Ligand- and Structure-based Virtual Screening Studies for the Discovery of Selective Inhibitors

摘要

New drug design research using computer studies the relationship between 2D/3D quantitative structure and activity of target protein, derives pharmacophore model, which is essential for pharmacological activity that substrate should have, proceed screening to select the optimal inhibitor. Computer-aided drug design (CADD) studies are divided into structure- and ligand-based drug design. Structure-based drug design is to design a pharmacophore model for amino acid residues in active sites based on the 3D structure of the target protein under study. Ligand-based drug design predicts the activity of a new candidate compound by designing a pharmacophore model and a quantitative structure-activity relationship model using information of compounds known to be active against a target protein. In this study, selective inhibitors were designed using both methods. Ligand-based pharmacophore model consisting of ring aromatic, positive ionizable, negative ionizable, and hydrogen bonding receptors were generated from known active inhibitors. In addition, by analyzed the binding modes of known target proteins and ligands, we generated a structure-based pharmacophore model with the addition of hydrogen bonding donors. To increase bioavailability from a total of 1,890,602 commercial drug candidates databases, Lipinski's rule of five and ADMET analysis were used to filter 178,161 chemical compounds for virtual screening. After that, 20 selective inhibitors were finally selected through the process of selecting compounds with a fit value of 3.54 or higher and eliminating duplication.