An integrated analysis of copy number alteration and global gene expression profiling reveals potential oncogenes underlying gastric cancer

Abstract

Gastric cancer is the most common cancer in Asia and remains the second leading cause of cancer mortality in the world. Gastric cancer is usually diagnosed at a very advanced stage. The median survival is ranges from 6-10 months, and 5-year survival rates are <10% in patients with advanced Gastric cancer. Although various studies had been performed to identify the causative genes involved in gastric cancer development, a deeper understanding of the oncogenes that could be characteristic of a variety of histological subtypes and stages of gastric tumor is required. Herein, we report an integrated analysis of genome-wide copy number alteration and global mRNA expression of 40 gastric cancer tissues from 19 elderly and 21 young patients using array-based comparative genomic hybridization (aCGH) and expression microarray. We showed that gastric cancer genomes showed recurrent DNA copy number alterations such as, amplifications at 3q26, 7p11-7p22, 8p11-8q36, 20p11-20p13, and 20q11-20q13, and deletions at 4p11-4p16, 4q11-4q35, 1p35-1p36, 3p21, 8p12, 9p24, and 19p12-19p13. The correlating of transcriptional expression profiling with the aCGH analysis revealed 506 candidate oncogenes showing both overexpression and frequent amplification, and 60 candidate tumor suppressor genes showing both down-regulation and frequent deletion. We identified two regions containing noble target genes. In addition to analyzing Heebo chip and real time PCR, we selected specific cell lines from 14 gastric cell lines. Among our candidate oncogenes, siRNA-based knockdown of CAPZA2, C17orf37, PERLD1, STARD3 and GRB7 within the MET and ErbB2 amplicons in gastric cancer cell lines resulted in a significant suppression of cell proliferation and migration, strongly suggesting that they could be potential oncogenes in gastric cancer. In this study, we suggest that oncogenes and tumor suppressor genes selected by aCGH and microarray, could be potentially used as therapeutic targets.