Sixteen semen samples coming from separate healthy Piétrain boars (2–3 years old) were provided by a commercial farm (Servicios Genéticos Porcinos, S.L. Roda de Ter, Spain). Animals were housed under climate-controlled conditions, fed with a standard diet and provided with water ad libitum. Ejaculates were collected through the gloved-hand method by specialized staff from the farm. The obtained sperm-rich fractions were immediately diluted at 17°C to a final concentration of 2×10⁷ sperm/mL in a commercial extender (Duragen^®, Magapor, S.L.; Ejea de los Caballeros, Zaragoza, Spain). Diluted samples were then split into 90-ml artificial insemination doses; two of these doses were transported at 17°C to our laboratory. Transport of samples from the commercial farm to our laboratory lasted as much as 60 min and experimental procedures started upon arrival.

All experiments were performed using eight independent biological samples, each resulting from two separate seminal doses. These samples came from eight different boars, all from the Pietrain breed. Samples were first checked to fulfil the following quality standards: ≥80% membrane-intact spermatozoa (SYBR14⁺/PI⁻), ≥85% morphologically normal spermatozoa, and ≥80% total motile spermatozoa. Thereafter, samples were split into four separate 1.5-ml aliquots and were either incubated for 10 min at 20°C in the dark (Control) or irradiated with red LED light at a wavelength range of 620–630 nm (PhastBlue^®, IUL, S.L.; Barcelona, Spain) for 1 min (1′), 5 min (5′) or 10 min (10’). In all cases, the temperature within the PhastBlue^® system was maintained at 20°C. In addition to the aforementioned, samples were exposed to the same four protocols (i.e., control, or irradiation for 1, 5 or 10 min) in the presence of 100 nM of PKC 20–28^® (PKCi; Sigma Aldrich; St. Louis, MO, United States, catalog number 476480), 1 µM antimycin A (Sigma Aldrich, catalog number A8674), or both inhibitors together. Following this, motility; plasma membrane and acrosome integrity; mitochondrial membrane potential; intracellular levels of ROS, calcium and ATP; O₂ consumption rate; and activities of PKC and cytochrome C oxidase (CCO) were determined.

In the case of intracellular ATP levels and activities of PKC and CCO, samples were first centrifuged at 2,000×g and 20°C for 30 s and the resulting pellet was frozen in liquid N₂. Subsequently, samples were stored at -80°C until use, which did not last more than 1 month. For the evaluation of O₂ consumption, samples were also incubated with either 5 µM oligomycin A, a specific inhibitor of ATP synthase activity (Grüber et al., 1994), or 5 µM carbonyl cyanide-4 (trifluoromethoxy) phenylhydrazone (FCCP), an uncoupling agent of the electron chain (Grasmick et al., 2018), alone (control) or together with PKCi and/or antimycin A. The use of oligomycin A and FCCP aimed to gain further insights into how light stimulation affects O₂ consumption rate.

A first step was to determine total PKC activity to ascertain the ability of PKCi to induce its inhibitory effect. The description of this technique is provided in Supplementary File S1.

Because motility is one of the most relevant features of sperm, a careful analysis of this parameter was needed. As in the case of PKC activity, the description of this technique is shown in Supplementary File S1.

Flow cytometry analysis was performed using a CytoFLEX cytometer (Beckman Coulter, California, United States). Sperm viability (SYBR-14/PI), mitochondrial membrane potential (JC-1), total intracellular ROS levels (H₂DCFDA/PI), superoxides (HE/YO-PRO-1), acrosome membrane integrity (PNA-FITC/PI) and intracellular calcium levels (Fluo3-AM/PI) were evaluated. In all cases, sperm were previously diluted with pre-warmed PBS to a final concentration of 1×10⁶ sperm per mL before they were stained with the corresponding protocol. SYBR-14, Fluo3-AM, YO-PRO-1, PNA-FITC, H2DCFDA and JC-1 monomers were excited with the 488 nm laser and their fluorescence was detected by the FITC channel (525/40). HE and JC-1 aggregates were excited with the 488 nm laser and its fluorescence was detected through the PE channel (585/42). PI was excited with the 488 nm laser and its fluorescence was detected by the PC5.5 channel (690/50).

