A combination of pharmacophore modeling, virtual screening, and molecular docking studies for a diverse set of colchicine site inhibitors

In the present study, 3D QSAR pharmacophore models were undertaken from a set of 16 colchicine binding site inhibitors (CSIs). The best pharmacophore model possessing two chemical features (hydrogen-bond acceptor and hydrophobic) showed an excellent correlation coefficient for the training (r(training) - 0.96) and a fair correlation coefficient for the test set (r(test) - 0.88) molecules. Considering the statistically significant results of the best pharmacophore model, the hypothesis was selected as a 3D structural query to screen the Maybridge and MiniMaybridge database. Ultimately, three of the hit molecules satisfied all of these conditions, which were then submitted to molecular docking studies to evaluate their optimal orientations and their interactions with the critical residues of tubulin. The molecules showed strong hydrogen-bond interactions as well as hydrophobic contacts with critical residues such as beta:Tyr376, beta:Ile378, Tyr48, and Ser178. As indicated above, the hit three molecules can be good candidates for the CSIs. And the developed pharmacophore model can be used to determine the essential structural requirements, thus as a valuable tool to select the novel compounds based on virtual screening approach.