Molecular basis for the host range function of the poxvirus PKR inhibitor E3.

The antiviral protein kinase R (PKR) is activated by viral double-stranded RNA and phosphorylates translation initiation factor eIF2α, thereby inhibiting translation and virus replication. Most poxviruses contain two PKR inhibitors, called E3 and K3 in vaccinia virus (VACV), which are determinants of viral host range. The prevailing model for E3 function is that it inhibits PKR through the non-specific sequestration of double-stranded (ds) RNA. Our data revealed that Syrian hamster PKR was resistant to E3, which is at odds with the sequestration model. However, Syrian hamster PKR was still sensitive to K3 inhibition. In contrast, Armenian hamster PKR showed opposite sensitivities, being sensitive to E3 and resistant to K3 inhibition. Mutational analyses of hamster PKRs showed that sensitivity to E3 inhibition was largely determined by the region linking the dsRNA-binding domains and the kinase domain of PKR, whereas two amino acid residues in the kinase domain (helix αG) determined sensitivity to K3. Expression of PKRs in congenic cells showed that Syrian hamster PKR containing the two Armenian hamster PKR residues in helix-αG was resistant to wild type VACV infection, and that cells expressing either hamster PKR recapitulated the phenotypes observed in species-derived cell lines. The observed resistance of Syrian hamster PKR to E3 explains its host range function and challenges the paradigm that dsRNA-binding PKR inhibitors mainly act by the sequestration of dsRNA. Significance: The molecular mechanisms that govern the host range of viruses are incompletely understood. A small number of poxvirus genes have been identified that influence the host range of poxviruses. We show that the host range functions of E3 and K3, two host range factors from vaccinia virus, are a result of species-specific interactions with the antiviral protein kinase R (PKR) and that PKR from closely related species displayed dramatic differences in their sensitivities to these viral inhibitors. While there is a substantial body of work demonstrating host-specific interactions with K3, the current model for E3-mediated PKR inhibition is that E3 non-specifically sequesters dsRNA to prevent PKR activation. This model does not predict species-specific sensitivity to E3; therefore, our data suggest that the current model is incomplete, and that dsRNA sequestration is not the primary mechanism for E3 activity.