Boar seminal plasma components and their relation with semen quality

Abstract

Select boar seminal plasma (SP) components and their relation to semen quality were investigated. Thirty nine boars from three artificial insemination (AI) centers were divided into group A (GA: > 80% normal sperm and >70% motility) and group B (GB: < 80% normal sperm and < 70% motility). Each ejaculate was collected and semen volume, concentration, sperm motility (computer aided semen analysis; CASA), morphology, and vitality (both eosin nigrosin staining) were investigated. The SP was separated and analyzed for aspartate-amino-transferase (AST), γ-glutamyl-transferase (GGT), alkaline phosphatase (ALP) activity, and the concentrations of sodium (Na), potassium (K), chloride (Cl), calcium (Ca), phosphate (PO₄^3-), magnesium (Mg), selenium (Se) and zinc (Zn) were assessed. Repeated measures (2 months interval) were conducted in eight boars of GA from one AI center. The activity of GGT (r = -0.482) and ALP (r = -0.459) was moderately associated (p < 0.05) with ejaculate volume and strongly associated with concentration (r = 0.580 and r = 0.618, respectively; p = 0.000). Moderate associations (p < 0.05) were found between ALP (r = 0.439), GGT (r = 0.387), Na (r = -0.428), K (r = 0.354), and Se (r = 0.354) with progressive motility. The SP concentration of Na (r = -0.401), Cl (r = -0.521), and K (r = 0.350) was associated (p < 0.05) with normal morphology. Only Mg was associated (p < 0.05) with membrane damage (r = -0.335). The concentration of Na, Cl, and Zn (1681.0 vs. 1701.0 µg/dL) was different between groups (p < 0.05). Repeated measures showed significant differences in time but only for Na, Mg, and Zn (p < 0.05). In conclusion, several biochemical components of SP were related to semen quality. The analysis of biochemical parameters could provide extra information about reproductive health of AI boars.

Keywords:

• alkaline phosphatase
• boar
• minerals
• seminal plasma

Introduction

The use of artificial insemination (AI) in commercial pig herds has increased significantly in the last decade [ Maes et al. 2011; Riesenbeck 2011]. To provide high quality insemination doses, routine assessment of the semen quality of the ejaculates is necessary. Although some AI centers may implement CASA in their routine semen assessment, semen quality analysis is mainly based on conventional techniques, i.e., concentration by photometer and visual evaluation of motility and morphology [Knox et al. 2008; Vyt et al. 2007]. The relation of these parameters with in vivo fertility is still under discussion [Tsakmakidis et al. 2010; Vyt et al. 2008]. The mechanisms underlying boar infertility are poorly understood mainly because boars are rapidly culled without any further analysis other than the above described conventional methods [Úbeda et al. 2010].

In human reproduction, assessment of SP including the analysis of various biochemical components is advised by the World Health Organization (WHO) for the routine seminogram. Some of its components are indicators of prostatic (dys)function and/or are related to infertility in men [Chia et al. 2000; WHO 2010]. Analysis of SP might therefore also be of significant clinical relevance in studying boar infertility.

The SP is a mixture of fluids from the cauda epididymidis and the accessory sex glands [Davies et al. 1975]. In pigs, it has been proposed that the SP can be substituted by extender without remarkable detrimental effect, but recent studies have shown that SP has different functions for sperm metabolism [Rodriguez-Martinez et al. 2011]. Additionally SP plays an important role during sperm capacitation and it stimulates the female immune system to remove pathogens and to tolerate spermatozoa and embryos [Rodriguez-Martinez et al. 2010; Rodriguez-Martinez et al. 2011]. The activity of enzymes such as γ-glutamyl-transferase (GGT) or alkaline phosphatase (ALP) is related to semen quality and membrane function and they participate in different metabolic processes during sperm maturation [Chikhi et al. 1999; King and Macpherson 1966; Kohdaira et al. 1986; Seligman et al. 2005]. Additionally low ALP activity has recently been described in one azoospermic boar [Clements et al. 2010].

