Divergent effect of fast- and slow-releasing H2S donors on boar spermatozoa under oxidative stress

Hydrogen sulphide (H2S) is involved in the physiology and pathophysiology of different cell types, but little is known about its role in sperm cells. Because of its reducing properties, we hypothesise that H2S protects spermatozoa against the deleterious effects of oxidative stress, a condition that is common to several male fertility disorders. This study aimed i) to determine the total antioxidant capacities of Na2S and GYY4137, which are fast- and slow-releasing H2S donors, respectively, and ii) to test whether H2S donors are able to protect spermatozoa against oxidative stress. We found that Na2S and GYY4137 show different antioxidant properties, with the total antioxidant capacity of Na2S being mostly unstable and even undetectable at 150 µM. Moreover, both H2S donors preserve sperm motility and reduce acrosome loss, although the effects were both dose and donor dependent. Within the range of concentrations tested (3–300 µM), GYY4137 showed positive effects on sperm motility, whereas Na2S was beneficial at the lowest concentration but detrimental at the highest. Our findings show that Na2S and GYY4137 have different antioxidant properties and suggest that both H2S donors might be used as in vitro therapeutic agents against oxidative stress in sperm cells, although the optimal therapeutic range differs between the compounds.

. Total antioxidant capacity and stability of the H 2 S donors Na 2 S and GYY4137. The H 2 S donors were incubated in phosphate-buffered saline solution at 38 °C in a water bath. Total antioxidant capacity is expressed as Trolox equivalents (mM). White histograms: Na 2 S; grey histograms: GYY4137. Upper panel: 20 minutes of incubation; middle panel: 120 minutes of incubation; lower panel: 210 minutes of incubation. Different letters indicate significant differences (p < 0.05) between H 2 S donors at the same concentration and incubation time. Data are shown as the mean±standard error of four replicates. and infertility 16,17 . Previous studies have shown that H 2 S is able to alleviate the effects of oxidative stress on testicular functions 4,8,18 , but the in vitro effects of this gasotransmitter on sperm cells under a ROS-generating system still need to be elucidated.
The aim of this study was to evaluate the total antioxidant capacity and stability of the H 2 S donors Na 2 S and GYY4137 under standard conditions (38 °C, pH≈7) and at different times (i.e. 20, 120, and 210 minutes) during the incubation (experiment I). Because sperm motility under a ROS-generating system may drop in a few hours 19,20 and based on the opposite modalities of H 2 S release by Na 2 S and GYY4137 (i.e. fast and slow release, respectively), these incubation times were chosen to determine the dynamics of the antioxidant activity of each donor during the early, mid, and late stages of incubation. Based on the results from experiment I and the physiological total antioxidant capacity of boar seminal plasma 21 , we then established a suitable range of concentrations of Na 2 S and GYY4137 to be tested in boar sperm samples under a ROS-generating system (experiment II). Although some Na 2 S and GYY4137 concentrations used in experiment II show a total antioxidant capacity that is below the range of detection by spectrophotometry, they were included in our experimental design because increasing evidence suggests that in vivo H 2 S levels range from low µM to high nM 22 . Next, we evaluated the effects of both donors on sperm motility, mitochondrial activity, plasma membrane integrity, acrosomal status, and lipid peroxidation. The results from this study elucidate the role of H 2 S donors in sperm samples under oxidative stress and the possible therapeutic implications of these compounds for alleviating the negative effects of ROS on sperm function.

Results
Experiment I. Total antioxidant capacity and stability of H 2 S donors. As can be seen from Table 1, Na 2 S and GYY4137 showed different total antioxidant capacities and stabilities during the incubation. Overall, the total antioxidant capacity of Na 2 S significantly decreased between 20 and 210 minutes of incubation, whereas that of GYY4137 tended to increase during this period and was significantly higher after 210 minutes than after 20 minutes of incubation at 2,400 and 1,200 µM (p < 0.05). Moreover, GYY4137 showed detectable levels of total antioxidant capacity at all concentrations tested, whereas Na 2 S was unstable within the range of 300 to 1,200 µM and was undetectable at 150 µM. Irrespective of the concentration considered, GYY4137 showed greater total antioxidant capacity than Na 2 S (Fig. 1).

