Zaire Ebola VP35 Protein

Kye Duren '16 and Rhea Le '16


I. Introduction

       The Zaire ebola virus currently designated as the type strain of Ebolavirus(EBOV),has been indicated as the cause of Ebola hemorrhagic fever outbreaks in humans with a fatality rate of 90%. The Ebola virus is a part of the Filoviridae virus family that have single-stranded negative-sense RNA. The viral genome encodes eight different proteins. More specifically, the viral protein (VP35) could play a major role in the Ebola virus pathogenesis due to its role as an antagonist of type I interferons (IFN). Type I IFNs are produced by the innate immune system as a response to viral infection. These interferons can interfere with viral replication by binding to specific receptors on uninfected host cells, inducing the intracellular production of two different classes of proteins so double-stranded RNA activated endoribonucleases that have the ability to cleave viral RNA. VP35 acts as a type I IFN antagonist by binding to dsRNA that activates these endoribonucleases. Therefore, by being a type 1 IFN antagonist, VP35 aids in the evasion of the host immune system and contributes to the pathogenicity of the Ebola virus.

II. General Structure

        VP35 interacts with dsRNA. Here, we show only one RNA strand from the duplex RNA complex included in the crystal structure. There are four VP35 polypeptides that directly interact to dsRNA. This structure exhibits two-fold noncrystallographic symmetry, where molecule A is equivalent to C and B is equivalent to D.Molecule A interacts with molecule B in a head to tail orientation where residues on the C terminus of Molecule A directly interact with residues on the N terminus of Molecule B . Molecule A is involved in protein-protein interactions while molecule B directly binds to dsRNA . VP35 consists of the N-terminal coiled-coil domain and the C-terminal interferon inhibitory domain (IID). The coiled coil domain is required for viral replication and nucleocapsid formation. The C-terminal IID contains a double-stranded RNA binding domain that is important for IFN inhibition. Two crucial regions within the IID are the central basic patch and the end-cap domain. The central basic patch is involved in protein to protein interactions and the end-cap region is used by VP35 to directly bind to dsRNA. We will be discussing specific residue interactions within these to regions so it will be helpful to know the nomenclature of these atoms. More specifically, these atom nomenclatures will be used throughout this page:
        Furthermore, VP35 proteins that contain F239A, R312A,R322A, K339A and mutations are unable to bind to dsRNA and are also unable to suppress IFN-B promoter activation, compared to WT VP35 . The F239A mutation is located in the central basic patch and R321A, R322A, and K339A mutations are located in the end-cap.  Additionally, this complex is stabilized by chloride and magnesium ions.

III. Central Basic Patch


        VP IID contains a central basic patch that are involved in binding to other proteins and nucleic acids. Protein-protein interactions between molecules A and B occur when Arg 312 and Arg322 on molecule A interacts with various residues from molecule B. Arg 312 of molecule A makes hydrogen bonds with residues Gly 270 O, Asp271 OD1, and Glu269 O of molecule B . Additionally, Arg322 from molecule A forms hydrogen bonds with Glu262 OE1 and OE2 from molecule B . Residues Arg 312, Arg322, and Lys339 of molecule A are believed to be required for RNA binding while residues Arg305, Lys309, and Lys319 enhance RNA binding . Arg305 and Lys319 mutations to alanine results in a 3 to 5 fold reduction in the binding affinity compared to the wildtype protein. Whereas, mutants exhibiting R312A, 322A, or K339A mutations are completely unable to bind to dsRNA. These results make sense regarding molecule B since Arg 312 and Arg322 from molecule B are involved in dsRNA binding with the phosphodiester backbone . However,  the Lys339 residue on molecule B do not participate in any dsRNA interactions . When the wild type VP35 IID was superimposed with VP35 IID containing either R312A or R339A, there exists and <0.5 change in the overall alpha-helix and beta-sheet structures. However, there is no significant difference in the backbone chain conformation between the mutants and the wild type.  

IV. End-cap Region

        The IID of VP35 also contains an end-cap region with hydrophobic residues. These residues directly form bonds with dsRNA but are not involved in protein-protein interactions.In vivo, (EBOV) VP35 protein can form end-cap interactions with the blunt ends of dsRNA, however with ssRNA we are limited to illustrating models of the actual interactions  with residues that are both similiar and proximal to the real target proteins. Hydrogen bonds are known to be between Gln274 NE2 and C1 O4' and between Ile340 OXT and C1 N4. However, due to the absence of the complementary RNA strand that contains the C1 involved in these interactions, we will be showing the hydrogen bonds between
Gln274 NE2 and C7 O4' and between Ile340 OXT and C7 N4 since C7 on the RNA strand included in the module is the most similar residue to the C1 on the complementary strand. Addtionally, there exists electrostatic interactions and van der Waals forces between the end cap and the nucleic acid. The end-cap is directly involved in binding to dsRNA since the F239A mutation results in the complete loss of dsRNA binding while the F235A mutation did not. The electrostatic interactions are between: Lys282 N and G8 O2P and Arg322 Ne and C7 O2P. 
The van der Waals contacts consist of: 
  • Phe239 CZ and CE2 to C6
  • Gln274 CG to C6 O4'
  • Ile278 CD1 to G8 N1  and C6
  • Gln279 CD to G8 O3'
  • Gln279 OE1 to G8 C2', C3', and O3'
  • Lys282 CD to G8 OP2 and O5'
  • Lys282 CE to G8