Hydroxy-selenomethionine as an organic source of selenium in the diet improves boar reproductive performance in artificial insemination programs

Abstract

This study aimed to compare different selenium (Se) sources in the diet on boar’s semen quality and fertility. For this, 28 boars aged 8 to 28 mo were fed with the following dietary treatments for 95 d: 0.3 mg Se/kg as sodium selenite (SS; n = 14) and 0.3 mg Se/kg as hydroxy-selenomethionine (OH-SeMet; n = 14). During this period, two experiments were carried out. In experiment 1, the semen of all boars was evaluated every 2 wk. Raw semen was initially evaluated for the processing of seminal doses, which were stored at 17 °C for 72 h, followed by sperm quality assessments. Furthermore, Se concentration and glutathione peroxidase (GPx) activity were measured in the seminal plasma. In experiment 2, 728 females were inseminated weekly with seminal doses from boars of the different experimental groups to further assess in vivo fertility and litter characteristics. Results demonstrated that boars fed OH-SeMet had more Se in their seminal plasma (P < 0.05), showing the greater bioavailability of the organic source in the male reproductive system. Moreover, boars fed OH-SeMet tended (P < 0.10) toward a higher total sperm count in the ejaculate (66.60 vs. 56.57 × 10⁹ sperm) and the number of seminal doses (22.11 vs. 18.86; 3 × 10⁹ sperm/dose) when compared with those fed SS. No effect of the dietary treatments was observed on GPx activity in seminal plasma (P > 0.05) as well as on raw and stored semen quality (P > 0.05). Under in vivo conditions, seminal doses from boars fed OH-SeMet tended (P < 0.10) toward a higher pregnancy rate at weeks 3, 5, and 8, and also resulted in a higher (P < 0.05) percentage of pregnant females in the overall period (99.30 vs. 97.00). In conclusion, the replacement of SS with OH-SeMet in boars’ diet can improve sperm production and results in better reproductive performance for them, bringing greater productivity and profitability to artificial insemination centers and commercial pig farms.

Introduction

Boar spermatozoa are especially sensitive to oxidation caused by the overproduction of reactive oxygen species (ROS), free radicals that accompany sperm from its development until the moment of fertilization. The high proportion of polyunsaturated fatty acids present in its plasma membrane is an excellent substrate for lipid peroxidation (Cerolini et al., 2000), which reduces sperm viability and fertility. In this way, selenium (Se) could be an essential modulator of semen quality due to its well-known antioxidant properties when added in vitro (Khalil et al., 2019) or even as a dietary supplement, exercising its biological roles after incorporation into selenoproteins by the cells (Qazi et al., 2019).

In boars, the essential role of Se during spermatogenesis and sperm maturation, as well as its impact on fertility, was demonstrated at the end of the 20th century (Marin-Guzman et al., 1997, 2000a, 2000b). Since then, specific roles were established for the phospholipid hydroperoxide glutathione peroxidase (GPx4), a selenoprotein that performs structural and enzymatic functions during sperm development in mammalian (Ursini et al., 1999; Pfeifer et al., 2001; Flohé, 2007). Besides that, the well-known participation of glutathione peroxidase (GPx) as ROS scavenger in the seminal plasma and the essential role of Selenoprotein P as a Se transporter, and likely also as a protective factor in the seminal fluid, have been reported in boars and males of other species (Koziorowska-Gilun et al., 2011; Michaelis et al., 2014; Burk and Hill, 2015; Qazi et al., 2019).

Based on the literature, adult boars’ Se requirement corresponds to 0.3 mg/kg of diet (NRC, 2012). It is known that selenomethionine (SeMet) delivered by organic Se forms are more bioavailable when compared with inorganic forms. Therefore, dietary supplementations from different sources at the same levels do not correspond to the delivery and use of Se with the same efficiency (Burk and Hill, 2015). The main advantage for the inorganic source is the SeMet ability to build and maintain Se reserves in the body, mainly in muscle tissues, and as part of the amino acid chains which are more available under stress conditions or even during a temporary absence of Se from the diet (Surai and Fisinin, 2015).

In this way, some studies evaluated the possible improvements in replacing the most inorganic form used, sodium selenite (SS), with a source of SeMet as an active form of Se (selenium-enriched yeast; SY), on boar reproduction. Several studies demonstrated benefits in favor of SeMet regarding sperm production, semen quality, and conception and in vitro fertilization rates (Speight et al., 2012; Martins et al., 2014, 2018; Petrujkić et al., 2014; Estienne and Whitaker, 2017). Other studies suggested that both sources are equivalent (Lovercamp et al., 2013; Martins et al., 2015) or even that SeMet can reduce some sperm motility parameters as well as their resistance to oxidative stress (López et al., 2010). It might be noted that the concentration of SeMet in commercial SY products is quite variable depending on the manufacturing process, where a range of factors, including yeast strain, selenite concentration, medium composition, and temperature, affect it (Surai et al., 2018). Therefore, differences can be expected when comparing studies using this organic source.