Experiment II. Effect of H 2 S donors on boar sperm parameters under induced oxidative stress. Sperm motility.
Overall, the effects of H 2 S on boar sperm motility under induced oxidative stress were dose-and donor-dependent (Table 2, Fig. 2). Thus, 3 and 30 µM GYY4137 and 3 µM Na 2 S preserved the sperm motility and kinetics under the ROS-generating system. Interestingly, all of the sperm kinetic parameters in these treatments did not differ from those of the control group without oxidative stress (CTR; p > 0.05). The results with both 3 and 30 µM GYY4137 showed higher percentages of total motility (TM) relative to that of the control group under oxidative stress (CTR-ox; p < 0.01). Moreover, a dose of 30 µM GYY4137 significantly increased the percentage of progressive motility (PM) over that in the CTR-ox group (p = 0.040). Although differences were not statistically significant, higher average path velocity (VAP) and straight-line velocity (VSL) were observed in samples treated with 3 µM GYY4137 than in those in the CTR-ox group (p = 0.071 and p = 0.064, respectively). On the other hand, the effects of Na 2 S were markedly dose dependent. At 300 µM, this fast-releasing H 2 S donor showed clear negative effects on sperm motility. No motile spermatozoa were observed in any replicate; therefore, no kinetics data could be provided. By contrast, at the lowest concentration, Na 2 S significantly increased the percentage of motile sperm cells relative to that in the CTR-ox group (p = 0.018). At a concentration of 30 µM, Na 2 S greatly decreased the TM, PM, VAP, and VSL (p < 0.05), although it did not affect the curvilinear velocity (VCL) and the remaining motion parameters in comparison with those of the CTR-ox group (p > 0.05). However, at this Na 2 S concentration, we observed some variability among the replicates with the percentage of motile spermatozoa ranging from 0 to almost 30%. There were no differences between the CTR-ox and H 2 S donor treatments in the amplitude of lateral head displacement (ALH), beat-cross frequency (BCF), linearity (LIN), straightness (STR), and wobble (WOB; p > 0.05).
Sperm mitochondrial status. At a concentration of 300 µM, Na 2 S showed clear negative effects on boar sperm mitochondrial status. In every replicate, there were no spermatozoa with active mitochondria (Table 3, Fig. 3). There were no differences between the CTR-ox group and the remaining treatment groups (p > 0.05).
Sperm plasma membrane integrity and lipid peroxidation. There was no significant effect of GYY4137 on sperm plasma membrane integrity at any of the concentrations used (p > 0.05, Table 3). On the other hand, 300 µM Na 2 S markedly impaired the plasma membrane integrity relative to the results obtained for the CTR-ox group (p < 0.001, Table 3, Fig. 3). No effects were observed at the remaining Na 2 S concentrations (p > 0.05). A similar pattern was observed for the sperm lipid peroxidation: higher values of malondialdehyde (MDA) per 10 8 spermatozoa were observed in samples treated with 300 µM Na 2 S than those in the CTR-ox group and the other treatment groups (p < 0.01, Fig. 4). No differences in the MDA levels were found between the CTR-ox group and the remaining treatment groups (p > 0.05).
Acrosomal status. We found that 300 µM Na 2 S impaired the acrosome integrity (normal apical ridge or NAR test) relative to that of the CTR group, both with and without oxidative stress (p < 0.001, Table 3, Fig. 3). No differences were observed in the NAR test results between the CTR-ox group and the other treatment groups (p > 0.05). On the other hand, all treatments showed lower percentages of acrosome-lost spermatozoa (evaluated with peanut agglutinin-fluorescein isothiocyanate, PNA-FITC) than that of the CTR-ox group (p ≤ 0.001, Table 3).

Discussion
In this study, we provide the first evidence, to the best of our knowledge, that Na 2 S and GYY4137 show different total antioxidant capacities and stabilities under standard conditions (38 °C, physiological pH) and after different periods (20, 120, and 210 minutes) of incubation. Our results also reveal that the total antioxidant capacity of Na 2 S is less stable than that of GYY4137, although the latter's total antioxidant capacity tends to increase over time. This phenomenon should be taken into account in studies entailing cell incubation at 38 °C and at physiological pH, because the release of H 2 S by GYY4137 is both pH and temperature dependent 12 . Moreover, GYY4137 shows higher total antioxidant capacity than that of Na 2 S after any incubation time, with the total antioxidant capacity of Na 2 S at 150 µM even being undetectable by spectrophotometry after 20 minutes of incubation. The patterns observed in the antioxidant capacities of these H 2 S donors may not reflect their H 2 S release, given that the inorganic salts Na 2 S and NaHS lead to a larger but shorter gas release than GYY4137 23,24 .
