Improvement of the Capripoxvirus, lumpy skin disease virus for use as a vaccine vector

Doctoral Thesis


Permanent link to this Item
Journal Title
Link to Journal
Journal ISSN
Volume Title
Lumpy skin disease (LSD) is a notifiable viral infection due both to its morbidity in cattle and its severe economic burden. The disease was confined to Sub-Saharan Africa but has in recent years spread to the Middle East and Europe. Vaccination is the only way of preventing LSD. Live attenuated lumpy skin disease virus (LSDV) has been used as a vaccine against LSD. The most successful LSD vaccine is the Neethling vaccine strain (nLSDV) from South Africa. There are however, reports of nLSDV being too attenuated or too virulent in different breeds of cattle. A South African produced vaccine strain of LSDV, Herbivac, was said to be more immunogenic than nLSDV (personal communication, Deltamune). Whole genomic sequence comparison of Herbivac with nLSDV revealed a single potentially significant change in open reading frame (ORF) 131. This ORF encodes a superoxide dismutase (SOD) homologue. The mutation identified in Herbivac is a 2bp deletion which causes a frameshift mutation that restores the SOD homologue to resemble the full-length SOD homolog encoded by the virulent field strain. This SOD homologue gene is truncated in nLSDV. Protein structural alignment of SOD homologues from LSDV and other characterised SOD homologues was done. Both the truncated and full-length SOD homologues lacked the catalytic arginine at position 142 which is involved in SOD activity. Some similarities with known SOD homologues which act as SOD decoys, such as the Leporipoxviruses myxoma and Shope Fibroma virus, were observed. Like Leporipoxvirus’ SOD homologues, the full-length Herbivac SOD homologue contained regions of homology with the copper chaperone for SOD (CCS). The putative SOD protein from Herbivac, and not nLSDV, contained 6 out of 8 metal binding residues. Unlike the SOD decoy from myxoma virus, the full-length SOD homolog from Herbivac did not include a cysteine molecule at position 56 that stabilizes the SOD-CCS heterodimer. The alignment suggests that all SOD homologs from LSDV are inactive as enzymes. Transcriptome analysis of messenger RNA from spleens of mice infected with nLSDV or Herbivac for 24hrs was carried out to determine the effect of Herbivac or nLSDV infection on host gene expression. Compared to the PBS control, nLSDV and Herbivac induced the differential expression of 98 genes in common, largely related to response to viral infection. nLSDV differed in the unique expression of 6 genes and Herbivac differed in the unique expression of 36 genes, including granzyme A and Poly (ADP-ribose) polymerase (Parp 9). Herbivac upregulated genes associated with pathogen pattern recognition, interferon response, immune response and cell death. More Gene ontology (GO) processes were enriched after Herbivac infection than nLSDV infection. Amongst these processes were immune response processes, the interferon and cell death related responses. To characterise the SOD homolog with respect to SOD activity, cell death and whether it plays a role in growth of viruses expressing it, recombinant viruses were constructed. The first set of recombinants included a SOD knock-out virus where the SOD gene from nLSDV was replaced with reporter gene GFP. A SOD knock-in virus was constructed with the SOD homologue gene altered to improve the stability of the gene. A reporter gene mCherry was also inserted. Histological examination of CAMs infected with nLSDVdSOD-M showed vacuolation and oedema to a greater degree than nLSDV, Herbivac and nLSDVSODis-M. Herbivac showed greater immune cell infiltration. SOD knock-in, nLSDVSODis-M showed increased epithelial hyperplasia and fibroplasia. Another set of recombinants without marker genes was made, nLSDVdSOD-UCT (SOD knock-out), and nLSDVSODis-UCT (SOD knock-in). Deleting the SOD homolog reduced virus yield by approximately ten-fold in MDBK cells. Histologically, the presence of the SOD homolog in nLSDVSODis-UCT caused greater inflammatory changes in the mesoderm after 5 days of infection compared to nLSDVdSOD-UCT. In vitro functional studies were done in MDBK cells which are permissible to LSDV infection. No difference in SOD activity could be detected amongst the different viruses. Differences could, however, be detected in induction and inhibition of apoptosis. There was increased induction of apoptosis following infection by all viruses containing full-length SOD compared to infection with truncated or SOD knock-out virus. Viruses expressing a full-length SOD showed greater inhibition of camptothecin induced apoptosis than nLSDV or the SOD knockout. Similarly, full-length SOD induced cell death by necrosis to a greater extent than the SOD knock-out or nLSDV; and all viruses inhibited camptothecin induced necrosis. These results suggest that the presence of a SOD homolog in a vaccine could be advantageous with respect to growth of the vaccine to higher titres and to improved immunogenicity of the vaccine. Further work is required to test this hypothesis in a bovine animal model.