Manipulation of Inflammatory Processes by a Poxvirus-Encoded Decoy Receptor

Poxviruses are large double-stranded DNA viruses, which encode an arsenal of secreted proteins that target crucial pro-inflammatory features of the host immune response and contribute significantly to virulence. The focus of my thesis work has been dedicated to the study of the cytokine decoy CrmD of ectromelia virus. CrmD has been shown to be integral to the control of inflammatory immune responses in the context of viral infection, and is able to bind and sequester two classes of ligands: TNF and chemokines. Herein, we identify the structural basis of TNF recognition by CrmD by determining the structure of CrmD alone and in complex with murine TNFα. Both structures show that the CrmD N-terminal domain has three Cysteine Rich Domains (CRDs) characteristic of all TNF receptors (TNFRs). The complex of CrmD with mTNFα shows that CrmD is able to engage the ligand in a similar fashion as human TNFR1 and TNFR2, with three CrmD molecules bound around one TNFα trimer, each at the monomer-monomer interface. Again, like the host TNFRs, the second and third CRDs in the N-terminal region are the only domains that participate in TNFα engagement. The C-terminal domain of CrmD is a β-sheet sandwich following the last CRD. The C-terminal β-sheet sandwich of CrmD swings out from the N-terminal domain and appears to have no role in TNF binding. In addition, comparison of the unbound CrmD structure with the TNF-bound CrmD reveals that the decoy receptor possesses a flexible hinge found in the third CRD allowing the C-terminal domain to undergo a rotation of 108°. We hypothesize that this molecular hinge may allow CrmD the plasticity to accommodate a wider range of ligands. Additionally, we have identified an entirely new target for poxviral-encoded decoy receptors, the human and mouse Fc-gamma receptor (FcR) family. CrmD engages functionally diverse FcRs with low to mid-nanomolar affinity allowing CrmD to bind to the surface of cells. To gain further insight into the structural features involved in the CrmD-FcR interaction, we determined the crystal structure of the CrmD C-terminal domain in complex with the human Fc-gamma receptor CD16b at 2.5 Å resolution. The structure reveals that CrmD binds the interdomain hinge of the FcR, the same interface involved in antibody Fc-region binding. Given that the N-terminal region of CrmD binds to TNFα and the C-terminal domain has now been implicated in both chemokine and FcR-binding, our findings describe the remarkable ability of CrmD to manipulate the effector functions of three unique host proteins. Structure-based mutagenesis enabled us to create a CrmD mutant that drastically decreases FcR-binding. This both corroborates our structure and provides a useful tool to dissect the functional implications of FcR engagement. Utilizing this mutant alongside our WT CrmD, we are able to show that the CrmD-FcR interaction serves to both block antibody binding, and to make CrmD a more efficient TNFα inhibitor, presumably by exploiting FcR-mediated internalization. We believe that CrmD may utilize the intramolecular flexibility shown to exist between its N- and C-terminal domains to engage multiple host proteins. This molecular flexibility in conjunction with high-affinity FcR interactions may enable CrmD to commandeer the endocytic pathway for the purposes of TNF and chemokine degradation, enabling ectromelia virus to potently diminish the inflammatory immune response. Further, we show that this new function may extend to related members of a larger family of poxvirus-encoded decoys. The research presented here opens up a new branch of study in poxvirus immune evasion and illustrates the therapeutic potential of this family of viral proteins