Our results show that both H 2 S donors partly palliate the damages provoked by oxidative stress in sperm cells, supporting the results found previously in other cells (neurons 13,25 , gastric cells 14 , testicular germ cells 4 ) and organs (lungs 15,26 , testes 18 ). In these studies, oxidative stress was induced by in vivo or in vitro ROS-generating systems, such as glutamate, hydrogen peroxide (H 2 O 2 ), heat, tobacco smoke, ovalbumin sensitisation, and ischaemia-reperfusion injury. In our study, we used Fe 2+ /ascorbate, which induces lipid peroxidation and catalyses the production of hydroxyl radicals ( • OH), the most powerful free radical known, by the Fenton reaction 27  confirm that the effects of H 2 S donors are not only dose but also donor dependent. Moreover, for the first time to the best of our knowledge, the in vitro effects of fast-and slow-releasing H 2 S donors were evaluated in sperm cells in the presence of a ROS-generating system. Overall, we found that the slow-releasing H 2 S donor GYY4137 did not show any cytotoxic effect. Moreover, 3 and 30 µM GYY4137 preserved the boar sperm motility against the detrimental effects of oxidative stress. Interestingly, at both concentrations, the percentage of motile sperm cells was almost 50% higher than that of the CTR-ox samples and no kinetic parameters differed from those of the CTR group without oxidative stress. Moreover, 30 µM GYY4137 samples showed a higher percentage of progressive motility than those in the CTR-ox group. However, in contrast to other studies 8,18 , we did not find any effect of GYY4137 on the levels of lipid peroxidation, which may be due to the different cell type and donor concentration used in our study. Our findings also show that, in comparison with GYY4137, the effects of Na 2 S were markedly dose dependent. At a concentration of 3 µM, this fast-releasing H 2 S donor preserves boar sperm motility (40.5% higher than that in the CTR-ox samples), whereas it shows clear cytotoxic effects at 300 µM. This www.nature.com/scientificreports www.nature.com/scientificreports/ pattern may reflect the well-known biphasic biological dose-response curve of H 2 S: it acts as an antioxidant at low concentrations, but a pro-oxidant at high concentrations 29 . Interestingly, we also found that 30 µM Na 2 S was not cytotoxic (i.e. no effect was observed on the mitochondrial activity, plasma membrane integrity, acrosome integrity, or lipid peroxidation), but it strongly inhibited sperm motility. As in our findings, Zhao et al. found that 25 µM Na 2 S inhibits boar sperm motility although it does not affect the viability and mitochondrial membrane potential 9 . Likewise, NaHS has been shown to inhibit the motility of mouse and human spermatozoa 8,30 . In our study, the inhibitory effects of Na 2 S on sperm motility were described by using a comprehensive set of kinetic parameters provided by Computer-Assisted Sperm Analysis (CASA). Overall, 30 µM Na 2 S remarkably decreased the sperm motility, VAP, and VSL, but it did not affect the remaining motion parameters. In spite of some variability among replicates, a small percentage of sperm cells showed very slow but progressive movement. With regard to the mechanism of action, Zhao et al. found that the inhibitory effects of Na 2 S on sperm motility are related to the adenosine 5′-monophosphate-activated protein kinase (AMPK) and protein kinase B (AKT) pathways 9 , whereas a more recent study revealed that H 2 S promotes the secretion of K + in the epididymis; this suppresses sperm motility and may contribute to keeping the sperm cells in a quiescent state before ejaculation 5 . Another plausible explanation for the reduced sperm motility elicited by Na 2 S can be provided by the inhibitory effect of H 2 S on cytochrome c oxidase (complex IV), the final component of the electron transport chain that plays a key role in aerobic respiration and adenosine triphosphate (ATP) generation 29 . Because mammalian spermatozoa devote most of the energy generated as intracellular ATP to motility 31 , a reduction in ATP levels would lead to an immediate decrease in sperm motility. In this way, for instance, cardiac cells treated with 25 µM Na 2 S showed over 80% decreased O 2 consumption rate relative to the baseline 32 . In contrast to the effects observed at low concentrations, high concentrations of Na 2 S showed clear detrimental effects: no motility or active mitochondria were observed in any sample treated with 300 µM Na 2 S. The toxic effects were acute and led to immediate failure of sperm motility and mitochondrial activity (personal observations). This phenomenon can be