Helicobacter pylori CagA—a bacterial intruder conspiring gastric carcinogenesis

Authors

  • Masanori Hatakeyama

    Corresponding author
    1. Division of Molecular Oncology, Institute for Genetic Medicine, Hokkaido University, Sapporo, Japan
    • Division of Molecular Oncology, Institute for Genetic Medicine, Hokkaido University, Kita-15, Nishi-7, Kita-ku, Sapporo 060-0815, Japan
    Search for more papers by this author
    • Fax: +81-11-706-7544


Abstract

Infection with cagA-positive Helicobacter pylori (H. pylori) is associated with the development of gastric adenocarcinoma. The cagA gene product CagA is delivered from the bacterium into the cytoplasm of the bacterium-attached gastric epithelial cell via the type-IV secretion system. Upon membrane localization and subsequent tyrosine phosphorylation by Src family kinases, translocated CagA functions as a scaffolding adaptor that interacts with a number of host proteins involved in cell signaling in both tyrosine phosphorylation-dependent and -independent manners. Of special interest is the interaction of CagA with the SHP-2 tyrosine phosphatase, of which gain-of-function mutations have recently been found in human malignancies. Through the complex formation, SHP-2 is catalytically activated and induces morphological transformation that is associated with increased cell motility. In addition to the perturbation of intracellular signaling, CagA disrupts the apical junctional complex that regulates the cell–cell contact and maintains the integrity of the epithelial structure. These CagA activities may collectively cause cellular dysfunctions that promote accumulation of multiple genetic changes involved in malignant transformation. Further elucidation of host cell signaling targeted by CagA should provide a new paradigm for ‘bacterial carcinogenesis’ and also give insights into general understanding of inflammation-mediated cancers. Clinically, detailed studies on the relationship between structural diversity and degree of pathogenic activity of CagA should make it possible to identify a high-risk group for gastric carcinoma among H. pylori-infected populations through cagA genotyping. © 2006 Wiley-Liss, Inc.

Gastric adenocarcinoma is the second leading cause of cancer-related deaths worldwide, with 876,000 estimated new cases and 405,000 estimated deaths in the year 2000.1 Gastric adenocarcinoma is histopathologically subdivided into 2 types, intestinal type and diffuse type. Intestinal-type gastric adenocarcinoma occurs in older people and is more common than the diffuse type that affects younger people with poor prognosis. Although a small subset of diffuse-type adenocarcinoma is of familial origin and is caused by mutation in the E-cadherin gene,2 most do not possess such specific mutations, and the molecular mechanisms that underlie the development of nonhereditary gastric carcinoma remain to be elucidated. Recent epidemiological studies as well as infection studies using animal models have indicated that chronic infection with Helicobacter pylori (H. pylori) in the stomach plays a key role in the development of gastric adenocarcinoma.3, 4, 5, 6, 7, 8

H. pylori is a spiral-shaped bacterium first described in 1984 by Marshall and Warren. H. pylori is thought to inhabit at least half of the world's human population. The H. pylori genome is highly dynamic, generating a huge diversity of strains, with each strain showing significant differences in their genome sequence.9 Even a single strain may generate multiple variants and select those that adapt to an individual host environment during long-term colonization.10 Reflecting this genomic plasticity, H. pylori exhibits significant genetic polymorphisms, which may contribute to the different degrees of pathogenicity among distinct clinical H. pylori isolates.

H. pylori CagA and cag pathogenicity island

Strains of H. pylori can be divided into 2 major subpopulations based on their ability to produce a 120–145-kDa immunodominant protein called cytotoxin-associated gene A antigen (CagA).11, 12 CagA is encoded by the cagA gene, which is localized at one end of the cag pathogenicity island (cag PAI), a 40-kb DNA segment that is thought to be incorporated into the H. pylori genome by horizontal transfer from unknown origin.13, 14 More than 90–95% of H. pylori strains isolated in East Asian countries such as Japan, Korea and China (hereafter called East Asian H. pylori strains) carry cag PAI and therefore cagA-positive. In contrast, ∼40% of H. pylori strains isolated in Western countries such as European countries, America and Australia (Western H. pylori strains) are cagA-negative. The cag PAI DNA contains 31 putative genes, including cagA and those encoding components of the type IV secretion system. The type IV secretion system acts as a molecular “syringe” through which macromolecules are delivered from the inside to the outside of the bacterium. Infection with a cagA-positive H. pylori strain causes gastric mucosal inflammation of higher grade than that caused by a negative strain and is associated with severe atrophic gastritis and gastric adenocarcinoma.15, 16, 17, 18, 19, 20, 21, 22

