Characteristics Of Equine Infectious Anemia Virus Integration In The Horse Genome

Equine infectious anemia virus(EIAV), which is closely related to human immunodeficiency virus 1(HIV-1), is a member of Lentivirus Genus in the Retroviridae Family. EIAV is an important etiological agent to the equine industry and EIAV infection can also be an animal model for HIV-1/AIDS researches. Last century, the Harbin Vet Res Institute of Chinese Academy of Agricultural Sciences had successfully developed an attenuated EIAV vaccine. This EIAV vaccine has been proven being able to induce immune protection to pathogenic strains, and is the only lentiviral vaccine that has been successfully applied in large scale. The research on biological characteristics for the EIAV strains with different virulence derived from different stages of the attenuation process of parental strain, particularly the attenuated EIAV vaccine strain, should provide valuable information for future studies on lentiviral vaccines. Previous study showed that after immune suppression with dexamethasone, the vaccine strain in vaccinated horses remained consistently low-level replication, suggesting the attenuated virus itself, rather than the host immune system, is majorly responsible for the low-level replication of EIAV attenuated vaccine strain in vivo. It is known that integration of retroviral c DNA into the host cell chromosome is an essential step in its replication. Previous studies have showed that the site of proviral DNA insertion is critical for the virus, as it can influence the rate of viral transcription and replication. How the integration site selection characteristics of EIAV, in particular the EIAV attenuated vaccine strain, in the horse genome? Are integration site characteristics in the horse genome associated with the low-level replication of EIAV attenuated vaccine strain? Answers to these questions will contribute to the further investigation of the mechanisms of the immune protection provided by EIAV attenuated vaccine strain.In this study, we selected three key viral strains of the attenuating process of EIAV pathogenic strain, including the donkey leukocyte virulent strain(EIAVDLV34), the donkey leukocyte-attenuated vaccine strain(EIAVDLV121) and the fetal donkey dermal cell-attenuated vaccine strain(EIAVFDDV13) for investigation of their integration characteristics in the horse genome. In addition, the integration characteristics of EIAVDLV34 and EIAVDLV121 were compared and analyzed for the possible contribution to the attenuated pathogenicity of the vaccine strain. To isolate host genomic DNA fragments containing integration sites, genomic DNA was isolated from equine monocyte-derived macrophages(e MEMs) or fetal equine dermal cells(FEDs) infected with either of the EIAV strains, cleaved with restriction enzymes(Ssp I or Dra I), then ligated to DNA linkers. Integration sites were amplified by linker-mediated PCR(LM-PCR), and then amplification products were cloned and sequenced. Based on the criteria for verifying the authenticity of integration site sequences, we obtained 1,103 integration sites, including 101 and 525 integration sites for EIAVDLV34-or EIAVDLV121-infected e MEMs, respectively, and 477 integration sites for EIAVFDDV13-infected FEDs. Sequences rounding the integration sites were analyzed for ① chromosomal locations, ② the frequency within the genes, ③ the orientation between provirus and the transcriptional direction of host cellular genes where EIAV provirus was integrated, ④ distance from the transcription start site, ⑤ repetitive sequences, and ⑥ function of genes in which EIAV provirus were integrated.Our analyses indicate that the integration site selection of the three EIAV trains in the horse genome is not random. Furthermore, integration site characteristics for these strains have some similarities and differences. The similarities are, ① EIAV integration has a preference for regions with AT-rich regions or with 21~40 genes/2Mb; ② a highly significant preference for integration within transcription units, in particular introns(P<0.01); ③ a weak palindromic sequence was seen around EIAV integration sites; ④ EIAV provirus integration disfavor transcription start site; ⑤ in integrations in repetitive elements, strongly favor regions of long interspersed elements(LINEs)(P<0.01), rather than short interspersed elements(SINEs), which were strongly favored by HIV-1 for integrating in the human genome. In addition, the gene ontology(GO) of equine genes hosting EIAV integration sites was analyzed. Gene clusters hosting integration sites of all these three viral strains include those participate in regulation of transcription, ion binding and ATP binding.Some features of integration appeared in only one or two of the three EIAV strains were also identified. These are: ① except LINE1, some other types of repetitive elements remarkably favor EIAV integration, which include DNA transposon favor EIAVDLV121 and EIAVFDDV13(P<0.05); ② the regions with low-gene density favor the integration of EIAVDLV121 and EIAVFDDV13; ③ GO analysis suggest that some other biological gene clusters also been frequently inserted by EIAVDLV121 or EIAVFDDV13. These genes include those responding to DNA damage stimulus and involving in adaptive immune response, leukocyte differentiation, apoptosis, lipid biosynthetic process and oxidation. In addition, combined with published studies on EIAV vector integration in the human genome, our results indicate that, ① EIAV integration is independent of the way of virus entrance(either binding authoritative EIAV receptor by the EIAV envelope protein or via the VSV-G envelope of the pseudotyped EIAV vector); ② EIAV have a significant preference for integration within transcription units in both horse and human chromosome.As shown in this study, EIAV proviruses have a significant preference for integrating in LINEs and DNA transposon, which can impact the transcription and expression of nearby genes. Thus, the LINEs hosting EIAV provirus may influence the replication of integrated provirus, and may be associated with attenuated virulence of the EIAV vaccine strain. In addition, some genes hosting EIAV attenuated vaccine strain provirus participate in immune responses, which may be associated with the induction of protective immunity provided by the EIAV vaccine strain. In conclusion, experimental data obtained in this study contribute to better understanding of the biological characteristics of the Chinese attenuated EIAV vaccine. Meanwhile, the study on EIAV integration features in the horse genome will extend the knowledge of interaction between lentiviruses and hosts, as well as lentiviral vectors for Gene Ther.