Abstract
The roundworm Trichostrongylus colubriformis is one of the most important parasites in sheep, impairing feed conversion and reducing growth. However, the molecular mechanism of host resistance to this important species remains elusive. In this study, we compared gene interaction networks manifested in the jejunum transcriptome between sheep bred for parasite resistance (RES) and susceptibility (SUS) in response to a T. colubriformis challenge infection. Selections for host resistance compromised parasite establishment and resulted in a 98.8% reduction in worm burden (P < 0.00001; N = 20 per group). 319 genes displayed a significant difference in transcript abundance between RES and SUS phenotypes. The gene encoding lectin, galactoside-binding, soluble, 15 (LGALS15) had significantly higher expression in RES than SUS lambs. Enriched pathways included antigen processing and presentation and Fc gamma R-mediated phagocytosis. Two algorithms, weighted gene co-expression network analysis (WGCNA) and differential gene correlation analysis (DGCA), were applied to infer gene interaction networks. Both algorithms demonstrate that genes in SUS lambs appear to be more closely correlated than in RES lambs. WGCNA identified a module that was positively correlated with worm counts only in SUS animals (R = 0.67; P = 0.001). DGCA detected approximately four times more unique gene correlation pairs in SUS than in RES lambs. Gene Ontology analysis among the genes with gain-of-correlations shows marked differences in cell division, such as mitotic cytokinesis, sister chromatid segregation, and kinetochore, between the two phenotypes. Correlated genes unique to RES can be used for the development of alternative biomarkers to facilitate breeding. Moreover, dietary approaches to disrupt close gene interactions within key modules may have the potential to reduce worm burden. It is conceivable that feeds, phytochemicals or feed additives that promote specific gene interactions can be used for sustainable parasite controls.