In this issue of Clinical Cancer Research, Leung et al. (1) present a comprehensive analysis of the effects of Helicobacter pylori eradication on the methylation status of the E-cadherin gene.
Germ line mutations within the E-cadherin-encoding gene (CDH1) were first identified in three Maori families predisposed to diffuse gastric cancer (2). Subsequently, several gastric cancer families from a variety of ethnic backgrounds, including Caucasian, Japanese, Korean, and African-American, have been found to carry CDH1 mutations, which led to the recognition of a distinct family gastric cancer syndrome termed hereditary diffuse gastric cancer (2, 3). E-cadherin is a 120 kDa calcium-dependent adhesion molecule that forms intercellular connections through homophilic interactions and maintains the organization of epithelial tissues. E-cadherin also regulates the normal cellular localization and function of catenins (α, β, and γ). Consistent with the concept that loss of normal cell-to-cell contact and dysregulation of catenin function promotes cancer, down-regulation of E-cadherin in tumors is associated with a poor prognosis and metastasis. Although biallelic inactivation of CDH1 can occur through somatic mutations or allelic imbalance events (loss of heterozygosity), aberrant methylation of the CDH1 promoter has also been shown to epigenetically inactivate E-cadherin in gastric cancer cases (4, 5).
In 1994, the IARC recognized H. pylori as a carcinogen for cancer of the stomach (6). The same designation has been given to human papilloma viruses and hepatitis viruses, and, collectively, these infections account for a major portion of the global neoplastic human health burden (7). There is great interest in unraveling the mechanisms of carcinogenesis used by these pathogens in the hope that it may provide insights into the carcinogenic process in general and offer the opportunity to design strategies for prevention. Establishment of H. pylori as a risk factor for cancer of the stomach has permitted an approach that identifies persons at increased risk; however, infection with this organism is extremely common and most colonized persons never develop cancer. Therefore, techniques to identify high-risk subpopulations must use other biological markers.
One H. pylori virulence constituent is the cag pathogenicity island, a multigene locus present in 60% to 70% of U.S. strains, and carriage of cag+ strains augments the risk for gastric adenocarcinoma compared with that incurred by cag− strains (Fig. 1; ref. 8). Several cag genes encode products that bear homology to components of a type IV bacterial secretion system that functions as a molecular syringe to export proteins, and the product of the terminal gene in the island (CagA) is translocated into host cells after bacterial attachment (9). Although cag+ strains are overrepresented among hosts who develop serious sequelae of infection, most persons colonized by these strains remain asymptomatic (8); this has highlighted the need to also define host factors that influence disease outcome. Polymorphisms within immune response genes, such as IL-1β, TNFα, and IL-10, that are associated with increased proinflammatory gene expression heighten the risk for gastric adenocarcinoma among H. pylori–infected persons (Fig. 1; refs. 10, 11). Further, a synergistic effect on cancer risk is present when virulence determinants of infecting H. pylori isolates are examined in conjunction with the polymorphism status of their cognate hosts (11, 12). These findings have helped construct a framework from which physicians can more appropriately focus H. pylori diagnostic and eradication strategies on targeted high-risk populations to optimize prevention of subsequent neoplastic events.
Progression to intestinal-type gastric cancer. H. pylori infection leads to superficial gastritis over a period of weeks. The presence of proinflammatory host polymorphisms and the H. pylori cag pathogenicity island increase the risk of developing gastric atrophy, intestinal metaplasia, and gastric adenocarcinoma. Epigenetic inactivation of E-cadherin via promoter hypermethylation may also contribute to intestinal-type gastric cancer.
Progression to intestinal-type gastric cancer. H. pylori infection leads to superficial gastritis over a period of weeks. The presence of proinflammatory host polymorphisms and the H. pylori cag pathogenicity island increase the risk of developing gastric atrophy, intestinal metaplasia, and gastric adenocarcinoma. Epigenetic inactivation of E-cadherin via promoter hypermethylation may also contribute to intestinal-type gastric cancer.
