Abstract
OBJECTIVE: Herbal medicine discovery is a complex and time-consuming process, while pharmacophore modeling and molecular docking methods enable simple and economic studies. The pharmacophore model provides an abstract description of essential intermolecular interactions between chemical structures, and the molecular docking technology can identify novel compounds of therapeutic interests and predict the ligand-target interaction at the molecular level. This study was based on the two methods to elucidate the mechanism of dehydrovomifoliol, an active ingredient extracted from Artemisia frigida willd, in nonalcoholic fatty liver disease (NAFLD). METHODS: Bioinformatics analysis was performed to screen target genes of dehydrovomifoliol in NAFLD treatment, which were thus intersected with NAFLD-related differentially expressed genes (DEGs) and NAFLD-related genes. Venn diagram was used to identify candidate DEGs. A pharmacophore model was then generated, and molecular docking was performed. A protein-protein interaction (PPI) network was constructed to identify core genes, which were evaluated using GO and the KEGG enrichment analyses. RESULTS: Seven target genes of dehydrovomifoliol in NAFLD treatment were screened out, namely E2F1, MERTK, SOX17, MMP9, SULT2A1, VEGFA, and BLVRA. The pharmacophore model and molecular docking of candidate DEGs and dehydrovomifoliol were successfully constructed. E2F1 was identified as a core gene of dehydrovomifoliol in NAFLD treatment. Further enrichment analysis indicated the regulatory role of E2F1 in fat metabolism was associated with the regulation of the AKT/mTOR signaling pathway. CONCLUSION: Overall, this study illustrates the anti-NAFLD mechanism of dehydrovomifoliol, which could be a useful compound for developing novel drugs in the treatment of NAFLD.