Translocation and tyrosine phosphorylation of H.pylori CagA

Recent studies have revealed a novel mechanism by which cagA-positive H. pylori attacks gastric epithelial cells. In the stomach, a fraction of the bacteria bind specifically to the surfaces of the gastric epithelial cells. Upon attachment, cagA-positive H. pylori directly injects the CagA protein into the host gastric epithelial cell via the cag PAI encoded-type IV secretion system.23, 24, 25, 26, 27 Translocated CagA localizes to the inner surface of the host plasma membrane, where it undergoes tyrosine phosphorylation by Src family kinases (SFKs).28, 29 Tyrosine phosphorylation of the CagA protein occurs rapidly after translocation into the AGS gastric epithelial cells, indicating that at least some SFKs are constitutively active in this cell type.23, 24, 25, 26, 27 Reversible protein tyrosine phosphorylation is a major signaling mechanism by which proliferation, survival, differentiation and migration of mammalian cells are regulated. Furthermore, deregulation of kinases and/or phosphatases that control the level of tyrosine phosphorylation is fundamentally associated with cellular transformation. This fact raises the possibility that upon tyrosine phosphorylation the bacterial protein deregulates intracellular signaling, which directly or indirectly contributes to gastric carcinogenesis.

CagA is tyrosine-phosphorylated at Glu-Pro-Ile-Tyr-Ala (EPIYA) motifs, which are present in multiple numbers in the carboxy-terminal region of the protein (Fig. 1).28, 29, 30, 31 From the sequences surrounding these EPIYA motifs, 4 distinct EPIYA segments, EPIYA-A, -B, -C and -D, each of which contains a single EPIYA motif, have been identified in the CagA protein.31, 32, 33 The representative CagA proteins of Western H. pylori strains (Western CagA) each possess a 32-amino-acid (AA) EPIYA-A segment and a 40-AA EPIYA-B segment, followed by a 34-AA EPIYA-C segment (A-B-C type CagA) (Fig. 1). The EPIYA-C segment variably duplicates among different Western CagA species because of homologous recombination of a 102-bp cagA gene fragment, which encodes the EPIYA-C segment. The prevalent CagA protein of East Asian H. pylori isolates (East Asian CagA) also contains the EPIYA-A and EPIYA-B segments but not the repeatable EPIYA-C segment. Instead, the protein has a distinct EPIYA-containing sequence, termed the EPIYA-D segment, which is unique to East Asian CagA (Fig. 1).31 East Asian CagA is therefore regarded as A-B-D type CagA. The tyrosine residues within the EPIYA-C and EPIYA-D sites are the major sites of phosphorylation in Western and East Asian CagA, respectively. In contrast, the EPIYA-A and EPIYA-B sites are only weakly tyrosine-phosphorylated in cells.31 It has recently been reported that the EPIYA sequence also acts as a membrane-targeting motif of CagA, regardless of tyrosine phosphorylation.33 Thus, the EPIYA motif has a dual function in membrane association and tyrosine phosphorylation of CagA.

Figure 1.

Structure of the EPIYA-repeat region of CagA. Prevalent Western CagA proteins contain 3 EPIYA motifs, which are designated EPIYA-A, EPIYA-B and EPIYA-C sites based on the sequence surrounding each of the EPIYA motifs. Accordingly, the EPIYA-repeat region of CagA can be subdivided into the EPIYA-A segment (blue box), the EPIYA-B segment (green box) and the EPIYA C segment (orange box). The EPIYA-C segment variably multiplies (mostly from one to three times) among different Western CagA species. East Asian CagA does not have the EPIYA-C segment, but instead possesses the EPIYA-D segment (purple box). EPIYA-C and EPIYA-D sites are major tyrosine phosphorylation sites of Western CagA and East Asian CagA, respectively. EPIYA-D site of East Asian CagA binds SHP-2 more strongly than does EPIYA-C site of Western CagA. Among Western CagA species, those having larger numbers of EPIYA-C exhibit stronger SHP-2 binding activity and greater activity to induce the hummingbird cells than those having lesser numbers of EPIYA-C do.

