Semen from 19 boars, of which 8 were of Finnish Landrace and 11 of Yorkshire breed, housed at the same AI station, was used in this trial. The average age of the boars was 26.8 ± 8.5 months (range 13 to 50 months). Three semen samples were collected from each boar using gloved-hand technique, within the regular collecting schedule - once a week - at the boar station. Fresh ejaculated semen was diluted to approximately 35 × 106 spermatozoa/ml, with X-cell extender (IMV Technologies, L'Aigle Cedex, France) and placed in 90 ml plastic tubes. One AI dose from each ejaculate was transported to the laboratory and stored at 17°C in a Unitron climate box (Unitron Skandinavia S/A) in closed plastic tubes until examination. The rest - were used for AI in commercial farms where sows were inseminated with fresh semen generally twice per oestrus 15 - 24 hours apart. Analysis of sperm motility, morphology, and plasma membrane integrity and fluorescence intensity was conducted at 24 h following semen collection and processing.
Fertility data were obtained from the Agricultural Data-Processing Center Ltd. (Vantaa, Finland). Each of 19 boars was used for at least 50 first inseminations in recorded herds and had ≥ 12 litters of recorded farrowings. In total, 2296 first inseminations and 1114 litters were recorded from all inseminations in approximately 110 commercial farms. Nonreturn rate within 60 days of first insemination (NR %) and litter size (total number of piglets born) of multiparous farrowings were used as fertility measures.
At the semen laboratory the sperm morphology was evaluated in air-dried Giemsa stained smears according to Watson . In total, 200 spermatozoa were examined. All abnormalities on any given spermatozoon were counted and then were divided into four groups according Blom , that is: normal spermatozoa, spermatozoa with major sperm defects (abnormalities of sperm head and acrosome, coiled sperm tail, etc.), spermatozoa with proximal droplets, and spermatozoa with minor sperm defects (simple bent tail, loose sperm head, etc.). Morphological abnormalities were expressed as a percentage of the total number of all counted spermatozoa.
Sperm motility was evaluated both subjectively and using a computer-assisted semen analyzer (CASA) (Sperm Vision Minitube™ of America, Inc., 2002). For the analysis, a 300-μl aliquot of the thoroughly but gently mixed semen sample was placed into an open 3-ml tube. The tube was kept in a 35°C water bath (Grants Instruments Ltd., Cambridge, UK) for 5 min before semen analyses. A 5-μl aliquot was placed on a pre-warmed 38°C microscope slide, covered with a coverslip (24 × 24 × 1.5 mm) and the proportions of total motile spermatozoa were recorded.
Calcein AM (CAM), propidium iodide (PI), Hoechst 33258 (H258) and Hoechst 33342 (H342) dyes were purchased from Molecular Probes Inc. (Eugene, OR, USA). One milligram of CAM was diluted in 1 ml of dimethyl sulfoxide (DMSO) (Mallincrodt Bacer B.V.), mixed for 10 min, kept in the dark, and then stored in 10-μl aliquots at -20°C. Twenty milligrams of PI were diluted in 1 l BTS (Beltsville Thawing Solution, Kubus S.A., Spain) and stored in 3-ml aliquots at -20°C. Six milligrams of H258 were diluted in 200 ml of BTS, mixed for 30 min in the dark, and stored in 2-ml aliquots at -20°C. Six milligrams of H342 were diluted in 200 ml of BTS, mixed for 30 min in the dark, and stored in 2-ml aliquots at -20°C. Before use, the dyes were thawed in a dark chamber at 35°C (Thermax, B8000, Bergen, Norway).
Assessment of plasma membrane integrity
Microscopic evaluation of plasma membrane integrity was carried out with a combination of two fluorescent stains, CAM and PI, according to Januskauskas and Rodriguez-Martinez , but using PI instead of Ethidium homodimer-1. Briefly, 10 μl of CAM (1 mg/ml) were mixed with 500 μl of BTS and added to 500 μl of PI (0.02 mg/ml) in BTS. For staining, 100-μl aliquots of semen were placed in 3-ml tubes, and 100 μl of CAM/PI solution was added. Each sample was further incubated for 10 min in the dark at 35°C. Sub-samples of 5 μl of the stained suspension were placed on clean microscope slides and overlaid carefully with coverslips. The slides were then evaluated under an epifluorescence microscope (Olympus BH2 with epifluorescence optics, Olympus Optical Co., Ltd., Japan) using ×500 magnification. In each slide 200 spermatozoa were categorized to CAM-stained green (live) and PI-stained red (dead) and the percentage of viable spermatozoa was then calculated.
Assessment of the fluorescence of the sperm nucleus
H258 and PI were used to measure the fluorescence intensity of the sperm nuclei in artificially membrane-ruptured spermatozoa. In contrast, membrane-permeant H342 was used to measure fluorescence intensities of intact spermatozoa. Fluorescence outputs were recorded in a fluorometer (Fluoroscan Ascent, Thermo Labsystems Oy, Vantaa, Finland) at 32°C. In order to estimate fluorescence intensities of given semen samples, sperm membranes ought to be disrupted. For this, 500-μl aliquots of semen were placed in 3-ml tubes and subjected to rapid freezing and slow thawing which induced membrane damage. The tubes were rapidly frozen by immersion directly into liquid nitrogen for 1 min. Thereafter the tubes were kept at room temperature for 30 sec, before being placed in a 35°C water bath for 3 min., as described previously by Alm et al. .
For the analysis, 50-μl aliquots of the artificially membrane-ruptured spermatozoa, plus 50 μl of PI or H258 were dispensed into the wells of a microtiter plate (Black Microtiter Plate 96 wells, Thermo Labsystems Oy, Vantaa, Finland) in three replicates. Blanks containing 50 μl of X-cell extender (IMV Technologies, France) and 50 μl of PI or H258 solution were dispensed in four replicates. The plate was then gently shaken for 2 min and further incubated in the fluorometer for 8 min at 32°C. Eleven samples and their corresponding blanks were analyzed at each assessment session. Semen samples of each boar were also stained with H342 in the same manner as with H258 except that membranes were not disrupted. The interference filter at the excitation path and the emission filter had maximum transmissions at 544 nm and 590 nm for PI, and 355 nm and 460 nm for H258 and H342. Sperm concentration in each AI dose was confirmed in a Bürker haemocytometer chamber (Fortuna, Germany). The results were expressed as fluorescence value/million spermatozoa.
Fluorometer-based assessment of membrane integrity
The fluorescence - based viability was assessed according to Alm et al. . Briefly, fluorescence intensities of PI and H258 were recorded in artificially killed and live semen samples. Percentage of viable spermatozoa was calculated based on the ratio of fluorescence output of intact and of killed subsamples, corrected in relation to background fluorescence (blank) .
Statistical analyses were carried out using SPSS software (version 13.0 for Windows, SPSS Inc., Chicago, IL, USA). Descriptive statistics, two-sample analysis, and Spearman rank correlations were calculated. The Spearman rank correlations were used to calculate the relationships between the sperm quality traits and fertility. Relationship between sperm viability and fertility results was represented by scatter diagram. Values are presented as means ± standard deviation (SD) and were considered statistically significant when P < 0.05.