Targeting Viral Proteins to Sites of Genome Replication: Discovery and Characterization of Viral Phosphoinositide-Binding Domains

• Main Author: Shengjuler, Djoshkun
• Format: Dissertation
• Language: English
• Published: ProQuest Dissertations & Theses 01-01-2017
• Subjects: Molecular biology; Virology
• ISBN: 9798678112293

How viral proteins are targeted specifically to sites of genome replication is a longstanding question in virology. All positive-strand RNA viruses reorganize host intracellular membranes in order to form replication organelles (ROs), which are platforms where viral genome replication takes place. These ROs protect the viral RNA from degradation by cellular RNases and detection by RNA sensors that trigger antiviral responses. Additionally, ROs can be used to concentrate viral/host factors for efficient replication. One of the most exciting advances has been the discovery of phosphatidylinositol 4-phosphate (PI4P) as an essential lipid component of ROs. In the cell, PI4P and other phosphoinositides serve many functions, but most importantly they function as docking sites for proteins that contain one or more phosphoinositide-binding domains. This observation inspired the hypothesis that a cellular lipid might be the platform used to recruit viral proteins to the sites of replication. In this dissertation, I showed that poliovirus proteins 3C, 3D, and 3CD proteins, which are intimately involved in the viral genome replication process, have the capacity to interact with phosphoinositides, including PI4P. Using computational and biochemical/biophysical approaches, I showed that poliovirus 3D (viral polymerase) domain interacts with PI4P and other phosphoinositides. The mutational analysis allowed me to identify a 3D derivative with diminished PI4P-binding activity. Because 3CD hijacks he cellular PI4P biogenesis pathway, further analysis of the 3D variant in the context of 3CD suggested that the 3D domain plays a role in a pre-Arf1 activation step. These findings are consistent with the hypothesis that 3CD needs to interact with PI4P-containing membranes in order to hijack the PI4P biogenesis pathway. Results from computational docking analysis suggested that the 3C domain of 3CD is also capable of interacting with PI4P. Empirical evidence for the docking studies was provided by NMR-HSQC experiments, which further confirmed that the binding site is near the N-terminus of 3C. Fluorescence polarization and supported lipid bilayer-based phosphoinositide-binding experiments both showed that 3C binds to phosphoinositides, with broad specificity. My work allowed us to further characterize the PI4P biogenesis pathway induced by poliovirus 3CD and show evidence that PI4P is used as a zip code for localization to ROs. Additionally, these findings increased the functionality assigned to each viral protein/domain, which is typical for viruses that have a limited coding capacity and evolved to encode multifunctional proteins to keep their genome size small. Finally, this work led to the development of a novel tool to assess phosphoinositide-binding in vitro, which will be instrumental in identifying many more viral phosphoinositide-binding proteins in the near future. Either used for subcellular localization and/or activation upon binding, determining the viral phosphoinositide-interactome will not only help us better understand the viral life cycle but also provide novel tractable targets for the design of antiviral agents.