Minerals, such as zinc (Zn) or selenium (Se; structural component of glutathione peroxidase, GPx), have been associated with semen quality in humans due to their antioxidant properties [Bedwal and Bahuguna 1994; Bjorndahl and Kvist 2010; Chia et al. 2000; Gavella et al. 2000; Powell 2000]. Furthermore, Zn is included in the WHO manual as an indicator of prostatic function [WHO 2010] and low levels of Zn have been found in SP of infertile men [Chia et al. 2000; Colagar et al. 2009]. In boars, some changes in mineral composition are related to season and heat stress [Larsson and Einarsson 1984; Murase et al. 2007]

Although several studies are available on these enzymes and minerals in boar semen, their association with semen quality was not clearly demonstrated [Boursnell et al. 1973; Ciereszko and Strzezek 1989; Einarsson et al. 1970; Lavon and Boursnell 1975; Strzezek et al. 1995; Strzezek et al. 1998; Wysocki and Strzezek 2006]. Moreover, many of these studies are rather dated. Several new sperm assessment techniques, such as CASA systems were introduced and validated in boars which allow a much more objective and detailed motility analysis. Furthermore to study the association of a given parameter with semen quality, boars with poor semen quality should be included in the study [Gadea et al. 2004] which has not always been the case in previous studies.

Based on the studies performed in other animal species, in human, and in boars, detection of several biochemical parameters could be of clinical relevance to detect, for instance, testicular or glandular pathologies and to monitor health. Therefore, in the present study, we investigated several SP parameters in boars with good and poor semen quality from three commercial AI centers. In addition to providing reference values, the association of these components with various semen quality parameters was investigated.

Results

Semen quality

The number of boars per AI centre, the different groups, and the methods used in each AI center to determine semen quality are summarized in Table 1. There were no significant differences between the groups (GA vs. GB) for ejaculate volume, sperm concentration, sperm count, and the percentage of sperm with intact membrane (Table 2). The percentage of motile sperm and the percentage of progressive motility were lower in GB (p < 0.05; Table 2). The number of all morphological sperm abnormalities was higher in GB (p < 0.05; Table 2). The interaction terms group x AI center were not significant for any of the semen quality parameters (p > 0.05).

Table 1. Descriptive data of the boars included in the present study and the semen quality analysis procedures used in the three artificial insemination (AI) centers to classify ejaculates in group A (>80% normal sperm morphology and sperm motility > 70%) and group B (< 80% normal morphology and motility < 70%).

	AI Centre 1		AI Centre 2		AI Centre 3
	Group A	Group B	Group A	Group B	Group A	Group B
Number of valid boars	7	3	8	8	9	4
Boar age (months ± SD)	49.8 ± 21.6	26.6 ± 7.2	33.0 ± 15.0	32.5 ± 24.2	41.9 ± 19.6	54.5 ± 10.4
Days to previous Collection ± SD	6.7 ± 0.5^a	5.7 ± 1.5^a	5.5 ± 1.1^a	5.2 ± 3.7^a	12.1 ± 3.0^b	10.7 ± 2.9^b
Sperm quality analysis
Concentration	Photometer		Photometer		CASA
Motility & morphology	Visual		Visual		CASA

^a,bdifferent superscripts within a row indicate significant differences.

Table 2. Average values (mean ± SD) of semen quality parameters of boars with good (Group A, > 80% normal sperm morphology and > 70%, sperm motility; n = 24) and poor (Group B, < 80% normal morphology and < 70%, motility; n = 15) semen quality.

