and AP-1), Rho-GTPases (Rac1 and Cdc42) and stress response kinases (JNK and p38) including the IKK kinases. Independent of the type IV secretion system, H. pylori regulates the promoters of genes that are involved in acid secretion (histidine decarboxylase) and prostaglandin synthesis (cyclooxygenase-2), transcription factors (BP1 and 2, USF1 and 2, and CREB), proliferation-associated GTPases (Ras and Rap1) and Map-kinases (MEK and ERK). b Abbreviations: AP-1, activator protein 1; BP1 and 2, beta protein 1 and 2; Cdc42, cell division cycle protein 42; CREB, camp responsive element-binding protein; ERK, extracellular signal-regulated kinase; IKK, IkappaB kinase; JNK, c-jun N-terminal kinase; MEK, MAP kinase kinase; NF-kB, nuclear factor kappa; USF1 and 2, upstream stimulatory factor 1 and 2.

4 32 Review TRENDS in Microbiology Vol.12 No.1 January 2004 in gastric epithelial cells in vitro in a time-dependent manner, H. pylori induces caspases 3, 7 and 8, and also anti-apoptotic proteins c-iap1 and c-iap2. Whereas, the proteins Bax, Bak and Bcl-X L are not changed [34]. In recent studies, activation of the peroxisome proliferatoractivated receptor g (PPARg) suppressed NF-kB-mediated apoptosis in vitro [35]; this inhibition was independent of cag PAI status. When studying colonic T84 epithelial cells, LeNegrate et al. [36] showed that H. pylori triggers apoptosis through a Fas-dependent pathway, which depends on the expression of the cag PAI. In another study, apoptosis of gastric epithelial cells was mediated by elevated levels of Smad5 as a result of cag PAI-dependent H. pylori infection [37]. From the limited clinical data available, the effects of bacterial virulence factors, such as the cag PAI, on gastric epithelial apoptosis in vivo is unresolved. Exposure of epithelial cells to H. pylori alters cell cycle control both in vitro and in vivo. Mucosal expression of cyclin D1, the tumour-suppressor p53 and the cell cycle inhibitor p21 was significantly higher in H. pylori-infected patients with intestinal metaplasia [38], but p53 mutations and p21 expression were not analysed in detail. A clear effect of H. pylori on cell cycle progression has been described in infected patients with intestinal metaplasia that overexpress cyclin D2 and show reduced expression of the cell cycle inhibitor p27 [39]. Cdx2 (caudal-related homeobox 2) plays an important role in differentiation and maintenance of intestinal epithelial cells [40]. The role of H. pylori virulence factors and the molecular mechanisms involved in contributing to malignant transformation in the gastric mucosa remain to be clarified. However, there is marked variability in the expression of cell cycle regulatory proteins, such as p27, cyclin D1, cyclin E Cdk2 activity and cyclin B1, that are induced by H. pylori in different gastric epithelial cell lines [41,42], making extrapolation of in vitro studies to events in vivo difficult. Epithelial cell signalling under the direct control of H. pylori The ability of a cell to respond to its extracellular environment involves a complex and highly organized series of events referred to as cellular signalling. These signalling processes regulate fundamental cellular responses and their abrogation can lead to the development of various human diseases, such as cancer. Despite numerous data, the genetic basis of gastric cancer remains unknown. None of the gene errors identified to date in gastric cancer are totally specific or unique, which indicates that gastric cancer is not a monomorphic entity. Multiple molecular and genetic alterations have been identified in gastric cancer, including activated oncogenes, growth factors and growth-factor receptors. In the following, the current knowledge of molecular signalling processes and the cell biology in H. pylori-induced epithelial cell responses will be reviewed. Epithelial cell signalling and inflammation Recently, the concept that inflammation is a crucial component of tumour progression has expanded. Cancer could arise from sites of infection, chronic irritation and inflammation. The tumour microenvironment, which largely accumulates inflammatory cells, is an indispensable participant in the process of cancer development. The physical contact between H. pylori and gastric epithelial cells leads to the activation of signal transduction pathways, directing the induction of immediate early response transcription factors NF-kB and activator protein 1 (AP-1) [43]. These transcription