Influenza virus infection continues to be a major healthcare issue and the combination of the strain-specificity of vaccines with the increased circulation of therapeutic resistant strains drives the need for new approaches for the prevention and treatment of infection. Arbidol is a broad-spectrum antiviral with demonstrated activity against both group 1 and 2 influenza hemagglutinin subtypes. While the Arbidol binding site has been determined for group 2 type hemagglutinins, it remains unidentified for the group 1 type. Structural differences in conjunction with differential binding affinities make it unclear if Arbidol binds in a group-specific manner, hindering its further development as a specific and optimized influenza therapeutic. Recent work in our lab has led to the construction of atomically detailed, three-dimensional physical models of multiple variants of the influenza viral coat using integrative modeling that combines high-resolution x-ray crystallography, homology modeling, and protein-protein docking of the membrane-bound glycoproteins with cryo-electron tomography. The 2009 H1N1 model was selected to be studied in greater detail and was simulated for ~120 ns using all-atom molecular dynamics (MD) simulations. This system is one of the largest simulated to date, at ~100 nm in diameter with ~160 million atoms. From this simulation, 234 individual HA trajectories have been extracted resulting in a total of ~28 µs of HA simulation. In this study, we present the first structural and kinetic basis for the probable group 1 HA Arbidol binding site, identified and characterized by the construction of a cell-scale, ensemble based Markov state model (MSM). Cell-scale ensemble based MSMs differ from the standard approach in that, in place of utilizing many distinct copies of single protein simulations, we instead draw our statistical sampling directly from the many copies of each substituent as simulated in the ‘mesocape’. This approach then enables the theoretically rigorous extraction of long timescale dynamics of particular cellular components in the context of biological complexity, thus the effects of protein crowding and correlated effects of neighboring components are taken into account.

Structural basis for group 1 influenza fusion inhibition by Arbidol characterized with a cell-scale, ensemble based Markov state model

2019 Fall National Meeting

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.