Sequence-based GWAS reveals genes and variants associated with predicted methane emissions in French dairy cows

Due to their contribution to global warming, methane emissions from ruminants have been the subject of considerable scientific interest. It has been proposed that such emissions might be reduced using genetic selection; proposed phenotypes differ in the measurement methods used (direct or predicted methane emissions) and in the unit under consideration (g/d, g/kg of milk, g/kg of intake, residual methane emissions). Identifying the quantitative trait loci (QTLs) and candidate genes responsible for genetic variation in methane emissions allows a better understanding of the underlying genetic architecture of these phenotypes. Therefore, the aim of this study was to identify the genomic regions associated with six methane traits predicted from milk mid-infrared (MIR) spectra (0.33 ≤ R² ≤ 0.88) in French Holstein dairy cows using genome-wide association studies at the whole-genome-sequence level.

Six methane emission traits-in g/d, in g/kg of fat- and protein-corrected milk, and in g/kg of dry matter intake-were predicted from milk MIR spectra routinely collected by French milk recording companies. A genome-wide association study of the predicted methane emissions of 40,609 primiparous Holstein cows was conducted using imputed whole-genome-sequence data. This analysis revealed 57 genomic regions of interest; between 1 and 8 QTLs were identified on each of the autosomes except 4, 12, 21, 24 and 26. We identified multiple genomic regions that were shared by two or more predicted methane traits, illustrating their common genetic basis. Functional annotation revealed potential candidate genes, in particular FASN, DGAT1, ACSS2, and KCNIP4, which could be involved in biological pathways possibly related to methane production.

The methane traits studied here, which were predicted from milk MIR spectra, appear to be highly polygenic. Several genomic regions associated with these traits contain candidate genes previously associated with milk traits. Functional annotation and comparisons with studies using direct methane measurements support some potential candidate genes involved in biological pathways related to methane production. However, the overlap with genes influencing milk traits highlights the challenge of distinguishing whether these regions genuinely influence methane emissions or reflect the use of milk MIR spectra to predict the phenotypes.