factors contribute to the activation of proinflammatory C-X-C chemokines, which in vivo attract neutrophils towards the colonized epithelium, and other innate defences [25,27]. Therefore, gastric epithelial cells actively participate in inflammation and mucosal immunity during initial and persistent H. pylori infection. H. pylori stimulates the transcription factor NF-kB, which involves the activity of the kinases IKKa (IkappaB kinase alpha) and IKKb (Table 1). These kinases are active within a complex with the essential modulator IKKg [44]. AP-1 activation involves c-jun N-terminal kinase (JNK) activity [45]. The bacterial effector injected by the cag PAI type IV secretion system is peptidoglycan. This is recognized by the intracellular nucleotide-binding oligomerization domain protein (Nod1) receptor molecule [46], which directly activates NF-kB. Nod1 belongs to a family that includes multiple members with NOD and leucine-rich repeats and recognizes peptidoglycan derived primarily from Gram-negative bacteria [46]. Besides stimulating innate defences in gastric epithelial cells, H. pylori induces arachidonic acid synthesis, which is required for the production of prostaglandins. The release of arachidonic acid involves activation of cytosolic phospholipase A 2 (PLA 2 ), pertussis toxin-sensitive heterotrimeric Ga I and Ga o proteins and the p38 kinase [47]. Activation of p38 in epithelial cells was only observed following culture with H. pylori strains that carry a functional cag PAI [47,48]. Therefore, certain signalling cascades that lead to the activation of the IKK complex, JNK kinase and p38 kinase, are only activated by H. pylori strains carrying the active cag PAI (Table 1). Whereas, the cag PAI-encoded CagA protein, which is translocated into the gastric epithelial cell via the type IV secretion system [43], is dispensable. Proliferation-associated signalling cascades Cell growth and differentiation in response to extracellular stimuli is mediated through various intracellular signal transduction pathways. The mitogen-activated protein kinase (MAPK) pathway is a major player in this kinasesignalling cascade from growth factors to the cell nucleus. The pathway involves kinases at two levels: MAP kinases, also known as extracellular signal-regulated kinases (ERKs); and MAP kinase kinases, also known as MEKs or MAPK ERK kinases. MEK is activated by the phosphorylation of two serine residues by upstream kinases, such as members of the raf gene family. When activated, MEK catalyzes the phosphorylation of threonine and tyrosine residues of ERK. The activated ERK then phosphorylates and activates transcription factors in the nucleus, such as the ternary complex factor (TCF) Elk-1, which regulate early genes including c-myc, fos and jun. Because MEK and ERK enzymes are known to be essential for normal cell proliferation and differentiation,

5 Review TRENDS in Microbiology Vol.12 No.1 January deregulation (overexpression, hyperactivity or gene mutation) of the MAPK signal transduction pathway might lead to proliferative diseases, such as cancer [31]. Therefore, cancer can be considered as a disease of communication at the molecular level. H. pylori activates ERK1 and 2 and MEK1 and 2, in AGS gastric epithelial cells, in a cag PAI-independent manner [48,49] (Table 1). With regards to the intensity of activation of these MAP-kinases, the cag PAI positive strains can induce moderately stronger ERK and MEK phosphorylation than cag PAI-negative strains [48]. Supernatants from H. pylori cultures also induce, using unknown secreted factors, activation of ERK and MEK kinases leading to regulation of the histidine decarboxylase (HDC) promoter (Table 1). HDC is the key enzyme involved in histamine biosynthesis and converts L-histidine into histamine in enterochromaffin-like cells of the corpus mucosa. Histamine stimulates the parietal cells that are responsible for acid secretion in the corpus of the stomach. The signalling cascade that leads to activation of the HDC promoter in H. pylori-infected gastric epithelial cells comprises activation of the kinase B-Raf, which is regulated by the Ras-like GTPase Rap1. The activity of Rap1 is triggered by the accumulation of cyclic-adenosinemonophosphate (camp) and is under the control of the heterotrimeric G-protein Ga s [50]. Another target-gene induced by H. pylori using ERK and MEK kinases is cyclooxygenase-2 (COX-2) [51]. (Table 1) Enhanced transcription of COX-2 in