Discovery of novel 5′-methylthioadenosine nucleosidase inhibitors by virtual screening and bioassays

5′-methylthioadenosine nucleosidase (MTAN) is a recognized target for potential antibacterial agents. In this study, a virtual screening method based on molecular docking was used to discover new potential active compounds for inhibiting MTAN. AutoDock Vina was used to establish a virtual screening methodology based on molecular docking. We used this system to generate a docking model with strong enrichment ability for active molecules, and screen a natural product library of natural product compounds.Targets with a binding energy lower than the threshold that met the five principles of drug-like compounds were retained for molecular dynamics simulation and in vitro antibacterial experiments. The molecular docking model of the 4YML target protein showed the strongest enrichment ability for the active molecules, and six natural product compounds with potential antibacterial activity were obtained by virtual screening. Molecular dynamics simulation showed that bis-demethoxycurcumin with the lowest binding affinity bound to the target protein well, which was consistent with the results of molecular docking. In vitro antibacterial experiments showed that two of the compounds exhibited significant effects against Escherichia coli activity. Based on virtual screening and bioassays, the effective components with inhibitory activity of MTAN can be found, which provide important theoretical guidance and experimental basis for the development of new MTAN inhibitors.