Resistance to nirmatrelvir, the SARS-CoV-2 main protease (Mpro) inhibitor approved for COVID-19 treatment, is threatening its long-term efficacy. Resistant mutations of Mpro have been identified in vitro selection, circulating strains and in patients. Despite significant efforts, the molecular mechanisms of drug resistance remain unclear. This is largely due to a limited understanding of the dynamic drug binding process, which hinders the development of next-generation Mpro inhibitors. To address the challenges of experimentally resolving transient drug-target intermediates and achieving sufficient sampling in molecular modeling, we combine enhanced sampling molecular dynamics (MD) simulations with solution-state protein NMR, leveraging their complementary strengths to construct high-resolution dynamic models of nirmatrelvir binding to Mpro. Specifically, the weighted ensemble (WE) strategy enables access to the timescales necessary to capture complete drug binding pathways while yielding unbiased kinetic estimates. Focusing on Mpro E166V, the most resistant mutant, our simulations revealed distinct conformational spaces, including unique binding pathways and intermediate states compared to the wild type. We also identified key residues mediating encounter complex formation and ligand-induced fit during the binding process. Subsequent analysis of the binding free energy landscapes and kinetic profiles will help better elucidate the dynamic resistance mechanism. Insights from this study lay the foundation for developing inhibitors with improved resilience to emerging mutations, ultimately advancing therapeutic strategies against COVID-19 and future coronavirus threats.
Yao Fu (2026) studied this question.