Computational and Biophysical Analysis of the Filovirus Matrix Protein System

Project Summary Ebola (EBOV) and Marburg (MARV) filoviruses cause severe hemorrhagic fever in humans with up to 90% mortality rates. Their genome contains only seven genes including the viral matrix protein VP40 which, when expressed in mammalian cells, is sufficient to produce virus-like particles (VLPs) that are essentially indistinguishable from live virions. VP40 forms dimers, hexamers and octamers mediated by different protein-protein (PPI) and protein-lipid (PLI) interactions that fulfill different and essential roles in the viral lifecycle, making VP40 a “swiss army knife” of proteins. The fascinating dynamic equilibria of VP40 and the availability of VLPs as a model system for direct observations outside of a BSL4 laboratory make VP40 a unique system to rigorously study the biophysical basis for viral budding as well as PPIs and PLIs in general. The significance of these studies is further increased because VP40 is the most conserved protein upon virus passage through humans, but exploiting VP40 as a potential drug target is unlikely to succeed without understanding the physical basis for oligomerization and function of VP40. The Stahelin and Wiest laboratories, building on established collaborations with each other and several other collaborators supplying specific expertise, will use computational, experimental and structural biophysics methods to investigate the central hypothesis of this grant: that interdomain interactions of VP40 are key regulators of VP40 structures during the viral life cycle. In two specific aims, we will (i) Determine the biophysical mechanisms by which VP40 dimer, hexamer and octamers form in silico, in vitro and in human cells and (ii) determine how mutations of VP40 that arise in humans during the course of an outbreak as well as in animals during passage of virus contribute to VP40 conformational change and rearrangement into its separate oligomeric forms. These questions will be studied using a tightly integrated approach using multiscale molecular dynamics simulations on the µs timescale and free energy perturbation methods on the computational side and hydrogen-deuterium exchange, cellular imaging of VLPs as well as more traditional biophysical experiments such as ultracentrifugation and SPR to determine the binding constants of wildtype VP40 from EBOV and MARV as well as pertinent mutants. This innovate and integrated approach will not only provide careful validation of the results, but also provide detailed insights into the PPIs and PLIs governing the oligomerization equilibria across many time- and lengths scale, thus enabling a rigorous understanding of the biophysical principles for a biomedically very important filovirus protein that will have a significant impact on understanding other PPIs and PLIs.