Aberrant DNA methylation precedes loss of heterozygosity on chromosome 16 in chronic hepatitis and liver cirrhosis

Abstract

The aim of this study was to examine the significance of aberrant DNA methylation, the participation of which in genetic instability is controversial, in hepatocarcinogenesis. The DNA methylation status of the region around the promoter of the E-cadherin tumor suppresser gene, which is located on 16q22.1, and the allelic status at the D16S421 locus, which is adjacent to the E-cadherin locus, were examined using microdissected liver specimens from 38 hepatocellular carcinoma (HCC) patients. Almost all of the non-cancerous liver tissues showed histological findings compatible with chronic hepatitis and cirrhosis, which are considered to be precancerous conditions. DNA hypermethylation was detected in 61% of the non-cancerous liver tissues. The incidence of DNA hypermethylation in the non-cancerous liver tissues of patients with HCCs also showing DNA hypermethylation (72%) was significantly higher than that of patients without DNA hypermethylation in their HCCs (53%, P<0.05). Loss of heterozygosity (LOH) at the D16S421 locus was detected in 35% of the non-cancerous liver tissues. The incidence of LOH in the non-cancerous liver tissues of patients with HCCs also showing LOH was 78%, whereas LOH was not detected in non-cancerous liver tissues of patients without LOH in their HCCs. Fifty-two percent of the non-cancerous liver tissues showed both or neither of DNA hypermethylation and LOH; the incidence of DNA hypermethylation alone in non-cancerous liver tissue was 41%. The incidence of LOH alone in non-cancerous liver tissue (7%) was significantly lower compared to those of the former two cases (P<0.0001). These data suggest that aberrant DNA methylation participates in the precancerous stage of hepatocarcinogenesis by preceding, or causing, LOH.

Introduction

Aberrant DNA cytosine methylation is one of the most consistent epigenetic changes in human cancers [1]. The global DNA methylation level is generally lower in cancer cells than in normal cells [2], but increased expression and activity of maintenance DNA (cytosine-5) methyltransferase (EC 2.1.1.37) are usually seen in cancer cells [3]. Some loci tend to show increased DNA methylation [4], [5], [6], [7], [8], [9], [10], whereas others are often hypomethylated [11]. DNA methylation may play a role in carcinogenesis, since DNA cytosine methylation facilitates gene mutation through deamination of 5-methylcytosine to thymine [12] and DNA methylation near the gene promoters affects the transcription of specific genes [13], [14], [15], [16], [17].

Moreover, an association between aberrant DNA methylation and chromosomal instability during carcinogenesis has been noted recently. Treatment with a DNA methyltransferase inhibitor, 5-aza-cytidine [18], and inactivating the DNA methyltransferase gene [19] each result in mitotic recombination and chromosomal deletion, probably through low global DNA methylation and chromatin configuration changes. Makos et al. suggest that DNA hypermethylation at the D17S5 locus, the location of a candidate tumor suppresser gene called HIC-1 (hypermethylated-in-cancer, [20]), predisposes the same locus to allelic loss in renal cancers [5] and neural tumors [6]. Their suggestion is based on two findings: (a) the incidence of DNA hypermethylation at the D17S5 locus exceeded that of loss of heterozygosity (LOH) at the same locus and (b) all of the cancers with LOH also showed DNA hypermethylation at the same locus [5], [6]. Most of our previous data on DNA methylation and allelic loss at the D17S5 locus in lung [8] and stomach [9] cancers and hepatocellular carcinoma (HCC; [10]) support their hypothesis.

In addition to DNA hypermethylation at the D17S5 locus, we reported previously that the DNA methyltransferase mRNA expression level is significantly higher in non-cancerous liver tissues showing chronic hepatitis and cirrhosis, which are considered to be precancerous conditions [21], [22], than in liver tissues with no remarkable histological findings [23]. The DNA methyltransferase mRNA expression level was even higher in HCC than in chronic hepatitis and cirrhosis [23]. Aberrant DNA methylation on chromosome 16, a hot spot for LOH in HCC [24], is observed frequently in chronic hepatitis and cirrhosis [7]. The degree of aberrant DNA methylation is higher in HCC than in chronic hepatitis or cirrhosis so DNA hypermethylation may precede or cause LOH on chromosome 16 [7]. Moreover, CpG methylation around the promoter region of the E-cadherin tumor suppresser gene, which is located on 16q22.1 [25] and encodes a Ca²⁺-dependent cell adhesion molecule [26], significantly correlates with reduced E-cadherin expression, resulting in loss of intercellular adhesiveness and destruction of tissue structure in HCC [17]. Taken together these findings suggest that aberrant DNA methylation promotes hepatocarcinogenesis, even during the precancerous stage, by predisposing some chromosomes to LOH and silencing some specific genes.

Recently, microdissection techniques and methodologies have been developed for detecting LOH in small numbers of cells from paraffin-embedded tissues. LOH has been observed in microdissected specimens from non-cancerous lesions, e.g. hyperplastic or dysplastic lesions accompanying non-small-cell lung cancer [27] and proliferative lesions accompanying breast cancer [28]. In cases of HCC, cells excised from cirrhotic foci frequently show LOH at the 8p [29], Rb [30] and p53 [31] loci. To the best of our knowledge, however, LOH on chromosome 16 has not been reported in the precancerous stage of HCC. In our previous study of DNA methylation and allelic loss on chromosome 16, LOH was not detected in non-microdissected specimens showing chronic hepatitis or cirrhosis [7].

To examine the significance of aberrant DNA methylation in hepatocarcinogenesis, we assessed the association between DNA methylation around the promoter of the E-cadherin gene and allelic loss at the D16S421 locus, which is adjacent to the E-cadherin locus, using microdissected specimens of both non-cancerous liver tissues and HCCs.