Computational design and optimization of INHA inhibitors based on 5-amino-1 H-pyrazole-4-carbonitrile for anti-tuberculosis applications

Abstract

Abstract Tuberculosis (TB) remains a persistent global health challenge, increasingly complicated by drug-resistant Mycobacterium tuberculosis . Direct inhibition of enoyl-acyl carrier protein reductase (InhA) offers a promising strategy to bypass the prodrug activation limitations of isoniazid. Here, we present an integrated computational framework for the design and optimization of 5-amino-1 H-pyrazole-4-carbonitrile derivatives as direct InhA inhibitors, explicitly retaining the NADH cofactor throughout all structure-based analyses to preserve catalytic site fidelity. A validated e-pharmacophore model (AUC = 0.84; GH = 0.71) guided screening, followed by QSAR modeling (R² = 0.878; Q² = 0.845). Docking and post-docking MM-GBSA identified promising candidates, which were subsequently validated through 200 ns molecular dynamics simulations. Crucially, trajectory-based MM-GBSA calculations revealed averaged binding free energies of − 54.6 ± 3.2 kcal/mol (DC26) and − 57.8 ± 2.9 kcal/mol (DC28), highlighting deviations from single-snapshot estimates. Per-residue energy decomposition identified Tyr158, Met199, and NADH as dominant energetic contributors, with distinct interaction fingerprints differentiating DC26 and DC28. Principal component analysis confirmed stable conformational sampling with convergence into dominant binding basins. Collectively, DC28 emerges as the more energetically favorable inhibitor, while DC26 exhibits complementary interaction stability. These findings refine the energetic and mechanistic understanding of pyrazole-based InhA inhibition and provide a quantitatively validated framework for future anti-TB drug development.