Construction Of A CeRNA Regulatory Network And Analysis Of Characteristic CeRNA Expression In Hepatocellular Carcinoma

Background:Hepatocellular carcinoma(HCC)is a common malignancy that poses a major threat to public health.Because of the insidious pathogenesis and rapid progression of HCC,exploring its regulatory mechanisms and finding molecular targets are crucial for the treatment of HCC.The hypothesis of competing endogenous RNAs(ceRNAs)suggests that mRNAs and non-coding RNAs can participate in the posttranscriptional regulation of protein expression and form a complex regulatory network by competitively binding miRNA response elements.However,the HCC-related ceRNA regulatory network has not been studied deeply enough.Methods:The differential expression genes were obtained by analyzing the GSE77314 data in GEO dataset.A protein-protein interaction(PPI)network was established by STRING database,and the CytoHubba tool was used to identify hub genes.KEGG pathway enrichment and GO function enrichment analyses were performed on the analyzed hub genes to explore their biological functions,cellular localization and the biological processes involved.The miRTarBase database and miRNet database were used to predict the miRNAs and lncRNAs associated with the hub genes,respectively.Expression difference analysis and survival curve analysis were performed by GEPIA and Kaplan-Meier plotter analysis tools for mRNA,miRNA and lncRNA,respectively,to screen out RNAs with significant expression differences and also significant impact on the prognosis of patients with hepatocellular carcinoma.Finally,the ceRNA network was constructed and Cox regression analysis was performed on the RNAs in the ceRNA network to identify whether they could be independent prognostic factors.Serum samples from 40 hepatocellular carcinoma patients and 9 surgically resected hepatocellular carcinoma and paraneoplastic tissues were collected and validated by RT-PCR to clarify whether there were differences in expression levels.Results:574 up-regulated genes and 274 down-regulated genes were identified to be relevant to the occurrence of HCC(P<0.05).The top 30 hub genes of the PPI networks were analysed using survival curve,and 16 hub genes that closely associated with the prognosis of HCC were found.The results of GO functional enrichment analysis included cell nuclear division and mitosis,and the results of KEGG pathway enrichment analysis included cell cycle and P53 signaling pathway.After prediction by miRTarBase database,366 miRNAs interacted with these hub genes were also identified,among which 8 miRNAs had significant influence on the prognosis of HCC.Further prediction using the miRNet database identified 133 lncRNAs targeting key miRNAs,and 6 of these lncRNAs were analyzed and confirmed to be important components of the ceRNA regulatory network.Lasso regression and Cox proportional hazards model of the ceRNA network were performed,and the final prognostic model was constructed using LINC01184 and miR-148a.RT-PCR results showed that miRNAs were downregulated in hepatocellular carcinoma serum and mRNAs and lncRNAs were upregulated in hepatocellular carcinoma tissues.Conclusion:Sixteen mRNAs,eight miRNAs and six lncRNAs were found and identified to be associated with prognosis of hepatocellular carcinoma,and the construction of a novel hepatocellular carcinoma ceRNA regulatory network was completed,which is important for finding potential prognostic biomarkers for hepatocellular carcinoma.The RNA molecules in the ceRNA network were evaluated by Cox risk model,and LINC01184 was obtained as an independent prognostic factor.RT-PCR validation experiments on ceRNA molecules demonstrated the potential of these RNA molecules as hepatocellular carcinoma biomarkers.