H. pylori-infected gastric epithelial cells is regulated through a proximal CRE-Ebox (camp responsive element-ebox) enhancer element by activation of the upstream stimulatory factor 1 and 2 (USF-1 and 2) and CREB (camp responsive elementbinding protein) transcription factors [51]. Furthermore, H. pylori induces upstream activators of ERK and MEK in Ras-dependent and Ras-independent pathways [52] (Table 1). Activation of ERK and MEK kinases appears to be through tyrosine kinase receptors [EGFR (epidermal growth-factor receptor), Her2 Neu and c-met], which are activated in a cag PAI-independent manner [53,54].In contrast to Wallasch et al. [53], the report by Keates et al. [52] describes EGFR activation in a cag PAI-dependent manner. Therefore, ERK and MEK kinases are regulated during H. pylori infection using secreted factors, and also in an epithelial contact-dependent manner in which cag PAI-positive strains exert the ability to induce higher levels of ERK and MEK phosphorylation. Activation of tyrosine kinase receptors Tyrosine kinase receptors control a diverse array of cellular responses including growth, proliferation, differentiation and migration. Tyrosine phosphorylation of target proteins is a reversible, dynamic process controlled by the activities of the kinases, including the tyrosine kinase receptors and the competing actions of the protein tyrosine phosphatases. For some of the tyrosine kinase receptors, an important role in carcinogenesis has been described [55]. Tyrosine kinase receptors have an important role in gastric carcinogenesis [55]. Recent studies have demonstrated that H. pylori activates EGFR, HER2 Neu (ErbB-2) and c-met in gastric epithelial cells [52 54] (Figure 2). The EGFR activation is dependent on extracellular transmembrane metalloprotease cleavage of proheparin binding epidermal growth factor (prohb-egf) and signalling by mature HB-EGF [53]. The upregulation of HB-EGF gene transcription by H. pylori requires metalloprotease, EGFR and MEK1 activities [53], indicating the involvement of the triple membrane-passing signal (TMPS) for EGFR transactivation [56] (Figure 3). Disruption of epithelial tight junctions by the interaction of translocated CagA with the scaffolding protein ZO-1 [57] and VacA-mediated phosphorylation of G protein-coupled receptor kinase-interactor 1 (Git1) [24] probably promotes binding of EGF ligands to the EGFR located on basolateral membranes of the epithelial cells (Figure 4). Overexpression of key elements of the TMPS cascade in those patients with gastric cancer or atrophic gastritis suggests that the EGFR autocrine paracrine signalling pathway induced by H. pylori is of pathophysiological relevance. H. pylori infection in humans is associated with increased gastric mucosal levels of EGF protein and EGFR transcripts [58]. The proteases that are involved in the processing of transmembrane prohb-egf remain to be identified. H. pylori increases matrix metalloprotease-3 (MMP-3), MMP-7 and MMP-9 in gastric carcinoma cell lines [59 61], and the bacterium itself has also been reported to have MMP-3 activity [59]. H. pylori also upregulates several ADAM (adisintegrin and metalloproteinase) genes in cultured gastric epithelial cells [26,62]. ADAM proteins are cell-surface glycoproteins responsible for ectodomain shedding of membrane proteins, including growth factors, cytokines and their receptors, and ligands involved in apoptosis. Expression of ADAM10 and ADAM17 is increased in the gastric mucosa of H. pyloriinfected patients [62]. Additionally, expression of several ADAM genes is strongly increased in gastric cancer [62], CagA Motogenic response H. pylori c Met EGFR Her2 Neu Proliferation TRENDS in Microbiology Figure 2. Helicobacter pylori activates receptor tyrosine kinases. H. pylori induces the activity of receptor tyrosine kinases, such as EGFR (epidermal growth-factor receptor), Her2 Neu (ErbB-2) and c-met in epithelial cells by unknown bacterial effectors [52 54]. The H. pylori effector protein CagA targets the c-met receptor intracellularly [54]. Activation of, and alterations in, tyrosinekinase receptor signalling result in proliferation-associated processes and the motogenic response. Preferentially, the c-met receptor promotes epithelial cell survival and cell scattering, where it stimulates the dissociation and dispersal of epithelial cells and the transition to a fibroblastic morphology, which occur during crucial phases of development and tumour progression.

