Pharmacophore modeling, atom-based 3D-QSAR and molecular docking studies on N-benzylpyrimidin-4-amine derivatives as VCP/p97 inhibitors

Abstract

Valosine containing protein (VCP/p97), which involves in several important cellular functions and plays an important role in ubiquitin-mediated misfolding protein degradation, has been found to be a novel target for the treatment of a range of cancers, such as lung cancer and breast cancer. In this study, the atom-based three-dimensional quantitative structure–activity relationship (3D-QSAR) and docking models were developed using PHASE and GLIDE modules of Schrodinger software, respectively. The theoretical models were generated from 38 N-benzylpyrimidin-4-amine inhibitors of p97. An AADRRR model consisting of two hydrogen bond acceptors (A), one hydrogen bond donor (D), and three aromatic rings (R) was obtained. Thirty eight derivatives were divided into a training set with 27 molecules to generate 3D-QSAR models and a test set with 11 molecules to validate 3D-QSAR model. A robust QSAR model with good prediction in internal and external verification was constructed, where R2, Q2, and Pearson-R were 0.924, 0.701, and 0.8783, respectively. QSAR model showed the hydrogen bond donor, electron-withdrawing group, and hydrophobic characteristics affecting the p97 activity. Molecular docking studies indicated that the H-bond and hydrophobic interactions existed between the inhibitors and p97, which was consistent with the results of 3D-QSAR. These results provided some useful information for designing new and effective p97 inhibitors.