Gene Methylation in Non-Neoplastic Mucosa of Gastric Cancer: Age or Helicobacter pylori Related?

To the Editor-in-Chief:

We read with interest the article of Waki et al 1 on the promoter methylation status of E-cadherin, hMLH1, and p16 genes in non-neoplastic gastric epithelia. We have also recently studied the methylation of the E-cadherin gene in non-neoplastic mucosa of patients with gastric cancer. 2 The methylation of E-cadherin was assessed by two different primer sets as designated by Herman et al 3 and Graff et al 4 to confirm the specificity. We observed that E-cadherin methylation was associated with age, the presence of chronic gastritis, and Helicobacter pylori (H. pylori) on univariate analysis. However, H. pylori infection was the only independent factor associated with E-cadherin methylation on multivariate analysis.

Hypermethylation at the promoter region of tumor suppressor gene has been shown to be an important mechanism in gene silencing. However, gene-acquired methylation may not necessarily be associated with neoplastic transformation. Toyota et al 5 proposed that two types of methylation might exist: type A methylation (such as the estrogen receptor gene) which is age-related and may be present in normal mucosa and type C methylation which is cancer-related and not present in normal mucosa. Furthermore, recent studies have shown that gene methylation may be present in non-neoplastic colorectal mucosa in patients with inflammatory bowel disease, 6,7 esophageal mucosa in patients with Barrett’s esophagitis, 8,9 and in liver tissues in chronic hepatitis. 10 In the stomach, methylation at THBS-1 and TIMP-3 in chronic gastritis was found to be present in 10.1% and 14.5% of cases respectively. 11 These findings suggested that gene methylation could be a result of chronic inflammation.

In the stomach, the interplay between age and chronic inflammation is complicated by the presence of H. pylori infection. The prevalence of H. pylori increases with age. In Western countries, about 50% of people over 60 years of age are infected, but only 20 to 30% below the age of 30 are infected. 12,13 In addition, the presence of H. pylori is almost invariably associated with gastritis. In a meta-analysis, Blaser 14 documented that 75% to 100% of symptomatic patients or asymptomatic volunteers with endoscopically proven gastritis were colonized with the organism. When a person is infected with H. pylori, a superficial gastritis results within hours or days and can progress to chronic gastritis. The prevalence of H. pylori is higher in China, Japan, and Korea than in Western countries. This fact might explain why E-cadherin methylation was absent in non-neoplastic gastric mucosa when tested in Western patients, 3,15 but present in studies performed in patients from Japan, Hong Kong, and Korea. 1,2,11 Hence, the issue of whether methylation in the non-neoplastic gastric mucosa is due to aging, chronic gastritis, or H. pylori infection should be addressed carefully.

Experimental data from in vitro studies support our contention that E-cadherin methylation might be related to H. pylori infection. El-Omar et al 16 reported that interleukin-1β polymorphism that led to up-regulation of interleukine-1β with H. pylori infection was associated with increased risk of gastric cancer. Furthermore, Hmadcha et al 17 found that interleukin-1β, through the production of nitric oxide and the subsequent activation of DNA methyltransferase, might induce gene methylation. It is thus possible that H. pylori induces methylation through the production of interleukin-1β.

Gen Tamura

Author’s Reply:

The authors appreciate the comments by Dr. Chan et al on our paper describing gene promoter methylation in non-neoplastic gastric epithelia. 1 A number of tumor suppressor and tumor-related genes exhibited promoter methylation in both neoplastic and non-neoplastic gastric epithelia, although certain gene methylation was found to be cancer-specific. 2,3 Such differences in methylation patterns among genes may correspond to type A (aging-specific) and type C (cancer-specific) methylation previously described for colorectal and gastric epithelial cells. 4,5 Type A methylation arises as a function of age in normal cells, potentially affecting genes that regulate the growth and/or differentiation of these cells, and could account, in part, for the hyperproliferative state that is thought to precede tumor formation. 4 In contrast, type C methylation affects only a subset of tumors, which then evolve along a global hypermethylation pathway. 4 However, hMLH1 methylation, once thought to be cancer-specific, 4 was found to be a common age-related event in normal colonic cells when the entire hMLH1 promoter ∼700-bp region was analyzed. 6 hMLH1 methylation is partial in normal colonic cells and increases with age, spreading to reach a threshold, and ultimately shutting down protein expression. 6 Therefore, these contradictory results might be due to the analysis of different CpG sites in these studies. If critical CpG sites for each indication of gene silencing are more precisely analyzed, age-related methylation may be found to be cancer-specific. This hypothesis is supported by the observation that DAP-kinase methylation was present in virtually every tumor and normal gastric and colorectal cell when the edge of a CpG island was examined, yet on analysis of the central region of the CpG island, the methylation was determined to be a more infrequent, cancer-specific phenomenon. 7 Thus, the differences of methylation patterns may not provide any evidence for the existence of any CpG island methylator phenotype, and some (or cancer cells) are simply more hypermethylated than others (or non-neoplastic cells). 8 Alternatively, age-related methylation may not be immediately oncogenic, but gradually spreads to inactivate gene function. In fact, age-related methylation of several tumor suppressor and tumor-related genes was observed in the small intestine where tumor evolution is extremely rare. 3