Sperm viability was determined by analyzing plasma membrane integrity following incubation with SYBR-14 (final concentration: 31.8 nM) and PI (final concentration: 7.6 µM) at 38°C for 10 min, as described by Garner and Johnson (1995). Sperm with an intact plasma membrane (SYBR14⁺/PI⁻) were distinguished from those showing different degrees of plasma membrane alterations (SYBR14^-/PI⁺ and SYBR14⁺/PI⁺).

Mitochondrial membrane potential was evaluated following a modified protocol from Ortega-Ferrusola et al. (2007). All samples were incubated with JC-1 (final concentration: 750 nM) at 38°C in the dark for 30 min. High mitochondrial membrane potential leads to the formation of JC-1 aggregates that emit orange fluorescence, whereas low mitochondrial membrane potential keeps JC-1 as monomers that emit green fluorescence.

Total ROS levels were determined through co-staining with H₂DCFDA (final concentration: 100 µM) and PI (final concentration: 6 µM) at 38°C for 20 min, as described by Guthrie and Welch (2006). The percentage and geometric mean of fluorescence intensity of DCF⁺ in the subpopulation of viable sperm with high ROS levels (DCF⁺/PI⁻) were recorded.

The assessment of intracellular superoxide levels was performed by co-staining with HE (final concentration: 5 µM) and YO-PRO-1 (final concentration: 31.25 nM) at 38°C in the dark for 20 min, following a modification of the protocol of Guthrie and Welch (2006). The percentage and geometric mean of fluorescence intensity of E⁺ in the subpopulation of viable sperm with high superoxide levels (E⁺/YO-PRO-1^-) were recorded.

Acrosome membrane integrity was evaluated following a procedure modified from Nagy et al. (2003). Sperm samples were incubated with PNA-FITC (final concentration: 1.17 µM) and PI (final concentration: 5.6 µM) at 38°C in the dark for 10 min. This procedure allowed the identification of four subpopulations. One of these subpopulations corresponded to viable sperm with an intact acrosome membrane (PNA-FITC^-/PI⁻), and the other three comprised those sperm cells that had different degrees of alteration in plasma membrane and acrosome membranes (PNA-FITC⁺/PI⁻, PNA-FITC^-/PI⁺ and PNA-FITC⁺/PI⁺).

Intracellular calcium levels of sperm were determined following the protocol described by Harrison et al. (1996) as modified by Yeste et al. (2015). Samples were incubated 10 min at 38°C with Fluo3-AM (final concentration: 1.2 µM) and PI (final concentration: 5.6 µM) at 38°C in the dark for 10 min. Four sperm subpopulations were identified: 1) viable sperm with low calcium levels (Fluo3^-/PI⁻); 2) viable sperm with high calcium levels (Fluo3⁺/PI⁻); 3) non-viable sperm with low calcium levels (Fluo3^-/PI⁺); and 4) non-viable sperm with high calcium levels (Fluo3⁺/PI⁺).

Not only were sperm stained with JC-1 analyzed through flow cytometry, but they were also observed under a Leica TCS 4D confocal laser scanning microscope (Leica Lasertechnik; Wertrieb, Germany) adapted to an inverted Leitz DMIRBE microscope at 63× (NA = 1.4 oil) Leitz Plan-Apo Lens (Leitz; Stuttgarts, Germany) in which the light source was an argon/krypton laser (74 mW). Observations were carried out in 4-well plates, each well containing 20 µL of sample. Fluorescence was excited at 488 nm and detected at 530 nm for green emission and at 590 nm for the orange/red one. In addition, motility of stained sperm was recorded through sequential tracking at a velocity of one capture/s for 20 s (Blanco-Prieto et al., 2020). All biological replicates (n = 8) and treatments were examined, with a minimum of 200 sperm per sample.

Activity of COO was analyzed in mitochondria-enriched sperm fractions, as described in Mclean et al. (1993). For this purpose, frozen pellets were solubilized in 500 µl of ice-cold PBS and then sonicated under the same conditions as those described for PKC activity. Thereafter, 500 µl of a 1.055 mg/ml Percoll^® solution in PBS at 4°C was placed onto each homogenate. Samples were subsequently centrifuged at 3,000×g and 10°C for 45 min. As a result, a mitochondria-enriched fraction appeared as a tight cloudy band that separated the aqueous phase from the Percoll^®-enriched one. This band was carefully collected with a micropipette and transferred into 1.5-ml tubes, which were centrifuged at 12,000×g and 20°C for 2 min. As a final step, resulting pellets were resuspended in 100 µL PBS at 20°C. Mitochondria-enriched suspensions were finally split into two separate aliquots. The first aliquot was used to determine CCO activity through spectrophotometry with a commercial kit (Cytochrome C Oxidase Assay Kit; Sigma-Aldrich; catalogue number CYTOCOX1). The other aliquot, which had a volume of 10 μL, was used to determine the total protein content through a commercial kit based on the Bradford method (Bradford, 1976), as described above. Enzyme activity in each sample was calculated after normalizing the obtained spectrophotometric values against the total protein content.

