Integrative identification of Epstein-Barr virus-associated mutations and epigenetic alterations in gastric cancer

Qiaoyi Liang, Xiaotian Yao, Senwei Tang, Jingwan Zhang, Tung On Yau, Xiaoxing Li, Ceen-Ming Tang, Wei Kang, Raymond W M Lung, Jing Woei Li, Ting Fung Chan, Rui Xing, Youyong Lu, Kwok Wai Lo, Nathalie Wong, Ka Fai To, Chang Yu, Francis K L Chan, Joseph J Y Sung, Jun Yu

Research output: Contribution to journalArticlepeer-review

81 Citations (Scopus)

Abstract

BACKGROUND & AIMS: The mechanisms by which Epstein-Barr virus (EBV) contributes to the development of gastric cancer are unclear. We investigated EBV-associated genomic and epigenomic variations in gastric cancer cells and tumors.

METHODS: We performed whole-genome, transcriptome, and epigenome sequence analyses of a gastric adenocarcinoma cell line (AGS cells), before and after EBV infection. We then looked for alterations in gastric tumor samples, with (n = 34) or without (n = 100) EBV infection, collected from patients at the Prince of Wales Hospital, Chinese University of Hong Kong (from 1998 through 2004), or the First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China (from 1999 through 2006).

RESULTS: Transcriptome analysis showed that infected cells expressed 9 EBV genes previously detected in EBV-associated gastric tumors and 71 EBV genes not previously reported in gastric tumors. Ten viral genes that had not been reported previously in gastric cancer but were expressed most highly in EBV-infected cells also were expressed in primary EBV-positive gastric tumors. Whole-genome sequence analysis identified 45 EBV-associated nonsynonymous mutations. These mutations, in genes such as AKT2, CCNA1, MAP3K4, and TGFBR1, were associated significantly with EBV-positive gastric tumors, compared with EBV-negative tumors. An activating mutation in AKT2 was associated with reduced survival times of patients with EBV-positive gastric cancer (P = .006); this mutation was found to dysregulate mitogen-activated protein kinase signaling. Integrated epigenome and transcriptome analyses identified 216 genes transcriptionally down-regulated by EBV-associated hypermethylation; methylation of ACSS1, FAM3B, IHH, and TRABD increased significantly in EBV-positive tumors. Overexpression of Indian hedgehog (IHH) and TraB domain containing (TRABD) increased proliferation and colony formation of gastric cancer cells, whereas knockdown of these genes reduced these activities. We found 5 signaling pathways (axon guidance, focal adhesion formation, interactions among cytokines and receptors, mitogen-activated protein kinase signaling, and actin cytoskeleton regulation) to be affected commonly by EBV-associated genomic and epigenomic alterations.

CONCLUSIONS: By using genomic, transcriptome, and epigenomic comparisons of EBV infected vs noninfected gastric cancer cells and tumor samples, we identified alterations in genes, gene expression, and methylation that affect different signaling networks. These might be involved in EBV-associated gastric carcinogenesis.

Original languageEnglish
Article number1350-62.e4
JournalGastroenterology
Volume147
Issue number6
DOIs
Publication statusPrint publication - Dec 2014
Externally publishedYes

Keywords

  • Adenocarcinoma/genetics
  • Cell Line, Tumor
  • Cyclin A1/genetics
  • DNA Methylation/genetics
  • Epigenesis, Genetic/genetics
  • Epstein-Barr Virus Infections/genetics
  • Gene Expression Regulation, Neoplastic
  • Gene Expression Regulation, Viral
  • Genes, Viral
  • Genome-Wide Association Study
  • Herpesvirus 4, Human/genetics
  • Humans
  • MAP Kinase Kinase Kinase 4/genetics
  • Protein-Serine-Threonine Kinases/genetics
  • Proto-Oncogene Proteins c-akt/genetics
  • Receptor, Transforming Growth Factor-beta Type I
  • Receptors, Transforming Growth Factor beta/genetics
  • Stomach Neoplasms/genetics
  • Transcriptome

Fingerprint

Dive into the research topics of 'Integrative identification of Epstein-Barr virus-associated mutations and epigenetic alterations in gastric cancer'. Together they form a unique fingerprint.

Cite this