A triad of highly divergent polymeric immunoglobulin receptor (PIGR) haplotypes with major effect on IgA concentration in bovine milk

Abstract

The aim of this study was to determine a genetic basis for IgA concentration in milk of Bos taurus. We used a Holstein-Friesian x Jersey F2 crossbred pedigree to undertake a genome-wide search for QTL influencing IgA concentration and yield in colostrum and milk. We identified a single genome-wide significant QTL on chromosome 16, maximising at 4.8 Mbp. The polymeric immunoglobulin receptor gene (PIGR) was within the confidence interval of the QTL. In addition, mRNA expression analysis revealed a liver PIGR expression QTL mapping to the same locus as the IgA quantitative trait locus. Sequencing and subsequent genotyping of the PIGR gene revealed three divergent haplotypes that explained the variance of both the IgA QTL and the PIGR expression QTL. Genetic selection based on these markers will facilitate the production of bovine herds producing milk with higher concentrations of IgA.

Figure 1. Chromosome 16 QTL for colostrum and milk IgA content.

(A) Location scores (LRT = likelihood ratio test) obtained when scanning bovine chromosome 16 for QTL influencing IgA concentration (dotted lines) and yield (continuous line) in colostrum collected at the 2^nd milking (“2^nd colostrum”; light gray), colostrum collected at the 8^th milking (“8^th colostrum”; dark gray) and mid-lactation milk (black) using a mixed-model based approach that simultaneously extracts linkage and LD information. The red curve corresponds to the location scores obtained for IgA yield of 8^th colostrum (giving the strongest signal in single QTL analysis) when adding the effect of PIGR haplotype (I, II and III; cfr. Fig. 4) in the model. (B) Effect (± SEM) of the haplotypes of the six F1 sires on IgA yield of the 8^th colostrum (mg/milking). The haplotypes are labeled according to their corresponding PIGR genotype (I: blue; II: green; III: red; cfr. Fig. 4).

Figure 2. Effects of hidden haplotype states.

Bivariate effects (Y-axis: PIGR expression in liver; X-axis: IgA yield of 8^th colostrum) of the 20 Hidden Haplotype States at the most likely position of the 8^th colostrum IgA yield QTL (4,533,875). Hidden Haplotype States are labelled according to their PIGR haplotype (I: blue; II: green; III: red; cfr. Fig. 4).

Figure 3. Chromosome 16 eQTL for PIGR mRNA expression.

(A) Location scores (LRT = likelihood ratio test) obtained when scanning bovine chromosome 16 for QTL influencing PIGR transcript levels (black line) in liver using a mixed-model based approach that simultaneously extracts linkage and LD information. The red curve corresponds to the location scores obtained for PIGR mRNA expression level (liver) when adding the effect of PIGR haplotype (I, II and III; cfr. Fig. 4) in the model. The gray line corresponds to the location score obtained for IgA yield of the 8^th colostrum (cfr. Fig. 1A). (B) Effect (± SEM) of the haplotypes of the six F1 sires on PIGR expression level in liver. The haplotypes are labeled according to their corresponding PIGR genotype (I: blue; II: green; III: red; cfr. Fig. 4).

Figure 4. Sequence comparison of PIGR haplotypes.

(A) Variant positions at which the corresponding pair of PIGR haplotypes Differ (upper line; “D”), or are the Same (lower line; “S”). The positions of the PIGR exons are marked by the transparent gray boxes. (B) Schematic representation of the three major PIGR haplotypes with indication of the positions at which they differ or not. Within the body of the PIGR gene, haplotype I is appears as a recombinant between haplotype II and III, which differ on average every 109 nucleotides. Upstream of the gene, the three haplotypes differ on average every ∼200 nculeotides.

Figure 5. Effect of PIGR genotype on IgA yield and PIGR expression.

Phenotypic effects (± SEM) of the six possible PIGR genotypes (I/I, II/II, III/III, I/II, I/III, II/III – I: blue; II: green; III: red; cfr. Fig. 4) on IgA yield of the 8^th colostrum (A) and PIGR expression in liver (B). In general the phenotypic mean of the heterozygotes is intermediate between the corresponding alternate homozygotes, supporting additivity.