Comparative QSAR and pharmacophore analysis for a series of 2,4-dihydro-3H-1,2,4-triazol-3-ones derivatives as angiotensin II AT1 receptor antagonists

This article describes the development of a robust two-dimensional (2D), Group-based quantitative structure–activity relationship (G-QSAR), k-nearest neighbor (kNN), and pharmacophore models and the investigation of SAR analysis of 42 triazolinone derivatives reported for angiotensin II (Ang II) AT1 receptor antagonists using VLife MDS software. Comparing with partial least square (PLS) methodology coupled with various feature selection methods, viz., stepwise and simulated annealing (SA) are applied for the optimization and selection of suitable descriptors for the development of QSAR model for triazolinone derivatives. The study correlates activity measured as pIC50 values of 42 structurally related triazolinone analogs to several physicochemical parameters representing spatial, electrostatic, hydrophobic, hydrogen bond acceptor (HAc) and donor (HDr), thermodynamic, and topological fields. The statistically significant best 2D-QSAR model by SA coupled with PLS analysis method showed 88.7 % variation in biological activity and q2 = 0.7817 with pred_r2 = 0.8003 with the descriptors like SddsN (nitro) count, SdsCHcount, SsBrcount, SsNH2-index, and SssNHE-index. The statistically significant best G-QSAR model having correlation coefficient r2 = 0.7764 and cross-validated squared correlation coefficient q2 = 0.6729 with external predictive ability of pred_r2 = 0.7141 was developed by SA-PLS method. Molecular field analysis was used to construct the best 3D-QSAR model using SA-PLS method, showing good correlative and predictive capabilities in terms q2 = 0.7811 and pred_r2 = 0.7597. The results indicated that the steric and spatial parameters significantly influence antagonistic activity. Electron-withdrawing group at aryl moiety R1 and R3 positions was found to be an essential feature for Ang II AT1 receptor. In addition, the pharmacophore model well corroborated with kNN studies as the contours of latter were in good agreement with the 3D orientation of the pharmacophoric features. A five-point pharmacophore with three HAcs one HDr, and one AroC feature (aromatic) was obtained. This information is pertinent for the further design of new antihypertensive agent containing the triazolinone nucleus.