6 34 Review TRENDS in Microbiology Vol.12 No.1 January 2004 H. pylori GPCR Gα β γ? Metalloproteinase(s) Membrane-bound EGF-like ligand EGFR Proliferation TRENDS in Microbiology Figure 3. Helicobacter pylori stimulates epidermal growth-factor receptor (EGFR) transactivation via a triple membrane-passing signal (TMPS) cascade. The EGFR activation involves G-protein coupled receptor (GPCR) activity and the TMPS for EGFR transactivation, which is dependent on extracellular transmembrane metalloprotease cleavage of pro-heparin binding epidermal growth factor (prohb-egf) and signalling by mature HB-EGF [53]. raising the possibility that overexpression might promote amplification of TMPS signalling cascades and dysregulated EGFR transactivation. Cell cell and cell matrix interactions and the motogenic response Decreased cell cell or cell matrix interactions are common in gastric cancer and might be related to the tendency to produce metastasis [63]. In polarized epithelial cells H. pylori affects the scaffolding protein ZO-1 and the tight junctional adhesion protein (JAM) in a CagA-dependent manner, and disrupts junction-mediated epithelial barrier functions [57] (Figure 4). Among the many types of adhesion molecules, E-cadherin serves as a prime mediator of cell cell adhesion within the zonula adherens junctions. Downregulation of E-cadherin in antral biopsies of H. pylori-infected patients has been described [64]. The cytoplasmic domains of E-cadherin interact with catenins (a and b), and alterations in this system have been ascribed an important role in tumour initiation and progression [65]. Loss of E-cadherin expression in epithelial cells is associated with the acquisition of the mesenchymal phenotype. Epithelial-mesenchymal transition (EMT) occurs during crucial phases of development and tumour progression. The factor involved in this transition is represented by the hepatocyte growth factor or scatter factor (HGF or SF). HGF activates the c-met receptor and promotes epithelial cell growth and survival and also cell scattering, where it stimulates the dissociation and dispersal of colonies of epithelial cells and the acquisition of a fibroblastic morphology [55]. Interestingly, in gastric epithelial cells H. pylori activates c-met in a cag PAIindependent manner [54]. However, the translocated H. pylori effector protein CagA targets the c-met receptor intracellularly and enhances the cell scattering (motogenic response), which suggests that dysregulation of growthfactor receptor signalling could play a role in motility and invasiveness of cells [54]. Cell motility induced by H. pylori in epithelial cells needs actin-reorganization with rufflelike structures, which involves the activity of the Rho- GTPases Rac1 and Cdc42. Activation of these GTPases depends on the functional type IV secretion system, but appears independent of H. pylori CagA protein expression [66]. Therefore, c-met activation, its modulation by translocated CagA, and transient polarization of the H. pylori-infected AGS cells by the activity of Rho- GTPases contribute to a forceful motogenic response. The active c-met receptor recruits, intracellularly, adaptor proteins to form an intricate signalling complex [67]. Growth factor treatment can induce Gab1 tyrosine phosphorylation and its direct association with several signal transducers, including Grb2, PI3-K, PLCg and SHP-2 [68]. The interaction of the tyrosine phosphatase SHP-2 with translocated CagA of H. pylori has recently been described [69]. Furthermore, the C-terminal region of Src kinase (Csk) interacts with CagA [70], and it has been H. pylori VacA PTP-ζ Git JAM JAM CagA ZO-1 Actin Epithelial erosion Tight junction TRENDS in Microbiology Figure 4. Helicobacter pylori affects epithelial tight junctions. In polarized epithelial cells the type IV secretion system (indicated by a syringe) translocates CagA, which interacts with the scaffolding protein ZO-1 and the tight junctional adhesion protein (JAM) [57]. During this process, H. pylori disrupts junction-mediated epithelial barrier functions. The bacterial toxin VacA mediates phosphorylation of Git1 (G protein-coupled receptor kinase-interactor 1) via interaction with the protein tyrosine phosphatase receptor PTP-z in gastric epithelial cells [24]. VacA contributes to the epithelial erosion and ulcerogenesis in the stomach of mice. The disruption of tight junctions probably promotes binding of epidermal growth factor (EGF) ligands to the EGF receptor (EGFR) located on basolateral membranes of the epithelial cells.