Concomitant with the conduct of most of these studies, a pure synthesized organic Se form was launched on the market containing hydroxy-selenomethionine (OH-SeMet) as an active substance. Able to deliver >95% OH-SeMet in its molecule in comparison to 60% to 70% SeMet offered by SY, it has been identified as the most bioavailable organic form and so more efficient to supply the Se requirements of the animals (Surai and Fisinin, 2015; Surai et al., 2018). Furthermore, OH-SeMet has shown high stability under feed preparation and storage conditions, taking advantage of other pure Se forms such as SeMet or Zn-SeMet (Surai et al., 2018). Several studies have demonstrated the benefits of using OH-SeMet in farm animals, such as poultry and pigs. In general, the authors report a better tissue Se enrichment, besides finding higher Se transfer to the eggs in laying hens and better antioxidant status in weaned piglets when OH-SeMet was compared with SS or SY (Briens et al., 2013, 2014; Jlali et al., 2013, 2014; Couloigner et al., 2015; Chao et al., 2019). Recently, good results were also observed on litters from OH-SeMet-supplemented sows, including increased litter size, better transfer of passive immunity for piglets, and enhanced antioxidant capacity (Li et al., 2020; Mou et al., 2020).

Given its greater bioavailability and ability for building Se reserves, we hypothesize that boars fed an OH-SeMet-supplemented diet have greater Se availability to meet their reproductive functions, which might impact sperm production and quality, fertility ability, and litter size. Therefore, the present study is the first to evaluate the effects of OH-SeMet as a dietary supplementation for boars on raw semen characteristics and quality of seminal doses stored at 17 °C for 72 h as well as the reproductive performance achieved when used for artificial insemination (AI).

Materials and Methods

Experimental procedures were in agreement with the legal and ethical standards of the Ethics Committee on the Use of Animals at the School of Veterinary Medicine and Animal Science, University of São Paulo (Brazil), which approved this study under protocol 3955160419.

Animals and housing

A total of 28 purebred boars (Large-White and Landrace) and 728 purebred females (Large-White and Landrace) from DB Genética Suína were used for this study. Boars initiated the study aged 8 to 28 mo and the females with parities from 0 to 5. Boars were housed in a commercial AI center (Unit of Advanced Genetic Diffusion; DB Genética Suína, Minas Gerais, Brazil) and were managed under the same environmental, nutritional, and sanitary conditions. Boars were maintained in individual pens or crates in an environmentally controlled room (21 to 25 °C).

Females were housed in five farms, three of which were nucleus units and two multipliers (DB Genética Suína, Minas Gerais, Brazil). The farms presented different housing systems, feeders. Two farms had collective pens with electronic sow feeders in the gestation phase, while the others housed the females in individual crates, where animals were fed by automatic or manual feeders once a day. Regardless of the farm, females were housed in farrowing crates with automatic or manual feeders near the expected farrowing date. Three farms had temperature-controlled facilities (gestation and farrowing units), and two farms did not. All females received a specific diet for each reproductive phase (gestation, lactation), with the quantity supplied adjusted individually.

Experimental design

A previous ranking of 28 boars was based on the average semen quality (total sperm motility and sperm morphology) of two ejaculates per boar collected one per week. For this, a formula was used: Score = (3 × sperm major defects) + (2 × sperm minor defects) + (1 × sperm motility inverse) (CBRA, 2013; Martins et al., 2018). For animal ranking, the lower the animal’s score, the better its seminal quality was considered. In addition to boars’ score, their age and breed were used to provide a uniform distribution into the dietary treatments: 0.3 mg Se/kg as sodium selenite (SS, n = 14) and 0.3 mg Se/kg as hydroxy-selenomethionine (OH-SeMet, n = 14; Selisseo, Adisseo France S.A.S, Antony, France). The animals received the dietary treatments for 14 wk (95 d), and during this period, two experiments were carried out.

In experiment 1, the raw semen of all boars was collected weekly and evaluated every 2 wk for 14 wk. After semen collection, an initial analysis of raw semen was performed for processing seminal doses, while some aliquots were obtained and stored for further analysis. Seminal doses from boars were produced as routine in AI center, of which one of each boar was separated and kept at 17 °C for 72 h, and then evaluated for sperm quality, including physical and functional tests. Samples of the basal diet and seminal plasma were also collected during this period to assess the Se concentration.

In experiment 2, seminal doses from boars studied in experiment 1 were used weekly in AI programs throughout offering the dietary treatments (14 wk). At the end of the study, an extensive data analysis was carried out for reproductive parameters within each experimental group. Both experiments were carried out between April 29 and August 2 in 2019. In Brazil, this period corresponds to autumn and winter.

Preparation and provision of dietary treatments

Se from organic and inorganic sources was mixed with protein concentrate as a carrier and weighed 0.04 kg (premixtures). This amount was added daily to a basal diet (0.3 kg) and fed to the animals. Once it was consumed, the feed was supplied to complete a total of 2.7 kg/d. Boars were fed manually once a day in the morning. The basal diet was in according to the Nutritional Requirements of Swine (NRC, 2012), where Se requirements were obtained by the different dietary treatments. The Se levels of the basal diet were derived from the ingredients used (corn, soybean, and sugarcane yeast; Supplementary Table S1).

