Screening key miRNAs for human hepatocellular carcinoma based on miRNA-mRNA functional synergistic network.

The safety of miRNAs has been proven and the prophylactic use of miRNA-based approaches may be foreseen for patients with hepatocellular carcinoma (HCC). However, the underlying regulatory mechanism of miRNAs in HCC has not been fully clarified. Using bioinformatic analyses, we compared data of miRNA and mRNA expression profiling of HCC from Gene Expression Omnibus (GEO) database, respectively. Differentially expressed miRNAs and differentially expressed genes (DEGs) were identified. Based on the miRTarBase predictions, the miRNA-dependent regulatory network was constructed. In total, comparative analysis of five mRNA datasets and two miRNA datasets led to 1449 DEGs and 17 differentially expressed miRNAs, respectively. Based on the predictions, a global miRNA-mRNA regulatory network was then constructed, which encompassed 451 miRNA target gene pairs whose expressions were inversely correlated. Three miRNAs (miR-641, miR-507 and miR-501-5p) were the most connected miRNAs that regulated a large number of genes, among which miR-641 and miR-507 were novel miRNAs altered in HCC. We suggested that miR-501-5p will represent a powerful therapeutic target for HCC. Moreover, four up-regulated miRNAs (miR-769-3p, miR-941, miR-362-3p and miR-16-1) and one down-regulated miRNA (miR-581) may be involved in HCC. Additionally, two targets of MAPK8 and SRPK2 were also detected in this study, whose roles in HCC will be notable. In conclusion, we developed an integrative approach to construct an interactive global network of miRNA-mRNA, which can contribute to refine miRNA target predictions for developing new therapeutic approaches.