	Group A	Group B	p-value
Volume (mL)	247.3 ± 70.64	277.3 ± 136.4	0.622
Concentration (x10^6/mL)	414.2 ± 234.8	422.0 ± 225.6	0.548
Sperm count (x10^9/mL)	94.8 ± 37.5	105.1 ± 43.2	0.547
Motility (%)	80.8 ± 7.0	61.1 ± 16.8	0.000
Progressive motility (%)	34.1 ± 13.8	18.7 ± 12.0	0.001
Normal sperm (%)	90.7 ± 5.7	55.5 ± 21.4	0.000
Abnormal head (%)	1.7 ± 2.1	5.6 ± 7.0	0.001
Abnormal tail (%)	1.6 ± 1.7	13.5 ± 13.9	0.000
Proximal Droplet (%)	2.5 ± 2.7	8.3 ± 6.5	0.000
Distal Droplet (%)	3.5 ± 2.6	17.1 ± 12.3	0.000
Membrane intact (%)	75.5 ± 12.6	79.3 ± 7.4	0.374

Associations between SP parameters and semen quality

The associations between SP biochemical values and semen quality parameters are summarized in Table 3. The concentrations of GGT and ALP were negatively associated with volume of the ejaculate and positively associated with concentration. There was a moderate positive association between GGT and ALP with progressive motility. There was no association between SP enzymes and sperm with normal morphology or membrane integrity. However, trends were observed for the association of GGT (r = 0.366; p = 0.024) and ALP (r = -0.359; p = 0.027) with abnormal heads and distal droplets, respectively.

Table 3. Spearman Rank correlation between SP enzymes and minerals concentration, and semen quality. Boars with good (Group A: >80% normal sperm morphology and > 70% sperm motility n = 24) and poor (Group B: <80% normal morphology and < 70% sperm motility, n =15) semen quality are included in the analysis.

	Volume	Concentration	Motility	Progressive	Normal Sperm	Membrane Intact
AST (U/L)	−0.050	0.343^a	−0.043	0.011	−0.134	0.197
GGT (U/L)	−0.482^b	0.580^c	0.279	0.387^a	0.050	−0.158
ALP (U/L)	−0.459^b	0.618^c	0.311	0.439^a	0.241	−0.178
Ca (mg/dL)	−0.146	0.293	0.162	0.308	0.224	0.228
PO4^3- (mg/dL)	−0.356^a	0.549^c	0.23	0.359^a	−0.048	−0.198
Mg (mg/dL)	−0.003	0.079	0.213	0.278	0.261	0.335^a
Na (meq/L)	0.197	−0.229	−0.396^a	−0.428^a	−0.401^a	−0.271
K (meq/L)	−0.338^a	0.201	0.278	0.354^a	0.350^a	0.162
Cl (meq/L)	−0.256	−0.077	−0.348	−0.394	−0.521^a	−0.390^a
Se(μg/l (ppb))	−0.324^a	0.436^c	0.302	0.354^a	0.246	0.137
Zn (μg/dL)	−0.005	0.062	0.140	0.194	0.094	−0.140

AST: aspartate-amino-transferase; GGT: γ-glutamyl-transferase; ALP: alkaline phosphatase; Ca: calcium; PO₄^3-: phosphate; Mg: magnesium; Na: sodium; K: potassium; Cl: chloride; Se: selenium; Zn: zinc

^ap < 0.05; ^bp < 0.005; ^cp < 0.001

The mineral concentration was associated with semen quantity and quality. Phosphate, K, and Se were negatively correlated with semen volume and Se and PO₄^3- were correlated positively with semen concentration (Table 3). Furthermore, there was a moderate positive association of PO₄^3-, K, and Se and a negative association of Na with progressive motility (Table 3).

Higher levels of Na and Cl were associated with a decrease in the number of spermatozoa with normal morphology. Whereas Na seemed to be positively associated with abnormal tails (r = 0.412; p = 0.010), high levels of Cl appeared to be associated with an increased number of abnormal heads (r = 0.449; p = 0.041) and abnormal tails (r = 0.431; p = 0.051). Higher levels of Mg and Se were associated with less membrane damage (r = -0.335; p = 0.040) and proximal droplets (r = -0352; p = 0.033), respectively. There was a trend for a negative association between Zn concentration and the number of abnormal tails (r = -0.327; p = 0.051). In addition, a moderate association was found between AST (r = 0.481; p = 0.002), ALP (r = 0.353; p = 0.030), K (r = 0.421; p = 0.009), Cl (-0.500; p = 0.003), and Zn (r = -0.398; p = 0.016) with days to previous collection. Selenium was strongly associated with days to previous collection (r = 0.628; p = 0.000).

