190

Porcine endogenous retrovirus (PERV) Integration in the pig genome

Linda Scobie¹, Rashmi Wali¹, Chris Bauser², Claire Rogel-Gaillard³, Jonas Blomberg⁴, Göran Sperber⁵, Yasu Takeuchi⁶

¹Department of Biological and Biomedical Sciences, Glasgow Caledonian University, Glasgow, Glasgow, United Kingdom; ²Biotech AG, GATC, Konstanz, Germany; ³UMR de Génétique Animale et Biologie Intégrative, INRA, Jouy-en-Josas, France; ⁴Department of Medical Sciences, Uppsala University Academic Hospital, Uppsala, Sweden; ⁵Department of Neuroscience, Uppsala University, Uppsala, Sweden; ⁶Infection and Immunity, UCL, London, United Kingdom

Introduction: Porcine endogenous retroviruses (PERV) inherited as proviruses in the pig genome can infect human cells, posing a potential risk of zoonosis in pig-to-human xenotransplantation. Introduction of PERV into the pig genome is relatively recent and integration sites are highly polymorphic. It is predicted that only a limited number of PERV loci are active and can direct production of infectious viruses. Identification of such active PERV loci and determination of their distribution in the pig population would help identify, breed or genetically engineer pigs free from problematic PERV.

Methods: We have investigated PERV integration sites via a search for PERV integration in the swine whole genome sequence (SWGS) Build 10 assembly (Archibald et al. BMC Genomics. 2010 ;11:438). by the RetroTector program (Sperber et al. NAR 2007;35:4064)as wellas manual BLAT/BLAST analyses.

Results: About 20 loci were identified to harbour PERV proviruses of more than 7kb length in the Duroc sow genome. The fact that only 4 out of these 20 loci have been found to be shared by a Large White BAC library (Rogel-Gaillard et al. Cytogenet Cell Genet. 1999;85:205-11)highlights the highly polymorphic nature of PERV integration. None were found to have complete and/or intact open reading frames for PERV genes. Validation of each of these loci has been carried out and their distribution analysed in a number of different breeds.

Conclusion: Distribution of these integrations is clearly different in both individual pigs and between breeds. In addition, PERV sequence information in SWGS has its limitations in revealing the PERV integration pattern in pigs as a population, but supports previous estimations that majority of PERV proviruses are defective.