Identification and Characterization of Surfactant Binding Sites on the SARS-CoV-2 Spike Protein and Its Variants by Microseconds-Long Molecular Dynamics Simulations

Abstract While airborne coronavirus virions penetrate respiratory airways, lung surfactants (LS) adsorb on spike proteins and may hinder their binding to cell receptors. In this work, we utilize microsecond-long unbiased and adaptive sampling molecular dynamics simulations performed on the Anton 2 supercomputer to elucidate the molecular mechanisms and dynamics of the adsorption of lung surfactant lipids and cholesterol onto the receptor-binding domain (RBD) of the SARS-CoV-2 spike protein and its Delta and Omicron variants. From unbiased simulations of spontaneous multisurfactant binding, we show that surfactants tend to destabilize the protein and induce enhanced structural fluctuations that can influence the receptor-binding interface. We identify four surfactant-binding sites (SBS) that are conserved across the variants, one of which corresponds to the experimentally identified fatty acid binding site (FABS). By developing an approximate model, we analyze the energetics of multisurfactant binding and show that the binding of multiple surfactants at a site weakens individual surfactant–protein interactions, leading to potentially more dynamic binding behavior. Using adaptive sampling with the Tempered Binding approach, we demonstrate unbinding and rebinding of surfactants from selected binding sites, elucidating the structural dynamics and the binding enthalpy profiles. Identification and quantitative characterization of the binding sites' affinity and stability provide a better understanding of the specifics of lung surfactant–spike protein interactions and may support future investigations into the potential role of these interactions in surfactant-based therapeutic strategies.