explained by the high levels of ROS induced by Na 2 S 9 , an explanation corroborated by the increased levels of lipid peroxidation and the low percentage of sperm with intact plasma membrane and acrosome (NAR) found in our study. In contrast to our findings, in testicular germ cells, Li et al. found that treatment with NaHS in the concentration range of 1-200 µM for 30 minutes does not affect cell viability, although cell injuries are induced at 1 mM 4 . This result might be because, unlike other male germ cells, spermatozoa have limited antioxidant defences. Spermatozoa lack the necessary cytoplasmic-enzyme repair systems, and their membranes are particularly rich in polyunsaturated fatty acids, which make these cells particularly susceptible to the damage caused by oxidative stress 16 . Another reason might be related to the use of open systems (e.g. four-well plates and Petri dishes) that lead to quick volatilisation of the gas during cell incubation 33 . It is known that Na 2 S volatilises very quickly 33 , and the use of closed systems, such as the one used in our study, delays this process and leads to more consistent results regarding the effects of H 2 S donors on cell biology. Our results also show that Na 2 S and GYY4137 partly reduce the damage induced by high ROS levels on acrosomal status by decreasing the percentage of acrosome-lost (PNA-FITC) spermatozoa relative to that in the CTR-ox group. However, neither of the two donors showed any positive effect on the acrosome integrity evaluated by the NAR test. This divergence could be explained by the different acrosomal attributes considered by these two methodologies: NAR evaluated in unstained samples by phase-contrast microscopy versus outer acrosomal membrane integrity assessed by PNA-FITC staining with epifluorescence microscopy 34,35 . Because acrosome integrity is a requisite for fertilisation and the ROS levels affect sperm function 36,37 , it remains to be tested whether Na 2 S and GYY4137 may preserve the fertilising potential of sperm cells under oxidative stress. Semen handling and storage decrease boar sperm quality and fertilising ability, probably because of oxidative stress 38,39 , so H 2 S donors may be useful for the optimisation of semen extenders used in artificial insemination programmes. In the porcine industry, higher efficiency of artificial insemination outcomes may have major economic implications because this assisted reproductive technology is the main tool for pig gene dissemination worldwide 40 . www.nature.com/scientificreports www.nature.com/scientificreports/ Several mechanisms are involved in the ROS scavenging properties of H 2 S. One of these mechanisms involves H 2 S itself by virtue of its reducing properties 13 . In this way, the total antioxidant capacities of Na 2 S and GYY4137 were tested in this study based on the compounds' abilities to reduce 2,2′-azino-bis(3-et hylbenz-thiazoline-6-sulfonic acid) (ABTS) previously oxidised with H 2 O 2 41 . Another mechanism that may explain the capacity of H 2 S donors to palliate the damages provoked by oxidative stress concerns the enhancement of the cellular antioxidant defences. Previous studies found that H 2 S increases glutathione synthesis, decreases ROS production, and stimulates the activities of superoxide dismutase, glutathione peroxidase, and glutathione reductase 4,9,13,25,26 . Moreover, H 2 S decreases the apoptosis rate, increases the protein expression ratio of Bax/Bcl-2, and stimulates Cyp19 gene expression, among other effects 4,18,42 . Because spermatozoa are transcriptionally inactive 43 , it is likely that H 2 S donors protect sperm cells against ROS damage through their proper reducing activity, as well as by increasing the ratio of reduced to oxidised glutathione and stimulating superoxide Figure 3. Detrimental effects of a high concentration of Na 2 S, a fast-releasing H 2 S donor, on the boar sperm mitochondrial status and the plasma membrane and acrosome integrities (normal apical ridge) under oxidative stress. Representative images of sperm cells assessed by epifluorescence microscopy (mitochondrial status was assessed by using rhodamine 123 and propidium iodide; plasma membrane integrity was assessed by using carboxyfluorescein diacetate and propidium iodide) or phase-contrast microscopy (the normal apical ridge was assessed after fixation with glutaraldehyde). dismutase and glutathione peroxidase activities, which represent the major antioxidant defence system of sperm cells 44 . Further studies are nevertheless required to investigate the mechanisms of action of H 2 S donors in sperm cells under oxidative stress.
