Network Pharmacology Integrated Bioinformatics Explores the Mechanism of Dahuang-Mudan Decoction for Colon Cancer and Construction of Prognostic Model

OBJECTIVE To identify the core compounds and targets of Dahuang-Mudan decoction for colon cancer, and construct a prognostic model to evaluate the overall survival of patients. METHODS Targets of Dahuang-Mudan decoction for colon cancer were screened from multiple databases. GO analysis and KEGG pathway enrichment analysis were performed to explore the possible mechanisms of targets. After screening for survival-related targets by univariate Cox analysis, LASSO regression were used to construct a prognostic model and calculate the scores of the model. KM survival analysis and ROC curve were adopted to verify and evaluate the model's feasibility. Protein protein interaction(PPI) network and Disease-Drug- Compound-Target network were constructed to screen the core target. HPA immunohistochemistry database and Image J software were employed to demonstrate the protein expression of the core target. Molecular docking and dynamics were evaluated to the binding activity of the core compounds and targets. RESULTS A total of 22 targets in Dahuang-Mudan decoction for colon cancer had been screened. These targets participated in regulating ketone and oxygen level reaction and were related to gastrin signaling and DNA injury response pathway. After removing the redundant factors, five genes including CHEK1, BIRC5, GSTM2,CCNB1 and VEGFA, which were screened by univariate Cox analysis, were used to construct a model by LASSO regression. Two core compounds including aloe-emodin and quercetin were screened by Disease-Drug-Compound-Target network. Four core targets including CCNB1, MYC, CCND1 and PCNA were screened via PPI network. One core target including CCNB1 was obtained by crossing the genes of prognostic model and four core targets. KM analysis showed significant difference in CCNB1 between high and low risk groups. Immunocytochemistry indicated that CCNB1 was differentially expressed in normal and colon cancer tissues. The difference was statistically significant(P<0.01). Molecular docking and dynamics demonstrated that the bindings of quercetin and aloe-emodin to CCNB1 were stable. CONCLUSION The role of Dahuang-Mudan decoction for colon cancer may be achieved by quercetin and aloe-emodin targeting CCNB1. The prognostic model have a good efficacy in assessing patients' prognosis.