Single-step genome-wide association study uncovers known and novel candidate genomic regions for endocrine and classical fertility traits in Swedish Red and Holstein dairy cows

In a study aiming to identify candidate genomic regions associated with endocrine and classical fertility traits in Swedish Red (SR) and Holstein cows, data on 3955 lactations in 1164 SR and 1672 Holstein cows were examined. The dataset comprised milk progesterone (P4) levels (n = 341,212) in 14 Swedish herds, automatically collected and analyzed in-line using the DeLaval Herd NavigatorTM. Endocrine traits studied were: days from calving to commencement of luteal activity (C-LA), first luteal phase length (LPL), length of first inter-luteal interval, length of first inter-ovulatory interval (IOI), luteal activity during the first 60 DIM, and proportion of samples with luteal activity during the first 60 DIM. Classical fertility traits based on insemination data were also investigated, such as days from calving to last insemination and calving interval. A total of 180 SR and 312 Holstein cows were genotyped with a low-density SNP chip and imputed to 50 K. Single-step genome-wide association (ssGWAS) was used to explore candidate genomic regions associated with fertility traits. A mixed linear single-trait animal model was fitted, considering season and parity as fixed effects and animal and permanent environment as random effects. The results revealed 990 and 415 SNPs above the threshold (-log (p-value) >= 4) for SR and Holstein cows, respectively. The breeds shared only eight SNPs significantly associated with fertility traits. Annotation analysis revealed 281 SNPs located in 241 genes. Functional enrichment analysis using DAVID tools reduced the number to 80 genes, which were mediated in various biological processes and KEGG pathways in multiple functions, including folliculogenesis, embryogenesis, uterine growth and development, immune response, and ovarian cysts. Of the 80 genes, 67 were associated with fertility traits in SR cows and 13 in Holstein. Most genes were associated with LPL and IOI in SR cows, but in Holstein the only association with an endocrine trait was with C-LA. Twenty QTL regions that embedded 40 genes were associated with fertility traits in both breeds. All the QTLs detected, except at BTA2 and BTA19 are novel QTL regions that were not reported previously. These novel QTL regions embedded the candidate genes that include ARHGAP20, PHLDB1, CACNA1D, ATG7, CCNE1, GPI, CDH13, ECT2, PLD1, FBN2, KIF3A, FGF12, KCNMB2, GJA1, MAN1A1, KCNN2, SMAD6, MAPK8IP1, PHF21A, LPXN, MMRN1, KCNIP4, NID2, PCDHGA8, GRIA1, PCDHGB4, PHLDB2, STXBP5L, PPP3CA, PTPRR, SRGAP1, SNX27, SPTA1, S100A10, TBC1D20 and ITCH. The candidate regions may help to improve genetic progress in female fertility if used in selection decisions. A challenge for future research is to determine why different regions seem relevant for different traits and breeds, and the practical implications for genomic selection.