Semen collection and quality assessment

The whole ejaculate of boars was collected using the semi-automatic system BoarMatic (Minitüb GmbH, Tiefenbach, Germany). After this, it was evaluated for volume, sperm concentration, and total sperm motility, these latter being assessed by Computer-Assisted Sperm Analysis (CASA) system (AndroVision, Minitüb GmbH, Tiefenbach, Germany). Total sperm motility was defined as the percentage of spermatozoa with an amplitude of lateral head displacement (ALH) > 4 µm and a beat-cross frequency (BCF) > 4 Hz. With these semen characteristics known, it was possible to calculate the total sperm count in the ejaculate and the number of seminal doses produced. Furthermore, even before the initial analyses were performed, two aliquots of semen were collected and properly stored for further analysis of sperm morphology (4 °C) and GPx activity (−196 °C).

Sperm morphology

For sperm morphology assessment, a drop of semen previously fixed in buffered formaldehyde saline solution (4%) was deposited on a slide, covered by a coverslip, and analyzed by a differential interference contrast microscopy (Nikon, Eclipse NI-U model). In this evaluation, 200 sperm per sample were counted and classified according to sperm morphology as normal, or abnormal (defects in the acrosome, head, neck, midpiece, and tail regions; the presence of proximal and distal cytoplasmic droplets; and teratological forms) according to Brazilian College of Animal Reproduction (CBRA, 2013).

GPx activity

The method used to determine the GPx activity in semen samples was based on the consumption of nicotinamide adenine dinucleotide phosphate (NADPH), according to Nichi et al. (2006). In this method, the reaction between a hydroperoxide and reduced glutathione (GSH) is induced. This reaction is catalyzed by the GPx together with the enzyme glutathione reductase (GR) and causes the conversion of glutathione disulfide (GSSG—glutathione oxidized) to GSH, which in turn consumes NADPH (measured with a spectrophotometer). For this, raw semen was initially centrifuged at 2,400 × g for 5 min to obtain 100 µL of seminal plasma, volume used by sample evaluation. The assay mixture consisted of NADPH (0.12 mM, 1 mL), GSH (1 mM, 100 µL), GR (0.25 U/mL, 20 µL), and sodium azide (0.25 mM, 20 µL). The spectrophotometer cell was brought up to a volume of 1.9 mL with phosphate buffer 143 mM, ethylenediaminetetraacetic acid (EDTA) 6.3 mM (pH 7.5), which was also used to dissolve the NADPH. The GSH was initially dissolved in 5% metaphosphoric acid. Sodium azide was used to inhibit the action of catalase. This reaction was initiated with the addition of 1.2 mM of tert-butyl hydroperoxide (100 µL). The NADPH consumption was detected at a wavelength of 340 nm for 10 min at 37 °C (measurements performed every 5 s). The results of GPx were expressed as units of GPx/mL of semen, and calculations used 6.22 mM⁻¹ cm⁻¹ as the extinction coefficient of NADPH.

Processing and transport of seminal doses

After initial assessment of raw semen, it was extended in a long-term preservation extender (VITASEM, Magapor; Zaragoza, Spain) to compose the seminal doses (1.5 or 3 × 10⁹ sperm). Seminal doses produced were initially stored at 17 °C at the AI center facilities until transported to the commercialization routes. One seminal dose of each boar was delivered to the Laboratory of Andrology and Technology of Swine Embryos, University of São Paulo, where it was kept at 17 °C for the total storage period of 72 h before being analyzed (experiment 1). The remaining seminal doses were delivered to the different DB farms, where they were used up to 3 d of storage in AI programs (experiment 2). Seminal doses were transported in a refrigerated vehicle (17 °C), ensuring no temperature fluctuation and leaving the AI center within 24 h (for DB farms) and 48 h (for the University of São Paulo).

Sperm quality after storage at liquid state

Motility characteristics

Sperm motility characteristics from samples of 72 h stored seminal doses were analyzed by the CASA system (SCA, Microptic S. L., Barcelona, Spain) mounted on the epifluorescence microscope (Nikon, Eclipse NI-U). The characteristics evaluated were total motility (TMOT, %), progressive motility (PMOT, %), average path velocity (VAP, μm/s), straight-line velocity (μm/s), curvilinear velocity (μm/s), the ALH (μm), BCF (Hz), straightness (STR, %), linearity (%), and wobble coefficient (%). With the aid of the Edit/Sort tool offered by the software, the percentage of hyperactivated sperm in each sample was also evaluated (Andrade et al., 2017; Pavaneli et al., 2017).

Total sperm motility was defined as the percentage of spermatozoa that showed a VAP > 15 m/s, and progressive sperm motility was defined as the percentage of spermatozoa that showed an STR > 45%.

Sperm morphology

For sperm morphology assessment in stored seminal doses, the same method and sperm classification used for raw semen samples were applied.

The integrity of plasma and acrosomal membranes

For assessment of sperm membrane integrity, a previous dilution of the stored seminal dose was carried out in the TALP medium (Bavister et al., 1983) to obtain a final concentration equal to 25 × 10⁶ sperm/mL. An aliquot of 150 µL of this simultaneously received the addition of 2 µL of Hoechst 33342 (40 µg/mL), 3 µL of propidium iodide (PI, 0.5 mg/mL), and 50 µL of Pisum sativum agglutinin conjugated to fluorescein isothiocyanate (FITC-PSA, 100 µg/Ml; Andrade et al., 2007; Celeghini et al., 2007). Afterward, the samples were incubated at 37 °C for 8 min. For analysis, a drop of 8 µL was placed between a pre-heated slide and coverslip, with immediate reading under epifluorescence microscopy (Nikon, Eclipse NI-U model) at 1000×. The percentage of sperm with the intact plasma membrane (PI negative) and intact acrosome (FITC-PSA negative) was determined.