Tyrosine phosphorylation-dependent biological activities of CagA

There has been an accumulation of evidence showing that tyrosine-phosphorylated CagA specifically interacts with SH2 domain-containing cellular proteins. This indicates that the bacterial protein functions as a scaffolding adaptor molecule, which functionally mimics mammalian proteins such as those of Gab or IRS family (Fig. 2).34

Figure 2.

Tyrosine phosphorylation-dependent biological activities of CagA. CagA is translocated from H. pylori into H. pylori-attached gastric epithelial cells via the bacterial type IV secretion system. The translocated CagA protein localizes to the inner surface of the plasma membrane and undergoes tyrosine phosphorylation by the SFK. Upon tyrosine-phosphorylation, CagA specifically binds to SH2 domain-containing tyrosine phosphatase SHP-2 and activates the phosphatase activity. CagA-activated SHP-2 dephosphorylates cellular substrates such as FAK, which is involved in the induction of hummingbird cells. CagA-activated SHP-2 also potentiates Erk MAP kinase activity, which regulates both cell growth and cell morphology. Tyrosine phosphorylated CagA also binds specifically to Csk and stimulates the kinase activity. As a result, CagA inhibits SFK activity via Csk activation and thereby attenuates tyrosine phosphorylation-dependent CagA activities. Tyrosine phosphorylated CagA also inhibits SFK kinase activities directly. Downregulation of SFKs by CagA results in the reduced levels of cortactin tyrosine phosphorylation, which may also play a role in induction of the morphological change by CagA.

CagA-SHP-2 interaction

Upon tyrosine phosphorylation by SFKs, CagA acquires the ability to specifically bind to SHP-2, a cytoplasmic Src-homology 2 (SH2) domain-containing protein tyrosine phosphatase.30 SHP-2 is composed of 2 tandemly repeated N-terminal SH2 domains (N-SH2 and C-SH2), a single catalytic protein tyrosine phosphatase (PTP) domain and a C-terminal tail.35 The complex formation between CagA and SHP-2 is detectable not only in cells in vitro infected or transfected with CagA but also in gastric mucosa of patients infected with cagA-positive H. pylori.36 Interactions of SHP-2 with Western CagA and East Asian CagA are mediated by the tyrosine-phosphorylated EPIYA-C and EPIYA-D sites, respectively.31 Furthermore, both of the N-SH2 and C-SH2 domains of SHP-2 are essential in the CagA-SHP-2 interaction. The result indicates that 2 of the tyrosine-phosphorylated EPIYA sites, either in cis or trans, are required for the stable complex formation between CagA and SHP-2. Crystallographic data of SHP-2 suggest that the N-SH2 domain interacts with the catalytic cleft of the PTP domain, thereby blocking substrate access.37 This closed structure represses basal phosphatase activity. Binding of tyrosine-phosphorylated CagA to the SH2 domains may therefore elicit a conformational change in SHP-2 that weakens the inhibitory interaction between the PTP domain and the N-SH2 domain, resulting in the activation of SHP-2 phosphatase.30

In most cases, SHP-2 functions as a positive regulator of signals generated by extracellular stimuli, including those for cell growth and cell motility.35 Consistently, gastric epithelial cells expressing CagA undergo a unique morphological change termed the “hummingbird” phenotype, which is characterized by the formation of needle-like cell protrusions.23, 30 Induction of the hummingbird phenotype requires deregulated SHP-2 phosphatase activity caused by the CagA-SHP-2 interaction.30, 38, 39 It was recently found that CagA-activated SHP-2 directly dephosphorylates focal adhesion kinase (FAK) at the activating tyrosine phosphorylation sites in gastric epithelial cells.40 Coexpression of constitutively active FAK with CagA inhibits induction of the hummingbird phenotype, whereas expression of dominant-negative FAK elicits an elongated cell shape characteristic of the hummingbird phenotype. Thus, inhibition of FAK kinase activity by CagA-activated SHP-2 plays a crucial role in the morphogenetic activity of CagA.