Seminal plasma reference values and intergroup comparison

Mean, median, and range (95% confidence interval) of the enzyme activity and mineral concentration in SP of GA and GB are summarized in Table 4. Most of the parameters except for GGT, Ca, PO₄^3-, Mg, and Na were affected by AI center (p < 0.05) but there was no group x AI center interaction.

Table 4. Reference values (mean, median, range) of seminal plasma parameters of boars with good (Group A: > 80% normal sperm morphology and > 70%, sperm motility n = 24) compared to the values obtained for boars with poor (Group B: < 80% normal morphology and < 70% motility, n = 15) semen quality.

	Group A			Group B			p-value
	Mean	Median	Range^1	Mean	Median	Range^1	(Group)
AST (U/L)	15.6	12.0	3.0−26.3	25.4	14.0	11.6−39.2	0.075
GGT (U/L)	6917.7	5715.0	5385.1−8772.4	5702.9	5361.0	3698.9−7706.8	0.586
ALP (U/L)	45586.3	41363.0	37944.9−54869.7	33611.0	32569.0	23598.1−43623.9	0.183
Ca (mg/dL)	3.1	2.9	2.6−3.4	2.7	2.4	2.3−3.2	0.250
PO4^3− (mg/dL)	1.4	1.3	1.1−1.5	1.3	1.1	1.0−1.5	0.767
Mg (mg/dL)	7.7	7.8	7.5−8.0	7.5	7.5	7.2−7.7	0.152
Na (meq/L)	105.8	106.0	100.7−111.6	116.7	120.0	110.3−123.1	0.015
K (meq/L)	15.7	15.2	14.7−17.0	14.5	13.9	13.1−15.8	0.228
Cl (meq/L)	107.2	97.4	93.8−120.5	136.4	138.8	112.9−159.9	0.022
Se (ppb)	22.6	20.0	18.0−28.6	18.7	13.0	12.4−24.9	0.871
Zn (μg/dL)	1927.5	1681.0	1690.4−2164.6	1718.8	1701.0	1438.3−1999.3	0.032

¹Range: 95% confidence interval; AST: aspartate−amino−transferase; GGT: γ−glutamyl−transferase; ALP: alkaline phosphatase; Ca: calcium; PO₄³⁻: phosphate; Mg: magnesium; Na: sodium; K: potassium; Cl: chloride; Se: selenium; Zn: zinc

From the selected parameters, LDH, Fe, Cu, and Vit E were in most of the samples below the detection limit of the test, and therefore they were not included in the statistical analysis. The concentration of Na, Cl, and Zn differed significantly between groups (Table 4). Sodium and chloride were higher in the GB whereas the concentration of zinc was lower in this group. No significant differences were observed between groups for the other minerals evaluated.

Repeatability of measurements of SP components

Values of SP parameters from 8 boars of GA from AI center 2 at two different time points are summarized in Table 5 together with the p-values for time effect of the repeated measures ANOVA. Most of the parameters varied numerically between samplings. Whereas AST, GGT, ALP, Na, K, and Se were lower in the second sampling, Ca, Mg, Cl, and Zn were higher. However, the time effect was only significant between time points for Mg, Na, and Zn (Table 5).

Table 5. Seminal plasma values of selected seminal plasma parameters of good boars (Group A; n = 8) from the same AI centre at 2 different time points (approximately 2 months between samples).