Most efforts to understand gastric carcinogenesis have focused on characterizing the accumulation of genetic mutations within epithelial cells that are involved in neoplastic transformation (13), following the classic Vogelstein model for colon carcinogenesis (14). The realization that epigenetic phenomena may play a crucial role in carcinogenesis is more recent. Like H. pylori infection itself, an increased frequency of methylation changes may have high sensitivity but low specificity; however, combinatorial analyses using different markers can enhance both sensitivity and specificity. To determine whether epigenetic inactivation of E-cadherin may play a role in H. pylori–induced gastric carcinogenesis, Leung et al. (1) performed a comprehensive analysis of E-cadherin methylation profiles before and after successful H. pylori eradication using gastric tissue harvested from infected subjects residing in Hong Kong, an area that has a high incidence of gastric cancer.
Summary of Article
Of 28 colonized patients, 25 had serial gastric biopsies obtained before and 1 year after successful anti–H. pylori therapy, and 18 (72%) of these 25 subjects had histologic evidence of gastric cancer precursor lesions (intestinal metaplasia). Compared with pretreatment samples, there was a mild reduction in the number of patients with E-cadherin methylation-positive gastric samples after antimicrobial therapy as determined by methylation-specific PCR; however, this did not reach statistical significance (1). When the investigators did a more comprehensive assessment of E-cadherin methylation (bisulfite sequencing of PCR-amplified CpG rich promoter and exon 1 regions of E-cadherin) on a subset of patients, a significant reduction in mean methylation density was found 1 year after eradication among the pooled samples. The intensity of gastric inflammation was higher in samples containing methylated E-cadherin versus samples without methylation, implicating the host inflammatory response in regulation of epigenetic E-cadherin inactivation (1). Unfortunately, the presence of E-cadherin methylation was not associated with alterations in E-cadherin expression as determined by immunohistochemistry and H. pylori did not induce E-cadherin promoter methylation in an in vitro model of bacterial-epithelial cell interaction.
Commentary and Discussion
The article by Leung et al. (1) in this issue of the journal adds new and valuable information linking a bacterial infection with methylation of the E-cadherin gene, and several important points emanate from the current study. First, identification of a potential gastric cancer biomarker, such as E-cadherin methylation, could permit H. pylori–infected subjects to be stratified into high versus low-risk groups, which may affect therapeutic decisions. Second, the finding that aberrant E-cadherin methylation may influence the propensity toward gastric cancer could lead to novel chemoprevention therapies. For example, in a small study of two HDGC families, half of the tumors showed methylation of the CDH1 promoter as the second hit; it is tempting to consider the potential value of prophylactic treatment of these mutation carriers with demethylating agents, such as 5-azadeoxycytidine (5). Finally, studies that identify potential biomarkers of gastric cancer, such as E-cadherin methylation, may also provide important insights into mechanisms that underpin H. pylori–induced carcinogenesis. As noted earlier, one of the main functions of E-cadherin is regulation of β-catenin (Fig. 2). Aberrant activation of β-catenin leads to its nuclear accumulation and the formation of heterodimers with the transcription factor LEF/TCF, resulting in transcriptional activation of genes that influence carcinogenesis (15). Nuclear accumulation of β-catenin is increased in gastric adenoma and dysplasia specimens, histologic stages that precede gastric adenocarcinoma (16). Recent data also indicate that H. pylori can activate β-catenin in model gastric epithelia in vitro via translocation of CagA, and nuclear accumulation of β-catenin is increased in gastric epithelium harvested from persons carrying cag+ versus cag− strains or uninfected persons (17). Whether these events are dependent on epigenetic inactivation of E-cadherin remains to be determined.
Hypothetical model of β-catenin signaling in unstimulated (A) or H. pylori–infected gastric epithelial cells (B and C). A, in the absence of H. pylori, β-catenin is bound by E-cadherin at the cellular membrane and low steady-state levels of free β-catenin are present in the cytosol or the nucleus. B, after H. pylori infection, the E-cadherin promoter becomes hypermethylated, leading to inactivation of E-cadherin transcription. C, epigenetic inactivation of E-cadherin allows β-catenin to translocate to the nucleus and initiate transcription of genes implicated in carcinogenesis.
Hypothetical model of β-catenin signaling in unstimulated (A) or H. pylori–infected gastric epithelial cells (B and C). A, in the absence of H. pylori, β-catenin is bound by E-cadherin at the cellular membrane and low steady-state levels of free β-catenin are present in the cytosol or the nucleus. B, after H. pylori infection, the E-cadherin promoter becomes hypermethylated, leading to inactivation of E-cadherin transcription. C, epigenetic inactivation of E-cadherin allows β-catenin to translocate to the nucleus and initiate transcription of genes implicated in carcinogenesis.
