Pharmacophore modeling, virtual screening, and 3D-QSAR studies on a series of non-steroidal aromatase inhibitors

Abstract

Aromatase inhibitors are the most important targets in treatment of estrogen-dependent cancers. In order to search for potent non-steroidal aromatase inhibitors (NSAIs) with lower side effects and overcome cellular resistance, Genetic Algorithm with Linear Assignment of Hypermolecular Alignment of Database was used to derive 3D pharmacophore models. The obtained best pharmacophore model contains one acceptor atom, one donor atom, and two hydrophobes, which was used in effective alignment of dataset. In succession, comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) were performed on 84 structurally diverse NSAIs to build 3D-QSAR models based on both pharmacophore and docking alignments. The CoMFA and CoMSIA models based on the pharmacophore alignment show better statistical results (CoMFA: q2 = 0.634, rncv2 = 0.986, rpred2 = 0.737; CoMSIA: q2 = 0.668, rncv2 = 0.926, rpred2 = 0.708). This 3D-QSAR approach provides significant insights that can be used to develop novel and potent NSAIs. In addition, the best pharmacophore model was used as a 3D query for virtual screening against NCI2000 database. The hit compounds were further filtered by docking, and their biological activities were predicted by the CoMFA and CoMSIA models, and six structurally diverse compounds with good predicted pIC50 values were obtained, which are expected to design novel NSAIs with new skeletons.