Evaluation of Stallion Frozen-Thawed Semen Using Conventional and Flow Cytometric Assays

Field evaluation of frozen-thawed stallion semen has been limited to tests such as post-thaw motility and morphology that are not only subjective but also evaluate only a small population of cells. Flow cytometry has provided a quick, repeatable, objective method of evaluating a large number of cells, including spermatozoa. Two experiments were designed to first validate the use of several flow cytometric tests on frozen-thawed stallion semen and then determine a model that may best explain variation in fertility. Comparing samples that were live and freeze-killed validated the flow cytometric tests.

In experiment one, six ejaculates were collected from each of three stallions. The semen from each ejaculate was centrifuged and frozen in 0.5 ml polyvinyl chloride straws. Two straws from each ejaculate were thawed and evaluated. Semen was evaluated for post-thaw motility, morphology, mitochondrial activity using Rhodamine 123 (R123), plasma membrane integrity using propidium iodide (PI) and ethidium monoazide (EMA), and chromatin structure using the sperm chromatin structure assay (SCSA). Data was recorded as percentages for all but the SCSA for both experiment one and two. The extent of chromatin denaturation was calculated using the SCSA and the alpha-t population [at = red/(red +green) fluorescence]. From the alpha-t population, statistics were calculated such mean (Xat), standard deviation (SDat), percentage of cells outside (COMPat) the main alpha-t population and the mean green fluorescence (mean green) of the population.

Results from experiment one demonstrated that all flow cytometric tests except EMA were able to distinguish between live and freeze-killed samples (p < 0.0001). Also the stallion accounted for most of the variation in samples when compared to ejaculate and straw within an ejaculate. Therefore two straws could be chosen at random from a stallion and evaluated in experiment two.

In experiment two, twenty-nine stallions were evaluated using the same tests as experiment one excluding EMA. Fertility data was obtained from the 1998 or 1999 breeding season. Multiple linear regression was used to determine the best-fit model to predict overall pregnancy rate. SCSA and R123-PI assays accounted for the largest amount of variation in fertility (R² = 0.65, p < 0.0004). Within SCSA and the R123/PI assays Xat and PI staining had the highest contribution to this variation in fertility (R² = 0.11, R² = 0.47) respectively. The best-fit model for predicting fertility included the assay combination listed above and the interactions between SDat and mean green staining as well as R123 and mean green staining. Post-thaw motility and morphology did not account for significant variation in fertility (p = 0.22, p = 0.46) respectively.

Based on this project post-thaw motility and morphology are poor predictors of fertility in frozen-thawed stallion semen. However, through the addition of SCSA and R123-PI to the routine evaluation of frozen-thawed stallion semen time and money may be saved in advance by identifying those stallions with poor post-thaw fertility.