	Time 1		Time 2		Time
	Mean	SD	Mean	SD	p-value
AST (U/L)	12.5	3.5	11.7	10.5	0.868
GGT (U/L)	9749.9	9965.3	6083.6	4181.9	0.118
ALP (U/L)	57775.9	47730.0	34394.9	21478.4	0.080
Ca (mg/dL)	2.7	1.1	3.1	1.8	0.562
PO4^3- (mg/dL)	1.1	0.3	1.2	0.4	0.754
Mg (mg/dL)	6.6	1.4	7.6	0.9	0.009
Na (meq/L)	121.6	12.1	107.8	15.6	0.003
K (meq/L)	16.5	1.7	15.5	1.9	0.084
Cl (meq/L)	102.7	13.8	115.7	33.0	0.497
Se (μg/l (ppb)	13.9	10.7	11.4	4.9	0.570
Zn (μg/dL)	1065.9	330.5	2262.0	839.3	0.013

AI: artificial insemination; AST: aspartate-amino-transferase; GGT: γ-glutamyl-transferase; ALP: alkaline phosphatase; Ca: calcium; PO₄^3-: phosphate; Mg: magnesium; Na: sodium; K: potassium; Cl: chloride; Se: selenium; Zn: zinc

Discussion

In the present study, selected biochemical components of boar SP and their association with various semen quality parameters were investigated. When SP parameters from all boars of GA and GB were plotted with semen quality parameters, interesting associations were observed. The AST, ALP, and GGT activity had a positive association with semen concentration and a negative association with semen volume. The testicular/epididymal origin for ALP has been previously demonstrated as ALP was not found in vasectomized boars [King and Macpherson 1966]. Several older studies based on immune histochemistry have shown the presence of ALP in testis of rats and low ALP in accessory glands of boars [Aitken 1960; Bern 1949]. A low activity of ALP could therefore be indicative of insufficient ejaculate due to obstruction of the ductuli efferentes or ductus deferentes. To our knowledge, there is only one recent report in which a single boar with low sperm counts and diagnosed of obstruction of the ductuli efferentes had low ALP activity in SP [Clements et al. 2010]. The observed ALP activity in that case report (8,100 U/L) was markedly below the observed values in our study for both GA and GB.

The association of ALP with sperm motility has been attributed to the role which ALP plays in the synthesis of fructose, one of the main energy sources for sperm movement. In this process, ALP appears to be involved in the dephosphorylation of glucose-6-phosphate [King and Macpherson 1966]. However, fructose seems to be synthesized in the seminal vesicles whereas ALP is of testicular/epididymal origin [Aitken 1960; Bern 1949; Mann 1946]. ALP may interact with fructose precursors released in the accessory glands. The activation of fructose precursors by ALP could be one of the triggers for spermatozoa activation when they come in contact with SP. A protective effect against oxidative stress during sperm maturation in the epididymis has been attributed to GGT which could explain the relationship observed between GGT and motility [Kohdaira et al. 1986; Seligman et al. 2005]. A numerical higher activity of AST was observed in poor ejaculates. These results are in agreement with previous studies which found an association between AST in seminal plasma and sperm damage. This is an intracellular spermatozoa enzyme and therefore its increase in seminal plasma is an indicator of cell damage [Larson et al. 1976; Strzezek et al. 1981]

It has been shown that addition of K to boar semen extenders helps to preserve motility and therefore it is routinely added to several commercial extenders [Johnson et al. 2000]. The role of PO₄^3- on motility has been explained by phosphorylation required to activate proteins involved in the initiation of sperm motility [Arrata et al. 1978; Tash and Bracho 1994]. Phosphate is also an important component of ATP and cAMP which are both necessary for motility. Se, is an important component of GPx, an enzyme that protects against lipid peroxidation resulting in improved sperm motility [Marin-Guzman et al. 1997].

In the present study higher ALP was associated with a reduction of distal droplets. It has been shown that boar sperm loses cytoplasmatic droplets during ejaculation and the number of sperm with these droplets increases with collection frequency [Pruneda et al. 2005]. Distal droplets could therefore indicate insufficient sperm maturation and it could be possible that ALP plays a role in sperm maturation during ejaculation. The association with tail droplets is in accord with the observed positive association of the enzyme with motility. The exact mechanism behind this association is not clear, but it has been suggested that fructose stimulates shedding of cytoplasmatic droplets of boar spermatozoa [Harayama et al. 1996].

