Dissecting active ingredients of Chinese medicine by content-weighted ingredient-target network.

Chinese medicine has been widely used in clinical practice, but its mode of action often remains obscure. This has seriously hindered further development and better clinical applications of Chinese medicine. Among the most critical questions to be addressed, the identification of active ingredients is an important one requiring more research. Existing methods are only concerned the potential pharmacological effects of the individual purified chemical ingredients without consideration of the contents of these ingredients, which is critical to the comprehensive effect of Chinese medicine. A novel approach was proposed here to integrate network pharmacology analysis and ingredient content in Chinese medicine to identify active ingredients. The therapeutic action of Xuesaitong (XST) injection on myocardial infarction was analyzed as an example in this study. Firstly, we built a cardiovascular disease (CVD) related protein-protein interaction (PPI) network. Secondly, the potential targets of the ingredients of XST were identified by integrating microarray data, text mining and pharmacophore model-based prediction. The target-ingredient relationships were then mapped to the network. Topological attributes related to the targets of these ingredients, together with the ingredients' contents, were combined to calculate a composition-weighted index for integrative evaluation of ingredient efficacy. Our results indicated that major active ingredients in XST were notoginsenoside R1, ginsenoside Rg1, Rb1, Rd and Re, which was further validated on myocardial infarction rat models. In conclusion, this study presented a novel approach to identify active ingredients in Chinese medicine.