Sequence and variant analysis of a pathogenic field strain of Equine Infectious Anemia Virus from Italian isolates

Introduction: Equine infectious anemia (EIA) is a persistent retroviral-based disease with a worldwide distribution significantly challenging the equine industry. In 2006 indeed, an outbreak of EIA occurred in Ireland and Italy, apparently as a result of iatrogenic transmission with a contaminated horse plasma infusion. This Irish and Italian episodes were characterized by cases of severe, sometimes fatal, disease [1,2]. Despite first clinical signs of EIA were reported as early as 1843, all the complete genomic sequences published to date are derived from just few viruses: the North American EIAV Wyoming, Chinese EIAV Liaoning, Japanese EIAV Miyazaki 2011-A and Irish EIAV IRE strains. Aim of study In the current study, we report for the first time the complete genomic sequences for Italian EIAV isolates. Unlike other authors, that previously sequenced EIAV genomes with Sanger methods, we used a strategy aimed to minimize PCR errors and chimeric clone amplification [3] and maximize low frequency mutations discovery through Next Generation Sequencing. This approach allowed for the identification of mutation accumulation in the EIAV genome that is known to correlate with the number of febrile peaks through time. Material and Methods DNA and RNA of 103 samples (blood or/and organs) derived from Italian outbreaks of EIA and EIAV seropositive asymptomatic Italian horses were screened for gag and pol viral genes. Positive samples were subjected to whole pro-viral genome amplification thorough a "Long Run PCR" strategy [3] resulting in about 8000 bp single amplicons. Two selected samples derived from symptomatic animals from the same outbreak but with different clinical history (one suddenly died after infection the other euthanatized after 5 months of several febrile peaks), were sequenced using the Roche 454 Flex platform. Consensus sequences were annotated and submitted to GenBank (KM247554, KM247555) Bioinformatics analyses were carried out with an “ad hoc” pipeline performing a variant calling to evaluate mutations with respect to an Irish reference sequence (JX480631). Variants called (61 for KM247555 and 66 for KM247554) were annotated with Annovar software: 37 of those are shared between the two genomes while 24 in KM247555 and 29 for KM247554 are genome specific. Conclusion In this study we report for the first time the complete genomic sequences for two Italian EIAV isolates; performing variant calling procedures, we evaluated the samples differences with a reference genome, moreover we had different time points of observation from the same outbreak allowing the identification of polymorphisms accumulation through time. Even if we should expect most of the variability in the RNA viral genome (which we have no data) we discovered significant mutations in the proviral one. The majority of them produce non-synonymous variants, especially in KM247554 (66 SNP) highlighting that the sample derived from the animal that showed most febrile peaks, has an amount of mutations that heavily affect the protein sequence. These variants are under deeper bio-informatics characterization in order to predict the effect of the mutations on virus proteins.