7 Review TRENDS in Microbiology Vol.12 No.1 January speculated that this interaction attenuates CagA phosphorylation. However, CagA fails to co-immunoprecipitate with Gab1 or Grb-2 [54,70]. In contrast to these reports, direct physical interaction between CagA and Grb2 in vitro has been described [71]. Therefore, CagA directly interacts with signal-transducing proteins and might play a role as an adaptor protein in growth-factor receptor signalling. Conclusions A characteristic of H. pylori infection in humans is gastritis, which persists for decades without causing serious damage in most cases. Therefore, the clinical complications of H. pylori infection, such as peptic ulcer disease and gastric cancer, appear to represent an imbalance in gastric homeostasis. It is tempting to speculate that H. pylori, together with environmental factors, contributes to the host responses by negatively regulating intracellular epithelial signalling pathways. Furthermore, the clinically apparent diseases probably result from the cumulative effect of multiple interactions between H. pylori and its host. Cancer is a multistep process, which involves alterations of several different factors including tumour-suppressor genes, dominant oncogenes and receptor tyrosine kinases. The induction of the motogenic response by H. pylori in epithelial cells represents an example of how a human microbial pathogen can activate growth-factor receptor tyrosine kinases, and modify signal transduction in the cell using translocated bacterial proteins. It will be essential to deepen our understanding of receptor crosstalk in H. pylori-infected epithelium and its contribution to EGFR, Her2 Neu and c-met activation. The study of signalling pathways that regulate EGFR, Her2 Neu and c-met expression and activity in H. pylori infection may identify promising therapeutic targets for suppression of transformation, and offer novel potential targets for the treatment and/or prevention of malignancies (Box 2). To downregulate the expression of anti-apoptotic genes is another potential therapeutic approach. In addition, it will be important in the future (Box 2) to characterize the functions of matrix metalloproteases (MMPs) and metalloprotease-disintegrins (ADAMs), and to identify the proteases involved in EGFR transactivation. These few examples show that the understanding of the pivotal role of H. pylori in modifying the cellular microenvironment is a high requirement to define prognostic markers and develop therapeutic strategies to avoid development of gastric tumours. Box 2. Future questions Do animal models help to understand the pathogenesis of gastric cancer in humans? What are the key steps in the signalling cascades that lead to gastric carcinogenesis? What is the therapeutic potential of inhibiting key signalling pathways on gastric carcinogenesis? Does the mucosal immune response to key virulence factors interfere with signalling? Acknowledgements Work in the laboratory of J.E.C. is supported by Yorkshire Cancer Research and the European Commission (contract ICA4-CT ). The work in the Institute of M.N. is supported by the Deutsche Forschungsgemeinschaft grants Na 292/5 2 and Na 292/6 2. Owing to the restrictions on the number of references, in many cases either a single paper or a review is cited. Therefore, we apologise to those researchers and colleagues whose work has not been cited in this review. References 1 Peek, R.M. and Blaser, M.J. (2002) Helicobacter pylori and gastrointestinal tract adenocarcinomas. Nat. Rev. Cancer 2, Uemura, N. et al. (2001) H. pylori infection and the development of gastric cancer. N. Engl. J. Med. 345, El-Omar, E.M. et al. (2003) Increased risk of noncardia gastric cancer associated with proinflammatory cytokine gene polymophisms. Gastroenterology 124, Jones, N.L. et al. (1997) Increase in proliferation and apoptosis of gastric epithelial cells early in the natural history of Helicobacter pylori infection. Am. J. Pathol. 