Comprehensive functional analysis reveals that acrosome integrity and viability are key variables distinguishing artificial insemination bulls of varying fertility

Abstract

In vitro methods of assessing bull semen quality in artificial insemination (AI) centers are unable to consistently detect individuals of lower fertility, and attempts to reliably predict bull fertility are still ongoing. This highlights the need to identify robust biomarkers that can be readily measured in a practical setting and used to improve current predictions of bull fertility. In this study, we comprehensively analyzed a range of functional, morphological, and intracellular attributes in cryopreserved spermatozoa from a selected cohort of Holstein Friesian AI bulls classified as having either high or low fertility (n = 10 of each fertility phenotype; difference of 11.4% in adjusted pregnancy rate between groups). Here, spermatozoa were assessed for motility and kinematic parameters, morphology, acrosome integrity, plasma membrane lipid packing, viability (or membrane integrity), superoxide production, and DNA integrity. In addition, spermatozoa were used for in vitro fertilization to evaluate their capacity for fertilization and successful embryo development. The information collected from these assessments was then used to phenotypically profile the 2 groups of bulls of divergent fertility status as well as to develop a model to predict bull fertility. According to the results, acrosome integrity and viability were the only sperm attributes that were significantly different between high-and low fertility bulls. Interestingly, although spermatozoa from low-fertility bulls, on average, had reduced viability and acrosome integrity, this response varied considerably from bull to bull. Principal component analysis revealed a sperm phenotypic profile that represented a high proportion of ejaculates from low-fertility bulls. This was constructed based on the collective influence of several sperm attributes, including the presence of cytoplasmic droplets and superoxide production. Finally, using the combined results as a basis for modeling, we developed a linear model that was able to explain 47% of the variation in bull field fertility in addition to a logistic predictive model that had a 90% chance of distinguishing between fertility groups. Taken together, we conclude that viability and acrosome integrity could serve as fertility biomarkers in the field and, when used alongside other sperm attributes, may be useful in detecting low-fertility bulls. However, the variable nature of low-fertility bulls suggests that additional, in-depth characterization of spermatozoa at a molecular level is required to further understand the etiology of low fertility in dairy bulls.