Discovery of potential Aurora-A kinase inhibitors by 3D QSAR pharmacophore modeling, virtual screening, docking, and MD simulation studies.

The Aurora-kinase family comprises of cell cycle-regulated serine/threonine kinases playing a vital role during mitosis. Aurora-A kinase is involved in multiple mitotic events in cell cycle and is a major regulator of centrosome function during mitosis. Aurora-A is overexpressed in breast, lung, colon, ovarian, glial, and pancreatic cancer. Hence, Aurora-A kinase is a promising target in cancer therapy. In our current study, a four-point 3D QSAR pharmacophore model has been generated using substituted pyrimidine class of Aurora-A kinase inhibitors. It had a fixed cost value 88.7429. The model mapped well to the external test set comprising of clinically active molecules, with a correlation coefficient  = 0.99. From the mapping, it was found that the hydrophobic features (HY) of a molecule play an important role for Aurora-A kinase inhibitory activity, whereas the ring aromatic feature provides geometric constraint for spatial alignment of different functional group. The hypothesis, with one hydrogen bond acceptor, two ring aromatic features, and one hydrophobic feature, was selected to screen miniMaybridge database. The screened ligands were filtered on the basis of activity, shape, and drug likeliness. This led to the identification of five top hits. These identified potential leads were further subjected to docking with the ATP-binding site of Aurora-A kinase. The molecular dynamic simulation studies of top lead molecules having diverse scaffolds endorsed that the identified molecules had distinctive ability to inhibit Aurora-A kinase. Thus, this study may facilitate the medicinal chemists to design promising ligands with various scaffolds to inhibit Aurora-A kinase. Communicated by Ramaswamy H. Sarma.