Na and Cl correlated negatively with normal morphology. It might reflect abnormal sperm leakage to the SP and that Na/K flux needed for sperm metabolism is altered. Contrary to our results, previous studies in rats, humans, and boars, have shown that a sodium blockade increases extracellular sodium ions improving sperm motility [Kong et al. 2009; Holt and Harrison 2002]. In these studies different compounds were added to the semen and this resulted in an increase in motility. The addition of these compounds may result in other biochemical changes like reduction of pH that may affect motility. Therefore it is difficult to compare the results of these studies with those presented in this manuscript. Although significant differences between groups were observed, we could not find any clear association of Zn with semen quality. Only a trend of a negative association for Zn with abnormal tails was observed. In addition to its antioxidant properties, Zn is involved in many aspects of spermatogenesis and sperm physiological function [Bedwal and Bahuguna 1994]. It is likely that it affects semen quality in different ways which would make it difficult to associate Zn with one single parameter. Similar to our results, a negative association of Zn with sperm abnormal morphology has been described in human semen [Colagar et al. 2009]. It is possible that the concentration of Zn could be an indicator of boar fertility but high measurement variations and the weak associations with semen quality that we observed limits its potential use for diagnostic purposes. Only Mg was associated with reduced membrane damage. In agreement with this finding, addition of magnesium to boar extenders has been shown to improve semen vitality [Szczesniak-Fabianczyk et al. 2003].

Interestingly, many of the studied SP components were associated with days to previous collection. It has been shown that high collection frequency affects the excretion and reabsorption patterns of epididymal epithelium [Pruneda et al. 2005]. Although the differences in days to previous collection were not significant between groups, there seemed to be a numerical difference and there was a significant difference between AI center 3 and the other 2. This could partially also explain differences in SP composition between groups and AI centers.

When comparing the 2 groups, only significant differences were found for Na, Cl, and Zn although some numerical trends were observed for the other parameters. Regarding Na and Cl, it seems that an appropriate osmolarity is required for sperm motility and that a balanced combination of these elements is necessary for sperm metabolism [Quinn et al. 1965]. With respect to Zn, it has been shown in human that low seminal plasma Zn is associated with low motility, low sperm vitality, and infertility [Chia et al. 2000]. In boars it has been shown that, when sperm cells are damaged, Zn accumulates in the sperm cells with a consequent reduction of Zn in SP [Westmoreland et al. 1967].

In the present study, reference values for SP parameters in boars were additionally provided. Many factors could explain the variations in the measurements. For example, we did not study the different fractions of the ejaculate and their composition may be different. However all ejaculates were handled in the same way and they were homogenized before a sample was collected. Consequently, this should not have biased the comparison between groups. A previous version of the device used for most of the analysis (Cobas 8000) has been shown to have a good precision and coefficients of variation ranged between 0.6% and 4.4% for routine chemistry [van Gammeren et al. 2008]. Accordingly, the method of analysis could be responsible for only a minor part of the observed variation. Variation can also be attributed to differences between AI centers as the statistical analysis showed differences between AI centers. Small differences in feed composition were observed that could result in different SP mineral compositions. However in all three AI centers feed composition complied with the NRC requirements making a deficiency that dramatically affected the results very unlikely. Other sources of variation could be the water or a possible cage biting behavior, since they are usually made of galvanized metal (Zn coated). Interval of collection to analysis may affect Zn analysis as it has been shown in human that samples analyzed 60 minutes after collection had, respectively, 29% and 17% lower Zn than samples analyzed 30 minutes or immediately after semen collection [Elzanaty and Malm 2007]. The authors hypothesized that after ejaculation zinc might bind to spermatozoa and thus removed from assay. This will require investigation. Because boar semen diluters contain minerals [Johnson et al. 2000], all samples remained undiluted to avoid interference of the diluters with the SP composition. However, the time from collection to submission of the samples was the same for both groups and similar among AI centers (collection in the morning, processing at midday, and submission to the laboratory within 12 hours) so this should not bias the comparison between groups.

