Pharmacophore Modeling for Qualitative Prediction of Antiestrogenic Activity

A ligand-based pharmacophore approach for the prediction of antiestrogenic activity to be used as an in silico screening tool for bioactive compounds including natural products was developed using Catalyst HypoGen. The generated pharmacophore hypothesis (HYPO-7) consisted of five features, namely, one hydrophobic (HY1), two hydrophobic aromatic (HY2), one hydrogen-bond acceptor (HBA), and one hydrogen-bond donor (HBD). HYPO-7 successfully predicted the lack of cytotoxicity of a number of new metabolites isolated from the red alga Laurencia glandulifera. Furthermore, a screening of the Asinex Gold Collection database was performed by coupling HYPO-7 with a docking filtration, which resulted in a restricted set of 12 new scaffolds to be investigated as potential SERMs. The inhibitory activity of these compounds was evaluated in vitro using MCF7 human breast adenocarcinoma cell line. Ten out of the twelve compounds exhibited inhibitory activity with IC(50) values between 26 and 188 microM. This result shows that application of HYPO-7 could assist in the selection of potentially active compounds, thus expediting the hit discovery process.