SHP-2 has been shown to positively regulate Erk MAP kinase activity by both Ras-dependent and -independent mechanisms.35 Consistently, expression of CagA in gastric epithelial cells results in prolonged activation of Erk MAP kinase activity.38 Sustained Erk activation has been suggested to play an important role in the progression of G1 to S phase.41 CagA may therefore predispose gastric epithelial cells to deregulated proliferation at least partly through a sustained Erk signal. Erk MAP kinase activity is also required for induction of the hummingbird phenotype by CagA,38 raising the possibility that Erk lies downstream of CagA-activated SHP-2.

CagA–Csk interaction

In addition to SHP-2, tyrosine-phosphorylated CagA has been shown to specifically bind to the SH2 domain of the C-terminal Src kinase (Csk).42 CagA–SHP-2 interaction requires the EPIYA-C or EPIYA-D site, whereas CagA–Csk interaction is mediated by the EPIYA-A or EPIYA-B site of CagA in gastric epithelial cells.40 Through the complex formation, CagA stimulates the kinase activity of Csk, which in turn phosphorylates SFKs at the C-terminal inhibitory tyrosine residue. As a result, SFK activity is downregulated in cells expressing CagA in a manner dependent on CagA phosphorylation. Given that SFKs phosphorylate CagA, this indicates the presence of a feedback mechanism that attenuates CagA-SHP-2 signaling, which may otherwise elicit excess toxicity in gastric epithelial cells.32, 42 This negative feedback regulation may ensure a long-term colonization of cagA-positive H. pylori in the stomach for decades, without causing acute and fatal damage to the host.

CagA-c-Src interaction

It has been reported that CagA directly inhibits SFK activity in a manner dependent on its tyrosine phosphorylation.43 Thus, CagA counteracts SFKs in both direct and indirect (through Csk) mechanisms. Considering that CagA itself is a substrate of SFKs, excess CagA may competitively inhibit tyrosine phosphorylation of physiological SFK substrates rather than directly inhibiting SFK kinase activity. CagA-mediated inhibition of SFK activity is associated with a decrease in the level of tyrosine-phosphorylated cortactin.43 Since cortactin is involved in the regulation of actin cytoskeletal rearrangement, a reduced level of cortactin tyrosine phosphorylation may also play a role in induction of hummingbird cells by CagA.

Phosphorylation-independent biological activities of CagA

In addition to the functions of CagA as a phosphorylation-dependent scaffolding adaptor, recent studies have revealed phosphorylation-independent activities of CagA, which also cause cellular dysfunctions (Fig. 3).

Figure 3.

Tyrosine phosphorylation-independent biological activities of CagA. Translocated CagA binds to Grb2 and activates the Ras-MAP kinase pathway, thereby potentiating cell proliferation as well as cell motility. CagA also interacts with the c-Met HGF receptor and deregulates HGF signaling that stimulates cell growth and cell motility. Association between CagA and ZO-1 leads to disruption of tight junctions and impairs cell–cell interaction. Furthermore, CagA induces translocation of cytoplasmic NFAT to the nucleus, where it transactivates a set of genes involved in cell growth and cell differentiation. These CagA activities are independent of CagA tyrosine phosphorylation. Activation of NFAT by CagA is counteracted by another H. pylori virulence factor VacA.