Because several factors may modulate age-related methylation, such as exogenous carcinogens, endogenously generated reactive oxygen species, and genetic differences in individual susceptibility, 9 Helicobacter pylori (H. pylori) infection and resultant gastritis may accelerate this process. We further explored methylation analysis in non-neoplastic gastric epithelia obtained at autopsy from 11 non-gastric cancer patients younger than 32 years old (range, 0 to 31 years; average, 14.0 years, excluding stillborn infants) and 25 non-gastric cancer patients greater than 42 years old (range, 43 to 87 years; average, 67.9 years) (Table 1) ▶ . Methylation was absent in younger individuals, except in APC (promoter 1A) (Table 1) ▶ . Methylation of one of the promoters (promoter 1A) is not oncogenic because the other (promoter 1B) is protected from methylation and thus APC is not inactivated. 10 Hence, APC methylation (promoter 1A), though present in younger individuals, does not contribute to gastric carcinogenesis. Methylation of other tumor suppressor and tumor-related genes was present at variable frequencies in non-neoplastic gastric epithelia from elderly individuals (Table 1) ▶ . There also were differences in methylation frequencies depending on the site in the stomach from which the sample was taken. The exact reasons for these phenomena are unclear. However, the antral location of gastric cancer is known to be susceptible to methylation of several tumor suppressor and tumor-related genes. 11 Intestinal metaplasia, which may be the consequence of H. pylori-associated gastritis, especially that of the incomplete type, commonly arises in the antrum and then expands toward the body of the stomach, and may be predisposed to promote methylation of several genes.

Table 1.

Frequencies of Gene Promoter Methylation in Nonneoplastic Gastric Epithelia

Gene	≦ 31 years of age			≧ 43 years of age
	U (n = 7)	M (n = 11)	L (n = 7)	U (n = 23)	M (n = 25)	L (n = 22)
APC	57%	36%	57%	91	96%	95%
DAP-kinase	0%	0%	0%	78%	80%	68%
E-cadherin	0%	0%	0%	78%	76%	64%
p16	0%	0%	0%	30%	16%	18%
RUNX3	0%	0%	0%	4%	4%	32%
RASSF1A	0%	0%	0%	4%	12%	5%
hMLH1	0%	0%	0%	4%	0%	14%
GSTP1	0%	0	0%	0%	0%	0%

RT-PCR analysis on IFN-γ-inducible expression of promoter III-derived (Pro III) and promoter IV-derived (Pro IV) CIITA (38 cycles of PCR) and GAPDH (28 cycles of PCR) in small cell lung cancer (SCLC), neuroblastoma (NB), and non-SCLC cell lines. Cultured cancer cells were treated with 500 U/ml IFN-γ for 24 hours.

U, upper third portion of the stomach; M, middle third portion of the stomach; L, lower third portion of the stomach.

E-cadherin methylation is apparently age-related in the stomach, and may be accelerated by H. pylori infection. However, this methylation also increased with age in the jejunum, ileum, and colon, where H. pylori-associated inflammation was unlikely, quite similar to the situation in the stomach. 1 To the best of this author’s knowledge and experience, there has been no evidence that there exist differences of susceptibility to age-related methylation between Western and Oriental patients (data not shown). Therefore, more important mechanisms than H. pylori infection must still underlie age-related E-cadherin methylation.