Classical approaches to African swine fever virus (ASFV) vaccine development have not been successful; inactivated virus does not provide protection and use of live attenuated viruses generated by passage in tissue culture had a poor safety profile. Current African swine fever (ASF) vaccine research focuses on the development of modified live viruses by targeted gene deletion or subunit vaccines. The latter approach would be differentiation of vaccinated from infected animals (DIVA)-compliant, but information on which viral proteins to include in a subunit vaccine is lacking. Our previous work used DNA-prime/vaccinia-virus boost to screen 40 ASFV genes for immunogenicity, however this immunization regime did not protect animals after challenge. Here we describe the induction of both antigen and ASFV-specific antibody and cellular immune responses by different viral-vectored pools of antigens selected based on their immunogenicity in pigs. Immunization with one of these pools, comprising eight viral-vectored ASFV genes, protected 100% of pigs from fatal disease after challenge with a normally lethal dose of virulent ASFV. This data provide the basis for the further development of a subunit vaccine against this devastating disease.

Live attenuated vaccines are considered to be the fastest route to the development of a safe and efficacious African swine fever (ASF) vaccine. Infection with the naturally attenuated OURT88/3 strain induces protection against challenge with virulent isolates from the same or closely related genotypes. However, adverse clinical signs following immunisation have been observed. Here, we attempted to increase the OURT88/3 safety profile by deleting I329L, a gene previously shown to inhibit the host innate immune response. The resulting virus, OURT88/3ΔI329L, was tested in vitro to evaluate the replication and expression of type I interferon (IFN) and in vivo by immunisation and lethal challenge experiments in pigs. No differences were observed regarding replication; however, increased amounts of both IFN-β and IFN-α were observed in macrophages infected with the deletion mutant virus. Unexpectedly, the deletion of I329L markedly reduced protection against challenge with the virulent OURT88/1 isolate. This was associated with a decrease in both antibody levels against VP72 and the number of IFN-γ-producing cells in the blood of non-protected animals. Furthermore, a significant increase in IL-10 levels in serum was observed in pigs immunised with OURT88/3ΔI329L following challenge. Interestingly, the deletion of the I329L gene failed to attenuate the virulent Georgia/2007 isolate.

African swine fever virus is currently present in all of the world’s continents apart from Antarctica, and efforts to control the disease are hampered by the lack of a commercially available vaccine. The Babraham large white pig is a highly inbred line that could represent a powerful tool to improve our understanding of the protective immune responses to this complex pathogen; however, previous studies indicated differential vaccine responses after the African swine fever virus challenge of inbred minipigs with different swine leukocyte antigen haplotypes. Lymphocyte numbers and African swine fever virus-specific antibody and T-cell responses were measured in inbred and outbred animals after inoculation with a low virulent African swine fever virus isolate and subsequent challenge with a related virulent virus. Surprisingly, diminished immune responses were observed in the Babraham pigs when compared to the outbred animals, and the inbred pigs were not protected after challenge. Recovery of Babraham pigs after challenge weakly correlated with antibody responses, whereas protective responses in outbred animals more closely correlated with the T-cell response. The Babraham pig may, therefore, represent a useful model for studying the role of antibodies in protection against the African swine fever virus.

PCR analysis of the genomes of 18 different African swine fever virus (ASFV) isolates showed that the I14L open reading frame (ORF) was present as either a long form or short form in all of the isolates. Sequencing of the ORF from eight isolates confirmed that both forms of the ORF were well conserved. Antisera raised against the I14L protein identified the long form of the protein as a 21 kDa protein expressed late during ASFV infection. Immunofluorescent analysis of transiently expressed haemagglutinin-tagged forms of the I14L protein showed that the long form of the protein localized predominantly to the nucleus and within the nucleoli. In contrast, although the short form of the protein was also present predominantly in the nucleus, it did not localize to the nucleoli. Deletion of the N-terminal 14 amino acids from the long form of the I14L protein, which includes a high proportion of basic Arg/Lys residues, abolished the specific nucleolar localization of the protein, although the protein was still present in the nucleus. Addition of this 14 amino acid sequence to beta-galactosidase or replacement of the N-terminal 14 amino acids of the I14L short form with those from the long form directed both of these modified proteins to the nucleolus. This indicates that this 14 amino acid sequence contains all the signals required for nucleolar localization.

We have compared the complete genome sequences of the vaccine strain of rinderpest virus and the virulent strain from which it was derived. Only 87 bases differed between the two genomes (0.55%). Possibly significant differences in amino acid sequence were found in the N, P, F, H and L proteins. A number of differences were also found in the leader region (3′ end of the genome), whilst the trailer region appears to be more conserved. In addition, the length of the genome was found in both cases to be 15882, an exact multiple of six, fulfilling predictions made earlier based on work with Sendai and measles viruses.

African swine fever virus causes a frequently fatal disease of domestic pigs and wild boar that has a high economic impact across 3 continents. The large double-stranded DNA genome codes for approximately 160 proteins. Many of these have unknown functions and this hinders our understanding of the virus and host interactions. The purpose of the study was to evaluate the role of two virus proteins, K145R and DP148R, in virus replication in macrophages and virulence in pigs. To do this, the DP148R gene, alone or in combination with the K145R gene, was deleted from the virulent genotype II Georgia 2007/1 isolate. Neither of these deletions reduced the ability of the viruses to replicate in porcine macrophages compared to the parental wild-type virus. Pigs infected with GeorgiaΔDP148R developed clinical and post-mortem signs and high viremia, typical of acute African swine fever, and were culled on day 6 post-infection. The additional deletion of the K145R gene delayed the onset of clinical signs and viremia in pigs by 3 days, but pigs showed signs of acute African swine fever and were culled on days 10 or 13 post-infection. The results show that the deletion of DP148R did not attenuate the genotype II Georgia 2007/1 isolate, contrary to the results obtained with the genotype I Benin97/1 isolate. Additional deletion of the K145R gene delayed clinical signs, but infected pigs reached the humane endpoint. The deletion of additional genes would be required to attenuate the virus.