Targeting Aurora A kinase: Computational discovery of potent inhibitors through integrated pharmacophore and simulation approaches.

Abstract

Cancer currently ranks as the second most common cause of mortality worldwide, primarily due to uncontrolled cell growth driven by aberrant mitotic processes. Aurora A kinase (AURKA), a key regulator of mitosis involved in centrosome maturation, bipolar spindle formation, and cytokinesis, has been identified as a promising anticancer target. This study employs a comprehensive computational approach to identify new AURKA inhibitors. Using MOE software, a ligand-based pharmacophore model was developed based on six potent AURKA inhibitors. The model, consisting of three features-Aro/HydA, Acc, and Don/Acc-at an 80 % threshold, demonstrated strong discriminative power with a sensitivity of 69.8 %, specificity of 63.6 %, and accuracy of 60.4 %. Screening of the ZINC database yielded 774 hits, from which A1 (ZINC63106872) and A2 (ZINC39272872) were identified as the top candidates, with superior docking scores (-9.24 and -8.97 kcal/mol) compared to the reference MK-5108 (-7.49 kcal/mol). These hits satisfied Lipinski's rule and exhibited favourable ADMET profiles. DFT analysis revealed higher dipole moments (A1: 6.15 D, A2:6.39 D) and narrower HOMO-LUMO gaps (A1: 0.33 eV, A2: 0.38 eV), indicating enhanced polarity and reactivity. MEP plots showed defined donor-acceptor zones for both compounds, having a balanced surface. Molecular dynamics simulations over 500 ns confirmed complex stability, with protein backbone RMSD around 2.8 Å and ligand RMSD of 4.0 Å (A1) and 6.0 Å (A2). RMSF values remained below 2.4 Å. The most favourable binding energy for A1 (-75.34 kcal/mol) in MM-GBSA analysis confirms its strong interaction and therapeutic potential.