There are also several issues raised by this study that will only be resolved with larger patient populations and more extended follow-up. Although chronic gastritis is the strongest known risk factor for gastric adenocarcinoma, eradication of H. pylori decreases gastric cancer incidence only in infected individuals without premalignant lesions, such as intestinal metaplasia (18). The vast majority of subjects in the current study (72%) had intestinal metaplasia at baseline and the prevalence of intestinal metaplasia did not change after H. pylori eradication. One year may be too premature to detect significant decreases in E-cadherin methylation patterns among individual patients, although a recent study reported that H. pylori eradication did result in a significant decrease in the number of subjects with E-cadherin promoter hypermethylation compared with untreated subjects, as early as 6 weeks after treatment (19). In that study, it was notable that infected patients had gastritis alone with no evidence of intestinal metaplasia at baseline (19). Gastric cancer prevention tends to follow a pattern that is in reverse parallel to the carcinogenesis process per se, as first described for lung cancer by Doll et al. (20). A recently completed chemoprevention trial in Colombia has shown that the prevention process, similar to gastric carcinogenesis itself, follows an exponential curve in which the first years of exposure (or lack of) have minimal measurable effects on markers of progression (21), but this is followed by greater effects in subsequent years (Fig. 3). The earlier the intervention (e.g., smoking for lung cancer, cure of H. pylori infection in gastric cancer prevention), the more effective the influence. Therefore, future studies should strive to enroll younger patients without detectable premalignant lesions who are evaluated at longer time points to determine if a higher response rate in regression of E-cadherin methylation may occur in this population. More extended follow-up intervals may also permit the discordant findings between E-cadherin methylation and E-cadherin expression patterns noted in the current study to be resolved.
Gastric histopathology scores in H. pylori–infected persons residing in a region with a high incidence of gastric cancer that were either successfully treated, untreated, or who failed antimicrobial therapy, by years of follow-up. Modified from ref. 21.
Gastric histopathology scores in H. pylori–infected persons residing in a region with a high incidence of gastric cancer that were either successfully treated, untreated, or who failed antimicrobial therapy, by years of follow-up. Modified from ref. 21.
Another question is whether aberrant methylation of E-cadherin is a specific consequence of H. pylori–induced gastritis or whether this also occurs in other forms of gastric inflammation and injury (e.g., Crohn's gastritis, nonsteroidal anti-inflammatory drug gastropathy). Several studies have implicated chronic infection with EBV as a risk factor for gastric cancer and EBV+ gastric cancer specimens contain a higher number of methylated genes than EBV-negative cancers (22, 23). The potential interactions between EBV, H. pylori, and host gene promoter hypermethylation add another layer of complexity to the molecular underpinnings of gastric carcinogenesis. Are there particular subsets of H. pylori–infected subjects who are more likely to harbor cells with aberrant E-cadherin methylation patterns (e.g., persons infected with cag+ strains)? The current study identified a significant association between the degree of inflammation and E-cadherin methylation, which raises the possibility that persons harboring high-expression immune response alleles may be at increased risk for epigenetic inactivation not only of E-cadherin but also of other tumor suppressor genes.
Summary and Future Implications
Studies such as the one from Leung et al. are valuable primarily due to the global burden of H. pylori–induced gastric adenocarcinoma. Although gastric cancer is a highly lethal disease, analytic tools now exist, including genome sequences (H. pylori and human), measurable phenotypes (CagA phosphorylation), and practical animal models, to discern the fundamental biological basis of H. pylori–associated neoplasia, which should have direct clinical applications. The rapidly unfolding story about E-cadherin and gastric cancer that has transpired in the last few years has shown that the power of genetics can alter our understanding and management of this disease; however, additional questions have now been raised that need to be answered in turn. It is important to gain more insight into the pathogenesis of H. pylori–induced gastric adenocarcinoma, not only to develop more effective treatments for this common cancer, but also because it might serve as a paradigm for the role of chronic inflammation in the genesis of other malignancies.