In conclusion, our study provides evidence about the antioxidant properties of two H 2 S donors, Na 2 S and GYY4137; this evidence will be useful for future studies aiming to test the antioxidant effect of this gasotransmitter. Our findings clearly show that Na 2 S has a shorter and less stable total antioxidant capacity than that of GYY4137; it is even undetectable by spectrophotometry at 150 µM. However, it is important to stress out that the antioxidant capacity of GYY4137 tends to increase over time. We also found that both H 2 S donors preserve sperm motility and protect the acrosomal status against the detrimental consequences of oxidative stress, although the effects were clearly both dose and donor dependent. Within the range of concentrations tested (3-300 µM), GYY4137 showed positive effects on sperm motility, whereas Na 2 S was detrimental at the highest concentration but beneficial at the lowest. Taken together, our results suggest that Na 2 S and GYY4137 may be used as in vitro therapeutic agents against oxidative stress in sperm cells, although the optimal therapeutic range varies between H 2 S donors.

Methods
Reagents were purchased from Sigma-Aldrich (Prague, Czech Republic), unless otherwise indicated.
Ethics statement. This study did not involve animal handling because the sperm samples were purchased as artificial insemination doses from a pig breeding company (Chovservis, Hradec Králové, Czech Republic). Experiment I. This experiment was designed to evaluate the total antioxidant capacity and stability of Na 2 S and GYY4137 at 20, 120, and 210 minutes during incubation at 38 °C in a water bath. The solutions were prepared shortly before the experiment and kept in microcentrifuge tubes tightly sealed with the attached cap (certified free of DNA, DNase, RNase, and endotoxins (pyrogens); material: virgin polypropylene; volume: 600 µl; Neptune Scientific, San Diego, CA, USA) during the whole incubation. For each concentration of H 2 S donor, analyses were performed on the same tube throughout the incubation period. Moreover, each microcentrifuge tube contained the same volume (i.e. 200 µl) of H 2 S donor or phosphate-buffered saline (PBS; blank) solution. The experiment was replicated four times.
Total antioxidant capacity of H 2 S donors. The total antioxidant capacity was determined by spectrophotometry (Libra S22, Biochrom, Harvard Bioscience Company, Cambourne, United Kingdom) at 660 nm by using the method described previously 41 . The principle of this assay is based on the antioxidant's capacity to reduce ABTS previously oxidised with H 2 O 2 . A standard curve was established by using known concentrations of 6-hydroxy-2 ,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox). The total antioxidant capacity was expressed as Trolox equivalents (µM or mM). The assay was run in duplicate for each sample. Experiment II. This experiment was designed to test whether H 2 S donors protect sperm cells against the deleterious effects of oxidative stress.
Sample collection and experimental design. Artificial insemination doses from 18 boars of different breeds were purchased from a pig breeding company. Sperm-rich fractions were collected by the gloved-hand method, diluted with Solusem ® extender (AIM Worldwide, Vught, Netherlands; pH≈7), and transported to the laboratory at 17 °C. The sperm motility and morphology were then checked, and only samples with at least 75% of motile and morphologically normal sperm were used for the experiments. Sperm samples from three boars were pooled to reduce the effect of male variability and were centrifuged at 167 g for 3 minutes at 17 °C to remove debris and dead sperm cells. The sperm concentration was then checked by using a Bürker chamber, adjusted to 30-40 × 10 6 spermatozoa/ ml with Solusem ® , and finally diluted 1:1 (v/v) with Solusem ® supplemented with 0.2% (w/v) of bovine serum albumin (BSA; ethanol-fractionated lyophilised powder). Thus, the final sperm and BSA concentrations were 15-20 × 10 6 spermatozoa/ml and 0.1%, respectively. Sperm samples were then randomly split into eight microcentrifuge tubes tightly sealed with the attached cap (certified free of DNA, DNase, RNase, and endotoxins (pyrogens); material: virgin polypropylene; volume: 2 ml; Neptune Scientific, San Diego, CA, USA): CTR, CTR-ox, and Na 2 S or GYY4137 at 300, 30, and 3 µM under oxidative stress. Oxidative stress was induced by adding a solution composed of 0.1 mM FeSO 4 and 0.5 mM sodium ascorbate (Fe 2+ /ascorbate) to the sperm samples. Because the effects of this ROS-generating system are clearly evident after 210 minutes of sperm incubation 19 , sperm analyses were performed at 0 hour (after 20 minutes of incubation for the CTR group only) and after 210 minutes of incubation at 38 °C in a water bath. The experiment was replicated six times with six independent semen pools.