151, Peek, R.M. et al. (1997) Helicobacter pylori caga þ strains and dissociation of gastric epithelial cell proliferation and apoptosis. J. Natl. Cancer Inst. 89, Moss, S.F. et al. (2001) Increased gastric epithelial cell apoptosis associated with colonisation with caga þ H. pylori strains. Cancer Res. 61, Lynch, D.A.F. et al. (1999) Correlation between epithelial cell proliferation and histological grading in gastic mucosa. J. Clin. Pathol. 52, Leung, W.K. et al. (2000) Interaction of Helicobacter pylori eradication and non-steroidal anti-inflammatory drugs on gastric epithelial apoptosis and proliferation: implications on ulcerogenesis. Aliment. Pharmacol. Ther. 14, Lynch, D.A. et al. (1995) Cell proliferation in Helicobacter pylori associated gastritis and the effect of eradication therapy. Gut 36, El-Zimiaty, H.M. et al. (2000) Sustained increases in gastric antral epithelial cell proliferation despite cure of Helicobacter pylori infection. Am. J. Gastroenterol. 95, Peek,R.M. et al. (2000) Helicobacter pylori alters gastric epithelial cell cycle events and gastrin secretion in Mongolian gerbils. Gastroenterology 118, Israel, D.A. et al. (2001) Helicobacter pylori strain specific differences in genetic content, identified by microarray, influence host inflammatory responses. J. Clin. Invest. 107, Court,M.et al. (2002) Gastric Helicobacter species infection in murine and gerbil models: comparative analysis of effects of H. pylori and H. felis on gastric epithelial cell proliferation. J. Infect. Dis. 186, Wang, T.C. et al. (1998) Mice lacking secretory phospholipase A2 show altered apoptosis and differentiation with Helicobacter felis infection. Gastroenterology 114, Court, M. et al. (2003) The effect of gender on Helicobacter felis mediated gastritis, epithelial cell proliferation and apoptosis in the mouse model. J. Pathol. 201, Fox, J.G. et al. (2003) Host and microbial constituents influence Helicobacter pylori-induced cancer in a murine model of hypergastrinaemia. Gastroenterology 124, Dockray, G.J. et al. (2001) The gastrins: their production and biological activities. Annu. Rev. Physiol. 63, Wang, T.C. et al. (2000) Synergistic interaction between hypergastrinaemia and Helicobacter infection in a mouse model of gastric cancer. Gastroenterology 118, Roth, R.A. et al. (1999) Cellular immune responses are essential for development of Helicobacter felis associated gastric pathology. J. Immunol. 163, Smythies, L.E. et al. (2000) Helicobacter pylori induced mucosal inflammation is Th1 mediated and exacerbated in IL-4 but not IFN gamma gene deleted mice. J. Immunol. 165, Sommer, F. et al. (2001) Lack of gastritis and of an adaptive immune response in interferon regulatory factor-1 deficient mice infected with Helicobacter pylori. Eur. J. Immunol. 31, Berg, D.J. et al. (1998) Rapid development of severe hyperplastic