CagA-c-Met/PLC-γ interaction

Infection with H. pylori leads to activation of the hepatocyte growth factor (HGF) receptor c-Met in gastric epithelial cells. H. pylori-activated c-Met undergoes autophosphorylation (more precisely trans-phosphorylation) and acquires the ability to interact with CagA regardless of the status of CagA phosphorylation.44 The CagA-c-Met interaction potentiates the c-Met receptor signaling, which stimulates both cell proliferation and cell motility. H. pylori infection also activates PLC-γ and induces association of CagA with PLC-γ, although the pathophysiological significance of this interaction remains to be elucidated.44

CagA–Grb2 interaction

CagA interacts with the Grb2 adaptor protein and activates the Ras-MAP kinase pathway.45 Although the CagA-Grb2 interaction requires the EPIYA-repeat region of CagA as well as the SH2 and SH3 domains of Grb2, it is independent of CagA tyrosine phosphorylation. The CagA-Grb2 interaction activates the Ras-MAP kinase pathway and thus promotes the growth of gastric epithelial cells. The interaction also enhances cell motility and induces the scattering phenotype.

Association of CagA with components of the apical-junctional complex

Beside its ability to deregulate intracellular signaling, translocated CagA perturbs the apical junctions of epithelial cells that regulate cell–cell adhesion and maintain the integrity of the cell barrier. Upon translocation and membrane association, CagA colocalizes with tight-junction proteins such as ZO-1 and JAM, causing an ectopic assembly of tight-junctional components at sites of bacterial attachment.46 CagA also recruits signaling molecules, such as SHP-2, in close proximity to the tight junction, which may alter apical-junctional complex function. Colocalization of CagA with these tight junctional proteins requires the NH2-terminal region of CagA,47 although it is not known whether CagA directly interacts with these junctional proteins. The altered cell–cell contacts by CagA are associated with a morphogenetic change in polarized epithelial cells that resembles an epithelial-to-mesenchymal transition.47

Activation of transcription factors by CagA

Genome-wide analysis of genes affected by H. pylori infection in polarized epithelial cells has revealed that more than 90% of the genes affected are cag PAI-dependent, reinforcing the importance of cag PAI in the H. pylori-host cell interaction.48 Among these cag PAI-dependent genes, 79% are CagA-dependent. Of these, 68% are tyrosine phosphorylation-dependent. The results indicate that CagA is capable of controlling transcription factors via both phosphorylation-dependent and -independent mechanisms. Indeed, CagA can activate serum responsive element (SRE)-dependent transcription, which is most likely mediated by Elk1, in a manner independent of CagA phosphorylation.49 Recent studies have also shown that CagA is capable of activating NF-κB via the Ras-MAP kinase pathway.50 Taken together with the finding that delivery of the H. pylori-derived proteoglycan into host cells via the type IV secretion system activates NOD1,51 which also stimulates NF-κB in gastric epithelial cells, it appears that H. pylori activates the NF-κB transcription factor through multiple distinct mechanisms and induces proinflammatory cytokines such as IL-8.

CagA has recently been found to stimulate the calcium-dependent serine/threonine phosphatase calcineurin and thereby induce translocation of the nuclear factor of activated T cells (NFAT) from the cytoplasm to the nucleus, where it transactivates NFAT-dependent genes, in gastric epithelial cells.52 Although the molecular mechanism by which CagA activates the calcineurin-NFAT system remains to be elucidated, the CagA activity is again independent of its tyrosine phosphorylation. It is possible that the reported CagA-PLCγ interaction triggers Ca2+ mobilization and subsequent activation of calcineurin.44 One of the NFAT-dependent genes activated by CagA in gastric epithelial cells is p21Cip1 cyclin-dependent kinase inhibitor.52 Accordingly, although CagA activates a growth-promoting signal via the SHP-2-MAP kinase pathway or the Grb2-Ras-MAP kinase pathway, it simultaneously inhibits cell proliferation through NFAT-dependent p21Cip1 induction. Intriguingly, the H. pylori-vacuolating toxin VacA counteracts the activity of CagA to stimulate NFAT.52 Thus, VacA has a role in determining the magnitude of NFAT deregulation in gastric epithelial cells expressing CagA. Such a functional interplay between CagA and VacA has already been suggested from the observation that secretion of VacA protein is associated with the presence of CagA despite the presence of vacA gene in all H. pylori strains.53

EPIYA-repeat polymorphism and biological activity of CagA

Based on the structure of the EPIYA-repeat region (see Fig. 1), most of the Western CagA proteins are categorized as the A-B-C type, in which the number of EPIYA-C sites varies from one to the next, mostly ranging from 1–3.11, 54 In contrast, prevalent East Asian CagA species belong to the A-B-D type. Notably, however, a small number of East Asian CagA species show complicated variations in the EPIYA-repeat region.31, 33

