RESUMEN
BACKGROUND: The aim of present study was to screen the novel and promising targets of curcumin in hepatocellular carcinoma diagnosis and chemotherapy. METHODS: Potential targets of curcumin were screened from SwissTargetPrediction, ParmMapper and drugbank databases. Potential aberrant genes of hepatocellular carcinoma were screened from Genecards databases. Fifty paired hepatocellular carcinoma patients' gene expression profiles from the GEO database were used to test potential targets of curcumin. Besides, GO analysis, KEGG pathway enrichment analysis and PPI network construction were used to explore the underlying mechanism of candidate hub genes. ROC analysis and Kaplan-Meier analysis were used to evaluate the diagnostic and prognostic value of candidate hub genes, respectively. Real-time PCR was used to verify the results of bioinformatics analysis. RESULTS: Bioinformatics analysis results suggested that AURKA, CDK1, CCNB1, TOP2A, CYP2B6, CYP2C9, and CYP3A4 genes served as candidate hub genes. AURKA, CDK1, CCNB1 and TOP2A were significantly upregulated and correlated with poor prognosis in hepatocellular carcinoma, AUC values of which were 95.7, 96.9, 98.1 and 96.1% respectively. There was not significant correlation between the expression of CYP2B6 and prognosis of hepatocellular carcinoma, while CYP2C9 and CYP3A4 genes were significantly downregulated and correlated with poor prognosis in hepatocellular carcinoma. AUC values of CYP2B6, CYP2C9, and CYP3A4 were 96.0, 97.0 and 88.0% respectively. In vitro, we further confirmed that curcumin significantly downregulated the expression of AURKA, CDK1, and TOP2A genes, while significantly upregulated the expression of CYP2B6, CYP2C9, and CYP3A4 genes. CONCLUSIONS: Our results provided a novel panel of AURKA, CDK1, TOP2A, CYP2C9, and CYP3A4 candidate genes for curcumin related chemotherapy of hepatocellular carcinoma.