Abstract
Dipeptidyl peptidase IV (DPP-IV) is an attractive target, and its launched inhibitors have made great significance to clinical therapy of type 2 diabetes. For developing high potent DPP-IV inhibitors, ligand- and structure-based approaches were applied for the optimization of cyanopyrrolidine in this study. Two statistically significant 3D-QSAR models (CoMFA with q2, 0.585; r2, 0.963; CoMSIA with q2, 0.678; r2, 0.949) were developed. In combination with the structure-based pharmacophore model, fundamental structural requirements and pharmacophoric features for R groups were determined. Suitable fragments for different R groups were retrieved for the generation of novel molecules. After being evaluated by molecular docking, QuaSAR descriptors filter and 3D-QSAR activity prediction, active DPP-IV inhibitors were found. The reliability indicated this workflow can be applied to facilitate lead optimization for DPP-IV and even for other drug targets.
Keywords: DPP-IV, 3D-QSAR, pharmacophore, fragment, type 2 diabetes.
Graphical Abstract