Petascale simulations to understand how viral membrane organization controls influenza entry

Peter Kasson, Georgia Institute of Technology

Usage Details

Enveloped viruses such as influenza, Ebola, SARS, and HIV enter and infect cells via a process of membrane fusion. Despite decades of research, much remains unknown regarding the biophysical mechanisms of viral membrane fusion, and viral entry has remained a difficult target for pharmacological inhibition. When faced with the potential for rapidly emerging and highly virulent infections, such an understanding of fusion becomes of particular importance to both fundamental biology and national health.

Experiments have not been able to directly address the mechanisms by which fusion proteins interact to bring about fusion due to several factors: the process involves multiple stochastic steps, the protein-lipid assemblies that mediate fusion are transient and dynamic, and these assemblies are heterogeneous multi-component structures that are challenging targets for high-resolution structure. We have recently used electron cryo-microscopy to show that cholesterol content can alter membrane organization in live influenza virus in a manner that correlates with changes to fusion kinetics and efficiency (1,2). Although these experiments clearly show that the spatial distribution of viral fusion proteins changes in response to the amount of cholesterol in the viral membrane, they cannot assign a detailed molecular mechanism.

Here, we use molecular dynamics simulation to model how membrane composition controls spatial patterning of the viral surface and thus fusion behavior. Even a reduced model system for portions of the influenza viral entry process involves simulating 300,000 to 1 million atoms, and multi-microsecond simulations are required to examine how protein components interact in these assemblies, as the proteins rearrange slowly in the membrane. The hybrid GPU/CPU architecture and fast networking fabric of Blue Waters will allow us to perform these long and computationally intensive simulations. Our goal is to predict how fusion protein components interact with membrane components and respond to membrane changes—we will then design experiments to test these interactions in our laboratory. Ultimately, we wish to build an understanding of how influenza fusion proteins work together to accomplish viral entry.