Structure-based virtual screening to identify inhibitors against Staphylococcus aureus MurD enzyme

Abstract

The MurD enzyme of Staphylococcus aureus is an attractive drug target as it is essential and ubiquitous in bacteria but absent in mammalian cells. In the present study, we performed in silico high-throughput virtual screening with small molecule library of 1.60 million compounds to identify potential hits. We used S. aureus modeled MurD protein for this purpose and to find the best leads, dock complexes were further subjected to the extra-precision docking and binding free energy calculations by MM-GBSA approach. It is evident that van der Waals and Coulomb energy terms are major favorable contributors while electrostatic solvation energy term strongly disfavors the binding of ligands to the S. aureus MurD enzyme. The inhibitory activity of two selected virtual hits H5 and H10 was performed against S. aureus MurD enzyme using malachite green assay. In in vitro antibacterial screening, compound H5 inhibited the growth of S. aureus NCIM 5021, S. aureus NCIM 5022, and methicillin-resistant S. aureus (MRSA strain 43300) at high concentrations while the other tested compound H10 was inactive against all the tested strains. In order to validate the stability of inhibitor-protein complex, compound H5 with the highest inhibitory against S. aureus MurD and lowest binding free energy was subjected to 30-ns molecular dynamics simulation. Further, ADMET predictions showed the favorable pharmacokinetic profile of compounds H5 and H10.