Sperm resistance to oxidative stress

Sperm susceptibility to lipid peroxidation was assessed by the thiobarbituric acid reactive substances (TBARS) assay, according to the methodology adapted by Nichi et al. (2006). This method indirectly quantifies the sperm resistance to oxidative stress by measuring malondialdehyde (MDA) concentration, an end product of lipid peroxidation. Firstly, samples were submitted to induction of lipid peroxidation through the incubation of 200 µL of semen with 50 µL of iron sulfate (4 mM) and 50 µL of ascorbic acid (20 mM) in the dry bath at 37 °C for 90 min. After induction, 600 µL of ice-cold trichloroacetic acid 10% was added to the sample following its storage (−20 °C) until the moment of analysis. Samples were then centrifuged at 20,800 × g for 15 min (5 °C) to precipitate proteins and debris. Then, 800 µL of the supernatant was recovered and incubated with 800 µL of thiobarbituric acid 1% at 95 °C in a water bath for 15 min. The reaction was stopped by placing samples on ice. In this assay, MDA and thiobarbituric acid react, producing a pink-colored complex, quantified in a spectrophotometer (Ultrospec 3300 Pro, Amersham Biosciences, USA) at a wavelength of 532 nm. The values obtained were compared with a standard curve for MDA concentration. The lipid peroxidation index was described in nanograms of TBARS/10⁶ sperm.

Se concentration in diet and seminal plasma

Samples of the premixtures containing the dietary treatments were checked for Se levels before the study. Samples of the basal diet were collected weekly throughout the experimental period (14 wk) and stored at −20 °C until laboratory analysis when a pool was used for Se quantification. Seminal plasma samples were collected before the experiment, during, and in the final period for the same assessment. To obtain this fluid, a small aliquot of raw semen was centrifugated at 2,400 × g for 5 min, and the supernatant was kept at −196 °C until the moment of use. For Se quantification, wet digestion was performed with a nitric-perchloric mixture of samples and subsequent reading by the fluorometric method of Olson et al. (1975) according to sensitization by diaminonaphthalene. For this assay, 0.5 g of basal diet and 2 mL of seminal plasma were used.

In vivo fertility assay

Seminal doses from 23 of the studied boars (SS, n = 11 and OH-SeMet, n = 12) were employed in AI programs during the experimental phase (14 wk), and a total of 728 females were inseminated: SS (n = 301) and OH-SeMet (n = 427). Each female was inseminated with seminal doses from a single male (homospermic AI) throughout estrus. According to needs, the commercial AI center chose boars to be used in the AI programs and the seminal doses’ distribution to the different farms. In the farms, gilts and sows were randomly inseminated according to each production unit (Supplementary Table S2). Sows were inseminated by post-cervical AI (1.5 × 10⁹ sperm; 45 mL) and gilts by the intracervical AI (3 × 10⁹ sperm; 90 mL). Only seminal doses with a maximum of 72 h of storage were used in the AI programs. Estrus detection was carried out twice daily. As soon as the signs of estrus were presented, females received the first AI immediately (gilts) or in the next period (sows). After that, they were inseminated at 24-h intervals until the end of estrus. Pregnancy diagnosis was carried out indirectly by the practice of estrus detection between days 17 and 25 after females were inseminated. The ratio between the number of inseminated females and the number of females considered pregnant corresponded to the pregnancy rate in each group studied. The number of piglets born in total, born alive, mummies, and stillborn was counted.

Averages of reproductive parameters were established for each experimental group (boars). Data were recorded from the first to the last day of the experimental phase.

Statistical analysis

Data were analyzed using the MIXED and GLIMMIX procedure (SAS 9.3; Institute Inc., Cary, NC, USA) according to a randomized block design with repeated measures. For the semen analysis and Se concentration, the analysis of variance (ANOVA) was conducted using the MIXED procedure of SAS. In these analyses, the interval between semen collections and the boars’ score were used as covariates, and the boars were considered random effects, and the treatment and the week were fixed effects. When the week factor was present, repeated measures in week were performed. Tukey–Kramer test was used to evaluate the effect of week.

For the pregnancy rate, parity and the boars’ score were used as covariates, and the farms were considered random effects and the treatment and the time as fixed effects. For litter size, parity, the boars’ score, and the total number of piglets born were used as covariates, and the farms were considered random effects and the treatment and the week as fixed effects. When the week factor was present, repeated measures in week were performed. In the statistical model, it was also considered the breed and the number of inseminations for each female for pregnancy rate and litter size variables.

Data from inseminated females that were culled or died during the study were removed before the pregnancy rate analysis. The pregnancy rate was considered a binary distribution and the percentage of stillborn, mummies, and teratological forms as a Poisson distribution. The binary and Poisson distributions were analyzed using the SAS GLIMMIX procedure.

Differences were considered significant when P < 0.05 and a statistical tendency when P < 0.10. All results were expressed as means and SEM.

