CoMFA and CoMSIA studies on 5-hydroxyindole-3-carboxylate derivatives as 5-lipoxygenase inhibitors: Generation of homology model and docking studies

Abstract

In this study, comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) were performed on a series of 2-substituted 5-hydroxyindole-3-carboxylate derivatives as potent 5-LOX inhibitors with IC50 values ranging from 0.031 to 13.4 μM. Two datasets of same molecules were prepared with two different partial atomic charges; one with Gasteiger-Huckel and another with the ESPFIT charges obtained from the gaussian package. CoMFA and CoMSIA models were generated for both the datasets and the results were analysed. With regard to the non-cross validated r2 values (rncv2) and cross-validated q2 values (qcv2) of the resulting QSAR models, the dataset with ESPFIT charges yielded higher values; hence it was further used in the study. The CoMFA and CoMSIA models have been further validated for their stability and robustness using group validation and bootstrapping techniques and for their predictive abilities using an external test set of ten compounds. The predictive power of the CoMSIA model was higher than the CoMFA model, the high predictive r2 values of the test set reveals that the models prove to be useful tools for activity prediction of newly designed 5-LOX inhibitors. The ESPFIT-derived charges yielded better models than those based on charges calculated from Gasteiger-Huckel charges. We generated a homology model for human 5-LOX and identified the key residues at the binding site. The 3D-QSAR models were compared with the interactions at the active site to further elucidate the accuracy of the models. The data generated from 3D-QSAR study was used to design potential 5-LOX inhibitors. © 2010 Elsevier Ltd. All rights reserved.