A combined pharmacophore modeling, 3D QSAR, virtual screening, molecular docking, and ADME studies to identify potential HDAC8 inhibitors

Abstract

The World Cancer Report 2014 shows that cancer is a leading cause of death globally, and cancer related death is likely to go up by about 70 % in the next two decades. Among several target receptors of cancer, histone deacetylases are the promising therapeutic target for many cancers. The current study deals with a primary goal of identification of novel non-hydroxamic acid histone deacetylase 8 inhibitors. In this context, six featured pharmacophoric hypotheses with hydrogen bond acceptors (A), hydrogen bond donor (D), and aromatic ring (R) were generated using 36 reported histone deacetylase 8 inhibitors. Virtual screening of database using two pharmacophore hypothesis AAAADR.161 and AAADDR.189 resulted 2000 hits each having fitness score >1.503. The pharmacophore AAADDR.189 yielded statistically significant three-dimensional quantitative structure-activity relationship model with training set (R2: 0.9756, SD: 0.0680, F: 151.6, N: 26) and test set (Q2: 0.6922, Pearson R: 0.8705, N: 10) molecules. The R2pred value of the model was 0.6951, which confirmed the good predictive ability of the model for external data set. Hits resulted from virtual screening and known inhibitors were subjected to molecular docking study to identify the binding affinity of inhibitors with active site amino acid residues. Finally, absorption, distribution, metabolism, and excretion study were undertaken to determine the drug likeness properties of identified hits. On the basis of fitness score, predicted activities, XP Glide score, interacting amino acid residues of known inhibitors and absorption, distribution, metabolism, and excretion properties, ten structurally diverse hits are reported in this paper as potential histone deacetylase 8 inhibitors which reduce the cost of histone deacetylase 8 inhibitor discovery and enhance the process prior synthesis.