Sperm mitochondrial status. Mitochondrial status was evaluated as previously described 45 , with minor modifications. Briefly, aliquots of sperm samples were incubated with rhodamine 123 (5 mg/ml, w/v, in dimethyl sulfoxide, DMSO) and propidium iodide (0.5 mg/ml, w/v, in PBS) for 15 minutes at 38 °C in the dark. Subsequently, samples were centrifuged at 500 g for 5 minutes, the supernatant was removed, and the sperm pellet was resuspended in PBS. Then, 200 spermatozoa were evaluated by using epifluorescence microscopy (40× objective; Nikon Eclipse E600, Nikon, Tokyo, Japan): the spermatozoa showing bright green fluorescence in the midpiece were considered to have active mitochondria.
Sperm plasma membrane integrity. The sperm plasma membrane integrity was evaluated as previously described 46,47 . Aliquots of sperm samples were incubated with carboxyfluorescein diacetate (0.46 mg/ml, w/v, in DMSO), propidium iodide (0.5 mg/ml, w/v, in PBS), and formaldehyde solution (0.3%, v/v) for 10 minutes at 38 °C in the dark. Then, 200 spermatozoa were evaluated by using epifluorescence microscopy (40× objective). The spermatozoa showing green fluorescence over the entire head area were considered to have intact plasma membrane.
Lipid peroxidation. Lipid peroxidation was assessed with the thiobarbituric acid reactive substances (TBARS) assay, as previously described 19,48 . At the end of each incubation period, sperm aliquots were collected and stored at −80 °C until analysis. The absorbance of each sample was then measured by spectrophotometry at 532 nm. A standard curve was established by using known concentrations of 1,1,3,3-tetramethoxypropane (MDA). The levels of lipid peroxidation are shown as µmol of MDA per 10 8 spermatozoa. The assay was run in duplicate for each sample.
Acrosomal status. Acrosome integrity was assessed after sample fixation in 2% (v/v) glutaraldehyde solution and by examination with phase-contrast microscopy (40× objective) 34 . For each sample, 200 spermatozoa were evaluated, and the percentage of sperm cells with NAR was determined. Acrosome loss was evaluated according to the protocol previously described 49 . Briefly, after methanol fixation and double washing with PBS, the samples were incubated with PNA-FITC (100 µg/ml, w/v, in PBS) for 10 minutes at 38 °C in the dark. Epifluorescence microscopy (40× objective) was used to evaluate 200 spermatozoa, and the cells showing no fluorescence over the acrosome were considered as acrosome-lost spermatozoa.
Statistical analysis. Data were analysed with the statistical program SPSS, version 20 (IBM Inc., Chicago, IL, USA). Shapiro-Wilk's and Levene's tests were used to analyse the normal distribution and the variance homogeneity of the data, respectively. The Mann-Whitney U-test was applied to check for differences between the total antioxidant capacities of Na 2 S and GYY4137 at the same concentration, whereas the repeated-measures Friedman test was used to compare the total antioxidant capacities of the H 2 S donors across the incubation times. The generalized linear model (GZLM) was performed to analyse the effects of the type and concentration of H 2 S donor on the sperm variables. The statistical significance was determined at p < 0.05. Data are shown as the mean±standard error.

Data availability
All data generated or analysed during this study are included in this article and its supplementary information file.