An integrated analysis of genome-wide DNA methylation and gene expression data in hepatocellular carcinoma

Despite progress in the treatment of hepatocellular carcinoma (HCC), 5-year survival rates remain low. Thus, a more comprehensive approach to explore the mechanism of HCC is needed to provide new leads for targeted therapy. We performed an integrated analysis to discover the relationship between DNA methylation and gene expression in hepatocellular carcinoma (HCC). DNA methylation and gene expression data for HCC were downloaded from The Cancer Genome Atlas (TCGA) database, and differential analysis was performed. Correlation analysis between DNA methylation and gene expression data was then performed in R language. Finally, we selected several crucial genes and evaluated their potential use as diagnostic biomarkers for HCC. In total, 1135 differentially DNA-methylated CpG sites (DMCs), 377 differentially methylated regions (DMRs), and 1194 differentially expressed genes (DEGs) were identified in HCC. Among the DEGs, 14 genes (ALX3, B4GALNT1, CTHRC1, DLX5, EMX1, IRX3, OTX1, SIX2, TLX1, VASH2, ZIC2, ZIC4, ZIC5, and ZNF695) exhibited changes in DNA methylation in terms of CpG sites or CpG island (CGI) level, of which TLX1 and ZIC4 had the most DMCs (12 and 13, respectively). Further analysis of CTHRC1, ZIC4, SIX2, VASH2, IL17D, TLX1, OTX1, and LART, examining alterations in both DNA methylation and gene expression level in HCC, showed their potential diagnostic value for HCC was better at the gene expression level than that the DNA methylation level. The DNA methylation status of CTHRC1, VASH2, and IL7D was significantly associated with HCC overall survival (P-value <0.05). This systemic analysis identified a group of novel gene signatures (CTHRC1, ZIC4, and OTX1) that may be regulated by DNA hypermethylation, which may be closely associated with HCC.