East Asian CagA (A-B-D type CagA) and Western CagA (A-B-C type CagA) utilize the EPIYA-D and EPIYA-C sites for their interaction with SHP-2, respectively.31 The N- and C-SH2 domains of SHP-2 bind with high affinity to the phosphotyrosine (pY)-containing consensus sequence, pY-(V/T/A/I/S)-X-(L/I/V)-X-(F/W).55 The sequence flanking the EPIYA-D site of East Asian CagA perfectly matches the consensus sequence. On the other hand, the sequence flanking EPIYA-C of Western CagA differs from the consensus sequence by a single amino acid at the pY + 5 position. Accordingly, whereas the A-B-C type East Asian CagA and A-B-D type Western CagA show equal levels of tyrosine phosphorylation, East Asian CagA exhibits stronger binding activity for SHP-2 and greater ability to induce the hummingbird phenotype compared with Western CagA.31 Among Western CagA species, the number of EPIYA-C sites is directly correlated with the levels of tyrosine phosphorylation, SHP-2 binding activity and morphogenetic activity of CagA.

Role of CagA in gastric carcinogenesis

H. pylori CagA exhibits biological activities that are reminiscent of mammalian oncoproteins, and signaling pathways deregulated by CagA are frequently associated with the development of human cancers. These findings raise the possibility that CagA is a bacterial oncoprotein that plays a key role in gastric carcinogenesis.

Multistep nature of gastric cancer development

Development of cancer is regarded as a multistep process that is characterized by progressive, multiple genetic/epigenetic alterations that stepwisely drive the transformation of normal cells into highly malignant cells. The gastric histopathological changes associated with chronic H. pylori infection typically include chronic gastritis with later development of intestinal metaplasia and gastric atrophy. Most cases of intestinal-type gastric carcinoma develop on gastric mucosa with multifocal atrophic gastritis, usually with extensive intestinal metaplasia. This finding suggests that intestinal metaplasia and gastric atrophy are premalignant lesions of the stomach. Based on these notions, Correa proposed a multistep model of gastric carcinogenesis: superficial gastritis—atrophic gastritis—intestinal metaplasia—dysplasia—adenocarcinoma.56 Because these mucosal lesions are intimately associated with chronic H. pylori infection, the pathological changes provide an additional link of H. pylori infection with gastric carcinoma.

Bifunctional role of CagA in cell proliferation

How can CagA contribute to the multistep gastric carcinogenesis? An intriguing biological feature of CagA is that it has a dual role in host cell proliferation; it stimulates growth-promoting molecules and at the same time induces a potent cell growth inhibitor, p21Cip1.30, 44, 45, 52 Accordingly, when expressed in AGS gastric epithelial cells, CagA causes G1-cell cycle arrest rather than proliferation, indicating that elevated p21Cip1 dominates over the growth-promoting activity of CagA.52 Thus, CagA may initially act as a p21Cip1-dependent suppressor of cell proliferation, which should contribute to the development of atrophic gastritis or peptic ulcers. Given that NFAT is involved in differentiation of cells of various lineages and that G1-cell cycle arrest is a prerequisite for differentiation programs to be triggered, one can speculate that CagA-activated p21Cip1 and NFAT cooperatively induce abnormal intestinal transdifferentiation of gastric epithelial cells known as intestinal metaplasia. The growth-inhibitory activity of CagA may also facilitate programmed cell death (apoptosis) and compensatory proliferation of residual epithelial (stem) cells. The increased cell turnover then predisposes cells to incur genetic/epigenetic changes, positively selecting variant cells such as those harboring alteration in p21Cip1 expression. Since CagA also stimulates growth-promoting molecules through multiple distinct mechanisms,30, 44, 45 p21Cip1 loss converts CagA from a cell-cycle inhibitor into a cell growth stimulator. Such a dual function of CagA may explain at least partly the gastric mucosal changes that lead to adenocarcinoma. Notably, however, exposure of CagA to the host cells is an epigenetic event as the cagA gene is not integrated into the host genome. Also, once established, gastric adenocarcinoma is independent of cagA-positive H. pylori for its autonomous growth as well as other malignant phenotypes. These facts indicate that the oncogenic role of CagA may be specifically confined to the early stages of gastric carcinogenesis. Consistent with this notion, H. pylori-injected CagA proteins are primarily detected in atrophic mucosa but not in the mucosa of intestinal metaplasia or gastric carcinoma because H. pylori cannot efficiently colonize mucosa with severe intestinal metaplasia or gastric carcinoma.36 Thus, CagA may play a specific role in the development and/or maintenance of precancerous gastric lesions, atrophic gastritis and intestinal metaplasia, from which gastric adenocarcinoma arises.