8 36 Review TRENDS in Microbiology Vol.12 No.1 January 2004 gastritis with gastric epithelial dedifferentiation in Helicobacter felisinfected IL-10 2 /2 mice. Am. J. Pathol. 152, Hahm, K.B. et al. (2002) Conditional loss of TGF-b signalling leads to increased susceptibility to gastrointestinal carcinogenesis in mice. Aliment. Pharmacol. Ther. 16 (suppl. 2), Fujikawa, A. et al. (2003) Mice deficient in protein tyrosine phosphatase receptor type Z are resistant to gastric ulecer induction by VacA of Helicobacter pylori. Nat. Genet. 33, Naumann, M. (2000) NF-kB activation and innate immune response in microbial pathogen infection. Biochem. Pharmacol. 60, Cox,J.M. et al. (2001) cdna array analysis of cag pathogenicity islandassociated H. pylori epithelial cell response genes. Infect. Immun. 69, Crabtree, J.E. (2001) Cytokine responses in Helicobacter pylori infection. In Helicobacter pylori: Molecular and Cellular Biology (Achtman, M. and Suerbaum, S., eds), pp , Horizon Press 28 Guillemin, K. et al. (2002) Cag pathogenicity island-specific responses of gastric epithelial cells to Helicobacter pylori infection. Proc. Natl. Acad. Sci. U. S. A. 99, Yu, J. et al. (2002) Relationship between Helicobacter pylori baba2 status with gastric epithelial cell turnover and premalignant lesions. Gut 51, Ogura, K. et al. (2000) Virulence factors of Helicobacter pylori responsible for gastric diseases in Mongolian gerbil. J. Exp. Med. 192, Hannahan, D. and Weinberg, R.A. (2000) The hallmarks of cancer. Cell 100, Cover,T.L.et al. (2003) Induction of gastric epithelial cell apoptosis by Helicobacter pylori vacuolating cytotoxin. Cancer Res. 63, Shibayama, K. et al. (2003) A novel apoptosis-inducing protein from Helicobacter pylori. Mol. Microbiol. 47, Maeda, S. et al. (2002) Analysis of apoptotic and antiapoptotic signalling pathways induced by Helicobacter pylori. Gut 50, Gupta, R.A. et al. (2001) Activation of peroxisome proliferatoractivated receptor g suppresses nuclear factor kb- mediated apoptosis induced byhelicobacter pylori in gastric epithelial cells. J. Biol. Chem. 276, Le Negrate, G. et al. (2001) Epithelial intestinal cell apoptosis induced by Helicobacter pylori depends on expression of the cag pathogenicity island phenotype. Infect. Immun. 69, Nagasako, T. et al. (2003) Up-regulated Smad5 mediates apoptosis of gastric epithelial cells induced by Helicobacter pylori infection. J. Biol. Chem. 278, Polat, A. et al. (2002) Expression of cell-cycle related proteins in Helicobacter pylori gastritis and association with gastric carcinoma. Neoplasma 49, Yu, J. et al. (2001) Effect of Helicobacter pylori eradication on expression of cyclin D2 and p27 in gastric intestinal metaplasia. Aliment. Pharmacol. Ther. 15, Satoh, K. et al. (2002) Aberrant expression of CDX2 in the gastric mucosa with and without intestinal metaplasia: effect of eradication of Helicobacter pylori. Helicobacter 7, Hirata, Y. et al. (2001) Helicobacter pylori activates the cyclin D1 gene through mitogen-activated protein kinase pathway in gastric cancer cells. Infect. Immun. 69, Sommi, P. et al. (2002) Helicobacter pylori releases factor(s) inhibiting cell cycle progression of human gastric cell lines by affecting cyclin E/cdk2 kinase activity and Rb protein phosphorylation through enhanced p27 KIP1 protein expression. Exp. Cell Res. 281, Naumann, M. (2001) Host cell signaling in Helicobacter pylori infection. Int. J. Med. Microbiol. 291, Foryst-Ludwig, A. and Naumann, M. (2000) p21-activated kinase 1 activates the nuclear factor kappa B (NF-kappa B)-inducing kinase Ikappa B kinases NF kappa B pathway and proinflammatory cytokines in Helicobacter pylori infection. J. Biol. Chem. 275, Naumann, M. et al. (1999) Activation of activator protein 1 and stress response kinases in epithelial cells colonized by Helicobacter pylori encoding the pathogenicity island. J. Biol. Chem. 274, Viala, J. et al. (2003) Nod1-dependent proinflammatory response to Helicobacter pylori infection in gastric epithelial cells. Gastroenterology 124 (suppl 1), A43 47 Pomorski, T. et al. (2001) Helicobacter pylori-induced prostaglandin E 2 synthesis involves activation of cytosolic phospholipase A 2 in epithelial cells. J. Biol. Chem. 276, Keates, S. et al. (1999) Differential activation of mitogen-activated protein kinase in AGS gastric epithelial cells by cag þ and cag - Helicobacter pylori. J. Immunol. 