3D-QSAR and Molecular Docking Approaches for the Identification of Novel Phyto-inhibitors of the Cyclin-dependent Kinase 4

Abstract

Cyclin-dependent kinase 4 (CDK4) is an important target in designing anti-cancer drugs. The activation of CDK4 results in phosphorylation of the retinoblastoma gene product. In this study, a total of one hundred and seventy-eight phytochemicals characterized from various anti-cancer plants were retrieved from the literature and screened against the orthosteric sites of CDK4. Lipinski's rule of five was used to determine the drug-likeness and the activities of the lead phytochemicals. Bioassay IC50 data for reported CDK4 inhibitors from the Chembl database were used to generate the 3D-QSAR model for CDK4 inhibition. The virtual screening showed catechin, kaempferol and quercetin as the lead phytochemicals. A positive correlation of 0.829 between the pIC50 and glide scores at p<0.01 revealed that computers can accurately predict experimental data. The ADME screening showed that naringenin, aporphine, catechin, coreximine and stepharine obey the Lipinski rules of five. The generated model was robust and thoroughly validated with a Pearson correlation R value of 0.934 and R² value of 0.872. The model with an adjusted R² value of 0.769 possesses good external validation. Aporphine, catechin, naringenin, stepharine and coreximine form important hydrogen bond interactions. These interactions are likely responsible for their inhibition of CDK4. The lead phytochemicals are drug-like compounds and potential inhibitors of CDK4.