Results

Only one variable among those evaluated showed a significant interaction between dietary treatment and week, while another one tended toward its effect. Interaction effect is illustrated in a figure format. For all other variables, the main effect analysis was performed. The tables exploring the week effect can be assessed in full in the Supplementary Data (Tables S3–S8). Although the interaction between dietary treatment and week was not significant for most of the variables studied in this study, some data showing how different treatments behaved over the weeks can also be assessed in the Supplementary Data (Tables S9–S11; Figures S1 and S2).

Sperm production and semen quality

An interaction between dietary treatment and week for sperm concentration was observed where boars fed OH-SeMet had higher values in the 8th week (P < 0.05) as well as in the 2nd, 4th, and 10th week (P < 0.10; Figure 1). Furthermore, these animals tended to have higher total sperm count in the ejaculate and number of seminal doses produced when compared with those fed a diet supplemented with SS (P < 0.10; Table 1). No effect of the dietary treatments was observed on volume, sperm TMOT, and morphology in raw semen (P > 0.05; Tables 1 and 2).

Figure 1.

Sperm concentration in raw semen of boars fed a diet supplemented with inorganic or organic selenium sources. Values represent the means and SEM. Asterisks indicate difference between dietary treatments at a given time point: *P < 0.05; **P < 0.10. Dietary treatments: SS, 0.3 mg selenium/kg as sodium selenite; OH-SeMet, 0.3 mg selenium/kg as hydroxy-selenomethionine.

Table 1.

Characteristics of raw semen of boars fed a diet supplemented with inorganic or organic sources of selenium for 95 d¹,²

	Dietary treatment, 0.3 mg Se/kg		SEM	P-value
	SS, n = 14	OH-SeMet, n = 14		T	W	T * W
Volume, mL	234.37	223.86	6.76	0.512	0.106	0.165
SC, × 106 sperm/mL	263.10	315.79	8.63	0.062	0.524	0.042
TSCE, × 109 sperm	56.57	66.60	1.73	0.090	<0.001	0.325
NSDP	18.86	22.11	0.58	0.094	<0.001	0.447
Total motility, %	88.21	89.16	0.36	0.400	0.155	0.599
GPx–SP, units/mL	38.61	42.59	1.92	0.290	<0.001	0.115
	(n = 9)	(n = 7)
Se–SP, ng/mL	44.70	52.38	1.22	0.027	<0.001	0.357

¹Values represent the means and SEM.

²SS, sodium selenite; OH-SeMet, hydroxy-selenomethionine; T, treatment; W, week; T * W, the interaction between treatment and week; SC, sperm concentration; TSCE, total sperm count in the ejaculate; NSDP, number of seminal doses produced per ejaculate (3 × 10⁹ sperm/dose); GPx–SP, glutathione peroxidase activity in seminal plasma; Se–SP, selenium concentration in seminal plasma.

Table 2.

Sperm morphology in raw semen of boars fed a diet supplemented with inorganic or organic sources of selenium for 95 d¹,²

	Dietary treatment, 0.3 mg Se/kg		SEM	P-value
	SS, n = 14	OH-SeMet,n = 14		T	W	T * W
Sperm defects, %
Acrosome	1.92	1.92	0.13	0.922	0.239	0.644
Head	1.47	1.23	0.08	0.406	0.143	0.730
Neck	0.07	0.02	0.01	0.124	<0.001	0.406
Middle piece	0.34	0.40	0.03	0.627	0.019	0.375
Proximal droplet	3.62	3.16	0.37	0.296	0.081	0.174
Distal droplet	2.03	1.58	0.12	0.185	0.212	0.182
Tail	19.51	18.27	0.74	0.983	0.009	0.895
Teratological forms	0.15	0.09	0.02	0.970	0.829	0.993
Normal sperm, %	62.53	66.40	1.25	0.880	0.003	0.817

¹Values represent the means and SEM.

²SS, sodium selenite; OH-SeMet, hydroxy-selenomethionine; T, treatment; W, week; T * W, the interaction between treatment and week.

For GPx activity in the seminal plasma, no dietary treatment affected this variable (P > 0.05; Table 1). Except for BCF, which was slightly higher for boars fed SS (P < 0.05), all other sperm motility characteristics evaluated in stored seminal doses were not affected by any Se source tested in this study (P > 0.05; Table 3). Also with regard to stored semen, no difference was observed for sperm morphology (P > 0.05; Table 4), resistance to oxidative stress (P > 0.05; SS: 1,767.59 vs. OH-SeMet: 1,911.74 ng TBARS/10⁶ sperm) and percentage of sperm with the intact plasma membrane and intact acrosome (P > 0.05; SS: 45.56 vs. OH-SeMet: 43.21). However, it was observed that regardless of the Se source used in the diet, the variables TMOT and PMOT and membrane integrity in stored seminal doses were improved at the end of the experiment compared with the first weeks (Supplementary Tables S6 and S8).

Table 3.

