Discovery of Neuraminidase Inhibitors based on 3D‐QSAR, Molecular Docking and MD Simulations

Abstract

AbstractNeuraminidase (NA) plays a critical part within the life cycle of influenza virus, which is a compelling target for anti‐influenza infection. In this study, 3D‐QSAR models were built up utilizing 48 compounds of 5‐benzyl‐4‐thiazolinone derivatives. The CoMFA model (q2=0.759, r2=0.994) and the CoMSIA (q2=0.512, r2=0.972) model showed good reliability and predictability. We docked all compounds to the active site, the amino acids around of the crystallographic ligand ZMR1002, to analyze their binding mode by molecular docking. Subsequently, 10 novel compounds were designed based on the counter maps and molecular docking. Their ADME/T properties and synthesizable properties were evaluated by web server. The results showed that the designed novel compounds had good pharmacokinetic properties and synthesizable properties. Molecular docking results showed that hydrogen bonding, van der Waal and Carbon Hydrogen Bond played important roles in their recognition, and the newly designed compounds had the same binding mode as the template compound 13. MD simulations revealed these residues play key roles for the inhibitory activity of NA in the active site including Glu276, Asp151, Arg371, Arg292, Ser246, Trp178, etc. The above results can provide useful guidance for the discovery of novel NA inhibitors.