The status of PRRSV in North America and the influence of vaccines

Authors: Faaberg, Kay; Anderson, Tavis; INDERSKI, BLAKE - Oak Ridge Institute For Science And Education (ORISE)

Submitted to: Meeting Abstract
Publication Type: Abstract Only
Publication Acceptance Date: 8/30/2019
Publication Date: N/A
Citation: N/A

Interpretive Summary:

Technical Abstract: The US Swine Pathogen Database (US-SPD) initiative, funded by the USDA-ARS-NADC-VPRU and National Pork Board of the US, has established a new and robust database of genetic sequences for significant swine pathogens. The US-SPD (https://swinepathogendb.org), is an open access web-based tool developed using the new USDA-ARS scientific high-perfomance computing cluster and network (SCINet) to assemble genetic sequence data, and link it with detection date and location, for the major pathogens of swine detected by the major US veterinary diagnostic laboratories. The laboratories include South Dakota Animal Disease Research & Diagnostic Laboratory (SDSU), Iowa State Veterinary Diagnostic Laboratory (ISU), Kansas State Veterinary Diagnostic Laboratory (KSU), and soon the Minnesota Veterinary Diagnostic Laboratory (MVDL). The database also contains over 12,000 PRRSV sequences collected primarily by SDSU and MVDL from 1989-2007. The US-SPD presently houses porcine reproductive and respiratory syndrome virus (PRRSV), Seneca virus A, and swine coronaviruses, and will eventually include circoviruses and others, except for swine influenza A virus which is housed at the Influenza Research Database. Additional viruses, including foreign animal diseases (i.e., African swine fever virus) and major bacterial coinfecting pathogens are planned. The US-SPD has incorporated all available GenBank files that have been fully annotated, along with convenient tools to retrieve, display, and select sequences for further analyses. At present, the database includes almost 31,000 PRRSV sequences, 5000 PEDV, and 224 Seneca virus A sequences. Using the US-SPD database all PRRSV nonstructural protein 2 (nsp2) and full-genome sequences detected around the world, including those detected by SDSU and ISU from 2014-2017, were assembled. In addition, all recent ORF5 sequences from SDSU, ISU and KSU were combined with a randomly selected set of 2000 GenBank ORF5 sequences and a 600-sequence reference gene set developed in 2010 describing PRRSV ORF5 diversity. We excluded high passage viruses, except for available vaccines, and synthetic sequences. Nucleotide alignments were completed for each set of sequences, which numbered 775 genomes, 808 nsp2 and 4918 ORF5 regions. We then inferred the best-known maximum likelihood phylogenetic tree for each alignment. Two statistically supported genetic clades were detected for the genome and nsp2 trees. The first clade consisted mostly of older sequences, including all available modified live vaccines. The second clade represented the majority of recent isolates collected in the US. The ORF5 gene tree, made up of more than 6-fold more nucleotide sequences than that of the genome and nsp2 phylogenies, had three statistically supported genetic clades. For the major genetic clades, we implemented a Bayesian analysis to quantify the variation in evolutionary rate and used these data to identify genetic sites that are undergoing diversifying selection. Our analyses demonstrate the dynamic landscape of PRRSV evolution with the cocirculation of multiple genetic clades, how the emergence of novel genetic types is associated with different evolutionary trajectories, and how data collated from all available data in the Western Hemisphere can be used to assess the potential efficacy of vaccine strains. These data demonstrate that PRRSV diversity is likely to limit the efficacy of current vaccines and the cocirculation of different genetic clades further generates heterogeneity among viruses that may emerge and spread in naïve host populations.