In silico identification of potential inhibitors for human aurora kinase b

Cell cycle progression through mitosis and meiosis involves regulation by serine/threonine kinases from the aurora family. Aurora kinase b (Aurkb) is mainly involved in the proper segregation of chromosomes during mitosis as well as meiosis. However, over expression of Aurkb leads to the unequal distribution of genetic information creating aneuploid cells, a hallmark of cancer. Thus, Aurkb can be used as an effective molecular target for computer-aided drug discovery against cancer. Existing Aurkb inhibitors are less efficient, hence an in silico work was carried out to identify novel potent inhibitors. Three published inhibitors azd1152, zm447439 and N-(4-{[6-methoxy-7-(3-morpholin-4-ylpropoxy) quinazolin- 4-yl] amino} phenyl) benzamide were subjected to high throughput virtual screening of over 1 million entries from a ligand info meta database, to generate a 1161 compound library. The crystal structure was optimized and energy was minimized applying an OPLS force field in Maestro v9.0. Molecular docking using Glide was performed to predict the binding orientation of the prepared ligand molecule into a grid of 20*20*20 A created around the centroid of the optimized human Aurkb protein. Nine lead molecules with good binding affinity with human Aurkb were identified. In silico pharmacokinetics study for these nine lead molecules has shown no ADME violation. Analysis of lead ‘1’- human Aurkb docking complex has revealed a XP Gscore of -10.20 kcal/mol with a highly stabilized hydrogen bond network with Asp218 and Ala157 and good Van der wall interactions. The docking complex coincides well with the native co- crystallized human Aurkb and inhibitor zm447439 complex. Thus, lead 1 would be highly useful for developing potential drug molecules for the treatment of cancer.