Animal Genomics Webinar

Professor Chris Tuggle and his group at Iowa State University in collaboration with Fios Genomics (UK), the Roslin Institute (UK), Affymetrix, and others, have identified immune response gene expression classes in pigs that have the potential to identify, prior to infection, whether animals are likely to become carriers of Salmonella. In United States swine herds, 53% of pigs are infected with Salmonella. Early identification of susceptible animals may reduce infection in herds and minimize risks to human food safety.

Depending on the severity of infection, pigs are characterized as either persistent fecal shedders (PS) or low fecal shedders (LS) of the bacterium. In collaboration with Affymetrix, Dr. Tuggle's team developed GeneChip Porcine Gene 1.0 ST Array–a second-generation array with gene sequences covering the entire porcine genome. The team used this array for whole-genome transcription analysis of these two types of pigs, aiming to identify gene classifiers that could be used to discriminate between PS and LS pigs prior to infection.

The team conducted gene expression studies on blood from PS and LS pigs prior to infection (Day 0). In endotoxin-stimulated samples in vitro, they identified a huge immune response in blood from PS pigs compared to blood from LS pigs. To further understand the immune response pathways, the team compared expression at Day 0 to expression two days post-infection (Day 2). The expression patterns in Day 2 samples correspondeded to the higher immune response in samples from PS animals at Day 0, suggesting that these upregulated genes might be used to predict susceptibility prior to infection.

To develop a predictive gene expression model, Dr. Tuggle's team analyzed in vitro endotoxin-stimulated blood samples from pigs of known shedder status, prior to infection. Using these known samples, they built a classifier algorithm to cluster sets of immune response genes that were differentially expressed in vitro in each sample type. They then applied the classifier to in vivo and in vitro samples at Day 0. The classifier identified 35 highly discriminatory genes in the in vivo samples, and 103 highly discriminatory genes in endotoxin-stimulated in vitrosamples. To minimize the complexity of further validation, they reduced the classifier set to 12 genes, with 6 clusters of 2 genes each from in vivo and in vitro endodoxin-stimulated samples.

The team continues to validate these gene clusters to assess the predictive value of the model against other challenges and to gain additional insights into the immune response pathways involved in bacterial infection. In addition to minimizing risks to pig health and human food safety, a model such as this might be used to identify superior animals for genetic selection and to inform our understanding of human immune response to Salmonella infection.