Genetic and epigenetic architecture of paternal origin contribute to gestation length in cattle

Abstract

The length of gestation can affect offspring health and performance. Both maternal and fetal effects contribute to gestation length; however, paternal contributions to gestation length remain elusive. Using genome-wide association study (GWAS) in 27,214 Holstein bulls with millions of gestation records, here we identify nine paternal genomic loci associated with cattle gestation length. We demonstrate that these GWAS signals are enriched in pathways relevant to embryonic development, and in differentially methylated regions between sperm samples with long and short gestation length. We reveal that gestation length shares genetic and epigenetic architecture in sperm with calving ability, body depth, and conception rate. While several candidate genes are detected in our fine-mapping analysis, we provide evidence indicating ZNF613 as a promising candidate for cattle gestation length. Collectively, our findings support that the paternal genome and epigenome can impact gestation length potentially through regulation of the embryonic development.

Fig. 1

Single-marker GWAS, gene-level fine-mapping, and GWAS signal enrichment of gestation length. a Manhattan plot of all the imputed sequence variants being tested. Red line denotes the genome-wide significance of P equals to 1.91e-08. All the candidate genes are determined with the posterior probability of causality >0.05 based on gene-level fine mapping (BFMAP). b GWAS signal enrichment on the basis of five gene annotation sources, including Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway, Reactome metabolism pathway, Medical Subject Headings (MeSH), and miRNA-target networks. The values at x axis are the number of genomic features (i.e., gene lists) being tested in the corresponding annotation sources. The red triangle denotes the top 10 items with the highest enrichments (i.e., −log₁₀P) in each annotation source. The names of these items are shown in the figure

Fig. 2

Genetic relationships of gestation length and other dairy traits of economic importance. a Genetic correlations between gestation length and 35 dairy traits in the U.S. Holstein population. The genetic correlations are approximately computed using the effects of all tested variants in GWAS. b GWAS signal enrichment of gestation length based on the Cattle QTLdb (https://www.animalgenome.org/cgi-bin/QTLdb/BT/index). Each dot denotes a list of genes that are associated with a complex trait in Cattle QTLdb. The top 10 traits with the highest enrichments (i.e., −log₁₀P) are shown in the figure

Fig. 3

Sperm DNA methylation alterations associated with gestation length. a Enrichment of differentially methylated regions (DMRs) across multiple genomic elements. The enrichment of DMRs on a genomic element is computed as the ratio between the observed density of DMRs in this particular genomic element and the expected density of DMRs in the entire genome. b GWAS signal enrichment of gestation length based on DMRs that were determined by comparing animals with high gestation length to those with low gestation length. DMRs are first defined based on five different q values cutoffs (i.e., 0.05, 0.01, 1e-5, 1e-8, and 1e-10) and the absolute value of methylation difference >5% (Both). DMRs were then divided into two subsets according to the sign of methylation difference, i.e., <−5% (Loss) and > 5% (Gain). c DMR-set enrichment analysis based on the Reactome pathway database

Fig. 4

Relationships of sperm methylation alterations associated with gestation length, sire calving ease (SCE), body depth (BDE), cow conception rate (CCR). a Intersections between differentially methylated regions (DMRs) of gestation length and DMRs of SCE, BDE and CCR, respectively. P-value is calculated using a Fisher-exact test. b KEGG pathway functional enrichment of genes that are overlapped with shared DMRs in gestation length & SCE, gestation length & BDE, and gestation length & CCR, respectively. c Frequency of shared DMRs with the same (+) or opposite (−) change directions in gestation length & SCE, gestation length & BDE, and gestation length & CCR, respectively. d Comparisons of genetic correlations of gestation length & SCE, gestation length & BDE, and gestation length & CCR within their shared DMRs of same (+) or opposite (−) change directions, and over the entire genome (O), respectively

Fig. 5

Genetic, epigenetic, and selection evidence implicating the ZNF613 gene on gestation length. a GWAS signals (point-plot) and differentially methylated regions (DMR, bar plot) around ZNF613 for gestation length, sire calving ease (SCE), body depth (BDE) and cow conception rate (CCR). Sperm-retained nucleosome is also shown below DMRs. b Methylation levels of the shared DMR (the second intron of ZNF613) in the two compared groups across these four traits. c Comparison of signs of variant effects (i.e., b, based on signal-marker GWAS) within ZNF613 between gestation length and 13 dairy traits, for which ZNF613 is also a candidate gene based on our previous fine mapping analyses. +denotes the same direction, and – denotes the opposite direction. d The changes of minor allele frequency (MAF) of the lead variant (chr18:58141989; P = 7.97e-84) within ZNF613 over the years from 1952 to 2012