For determination of ATP levels, pellets prepared as described previously were thawed and resuspended in 300 µL of ice-cold 10 mM 2-[4-(2-hydroxyethyl)piperazin-1-yl]ethanesulfonic acid (HEPES) buffer containing 250 mM sucrose (pH 7.4), following the protocol set by Chida et al. (2012). Solubilized pellets were then sonicated with a Bandelin Sonopuls HD 2070 system (Bandelin Electronic GmbH and Co.) at a frequency of 10 kHz (20 pulses); samples were kept on ice at 4°C to avoid specimen heating. Following this, samples were centrifuged at 1,000×g and 4°C for 10 min and supernatants were collected. Twenty µl of each supernatant were used to determine the total protein content through the Bradford method (Bradford, 1976), using a commercial kit (Bio-Rad laboratories). The remaining volume was thoroughly mixed with 300 µl of ice-cold 10% (v:v) trichloroacetic acid, and then incubated at 4°C for 20 s. Subsequently, samples were centrifuged at 1,000×g and 4°C for 30 s, and the resulting supernatants were carefully separated from the pellet. These supernatants were again centrifuged at 1,000×g and 4°C for 10 min, obtaining a small, white pellet. The pellet was discarded, and the supernatant was mixed with two volumes of 1 M Tris-acetate buffer (pH = 7.75). Concentration of ATP in these resulting supernatants was determined through an Invitrogen^® ATP Determination Kit (ThermoFisher Bioscientific; Waltham, United States; catalogue number: A22066), following the manufacturer’s instructions. Samples were transferred into a 96-well microplate for fluorescence-based assays (Invitrogen^®), and ATP levels were measured in an Infinite F200 fluorimeter (TECAN^®; Männedorf, Switzerland). Data were normalized against the total protein content determined by the Bradford method.

The O₂ consumption rate was determined with a SensorDish^® Reader (SDR) system (PreSens Gmbh; Regensburg, Germany), as described in Blanco-Prieto et al. (2020). For this purpose, 1-ml aliquots were transferred into a specifically-designed Oxodish^® OD24 plates (24 wells/plate). Plates were sealed with Parafilm^® and then placed into the SDR device, which was introduced in an incubator at 38°C. The O₂ consumption was evaluated once a min for 2 h and, when finished, data were exported to an Excel file. The O₂ consumption rate was calculated as the decrease of O₂ concentration in each well during the 2-h incubation period, normalized against the percentage of viable sperm (SYBR14⁺/PI⁻) determined by flow cytometry.

As mentioned in the Experimental design section and to delve further into how irradiation with red light in the presence of antimycin A and/or PKCi affects O₂ consumption, sperm were also incubated with 5 µM oligomycin A, a specific ATPase inhibitor, or 5 µM FCCP, a specific disruptor of the electron chain, as described in Blanco-Prieto et al. (2020). The inclusion of these additional samples allowed us to determine the following parameters (Supplementary Figure S1): 1) non-mitochondrial O₂ consumption (NMC), which corresponded to O₂ consumption recorded in the presence of 1 µM antimycin A; 2) basal respiratory rates (BR), which were the values obtained in the absence of any inhibitor/disruptor; 3) oligomycin-resistant O₂ consumption (OC), which corresponded to samples incubated with 5 µM oligomycin A; 4) Maximal respiration (MR), which corresponded to samples incubated with 5 µM FCCP; 5) H⁺ leakage (HL): OC-NMC; 6) ATP turnover (AT): BR-HL; and 7) Spare respiratory capacity (SRC): MR-BR. Total O₂ consumption rate resulted from the sum of NMC, BR and SRC, and, for all experimental conditions, the percentage of each of these components was calculated.