Interestingly repeatability measures analysis showed variability for all parameters in time although time effect was only significant for Na, Mg, and Zn. Possible reasons for the variation in the measurements have already been explained above. In addition, we should also consider a possible seasonal effect as samples were collected at the beginning of the summer and at the end of the summer. Some seminal plasma components like proteins or citric acid seem to increase during the winter [Trudeau and Sanford 1986]. However, the same study did not show season effect on ALP activity in SP. Season may affect sperm concentration and volume and steroid composition of the SP [Claus et al. 1983]. Boars with poor semen quality were not included in the analysis as this condition might affect the repeatability of the SP components.

In conclusion, reference values of selected SP components for boars are provided in the present study. A significantly different composition of Na, Cl, and Zn of the SP of good and poor quality ejaculates was observed. Different associations of biochemical components of the SP with semen quantity and quality were found. Analysis of SP components could therefore add valuable information to the increasing research performed on boar SP as well as to the clinical examination of AI boars with poor semen quality problems.

Materials and Methods

Study population and semen samples

The study was approved by the local sanctioning board. Forty eight boars from 3 different AI centers were included in the study. In each AI centre, the boars were housed within the same building in individual pens with straw bedding and received a commercial feed (2-3 kg feed/d, depending on age) and ad libitum drinking water from a deep pit. A feed sample was collected from each AI center to analyze the composition.

One ejaculate of each boar was collected using the gloved hand technique [Shipley 1999]. Immediately after collection, volume (weight) and concentration of each ejaculate was measured, and the total number of sperm cells per ejaculate was calculated (volume x concentration). The technician performing the collection was recorded as well as the number of days to previous collection (Table 1). Subsequently semen quality parameters were investigated by personnel of the AI center and ejaculates were divided in two groups. Ejaculates with semen quality complying with the AI center's cut-off values (minimum 80% normal sperm morphology and sperm motility of at least 70%) were considered as good ejaculates and were included in GA. The ejaculates with semen quality below these cut-off values were considered as poor ejaculates and were included in GB. The number of boars per AI center, the different groups, and the methods used in each AI center to determine semen quality are summarized in Table 1.

Prior to any dilution, 4-5 ml of raw ejaculated semen was collected into blood tubes to study the semen quality, and minerals and enzymes composition of the seminal plasma. The raw ejaculate samples were stored at room temperature until they were transported in isotherm boxes to the laboratory for sperm analysis at the Faculty of Veterinary Medicine, Ghent University, Belgium.

Sperm motility, morphology, and vitality

Upon arrival at the semen laboratory at the Department of Reproduction, Obstetrics and Herd Health, Ghent University, Belgium (3-4 h after collection), a subsample (1 mL) of the raw ejaculate was taken, placed in an eppendorf tube and used to study sperm motility and morphology.

Different sperm motility characteristics were measured objectively with a CASA system (HTR Ceros 12.3 semen analyzer, Hamilton-Thorne Research, Beverly, USA). Prior to the CASA analysis, a subsample of the raw ejaculate was extended in PBS (1:7) and warmed at 37°C in an incubator (IN, Memmert GmbH + Co.KG, Germany) for 30 min [Vyt et al. 2008]. Samples were subsequently submitted to CASA analysis in a random order and were consistently investigated by the same person. After gentle mixing, a 10 µl droplet of each sample was placed on a slide prepared according to WHO [2010] guidelines and analyzed by CASA. The slide was evaluated following the manufactureŕs protocol [Vyt et al. 2004]. Up to 1,000 tracks of sperm cells were obtained by analyzing 4 randomly selected fields that were measured 5 times each [Filliers et al. 2008; Rijsselaere et al. 2003]. The average of the measurements on these 4 fields was calculated and used for the statistical analysis. The software settings for the HTR Ceros 12.3 were those recommended by the manufacturer for analysis of boar sperm. Different motility parameters were determined but only MOTILE% (percentage of motile sperm) and PROGR% (percentage of progressively moving sperm: sperm cells with both VAP > 50 µm/s and STR > 70%) were used for the statistical analysis.

