Flavonoids as Structural Probes Reveal Conformationally Dependent Ligand Recognition in the SARS-CoV-2 JN.1 Spike Protein

Abstract

The SARS-CoV-2 JN.1 subvariant has attracted attention due to the accumulation of mutations in the Spike (S) glycoprotein, particularly within the receptor-binding domain (RBD), contributing to the structural heterogeneity of the Spike protein. However, the extent to which conformational dynamics influence ligand-recognition patterns across functional Spike states remains insufficiently characterized. In this study, we applied a hierarchical computational framework combining homology modeling, molecular dynamics simulations, and molecular docking to determine whether conformational sampling modifies the location, accessibility, and recurrence of ligand-interaction regions in the JN.1 Spike protein. Closed, semi-closed, and open conformations were modeled and subjected to triplicate simulations, including an initial 100 ns phase followed by extended 200 ns simulations. Representative conformations were obtained by trajectory clustering and used for docking analyses. Two flavonoids, hesperitin-7-O-rutinoside (H7R) and flavanone-7-O-glucoside (F7G), were employed as structural probes to assess how ligand interaction patterns vary across conformational states. Comparative analyses revealed that conformational sampling modifies the location, accessibility and spatial distribution of ligand-interaction regions while generating distinct docking poses and recurrent interaction patterns not fully captured by static structural models. These findings highlight the importance of dynamics-informed structural ensembles for docking analyses in flexible viral proteins. Rather than predicting absolute binding affinities, this study provides an exploratory computational framework for evaluating ligand-recognition behavior in structurally dynamic viral systems.