163, Wessler, S. et al. (2000) Helicobacter pylori activates the histidine decarboxylase promoter through a mitogen-activated protein kinase pathway independent of pathogenicity island-encoded virulence factors. J. Biol. Chem. 275, Wessler, S. et al. (2002) B-Raf/Rap1 signaling, but not c-raf-1/ras, induces the histidine decarboxylase promoter in Helicobacter pylori infection. FASEB J. 16, Jüttner, S. et al. (2003) Helicobacter pylori stimulates host cyclooxygenase-2 gene: critical importance of MEK/ERK-dependent activation of USF1/-2 and CREB transcription factors. Cell. Microbiol. 5, Keates, S. et al. (2001) cag þ Helicobacter pylori induce transactivation of the epidermal growth factor receptor in AGS gastric epithelial cells. J. Biol. Chem. 276, Wallasch, C. et al. (2002) Helicobacter pylori stimulated EGF receptor transactivation requires metalloprotease cleavage of HB-EGF. Biochem. Biophys. Res. Commun. 295, Churin, Y. et al. (2003) Helicobacter pylori CagA protein targets the c-met receptor and enhances the motogenic response. J. Cell Biol. 161, Thiery, J.P. (2002) Epithelial-mesenchymal transitions in tumour progression. Nat. Rev. Cancer 2, Prenzel, N. et al. (1999) EGF recptor transactivation by G-proteincoupled receptors requires metalloprotease cleavage of prohb-egf. Nature 402, Amieva, M.R. et al. (2003) Disruption of the epithelial apical-junctional complex by Helicobacter pylori CagA. Science 300, Wong, B.C. et al. (2001) Epidermal growth factor and its receptor in chronic active gastritis and gastroduodenal ulcer before and after H. pylori eradication. Aliment. Pharmacol. Ther. 15, Gooz, M. et al. (2001) Epithelial and bacterial metalloproteinases and their inhibitors in H. pylori infection of human gastric epithelial cells. Am. J. Physiol. Gastrointest. Liver Physiol. 281, G823 G Mori, N. et al. (2003) Helicobacter pylori induces matrix metalloproteinase-9 through activation of nuclear factor kappab. Gastroenterology 124, Wroblewski, L. et al. (2003) Stimulation of MMP-7 (matrilysin) by Helicobacter pylori in human gastric epithelial cells: role in epithelial cell migration. J. Cell Sci. 116, Yoshimura, T. et al. (2002) ADAMs (A Disintegrin and Metalloproteinase) messenger RNA expression in H. pylori-infected, normal and neoplastic gastric mucosa. J. Infect. Dis. 185, Werner, M. et al. (2001) Gastric adenocarcinoma: pathomorphology and molecular pathology. J. Cancer Res. Clin. Oncol. 127, Terres, A.J. et al. (1998) Helicobacter pylori infection is associated with downregulation of E-cadherin, a molecule involved in epithelial cell adhesion and proliferation control. J. Clin. Pathol. 51, Guilford, P. et al. (1998) E-cadherin germline mutations in familial gastric cancer. Nature 392, Churin, Y. et al. (2001) Pathogenicity island dependent activation of Rho GTPases Rac1 and Cdc42 in Helicobacter pylori infection. Mol. Microbiol. 40, Furge, K.A. et al. (2000) Met receptor tyrosine kinase: enhanced signaling through adapter proteins. Oncogene 19, Trusolino, L. and Comoglio, M. (2002) Scatter-factor and semaphorin receptors: cell signalling for invasive growth. Nat. Rev. Cancer 2, Higashi, H. et al. (2002) SHP-2 tyrosine phosphatase as an intracellular target of Helicobacter pylori CagA protein. Science 295, Tsutsumi, R. et al. (2003) Attenuation of Helicobacter pylori CagA- SHP-2 signaling by interaction between CagA and C-terminal Src kinase. J. Biol. Chem. 278, Mimuro, H. et al. (2002) Grb2 is a key mediator of Helicobacter pylori CagA protein activities. Mol. Cell 10,