Structural and dynamic properties of guanosine-analog binding to 2'-deoxyguanosine-II riboswitch: a computational study.

Antibiotic resistance poses a significant global health threat, necessitating the development of novel antibacterial strategies. Riboswitches, particularly those sensing purines, have emerged as promising targets for antibiotic development. In an unconventional approach, we explore the repurposing of antiviral compounds as potential antibacterial agents targeting riboswitches. This study employs a comprehensive computational framework to investigate the binding of 2'-deoxyguanosine (2'-dG) and 10 antiviral analogs to the 2'-dG-II riboswitch. We integrated structural similarity analysis, molecular docking, molecular dynamics simulations and molecular mechanics/generalized Born surface area (MM/GBSA) free energy calculations to elucidate the molecular basis of ligand recognition and assess the feasibility of antiviral compounds as riboswitch-targeting antibacterials. Our results revealed that the guanine moiety's binding orientation is conserved across all compounds, while interactions involving sugar or sugar-like moieties significantly influence binding stability and specificity. Key nucleotides U22, C51 and C78 play crucial roles in ligand recognition across all complexes. Notably, the antiviral compounds lagociclovir and valganciclovir demonstrated higher binding affinities than the native ligand, with unique interaction patterns. The MM/GBSA analysis provided binding free energies consistent with the experimental data for known compounds, validating our computational approach. This study offers detailed insights into the adaptability of the 2'-dG-II riboswitch binding pocket and identifies promising candidates for riboswitch-targeting antibiotics, contributing to the broader efforts to combat antibiotic resistance through computational drug discovery and repurposing strategies.