Identification of genes epigenetically silenced by CpG methylation in human gastric carcinoma

https://doi.org/10.1016/j.ejca.2008.12.027Get rights and content

Abstract

To identify novel methylation-silenced genes in gastric cancer, we carried out a genome-wide search for genes that are up-regulated after treatment with the demethylating agent, 5-aza-2′-deoxycytidine (5Aza-dC). When three gastric cancer cell lines (SNU-1,-601, and -719) were treated with 5Aza-dC, 143 genes were found to be upregulated by twofold or more using oligonucleotide microarrays. Six of these genes, i.e. TFPI2, GPX3, GPX1, IGFBP6, IRF7 and DMRT1, showed promoter hypermethylation in one or more gastric cancer cell lines, but were unmethylated in normal gastric mucosa by bisulphite sequencing and methylation-specific PCR analysis. The following percentages of these genes were found to be aberrantly methylated in gastric cancer samples; TFPI2 (80.9%), GPX3 (30.1%), DMRT1 (46.9%), GPX1 (16.7%), IGFBP6 (22.6%) and IRF7 (32.1%). Interestingly, the survival of patients possessing methylated alleles of TFPI2 (123/152, 80.9%) was poorer than that of patients with unmethylated alleles (p = 0.023). Multivariate analysis confirmed that TFPI2 methylation is a significant and independent prognostic factor in gastric carcinoma. Furthermore, altered TFPI2 expression, as demonstrated by immunohistochemistry in 566 consecutive gastric cancer tissues, was found to be significantly associated with sex (p = 0.003), WHO classification (p < 0.001), and a mixed subtype by Lauren’s classification (p < 0.001). Thus, the present study identified several novel genes, which were methylated in gastric cancer and among them, methylation of TFPI2 was an unfavourable prognostic marker.

Introduction

Globally, gastric cancer is the second most common cause of cancer death, 1 and although the incidence of gastric cancer is declining in developed countries, it remains a leading cause of cancer-related death in the majority of developing countries. Genetic and environmental factors, including Helicobacter pylori infection, are considered to contribute to the development of gastric cancer. Moreover, recent studies have shown that the silencing of tumour suppressor genes by epigenetic modification is a fundamental mechanism of the pathogenesis of human cancer.2 Gastric carcinogenesis is thought to be a multistep process that involves multiple genetic and epigenetic events, but its underlying mechanisms have yet to be clarified, although it is known that promoter hypermethylation plays an essential role in tumour suppressor gene loss of function. In terms of the molecular basis of gastric cancer, it is believed that the aberrant DNA methylation of known tumour suppressor genes is more common that their mutation.3 However, the roles played by many of the genes affected by methylation-silencing in gastric cancer remain un-revealed, and further studies are necessary to identify all the silenced genes.

The epigenetic alterations of CpG islands in the promoter region of genes by aberrant DNA methylation have been established to be a common mechanism for silencing tumour suppressor genes in cancer cells.4, 5 Aberrant DNA methylation affects important cellular pathways by silencing tumour-related genes.6 Generally, tumour cells have promoter methylation patterns that differ from those of normal cells.7 Furthermore, recently, DNA methylation has emerged as promising biomarker for cancer detection8, and a number of genes, e.g. APC, COX2, DAP-K, CDH1, p14, CDKN2A, THBS1, TIMP3, hMLH1, p15, MINT1, MINT2, MINT25, CHFR and CDH4, have been found to be aberrantly methylated in gastric cancer.9, 10, 11, 12, 13, 14 Nevertheless, many of the specific molecular events involved in the pathogenesis and progression of gastric cancer remains to be elucidated. A global approach to the identification of epigenetically silenced genes in gastric cancer cells could provide methylation signatures for early detection and prognostic stratification, identify novel targets for therapy and lead to a deeper understanding of the biology of gastric cancer.

Epigenetic silencing is not an irreversible process, for example, drugs like 5-aza-2′-deoxycytidine (5Aza-dC) cause generalised demethylation. 5Aza-dC is known to act by being incorporated into new DNA strands during replication, where it forms a covalent complex with methyltransferase active sites, thus reducing methyltransferase activity.4 On the other hand, chromatin is a DNA/histone complex, and histone deacetylation is also known to impair gene transcription.15 Furthermore, DNA methylation has been reported to promote histone deacetylation.16 Methyl-CpG binding protein 2 (MeCP2) appears to reside as a complex with histone deacetylase17, whereas DNA methyltransferase binds to histone deacetylase 2 (HDAC2) and a transcriptional co-repressor, DMAP1.18 Thus, densely methylated DNA associates with transcriptionally repressive chromatin, which is characterised by the presence of under-acetylated histones. Trichostatin A (TSA; a histone deacetylase inhibitor) has been reported to reverse the formation of transcriptionally repressive chromatin on methylated promoter templates.19 Thus, epigenetic alterations are dynamically linked, and synergy between DNA demethylation and histone deacetylase inhibition using TSA has been shown to reactivate genes silenced in carcinoma more robustly than 5Aza-dC alone.20, 21

