Pradeep Hanumanthappa, Mahesh K. Teli and Rajanikant G. Krishnamurthy Pages 436 - 451 ( 16 )
In the present report, 3D-QSAR analysis was executed on the previously synthesized and evaluated derivatives of isoquinolin-1-ones and quinazolin-4-ones; potent inhibitors of tumor necrosis factor α (TNFα). Statistically significant 3D-QSAR models were generated using 42 molecules in the training set. The predictive ability of models was determined using a randomly chosen test set of 16 molecules, which gave excellent predictive correlation coefficients for 3-D models, suggesting good predictive index. Pharmacophore prediction generated a five point pharmacophore (AAHRR): two hydrogen bond acceptor (A), one hydrophobic (H) and two ring (RR) features. This pharmacophore hypothesis furnished a statistically meaningful 3D-QSAR model with partial least-square (PLS) factors seven having R2 = 0.9965, Q2 = 0.6185, Root Mean Squared Error = 0.4284 and Pearson-R = 0.853. Docking study revealed the important amino acid residues (His 15, Tyr 59, Tyr 151, Gly 121 and Gly 122) in the active site of TNFα that are involved in binding of the active ligand. Orientation of the pharmacophore hypothesis AAHRR.25 corresponded very closely with the binding mode recorded in the active site of ligand bound complex. The results of ligand based pharmacophore hypothesis and atom based 3D-QSAR furnished crucial structural insights and also highlighted the important binding features of isoquinolin-1-ones and quinazolin-4-ones derivatives, which may provide guidance for the rational design of novel and more potent TNFα inhibitors.
3D-QSAR, Docking, Isoquinolin-1-ones and quinazolin-4-ones derivatives, Inflammatory diseases, Pharmacophore, TNFα, isoquinolin, cytokine, fragment-based, Hansch-Fujita models
School of Biotechnology, Coordinator, Bioinformatics Infrastructure Facility, National Institute of Technology Calicut, Calicut - 673601, Kerala (India).