Genome-Wide Association Study Reveals Quantitative Trait Loci and Candidate Genes Associated with High Interferon-gamma Production in Holstein Cattle Naturally Infected with Mycobacterium Bovis

Abstract

Mycobacterium bovis (Mb) is the causative agent of bovine tuberculosis (bTb). Genetic selection aiming to identify less susceptible animals has been proposed as a complementary measure in ongoing programs toward controlling Mb infection. However, individual animal phenotypes for bTb based on interferon-gamma (IFNɣ) and its use in bovine selective breeding programs have not been explored. In the current study, IFNɣ production was measured using a specific IFNɣ ELISA kit in bovine purified protein derivative (bPPD)-stimulated blood samples collected from Holstein cattle. DNA isolated from the peripheral blood samples collected from the animals included in the study was genotyped with the EuroG Medium Density bead Chip, and the genotypes were imputed to whole-genome sequences. A genome-wide association analysis (GWAS) revealed that the IFNɣ in response to bPPD was associated with a specific genetic profile (heritability = 0.23) and allowed the identification of 163 SNPs, 72 quantitative trait loci (QTLs), 197 candidate genes, and 8 microRNAs (miRNAs) associated with this phenotype. No negative correlations between this phenotype and other phenotypes and traits included in the Spanish breeding program were observed. Taken together, our results define a heritable and distinct immunogenetic profile associated with strong production of IFNɣ in response to Mb.

Keywords: Mycobacterium bovis; breeding; interferon-gamma; polymorphisms.

Figure 1

IFNγ production. Error plot of the levels of IFNɣ in stimulated blood samples from the 117 cows included in the study. The central line represents the mean of the group, and the whiskers represent the standard error. According to the interpretation criteria of the kit, 76 of the 117 cows included in the study were bTb-positive because the OD (bPPD-PBS) was >0.05 and the OD (bPPD-PBS) > OD (aPPD-PBS).

Figure 2

Results of the GWAS analysis. (A) Manhattan plot of the –log₁₀ of the p-values of the association test between each SNP and the IFNɣ levels. Each dot represents one SNP. Chromosome localization of the SNPs is indicated on the x-axis. The horizontal red line is drawn at –log₁₀ (5 × 10⁻⁷); (B) Genomic distribution of the 163 SNPs surpassing the threshold (p-value ≤ 5 × 10⁻⁷) according to the Ensembl Variant Effect Predictor (VEP); (C) Quantile–quantile plot comparing the observed distribution of –log (p-values) to the expected values under the null hypothesis.

Figure 3

Bubble plot displaying the QTL enrichment results for health traits. The darker the red shade in the circles, the more significant the enrichment. The area of the circles is proportional to the number of QTLs. The x-axis shows a richness factor obtained by the ratio of the number of QTLs annotated and the total number of each QTL in the reference database.

Figure 4

Protein-to-protein network analysis using the identified candidate genes. Individual nodes represent proteins with relationships represented by edges. The green lines represent gene neighborhood, red lines represent gene fusions, blue lines represent gene co-occurrence, black lines represent co-expression, pale blue lines represent homology, and purple lines represent experimentally determined interaction. The candidate proteins with no associations to other proteins were hidden.