Sperm motility characteristics in stored seminal doses from boars fed a diet supplemented with inorganic or organic sources of selenium for 95 d¹,²

	Dietary treatment, 0.3 mg Se/kg		SEM	P-value
	SS, n = 14	OH-SeMet, n = 14		T	W	T * W
TMOT, %	93.19	92.66	0.36	0.521	<0.001	0.265
PMOT, %	69.07	65.19	0.90	0.157	<0.001	0.375
VCL, µm/s	48.27	46.07	0.64	0.558	<0.001	0.657
VSL, µm/s	31.25	29.11	0.37	0.185	<0.001	0.729
VAP, µm/s	37.97	35.81	0.47	0.320	<0.001	0.698
BCF, Hz	6.13	6.00	0.02	0.038	0.020	0.869
ALH, µm	1.87	1.88	0.02	0.572	<0.001	0.131
STR, %	82.54	81.66	0.24	0.191	<0.001	0.232
LIN, %	65.22	63.97	0.42	0.273	<0.001	0.329
WOB, %	79.03	78.33	0.30	0.389	<0.001	0.518
HYP, %	1.40	1.27	0.07	0.656	<0.001	0.615

¹Values represent the means and SEM.

²SS, sodium selenite; OH-SeMet, hydroxy-selenomethionine; T, treatment; W, week; T*W, the interaction between treatment and week; TMOT, total motility; PMOT, progressive motility; VCL, curvilinear velocity; VSL, straight-line velocity; VAP, average path velocity; BCF, beat-cross frequency; ALH, the amplitude of lateral head displacement; STR, straightness; LIN, linearity; WOB, wobble coefficient; HYP, hyperactive sperm.

Table 4.

Sperm morphology in stored seminal doses from boars fed a diet supplemented with inorganic or organic sources of selenium for 95 d¹,²

	Dietary treatment, 0.3 mg Se/kg		SEM	P-value
	SS, n = 14	OH-SeMet, n = 14		T	W	T * W
Sperm defects, %
Acrosome	15.31	16.85	0.52	0.233	<0.001	0.344
Head	1.53	1.32	0.08	0.341	0.012	0.482
Neck	0.15	0.14	0.02	0.829	0.185	0.293
Middle piece	0.28	0.24	0.03	0.725	0.035	0.663
Proximal droplet	2.19	2.50	0.24	0.715	0.270	0.156
Distal droplet	2.34	1.89	0.15	0.238	0.032	0.615
Tail	4.55	5.35	0.21	0.525	<0.001	0.351
Teratological forms	0.21	0.12	0.02	0.178	0.431	0.830
Normal sperm, %	63.39	67.91	1.05	0.258	<0.001	0.066

¹Values represent the means and SEM.

²SS, sodium selenite; OH-SeMet, hydroxy-selenomethionine; T, treatment; W, week; T * W, the interaction between treatment and week.

Se concentration

The result obtained from Se quantification in the basal diet was 0.133 mg Se/kg, which indicates that this was the basal concentration ingested equally by all animals, regardless of the dietary treatments. For seminal plasma, it was observed that boars fed the diet supplemented with OH-SeMet presented a higher concentration of Se in their fluid when compared to those fed SS (P < 0.05; Table 1). Furthermore, our results showed that regardless of the Se source used in the diet, Se concentration increased in seminal plasma over the weeks (Supplementary Table S4).

In vivo fertility assay

From the 728 females inseminated in this study, 701 were pregnant and farrowed (Supplementary Table S2). Of the 27 females removed, 15 were from SS treatment and 12 from OH-SeMet treatment. The major causes were abortion (SS, 1.99%; OH-SeMet, 2.11%) and regular return to estrus (SS, 2.99%; OH-SeMet, 0.70%).

As a result of the in vivo fertility assay, a tendency to an interaction between treatment and week for pregnancy rate was observed (P < 0.10; Table 5). At weeks 3, 5, and 8, there was a higher percentage of pregnant females using the seminal doses from OH-SeMet treatment than SS (P < 0.05; Figure 2). Furthermore, the highest pregnancy rate in the overall period was found using semen from boars fed the OH-SeMet treatment (99.30 vs. 97.00, P < 0.05; Table 5). Regarding the variables of litter size, there were no effects of treatment or week (P > 0.05; Table 5).

Table 5.

Reproductive performance of boars fed a diet supplemented with inorganic or organic sources of selenium for 95 d¹,²

	Dietary treatment, 0.3 mg Se/kg		SEM	P-value
	SS	OH-SeMet		T	W	T * W
Fertility	(n = 301)	(n = 427)
Pregnancy rate, %	97.00	99.30	0.01	0.031	0.113	0.069
Litter size 3	(n = 286)	(n = 415)
Number born alive	14.40	14.64	0.12	0.338	0.863	0.460
Total number born	16.04	15.82	0.13	0.512	0.836	0.993
Stillborn, %	7.11	5.87	0.30	0.106	0.617	0.303
Mummies, %	2.25	1.90	0.14	0.535	0.987	0.502

¹Values represent the means and SEM.

²SS, sodium selenite; OH-SeMet, hydroxy-selenomethionine; T, treatment; W, week; T*W, the interaction between treatment and week.

³Nonpregnant females, those who had a regular return to estrus or miscarried, were removed for analysis of litter size.

Figure 2.

