How Does Glycosylation Affect Drug Binding on Influenza? the Roles of Electrostatics and Sterics Examined Through Brownian Dynamics Simulations

Influenza is one of the most well-known viruses, yet much remains to be discovered. Glycans, the ubiquitous polysaccharides that decorate the surface of influenza membrane proteins, play a complex part in influenza's life functions. They are known to modulate influenza's transmissibility, virulence, electrostatics, sterics, cooperativity, drug binding, dynamics, antibody response, and immune evasion, amongst other aspects, but their exact role is not always well-understood. Additionally, experimental influenza work is challenging to deconvolute due to complex and overlapping interactions, and due to influenza's mutability. These difficulties can be bypassed by creating these systems computationally. To examine how glycan and influenza strain differences affect the binding kinetics of Tamiflu and Relenza, we created in silico systems of neuraminidase from avian influenza (H5N1 A/Vietnam/1203/2004) and swine flu (H1N1 A/California/04/2009) with experimentally-derived, biologically-relevant glycoprofiles. We then benchmarked association rates of the drugs to the active site of neuraminidase with Brownian dynamics, a type of simulation evaluating the electrostatics and sterics of a system. Next, we modulated this association rate by removing glycans, to deconvolute the role of electrostatics and sterics in drug association to neuraminidase, and how this varies across strain and glycoprofile. These results increase our understanding of how the viral glycoprofile influences drug association, and has implications for vaccine optimization.