Computational insight into dengue virus NS2B-NS3 protease inhibition: A combined ligand- and structure-based approach

Abstract The NS2B-NS3 protease is essential for the replication process of Dengue Virus, which make it an attractive target for anti-virus drugs. Since a considerable number of NS2B-NS3 protease inhibitors have been reported so far, it is significant for the discovery of more effective antivirus compounds with the essential structure-activity relationship extracted from known inhibitors. In this perspective, the relationship between the chemical features of inhibitors and their biological activities was investigated with a combined ligand- and structure-based approach. Furthermore, 3D pharmacophore models were generated with the best selected, which consisted of five chemical features: one ring aromatic group, one hydrophobic group, one hydrogen bond donor and two hydrogen bond acceptors (RHDAA). Subsequently, molecular docking was employed to explore the specific allosteric site for non-peptidic inhibitors to bind, which was proved to be located behind the catalytic triad. Taken the results of both molecular docking and pharmacophore modeling into consideration, a model of receptor-ligand interaction was obtained with four essential chemical features including aromatic rings and hydrogen bonds. This research provided an accurate binding model for the discovery and optimization of NS2B-NS3 protease inhibitors.