A number of different approaches have been developed for the genome-wide identification of cancer-associated hypermethylated genes, such as, restriction landmark genome scanning, methylation-sensitive arbitrarily primed PCR and methylated CpG island amplification.22 An alternative approach is to treat cells with epigenetic modifying drugs to facilitate the expressions of hypermethylated/silenced genes and then to compare gene expression profiles by microarray analysis to identify putatively hypermethylated genes. The methylation status of such genes can then be confirmed in untreated cells by bisulphite sequencing or methylation-specific PCR analyses. Global demethylation and microarray analysis have been successfully used to identify a number of novel genes hypermethylated in colorectal and pancreatic cancer.21, 23 In the present study, we examined the global reactivations of epigenetically silenced genes using an oligonucleotide microarray in three gastric cancer cell lines. Genes that were markedly upregulated (>twofold) by 5Aza-dC treatment in these cancer cell lines were considered genes of interest. Subsequently, bisulphite sequencing and methylation-specific PCR analyses were carried out on these genes to confirm the presence of aberrantly methylated CpG dinucleotides. In the present study, we show that the methods employed preferentially selected epigenetically silenced hypermethylated genes in primary gastric carcinomas. Finally, six genes found to be hypermethylated in at least one of the three cell lines were investigated to determine whether their methylation statuses were correlated with clinicopathological parameters.

Section snippets

Cell lines and tissue samples

Three human gastric cancer cell lines, (SNU-1, -601 and -719) were obtained from the Korean Cell Line Bank (Seoul, Korea). All cell lines were grown in RPMI1640 supplemented with 10% foetal bovine serum (FBS; Hyclone, Lorgan, UT, USA) and antibiotics (100 U/ml penicillin G and 100 μg/ml streptomycin) at 37 °C in a humidified 5% CO2 incubator.

Formalin-fixed, paraffin-embedded samples of 566 gastric cancer tissue specimens, resected at Seoul National University Hospital from January 1995 to December

Identification of candidate genes reactivated by 5Aza-dC in gastric cancer cell lines

Global changes in gene expression profiles induced by 5Aza-dC in the three gastric cancer cell lines (SNU-1, -601 and -719) were determined using Affymetrix HG U133 2.0 oligonucleotide microarrays containing 22,277 transcripts representing 14,500 well-characterised human genes. A comprehensive survey for hypermethylated gene candidates associated with gastric cancer was performed by comparing mock-treated and 5Aza-dC-treated gastric cancer cells. A summary of our differential and epigenetic

Discussion

Methods of detecting DNA methylation are based on the differentiation of cytosine and 5-methylcytosine in DNA. Three strategies are currently used: (1) DNA digestion using a methylation-sensitive or -insensitive restriction endonuclease, (2) chemical modification of DNA by sodium bisulphite or metabisulphite and (3) immunoprecipitation of 5-methylcytosine to directly separate unmethylated and methylated genomic fractions. Recently, all three of these approaches have been coupled to

Conflict of interest statement

None declared.

Acknowledgement

This study was supported by a Grant (A080316) of the ’08 Good Health R&D Project, Ministry of Health & Welfare, Republic of Korea.

References (44)

  • A. Merlo et al.

    5′ CpG island methylation is associated with transcriptional silencing of the tumour suppressor p16/CDKN2/MTS1 in human cancers

    Nat Med

    (1995)
  • T. Ushijima

    Detection and interpretation of altered methylation patterns in cancer cells

    Nat Rev Cancer

    (2005)
  • D. Sidransky

    Emerging molecular markers of cancer

    Nat Rev Cancer

    (2002)
  • G.H. Kang et al.

    CpG island methylation in premalignant stages of gastric carcinoma

    Cancer Res

    (2001)
  • J.H. Lee et al.

    Frequent CpG island methylation in precursor lesions and early gastric adenocarcinomas

    Oncogene

    (2004)
  • A. Satoh et al.

    Epigenetic inactivation of CHFR and sensitivity to microtubule inhibitor in gastric cancer

    Cancer Res

    (2003)
  • G. Tamura et al.

    E-Cadherin gene promoter hypermethylation in primary human gastric carcinomas

    J Natl Cancer Inst

    (2000)
  • K.F. To et al.

    Promoter hypermethylation of tumor-related genes in gastric intestinal metaplasia of patients with and without gastric cancer

    Int J Cancer

    (2002)
  • P. Marks et al.

    Histone deacetylases and cancer: causes and therapies

    Nat Rev Cancer

    (2001)
  • S.G. Gray et al.

    Histone acetylation/deacetylation and cancer: an “open” and “shut” case?

    Curr Mol Med

    (2001)
  • X. Nan et al.

    Transcriptional repression by the methyl-CpG-binding protein MeCP2 involves a histone deacetylase complex

    Nature

    (1998)
  • M.R. Rountree et al.

    DNMT1 binds HDAC2 and a new co-repressor, DMAP-1, to form a complex at replication foci

    Nat Genet

    (2000)
  • Cited by (85)

    • Current ideas on the biology of IGFBP-6: More than an IGF-II inhibitor?

      2016, Growth Hormone and IGF Research
      Citation Excerpt :

      IGFBP-6 itself is down-regulated in many cancers and this may play a role in their progression [23]. Recently, epigenetic mechanisms involving methylation of the IGFBP-6 promoter have been implicated in this process [60]. A recent study in fish suggested a novel role for IGFBP-6 in vivo by showing upregulation of IGFBP-6 expression following infection [61]; the authors suggested that this increase is directly regulated by proinflammatory cytokines and contributes to suppression of IGF-mediated growth when energy is required for immune system activation to counter the infection.

    View all citing articles on Scopus
    View full text