Females’ pregnancy rate inseminated with seminal doses from boars fed a diet supplemented with inorganic or organic selenium sources. Values represent the means and SEM. Asterisks indicate the difference between dietary treatments at a given time point: *P < 0.05. Dietary treatments: SS, 0.3 mg selenium/kg as sodium selenite; OH-SeMet, 0.3 mg selenium/kg as hydroxy-selenomethionine.

Discussion

It is widely known that Se is essential for male reproduction, participating in events such as spermatogenesis and sperm maturation, and therefore acting as a modulator of seminal quality and fertility (Ahsan et al., 2014; Qazi et al., 2019). In this context, Se supplementation of the boar diet is already routinely performed in semen processing centers. However, given the intense regimen of semen collection for these animals, it has been supposed that a greater Se input could be necessary for ensuring more consistent sperm production, and so, organic Se forms could be the better choice due to their ability for building Se reserves in the body (Surai and Fisinin, 2015). In the search for increasingly efficient organic sources, several studies have demonstrated the advantages of offering OH-SeMet for pigs and poultry on their productivity (Briens et al., 2013, 2014; Jlali et al., 2013, 2014; Couloigner et al., 2015; Chao et al., 2019; Li et al., 2020; Mou et al., 2020). To our knowledge, this is the first study evaluating the effects of this organic Se form as dietary supplementation in boars on their semen production and quality, fertility, and litter produced. As expected, our work showed that the inclusion of OH-SeMet in the diet improves the Se availability for the male reproductive system and tends toward increasing sperm production while maintaining good semen quality before and after storage at 17 °C for 72 h. Furthermore, its use improves the reproductive performance of boars in already quite efficient AI programs about the pregnancy rate, although it does not impact litter size.

Organic Se forms (SeMet) have greater bioavailability than inorganic ones, which is explained by their better absorption via intestinal methionine transporters and the unique ability to enter the methionine pool in the body (Burk and Hill, 2015). In this way, the current study showed that Se concentration was greater in the seminal plasma of boars fed OH-SeMet when compared with those fed SS, different from that previously observed when the same type of biological sample was evaluated testing SY vs. SS, and no difference was found (López et al., 2010; Lovercamp et al., 2013; Martins et al., 2014). This result clearly shows that boars fed with OH-SeMet are better supplied with the mineral to meet their reproductive functions.

Once Se is more present in seminal plasma of boars fed OH-SeMet, and previous reports showed an increased GPx activity as a response to Se supply (Marin-Guzman et al., 1997), one could also expect more of this enzyme activity in the seminal fluid of these animals. However, our results did not show this direct relation between the factors, which was also observed by other authors testing SY when assessing the same parameters in the semen (Petrujkić et al., 2014). Previous work demonstrated that these two factors are not directly correlated in boar semen and are likely to be regulated by independent mechanisms (Lasota et al., 2004). It is important to note that not increasing the activity of this selenoprotein is not deleterious. On the contrary, the scenario shows us that there was probably no unbalance between the production of ROS and the antioxidant defenses (oxidative stress), a situation that would justify the increase in GPx activity. Herein, the results obtained for sperm resistance to oxidative stress (TBARS assay) in our study reinforce this assumption, where no difference was observed between the experimental groups for lipid peroxidation.

In addition to evaluating the Se availability for boars’ reproductive system, and its impact on the GPx activity in this environment, this study aimed to evaluate the direct effects of OH-SeMet on sperm production and the quality of raw and stored semen. Although our results did not find differences in semen quality, similar to that observed by other authors evaluating SY vs. SS (Lovercamp et al., 2013; Martins et al., 2014, 2015), they showed that at some timepoints over the study, boars fed OH-SeMet had higher sperm concentration than those fed SS, more significantly in the eighth week. Considering the duration of spermatogenesis + sperm maturation in boars (approximately 45 d), this would be the time when we would expect the main differences between treatments to be observed as we would be working with fully developed sperm within the experimental period (Parrish et al., 2017).

Besides that, boars fed OH-SeMet tended to have a higher total sperm count, which resulted in the production of +3.25 seminal doses for intracervical AI (3 × 10⁹ sperm) or even +6.5 units for post-cervical AI (1.5 × 10⁹ sperm) per ejaculate. Although this difference was not statistically significant here, it is undoubtedly a critical gain concerning the profitability of AI centers and the dissemination of superior genes. Our results are in accordance with López et al. (2010) and Martins et al. (2014, 2018) that published similar findings of the boar sperm concentration and total sperm count, respectively, while other studies did not demonstrate consistent differences in semen quantity parameters when SY was tested (Speight et al., 2012; Lovercamp et al., 2013; Petrujkić et al., 2014). In summary, our results allow us to suggest that sperm production was increased by the inclusion of OH-SeMet in the diet of boars and resulted in normal morphological cells and good quality semen by the tests performed in raw and stored samples.

Most of the variables analyzed in this study behaved differently over the weeks, regardless of the experimental group. Among them, it is worth commenting on the TMOT and PMOT and membrane integrity, assays widely used to assess sperm quality. Our work showed that such aspects were significantly improved in samples of seminal doses stored in a liquid state for 72 h when comparing the final period of the experiment with the first weeks of offering the treatments. It seems that regardless of the Se source offered to the boars, the demand for the mineral was met, at least to maintain good results in these analyses. It is important to remember that our study was carried out during the autumn–winter in Brazil, where seasonality seems to act in favor and not against the quality of boar semen, as in the spring–summer period (Zasiadczyk et al., 2015; Peña et al., 2019). Another factor is about the excellent conditions offered by the AI centers (environmentally controlled room; good nutrition and health). We believe that challenges were lacking so that the demands for Se by animals were exceeded, and the beneficial effect of the organic source OH-SeMet was maximized.

