In order to explore the interactions between flavones and P-gp, in silico methodologies such as docking and 3D-QSAR were performed. CoMFA and CoMSIA analyses were done using ligand based and receptor guided alignment schemes. Validation statistics include leave-one-out cross-validated R(2) (q(2)), internal prediction parameter by progressive scrambling (Q(*2)), external prediction with test set. They show that models derived from this study are quite robust. Ligand based CoMFA (q(2) = 0.747, Q(*2) = 0.639, r(pred)(2)=0.802) and CoMSIA model (q(2) = 0.810, Q(*2) = 0.676, r(pred)(2)=0.785) were developed using atom by atom matching. Receptor guided CoMFA (q(2) = 0.712, Q(*2) = 0.497, r(pred)(2) = 0.841) and for CoMSIA (q(2) = 0.805, Q(*2) = 0.589, r(pred)(2) = 0.937) models were developed by docking of highly active flavone into the proposed NBD (nucleotide binding domain) of P-gp. The 3D-QSAR models generated here predicted that hydrophobic and steric parameters are important for activity toward P-gp. Our studies indicate the important amino acid in NBD crucial for binding in accordance with the previous results. This site forms a hydrophobic site. Since flavonoids have potential without toxicity, we propose to inspect this hydrophobic site including Asn1043 and Asp1049 should be considered for future inhibitor design.
Copyright © 2011 Elsevier Masson SAS. All rights reserved.