Study on dynamic of sperm population in semen of stallion, boar and rabbit

Quintero-Moreno A.

Unit of Reproduction, Department of Animal Medicine and Surgery, School of Veterinary Medicine, Autonomous University of Barcelona, E-08193 Bellaterra, Spain

The first aim of this study was to test the presence of separate sperm subpopulations with specific motility characteristics in stallion, boar and rabbit ejaculates by using a computer-assisted semen analysis system (CASA). Sperm motility descriptors were analyzed thorough the clustering of variables based on a covariance matrix. This matrix selects the descriptors of sperm motility that better explain the spermatozoon kinetics. Sperm subpopulations were obtained by disjointing cluster analysis where the observations are divided into clusters by which each observation belongs to one specific cluster. This test showed that three separate sperm subpopulations with different motility characteristics in boar, and four in stallion and rabbit, coexist in the ejaculates. There were significant (P<0.001) differences in the distribution of these subpopulations among individuals in all of the studied species, but no clear relationship between motile subpopulation structure and fertility was obtained. A second aim of the study was to test the possibility for a precise estimation of the fertilizing ability of mammalian ejaculate based upon the results of semen analysis. For this purpose, we tested the mathematical combination of several parameters of the boar and rabbit semen quality analysis as predictive “in vivo” fertility tools. The main mathematical relations utilized among parameters were logistic and linear regressions. In boar, two mathematical models obtained by logistic regression involving osmotic resistance test, hyperosmotic resistance test and viability of fresh samples, showed a significant (P<0.05) relationships between semen characteristics and conception rate. However, none of the obtained models produced a significant relation between semen characteristics and litter size. In rabbits, logistic and linear regression analysis rendered two mathematic, significant (P<0.05) models, with related some semen characteristic (sperm viability and abnormalities) with “in vivo” fertility and litter size. In stallion, the study of subpopulations in ejaculates which showed confirmed fertilizing capacity showed that these ejaculates had the majority of their motile spermatozoa included in a subpopulation with high progressiveness and low linear velocity. Moreover, all the ejaculates with proven fertility which have a total sperm count >20×109 spermatozoa/ejaculate showed all of their motile sperm included in this subpopulation. Our results support that the use of the values obtained in a standard boar, rabbit and stallion semen quality analysis to predict life fertilizing ability of a single ejaculate can reasonably be achieved by applying logistic and linear regression analyses to the parameters included in this analysis. Thus, our methodology can explain in a systematic manner mammal semen quality, relating it to conception rate and litter size.