Computational design and evaluation of low-toxicity saquinavir analogues targeting the catalytic dyad and oxyanion-hole loop of SARS-CoV-2 Mpro: insights from ensemble docking, molecular dynamics, dynamic undocking, and ADMET analysis.

A myriad of therapeutic candidates targeting SARS-CoV-2 have entered clinical trials; however, the ongoing challenges in SARS-CoV-2 drug discovery, such as adverse effects associated with some therapeutic candidates, necessitate continuous efforts to identify novel therapeutic targets and strategies. This study leverages integrated in silico approaches, encompassing ensemble docking, molecular dynamics (MD) simulations, dynamic unbinding (DUck), and ADMET predictions, to identify novel saquinavir-related antiviral inhibitors targeting the catalytic dyad and oxyanion-hole loop of the SARS-CoV-2 main protease (Mpro). From a library of 33 saquinavir-related analogs, ensemble docking identified three high-affinity ligands (ΔG ≤ -9.8 kcal/mol). Subsequent MD simulations revealed stable Mpro-ligand complexes and significant structural perturbations within the catalytic dyad (His41-Cys145, ΔDdyad >1.0 Å) and the oxyanion-hole (Gly143-Ser144-Cys145, Δθoxy >5°). DUck simulations elucidated a stepwise dissociation mechanism, identifying key hotspot residues critical for ligand binding. Compounds CHEMBL3706523 and CHEMBL3706524 emerged as promising candidates, exhibiting robust interactions and slower dissociation rates (WQB >6 kcal/mol). These ligands stabilized the receptor and induced conformational changes that may hinder substrate binding, suggesting a potential 'block cluster' mechanism for inhibition. Favorable ADMET profiles further support their potential as drug candidates with low mammalian toxicity. This study provides a strong foundation for experimental validation and the subsequent development of effective antiviral therapies against SARS-CoV-2.