Abstract
The Poxvirus family is a group of large DNA viruses whose host range encompasses both vertebrate and invertebrate hosts. Their large genomes allow them to encode genes for replication and assembly as well as numerous accessory genes that aid in virulence and immune evasion. The Parapoxviruses comprise a genus of poxviruses that cause highly contagious, pustular skin lesions in ruminant animals and zoonoses in human. The prototypical parapoxvirus Orf virus (ORFV) encodes immunomodulatory genes that aid viral replication and control inflammation in the skin. The parapoxviruses are one of three genera of poxviruses that encode a chemokine binding protein (CBP) that sequester host chemokines preventing chemokine signalling and the recruitment of leukocytes to the site of viral replication. The key feature of CBPs is that they lack sequence and structural homology to any host chemokine receptor. The ORFV CBP has previously been shown to bind the CC- and XC-chemokine classes, in contrast to the CBPs from the other poxvirus genera that specifically target CC-chemokines. The CBP gene has subsequently been identified and cloned from three other species of parapoxvirus; Bovine papular stomatitis virus (BPSV), Pseudocowpox virus (PCPV) and Parapoxvirus of red deer in New Zealand (PVNZ) which encodes three unique CBPs. Interestingly, the parapoxvirus CBPs have a diverse amino acid sequence, sharing only 20 – 45 % identity. This study set out to further characterise the interactions between the ORFV CBP and its chemokine ligands and determine if the parapoxvirus CBPs are functional chemokine inhibitors. An indirect chemokine capture ELISA was used to determine the chemokine ligands for the parapoxvirus CBPs. Consistent with previous findings the ORFV CBP bound the CC- and XC-chemokines, and was also shown to bind the CXC-chemokines CXCL2 and CXCL4. The soluble version of the glycosaminoglycan (GAG) heparin did not inhibit the interaction of the ORFV CBP with the chemokine CCL2 indicating that the ORFV CBP was not binding the GAG binding domain of chemokines. However, in an immobilised heparin-BSA assay ORFV CBP was demonstrated to be removed from solution by heparin that together suggest chemokine and GAG binding by the ORFV CBP is similar to the M-T1 CBP from the Myxoma virus. Chemokine competition and displacement ELISAs were used to determine whether the ORFV CBP has one or multiple binding sites for the various classes of chemokines. The ORFV CBP preferentially bound the chemokine XCL1 over CCL2 and CXCL2, but its relative affinity of binding is greater for CXCL2 and CCL2 than XCL1. Mutational analysis of the ORFV CBP to determine regions of the ORFV CBP involved in ligand binding indicates the predicted Loop7 region is involved in high affinity ligand binding and the C-terminus contributes to the correct formation of the ORFV CBP. Together these findings support the working theoretical structural model of the ORFV CBP that predicts one region of the CBP to be involved in binding, but contains distinct regions involved in specificity and affinity. The BPSV and PVNZ gene 112.0 CBPs were shown to have an ORFV CBP-like binding spectrum inhibiting the same chemokines as the ORFV CBP with similar affinities. The PVNZ gene 112.3 CBP had a novel CXC-chemokine specific binding spectrum, binding an extended range of CXC-chemokines; CXCL2, CXCL4, CXCL8 with high affinity and CXCL10 with low affinity, and did not bind any CC- or XC-chemokines with high affinity. The PVNZ gene 112.6 CBP was only able to bind the chemokines CCL5 and CCL19 with high affinity and the PCPV CBP only bound CCL19 with high affinity. The parapoxvirus CBPs that bound the CC-chemokines had a higher affinity for the CC-chemokines than the Vaccinia virus (VACV) 35 kDa CBP. In summary this study has shown that ORFV CBP is able to bind a wider range of chemokine classes than previously reported and is also able to interact with the GAG heparin. Structural and preliminary mutational analyses indicate a single binding region of the ORFV CBP for chemokine binding and a separate region for heparin binding. These results also confirm that the parapoxviruses all encode functional CBPs despite the large variation in sequence.