Results of recent studies using a Helicobacter felis-infected mouse model have led to the conclusion that gastric adenocarcinoma originates from circulating bone marrow-derived stem cells (BMDC), not from resident stem cells.57 If this is also the case in humans, chronic mucosal inflammation provoked by cagA-positive H. pylori may exhaust gastric stem cells and eventually lead to depletion of the resident stem cell pool, resulting in recruitment of BMDC to the damaged gastric mucosa.

SHP-2, a cellular oncoprotein

Among the various CagA targets in gastric epithelial cells, SHP-2 may play an important role in gastric carcinogenesis because mutations of SHP-2 have recently been found in various human malignancies.58, 59 Most of the reported cases carry missense mutations in exons 3 and 8 in the PTPN11/SHP-2 gene, which encode segments of the N-SH2 domain and the PTPase domain, respectively. Molecular modeling of SHP-2 indicates that such mutations weaken the autoinhibitory interaction, resulting in the constitutive activation of phosphatase activity. Accordingly, deregulation of SHP-2 by CagA functionally mimics the gain-of-function mutation of SHP-2 that is associated with human malignancies. Normal cells are often induced into pathways that lead to cell growth arrest, senescence, and/or apoptosis in response to the acute expression of oncoproteins such as activated Ras or Raf.60, 61 Consistently, sustained hyperactivation of SHP-2 in gastric epithelial cells can induce apoptosis.42 Thus, deregulation of SHP-2 by CagA in conjunction with elevated p21Cip1 may cooperatively elicit cell growth inhibition and subsequent apoptosis, providing a pathophysiological situation that further enhances turnover of gastric epithelial cells.

CagA can transform primary gastric epithelial cells, which are immortalized by SV40 T antigen, through activation of the Erk MAP kinase pathway independent of Ras.62 This observation is consistent with the finding that CagA-stimulated SHP-2 induces sustained Erk MAP kinase activation, which is also independent of Ras.38 A potential role of SHP-2 in the development of gastric carcinoma has also been suggested by results of a recent study demonstrating that genetically engineered mice lacking the SHP-2 binding site on the IL-6 family coreceptor gp130 develop intestinal-type gastric adenocarcinoma.63

Impaired cell–cell contacts in cancer development

The pathophysiological effect of CagA on cell–cell interaction may also contribute to the development of gastric carcinoma.46, 47 The apical junctions create a polarized intercellular barrier between luminal and serosal fluid compartments and segregate luminal growth factors from their basal–lateral receptors.64 Breakdown of this barrier should allow access of luminal growth factors to the homologous receptors expressed on the basal–lateral cell membranes. This property, while important in the epithelial regeneration in response to tissue injury, may promote the development of cancer cells in premalignant epithelial tissues, in which the cell–cell junctions leak growth factors. Loss of cell–cell contact is also associated with morphological change that resembles the epithelial-to-mesenchymal transition, a pathological dedifferentiation that involves the loss of epithelial cell polarity and the acquisition of a variety of mesenchymal phenotypic traits.65 Evidence suggests that the epithelial-to-mesenchymal transition plays a role in the migration of cancer cells from a primary tumors into the circulation during metastasis. In this regard, the questing remains as to whether CagA can still play a role in the late stages of gastric carcinogenesis such as cancer metastasis because H. pylori does not attach to a mucosal lesion with severe atrophy or gastric carcinoma.

