X chromosome variants are associated with male fertility traits in two bovine populations

Abstract

BACKGROUND: Twenty-five phenotypes were measured as indicators of bull fertility (1099 Brahman and 1719 Tropical Composite bulls). Measurements included sperm morphology, scrotal circumference, and sperm chromatin phenotypes such as DNA fragmentation and protamine deficiency. We estimated the heritability of these phenotypes and carried out genome-wide association studies (GWAS) within breed, using the bovine high-density chip, to detect quantitative trait loci (QTL). RESULTS: Our analyses suggested that both sperm DNA fragmentation and sperm protamine deficiency are heritable (h2 from 0.10 to 0.22). To confirm these first estimates of heritability, further studies on sperm chromatin traits, with larger datasets are necessary. Our GWAS identified 12 QTL for bull fertility traits, based on at least five polymorphisms (P < 10-8) for each QTL. Five QTL were identified in Brahman and another seven in Tropical Composite bulls. Most of the significant polymorphisms detected in both breeds and nine of the 12 QTL were on chromosome X. The QTL were breed-specific, but for some traits, a closer inspection of the GWAS results revealed suggestive single nucleotide polymorphism (SNP) associations (P < 10-7) in both breeds. For example, the QTL for inhibin level in Braham could be relevant to Tropical Composites too (many polymorphisms reached P < 10-7 in the same region). The QTL for sperm midpiece morphological abnormalities on chromosome X (QTL peak at 4.92 Mb, P < 10-17) is an example of a breed-specific QTL, supported by 143 significant SNPs (P < 10-8) in Brahman, but absent in Tropical Composites. Our GWAS results add evidence to the mammalian specialization of the X chromosome, which during evolution has accumulated genes linked to spermatogenesis. Some of the polymorphisms on chromosome X were associated to more than one genetically correlated trait (correlations ranged from 0.33 to 0.51). Correlations and shared polymorphism associations support the hypothesis that these phenotypes share the same underlying cause, i.e. defective spermatogenesis. CONCLUSIONS: Genetic improvement for bull fertility is possible through genomic selection, which is likely more accurate if the QTL on chromosome X are considered in the predictions. Polymorphisms associated with male fertility accumulate on this chromosome in cattle, as in humans and mice, suggesting its specialization.