Systematic Elucidation of the Potential Mechanisms of Core Chinese Materia Medicas in Treating Liver Cancer Based on Network Pharmacology.

OBJECTIVE

In this study, the data mining method was used to screen the core Chinese materia medicas (CCMMs) against primary liver cancer (PLC), and the potential mechanisms of CCMMs in treating PLC were analyzed based on network pharmacology.

METHODS

Traditional Chinese medicine (TCM) prescriptions for treating PLC were obtained from a famous TCM doctor in Shenzhen, China. According to the data mining technique, the TCM Inheritance Support System (TCMISS) was applied to excavate the CCMMs in the prescriptions. Then, bioactive ingredients and corresponding targets of CCMMs were collected using three different TCM online databases, and target genes of PLC were obtained from GeneCards and OMIM. Afterwards, common targets of CCMMs and PLC were screened. Furthermore, a network of CCMMs bioactive ingredients and common target gene was constructed by Cytoscape 3.7.1, and gene ontology (GO) and signaling pathways analyses were performed to explain the mechanism of CCMMs in treating PLC. Besides, protein-protein interaction (PPI) analysis was used to identify key target genes of CCMMs, and the prognostic value of key target genes was verified using survival analysis.

RESULTS

A total of 15 high-frequency Chinese materia medica combinations were found, and CCMMs (including Paeoniae Radix Alba, Radix Bupleuri, Macrocephalae Rhizoma, Coicis Semen, Poria, and Curcumae Radix) were identified by TCMISS. A total of 40 bioactive ingredients (e.g., quercetin, kaempferol, and naringenin) of CCMMs were obtained, and 202 common target genes of CCMMs and PLC were screened. GO analysis indicated that biological processes of CCMMs were mainly involved in response to drug, response to ethanol, etc. Pathway analysis demonstrated that CCMMs exerted its antitumor effects by acting on multiple signaling pathways, including PI3K-Akt, TNF, and MAPK pathways. Also, some key target genes of CCMMs were determined by PPI analysis, and four genes (MAPK3, VEGFA, EGF, and EGFR) were found to be correlated with survival in PLC patients.

CONCLUSION

Based on data mining and network pharmacology methods, our results showed that the therapeutic effect of CCMMs on PLC may be realized by acting on multitargets and multipathways related to the occurrence and development of PLC.