It has also recently been reported that in vivo adaptation endows an H. pylori strain with the ability to induce gastric adenocarcinoma in Mongolian gerbils.66 The oncogenic H. pylori strain can specifically deregulate β-catenin, an essential component of the adherens junctions, in a CagA-dependent manner. Furthermore, nuclear accumulation of β-catenin is increased in gastric epithelium harvested from persons carrying cagA-positive H. pylori. Given the critical role of mutated E-cadherin, another component of the adherens junctions in hereditary diffuse-type gastric cancer,2 it is reasonable to speculate that H. pylori-induced deregulation of β-catenin, which impairs the adherens junction and at the same time induces β-catenin-dependent genes, plays a role in the development of sporadic gastric adenocarcinoma as well.

CagA polymorphism and gastric cancer

The incidences of gastric carcinoma in East Asian countries such as Japan, Korea and China are significantly higher than those in Europe, North America and Australia.32 In addition to the genetic differences among distinct ethnic groups as well as environmental factors, the diversity of H. pylori strains circulating in different geographic areas may contribute to the wide variation in the incidence of gastric carcinoma throughout the world. East Asian H. pylori and Western H. pylori possess CagA proteins with distinctly structured tyrosine phosphorylation/SHP-2-binding sites, EPIYA-D and EPIYA-C, respectively. The EPIYA-D site shows stronger SHP-2 binding and greater morphogenetic activity than does the EPIYA-C site.31 In accordance with the results of these in vitro studies, the degrees of inflammation, activity of gastritis and atrophy are significantly higher in patients infected with East Asian cagA-positive strains than in patients infected with Western cagA-positive strains.67, 68 Thus, populations infected with East Asian cagA-positive H. pylori may be at greater risk for gastric carcinoma than those infected with Western cagA-positive or cagA-negative strains. Among the Western CagA species, the number of EPIYA-C segments is directly correlated with the levels of tyrosine phosphorylation, SHP-2 binding activity and morphogenetic activity of CagA.31 Thus, H. pylori strains carrying Western CagA proteins with a greater number of EPIYA-C segments may be pathophysiologically more virulent and thus more carcinogenic than those carrying less EPIYA-C segments. This notion has been supported by the results of a recent work demonstrating that 5 of 6 Western H. pylori strains harvested from gastric carcinoma patients possessed multiple EPIYA-C sites, whereas 18 of 19 Western H. pylori strains isolated from noncancer patients possessed a single EPIYA-C site.54

Conclusions

Deregulation of host signaling by translocated bacterial proteins provides a new aspect of microbial-host cell interaction. Furthermore, H. pylori CagA-host cell interaction is introducing a new paradigm for “bacterial carcinogenesis.” The bacterial protein targets the tyrosine phosphorylation system in mammalian cells, the disturbance of which plays a crucial role in the development of cancer. Elucidation of the mechanism of H. pylori CagA-induced cellular dysfunction will not only uncover the molecular process leading to gastric carcinoma but also shed light on the pathophysiological framework underlying other infection/inflammation-associated cancers such as hepatocellular carcinoma, colorectal carcinoma, cervical carcinoma and lung carcinoma.

From a clinical point of view, identification of H. pylori and host factors that increase the risk of gastric cancer development will promise the development of new strategies for prevention as well as treatment of gastric carcinoma. It is becoming clear that polymorphisms in cytokine genes play an important role in gastric cancer development.69 Furthermore, genetic diversity of critical pathogenic elements such as the cagA gene appears to determine the degree of oncogenic potential of individual H. pylori strains. Although H. pylori infection is declining as the standard of living rises, about half of the world human populations have already been infected with H. pylori and are at a higher risk of developing gastric carcinoma in the short or medium term. Results of recent clinical studies have shown that chemointervention of H. pylori infection in patients lowers the risk of developing gastric carcinoma.70 Thus, it is important to identify the high-risk populations that are genetically more susceptible to gastric carcinoma and at the same time infected with more virulent/oncogenic H. pylori strains. Eradication of H. pylori from such high-risk populations would markedly reduce the incidence of gastric carcinoma.

Ancillary