3D-QSAR and molecular docking studies of hydroxamic acids as peptide deformylase inhibitors

Comparative molecular field analysis (CoMFA), comparative molecular similarity indices analysis (CoMSIA), and molecular docking study were conducted on hydroxamic acids as potent peptide deformylase (PDF) inhibitors. The optimal CoMFA model gave statistically significant results with q ² and r ² values of 0.568 and 0.956, respectively. The optimal CoMSIA model with combination of steric, hydrophobic and H-bond donor fields resulted in the best results with q ² and r ² values of 0.722 and 0.958, respectively. These two models were validated by an external test set of eight compounds with satisfactory predictive r ² values of 0.810 and 0.820, respectively. The contour plots of molecular fields indicated that electrostatic and bulky groups substituted at the R₁ position, and electropositive and small substituted at R₂ position were favorable for the inhibitory activity. In addition, FlexX docking was employed to investigate the binding mode between PDF and its inhibitors. It was found that hydrogen bond interactions might be an important factor for binding affinity of inhibitors in the hydrophobic cavity. Based on the optimal CoMSIA model and FlexX docking, a series of PDF inhibitors with high predictive activities have been designed. This work might provide valuable information in designing more promising PDF inhibitors.