CoMFA and docking studies of 2-phenylindole derivatives with anticancer activity

Three-dimensional (31)) quantitative structure-activity relationship (QSAR) and docking studies of 43 tubulin inhibitors, 2-phenylindole derivatives with anticancer activity against human breast cancer cell line MDA-MB 231, have been carried out. The established 3D-QSAR model from the comparative molecular field analysis (CoMFA) in training set shows not only significant statistical quality, but also satisfying predictive ability, with high correlation coefficient value (R-2 = 0.910) and cross-validation coefficient value (q(2) = 0.705). Moreover, the predictive ability of the CoMFA model was further confirmed by a test set, giving the predictive correlation coefficient (R-pred(2)) of 0.688. Based on the CoMFA contour maps and docking analyses, some key structural factors responsible for anticancer activity of this series of compounds were revealed as follows: the substituent R-1 should have higher electronegativity; the substituent R-2 should be linear alkyl with four or five carbon atoms in length; and the substituent R-3 should be selected to OCH3-kind group whereas should not be selected to CF3-kind group. Meanwhile, the interaction information between target and ligand was presented in detail. Such results can offer some useful theoretical references for understanding the action mechanism, designing more potent inhibitors and predicting their activities prior to synthesis. (C) 2008 Elsevier Masson SAS. All rights reserved.