Statistical analyses were conducted with a statistical package (SPSS^® Ver. 25.0 for Windows; IBM corp., Armonk, NY, United States). The first step was to test normality and homogeneity of variances through Shapiro-Wilk and Levene tests, respectively. When required, data were linearly transformed with √x or arcsin √x. The effects of light-stimulation and the presence/absence of antimycin A and PKCi on all variables (motility, plasma membrane and acrosome integrity, MMP, calcium levels, total ROS, superoxides, activities of PKC and CCO, ATP, and O₂ consumption rate, including NMC, BR and SRC) were determined through a two-way analysis of variance (ANOVA) followed by post-hoc Sidak test. In this model, factors were: 1) light-stimulation pattern (i.e. non-irradiation, irradiation for 1 min, irradiation for 5 min, and irradiation for 10 min); and 2) presence/absence of inhibitors (control, antimycin A and PKCi, as well as oligomycin A and FCCP in the case of O₂ consumption rate analysis).

On the other hand, motile sperm subpopulations were also determined as described in Luna et al. (2017). Kinematic parameters corresponding to each sperm cell (VCL, VSL, VAP, BCF, ALH, DNC, absMAD and algMAD) were used to run a Principal Component Analysis (PCA). The obtained matrix was rotated through the Varimax method and normalized with the Kaiser approach. The regression scores assigned to each sperm cell for the new PCA components were used to run a two-step cluster analysis (log-likelihood distance; Schwarz’s Bayesian Criterion). Three motile sperm subpopulations were identified (SP1, SP2 or SP3) and the proportions of each of these subpopulations were used to run a two-way ANOVA and post-hoc Sidak test.

In all analyses, the level of significance was set at p ≤ 0.05.

Neither light-stimulation, regardless of the pattern, nor the presence of antimycin A modified PKC activity. More details are provided in Supplementary File S1.

Total and progressive sperm motility were not affected by irradiation with red light; these results are further described in Supplementary File S1. Regarding the structure of motile sperm subpopulations, three sperm subpopulations (SP) were identified, and their descriptive parameters are shown in Supplementary Table S2. These SPs were named in a descending order according to their VCL values. Proportions of SP1 (the fastest, based on VCL), SP2 and SP3 (the slowest) in non-irradiated samples without any inhibitor were 28.6% ± 8.1%, 51.4% ± 6.7% and 20.1% ± 8.1%, respectively (Figure 1). While these proportions remained unaffected when samples were irradiated for 1 min, those of SP1 increased and those of SP2 decreased when they were exposed to red light for 5 min or 10 min (p < 0.05; Figure 1). Remarkably, pre-incubation of sperm with PKCi abolished the aforementioned increase in the proportions of SP1 observed when samples were irradiated for 5 min or 10 min (Figure 1).

Irradiation of sperm with red LED light did not affect plasma membrane or acrosome integrity. More information about the evaluation of these two sperm variables is provided in Supplementary File S1.

The percentage of sperm with high MMP, which in non-irradiated control samples was 71.0% ± 4.6%, did not vary when sperm were pre-incubated with PKCi (Figure 2A). Light irradiation induced a significant (P˂0.05) decrease in the percentage of sperm with high MMP at all tested irradiation times (Figure 2A). Preincubation with PKCi did not significantly modified the light-induced effect. Otherwise, pre-incubation with antimycin A, either alone or together with PKCi, led to an almost complete reduction of the percentage of sperm with high MMP in both control and irradiated samples, regardless of the irradiation time (Figure 2A).

In the absence of inhibitors, the ratio between JC-1 aggregates and JC-1 monomers (JC-1agg/JC-1mon) in the sperm population with high MMP significantly (p < 0.05) decreased after irradiation for 1 min (non-irradiated, control samples: 0.56 ± 0.01 vs. irradiation for 1 min: 0.77 ± 0.03; see Figure 2B). In both non-irradiated and irradiated samples, pre-incubation with antimycin A, either alone or together with PKCi, also induced a significant (p < 0.05) decrease in this ratio (Figure 2B). In contrast, pre-incubation with PKCi had no effect, either in non-irradiated or irradiated samples (Figure 2B).

When sperm stained with JC-1 were examined under a confocal laser scanning microscope, orange-stained mitochondria were mainly located at the two mid-piece poles in control samples and in those pre-incubated with PKCi (Supplementary Figure S4). Conversely, sperm pre-incubated with antimycin A did not show orange-stained mitochondria, but all were uniformly stained in green (Supplementary Figure S4). No irradiation pattern modified these specific distributions (data not shown).