Our in vivo fertility and litter size results confirm the excellent semen quality of the boars fed OH-SeMet and demonstrate essential insights about its use on their reproductive performance. Although the interaction between treatment and week was only a tendency for pregnancy rate, it was observed that seminal doses from boars fed OH-SeMet resulted in more pregnancies at some time points (third, fifth, and eighth weeks). Furthermore, boars fed OH-SeMet resulted in a greater pregnancy rate in AI programs where this index was already very satisfactory, even with SS (99.3% vs. 97%). So, an increase of 2.3% in the pregnancy rate in this scenario makes us believe that OH-SeMet contributes in any way to better maintenance of sperm fertilizing ability after semen collection, handling, processing, transport, and storage in the liquid state. More applicable, this gain would correspond to approximately an extra 52 farrows, and more than 1,200 weaned piglets in 1 yr, considering the rates recorded in good Brazilian farms and the average number of born alive obtained in our study when semen came from boars fed with OH-SeMet. The percentages of abortion and regular return to estrus observed for both groups were within limits considered acceptable for commercial swine farms.

Naturally, a doubt can arise about how better reproductive performance observed for boars fed OH-SeMet in the current study can be explained if their semen quality was the same as those fed SS. Here, some points must be raised and need further investigation. First, is that the tests carried out in this study are far from thoroughly evaluating the sperm, and finding the same semen quality for both experimental groups does not always correspond to the same results for their fertility in the field (Andrade, 2021). Although the tests are essential to predict good semen quality, they can be many times insufficient to predict spermatozoa’s fertility ability (Oliveira et al., 2012; Andrade et al., 2017). Indeed, it is known that sperm can suffer functional damage during the first 72 h of storage, which cannot be assessed by these conventional tests, and other ones would help evaluate their ability for fertilization (Henning et al., 2012). These events could include in vitro capacitation and acrosome exocytosis (Yeste et al., 2015; Pavaneli et al., 2019, 2020) or even in vitro sperm–oviduct binding assays (Henning et al., 2019).

Also, a deeper and more complete analysis of seminal plasma could tell us which form of selenoprotein is increased in boars fed OH-SeMet, and from then on, it might better clarify its function and effect on their fertility. Indeed, there are 25 selenoproteins identified in mammalian and avian species (Kryukov et al., 2003; Sun et al., 2019). At this point, Selenoprotein P, widely known for being a Se transporter into the body, has been pointed out also as a potential and stable biomarker of seminal quality in men (Michaelis et al., 2014). As our study did not measure this specific selenoprotein in boar’s seminal plasma, it is difficult to affirm that it was important here. Nevertheless, other authors have demonstrated that OH-SeMet was able to modulate selenoprotein gene expression differently and their activity in broiler chicks, including Selenoprotein P (Zhao et al., 2017), which leads us to think that something similar might have happened in this study.

Another point is about sperm epigenetics, which has been significantly studied in the last decade, mainly in humans. This term means that in addition to delivering the haploid genome to the oocyte at the time of fertilization, the spermatozoa also transfer its DNA methylation profile, DNA-associated proteins, nucleo-protamine distribution pattern, and non-coding RNA established during spermatogenesis, with each factor able to influence fertility, embryo and placental development, and health of the offspring (Carrell and Hammoud, 2010; Wang et al., 2013; Champroux et al., 2018; Galan et al., 2020; Pedrosa et al., 2021). In this way, it is well-known that sperm epigenetics is constantly changed by environmental exposures and male lifestyle, like the diet consumed (Schagdarsurengin and Steger, 2016; Andrade, 2021). At this point, the real impact of a different Se source in the diet of boars on their sperm epigenetics landscape is unknown until the present. However, considering the roles of Se via selenoproteins on epigenetic regulation (Speckmann and Grune, 2015), our study can suggest that a protective role could occur due to a better Se support offered by OH-SeMet and then, an improvement of the reproductive performance of the animals in AI programs.

In conclusion, our results indicate that the use of OH-SeMet as dietary supplementation for boars can improve pig breeding in two scenarios when compared with the inorganic form SS: tends toward increasing boar sperm production and so, being interesting economically for semen processing centers, and also improving the pregnancy rate in AI programs in commercial pig farms. These two aspects contribute to the continuous search for better reproductive performance and greater dissemination of superior genes.

Glossary

Abbreviations

AI

artificial insemination

CASA

computer-assisted sperm analysis

GPx

glutathione peroxidas

GR

glutathione reductase

GSH

reduced glutathione

GSSG

glutathione disulfide

LIN

linearity

MDA

malondialdehyde

PI

propidium iodide

ROS

reactive oxygen species

TBARS

thiobarbituric acid reactive substances

Conflicts of interest statement

Two of the authors (G.V.F-N and N.S.F.) are employed by Adisseo Brasil Animal Nutrition (São Paulo, Brazil), who owns the product tested (Selisseo®).