Integration of LC/MS-based molecular networking and molecular docking allows in-depth annotation and prediction of the metabolome: A study of Salvia miltiorrhiza Bunge

Plant secondary metabolites play essential roles in plant-environment interactions and biological communication. Although untargeted metabolomics is widely used in plant research, it suffers from limited annotated metabolites and relatively poor reliability. Herein, we report a technical route of untargeted metabolomics that involves using MS-DAIL to preprocess the data, using Cytoscape to construct a molecular network and quickly expand secondary metabolites annotation, using SIRIUS4 to construct a mass spectrum tree and summarize the fragmentation rules, and using bioinformatics analysis and molecular docking to predict the latent effect of new compounds. The new method annotates metabolites and finds new substances according to mass spectrometry similarity, and predicts its potential application value before separating new substances. 34 tanshinones and 31 phenolic acids were rapidly annotated by the technical route, and the structures of two new salvianolic acids with large molecular weights were deduced by the summarized mass spectrometry fragmentation rules. The results of molecular docking indicated that the two new substances may act on aldose reductase and plasminogen. The new technical route established in this study is helpful for the identification of plant secondary metabolites and the prediction of potential application value of new compounds. The method established in this study is helpful to promote plant metabolomics studies.