As shown in Supplementary Figure S5, the percentage of viable sperm with high calcium levels was low in non-irradiated, control samples (0.55% ± 0.20%). Neither irradiation nor antimycin A or PKCi affected that percentage or the geometric mean intensity of Fluo3⁺ in the subpopulation of viable sperm with high calcium (Fluo3⁺/PI⁻; Supplementary Figure S5B).

The percentage of viable sperm with high total ROS levels in non-irradiated, control samples was 15.1% ± 8.4% (DCF⁺/PI⁻; Figure 3A). No significant differences in this percentage or in the geometric fluorescence intensity of DCF⁺ in the subpopulation of viable sperm with high ROS were observed following irradiation or when samples were pre-incubated with PKCi. Sperm pre-incubated with antimycin A, however, and irradiated for 5 min or 10 min showed a significantly (p < 0.05) lower geometric mean fluorescence of DCF⁺-intensity than those irradiated for 1 min (Figure 3B).

The percentage of viable sperm with high superoxide levels was very low in non-irradiated, control sperm, and lower than that of viable sperm with total ROS (1.6% ± 0.6%; Figure 3C). While irradiation did not induce a significant change in this parameter, percentages of viable sperm with high superoxide levels in samples pre-incubated with antimycin A were significantly (p < 0.05) lower after irradiation for 1 min or 10 min than when not exposed to light (Figure 3C). Finally, the geometric mean fluorescence of E⁺-intensity in the subpopulation of viable sperm with high superoxide levels (E⁺/YO-PRO-1^-) was not affected by irradiation patterns or the presence of antimycin A or PKCi (Figure 3D).

As shown in Figure 4A, irradiation of sperm for 5 min significantly (p < 0.05) increased CCO activity (irradiation for 5 min: 65.2 mIU/mg protein±19.1 mIU/mg protein vs. non-irradiated, control sperm: 28.1 mIU/mg protein±9.6 mIU/mg protein). Yet, this increase was abolished when samples were previously incubated with PKCi. Moreover, no changes in CCO activity were observed in samples pre-incubated with antimycin A. Otherwise, neither light-stimulation nor antimycin A or PKCi altered intracellular ATP levels, which in non-irradiated, control samples were 15.0 nmol/mg protein±2.7 (Figure 4B).

Irradiation of sperm for either 5 min or 10 min induced a significant (p < 0.05) increase in total O₂ consumption rate (non-irradiated, control samples. 1.11 nmol O₂×h/10⁶ viable sperm±0.06 nmol O₂×h/10⁶ viable sperm vs. irradiation for 10 min: 2.09 nmol O₂×h/10⁶ viable sperm±0.04 nmol O₂×h/10⁶ viable sperm; Figure 5A, Figure 6). This increase in total O₂ consumption rate was based upon significant (p < 0.05) increases in basal and maximal levels of respiration, and ATP turnover rate, which were concomitant with a significant (p < 0.05) decrease in the reserve capacity (Figures 5B–D). In contrast, light-stimulation did not affect H⁺ leakage or non-mitochondrial O₂ consumption (Figure 5E). On the other hand, while pre-incubation with antimycin A, either alone or in combination with PKCi, almost suppressed mitochondrial O₂ consumption, this inhibition was not counteracted by red light irradiation (Figure 6). Pre-incubation with PKCi significantly (p < 0.05) increased O₂ consumption rate in non-irradiated samples (PKCi: 1.54 nmol O₂×h/10⁶ viable sperm±0.04 nmol O₂×h/10⁶ viable sperm vs. control: 1.11 nmol O₂×h/10⁶ viable sperm±0.06 nmol O₂×h/10⁶ viable sperm; Figure 6). Pre-incubation with PKCi, nevertheless, prevented the aforementioned increase in total O₂ consumption observed after irradiation for 5 min or 10 min (Figure 6). This effect occurred together with a decrease in SRC without affecting AT (Supplementary Figure S6A). When sperm were pre-incubated with PKCi and were subsequently irradiated for 1 min or 5 min, a decrease in BR and an increase in SRC were observed (p < 0.05; Supplementary Figure S6B). These effects were less apparent when samples were irradiated for 10 min, as only the increase in SRC was significant (p < 0.05; Supplementary Figure S6A). Finally, neither PKCi nor antimycin A had an impact on NMC (Supplementary Figure S6C).