Modeling the Expression Patterns of IFN-γ-Responsive Genes to Differentiate Pigs Shedding High and Low Levels of Salmonella

Investigating the porcine transcriptional response to Salmonella is important for understanding disease resistance and developing predictive tools for disease outcome. Our whole blood transcriptome analysis revealed many differentially expressed (DE) genes between pigs classified as persistent (PS) or low (LS) Salmonella shedders.  Pathway analysis of DE genes showed that IFN-γ-responsive genes represented the largest response network in PS pigs.  Here, we used qPCR to confirm in vivo patterns of 15 DE genes annotated as IFN-γ targets.  Further investigations verified that differential patterns observed in vivo in PS and LS pigs could be recapitulated in vitro. Whole blood from 3 healthy pigs was stimulated in vitro with various doses of IFN-γ with or without S. Typhimurium endotoxin (STE). The qPCR analysis revealed that the in vitro response to IFN-γ alone for five genes (CXCL10, IL10, MMP8, PSMB9, TMEM176) was dose-dependent.  Simultaneous stimulation of whole blood with STE and IFN-γ induced IFN-γ dose-dependent expression for these genes plus CASP4, CYBA, IRF1, JAK2, NCF1, SOD2, and TAP1. Cluster analysis revealed that in vivo gene expression patterns across all genes in PS swine clustered most closely with patterns in whole blood stimulated with the two highest IFN-γ levels plus STE. We conclude that quantitative differences in IFN-γ levels explain the expression of most tested genes, and that the IFN-γ regulon is a source of genes whose expression levels two days post-infection can predict shedding outcomes. Such genes can now be further evaluated as candidates for development of predictive assays for shedding outcome in swine.