Morphology was assessed using eosin-nigrosin stained slides following standard procedures [Dott and Foster 1972]. A total of 100 cells/sample were evaluated to determine the percentage of normal sperm and the percentage of sperm with abnormal heads, abnormal tails, and proximal and distal cytoplasmic droplets. In addition, up to 100 spermatozoa/sample were evaluated to establish the percentage of membrane damaged spermatozoa as evidenced by a pink to red color.

In order to clearly define good and poor ejaculates, doubtful ejaculates were excluded. Therefore, when an ejaculate was considered as good for the AI center (> 80% normal morphology and > 70% motility) or bad (< 80% normal morphology and < 70% motility) but the semen quality analysis at the sperm laboratory at Ghent University showed discordant results, the ejaculate was excluded from the study. From the initial 48 collected boars 39 boars were finally considered as valid and were included in the study (Table 1).

Biochemical analysis of seminal plasma

The raw ejaculates were centrifuged for 20 min at 1,000 × g at room temperature. Subsequently, the SP was separated from the sperm layer and the remaining SP was sent to an external diagnostic laboratory for biochemical analysis. Absorbance photometry (Cobas 8000, Roche, Basel, Switzerland) was used to measure the activity of AST, GGT, ALP, and LDH as well as the concentrations of Fe, Ca, PO₄^3-, and Mg [van Gammeren et al. 2008]. The concentration of Na, K, and Cl was measured with ion selective electrode system (Cobas 8000, Roche) [Oesch et al. 1986; van Gammeren et al. 2008] and Cu, Se, and Zn by atomic absorption (Graphite oven SA600, Perkin Elmer, Waltham, Massachusetts, USA). Vitamin E was assessed by high performance liquid chromatography (Perkin Elmer Flexar HPLC).

Repeatability

To study the repeatability of the biochemical parameters in the SP, 8 boars with good ejaculates from AI center 2 were sampled a second time 2 months after the first sampling. The ejaculates were processed in exactly the same way as described above. The semen quality analysis and seminal plasma parameters were also measured in the same way as for the first analysis.

Statistical analysis

Data were close to the normally distributed as determined by Kolmogorov-Smirnov test although the test was positive for some parameters. Because of the presence of outliers, Spearman rank correlation was used to investigate the associations between the biochemical and the semen quality parameters. Differences between groups in semen quality and in SP biochemical parameters were studied with ANOVA. Group and AI center were included as fixed factor and their interaction was investigated. When the interaction was not significant, it was excluded from the model and only the main effects were investigated. Differences were considered as significant if P-values were lower than 0.05 (2-sided test). Descriptive statistics of the biochemical parameters were additionally presented to establish baseline levels. The repeatability of the measurements of SP parameters was studied with repeated measures ANOVA. Statistical analyses were performed using the statistical software package SPSS version 19.00. Data are expressed as means ± SD, unless stated otherwise.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Author contributions: All authors participated in the elaboration of the protocol, experimental design, discussion of the data, and drafting of the manuscript. Performed the semen quality analysis: ALR; Performed the biochemical analysis of the seminal plasma: PV.

Abbreviations

SP:	=	seminal plasma
AI:	=	artificial insemination
GA:	=	group A
GB:	=	group B
CASA:	=	computer aided semen analysis
AST:	=	aspartate-amino-transferase
GGT:	=	γ-glutamyl-transferase
ALP:	=	alkaline phosphatase
Na:	=	sodium
K:	=	potassium
Cl:	=	chloride
Ca:	=	calcium
PO43-:	=	phosphate
Mg:	=	magnesium
Se:	=	selenium
Zn:	=	zinc
AI:	=	artificial insemination
WHO:	=	World Health Organization
LDH:	=	lactate-dehydrogenase
Fe:	=	iron
Na:	=	sodium
Cu:	=	copper