Oxidative profile and protease regulator potential to predict sperm functionality in donkey (Equus asinus)

Seminal plasma (SP) of donkey stallions was evaluated using various oxidative stress parameters as well as protease and protease inhibitor activities. SP was obtained by nine donkey stallions. In addition, one donkey stallion with non-obstructive azoospermia was enrolled in this study. Free radical scavenging activity (FRSA), the ferric reducing ability of plasma (FRAP), total antioxidant capacity (TAC), and total thiol level (TTL) were highly correlated with each other and with the protease inhibitor activity. However, only FRAP, TAC, and the nitrate/nitrite concentration (NOx) were significantly correlated with sperm concentration, production, and kinetics. Protease inhibitor activity was highly correlated with sperm concentration and production; however, it did not correlate with sperm kinetics. The azoospermic stallion produced a lower amount of semen than the normospermic stallions and its SP showed a lower antioxidant activity when evaluated with FRAP, TAC, and TTL as well as a higher NOx and a lower protease inhibitor activity. In conclusion, the evaluation of SP oxidative profile by FRAP, TAC, and NOx may provide reliable information on donkey sperm quality whereas protease inhibitor activity may play a role as a marker of the sperm concentration in this species.

www.nature.com/scientificreports/ However, the activity of the SP takes on a different meaning when the semen is collected, manipulated, and stored, thus deferring the timing of the insemination. Among the properties attributable to SP and useful for out-of-body semen preservation, the ability to counteract the oxidative damage to which spermatozoa are subject is undoubtedly included 14 . Sperm cells are highly sensitive to oxidative damage 15,16 due to their: (i) intense metabolic activity connected to movement 17 ; (ii) low cytoplasmic component and, therefore, to the inability to accumulate intracellular compounds capable of counteracting oxidative damage 14,18 ; (iii) high proportion of plasma membrane in relation to the cell mass containing a large quota of unsaturated fatty acids ensuring the membrane fluidity required for flagellar movement 19 ; (iv) low content of DNA repair mechanisms 20 . The oxidative stress (OS) affects sperm quality and brings to DNA fragmentation, lipid membrane peroxidation, lowering motility, and, ultimately, in the reduction of fertility 16 . To assess the degree of seminal impairment from oxidative damage and the ability of seminal plasma to counteract it, numerous approaches have been used in humans and animal species 21,22 .
Besides OS, protease activity (PA) represents a further threat affecting semen quality. Proteolytic enzymes, as acrosin and hyaluronidase, are contained within the sperm acrosome as inactive precursors 23 . They are activated following their release during fertilization as well as from dead and damaged spermatozoa, and, hence, the assessment of their activity represents a sperm quality assay 24 . To counter this danger, SP contains protease inhibitors 25 .
This study was aimed to detect the oxidative profile as well as the protease and anti-protease activities of SP in donkey stallions. Both evaluations, here analyzed for the first time in the asinine species, are of relevant diagnostic interest as alterations in the oxidative and proteolytic balance have been associated with sperm dysfunctions in humans and other animal species 16,26 . The oxidative profile has been tested by evaluating the substances resulting from oxidative and nitrosative processes, as lipid hydroperoxide levels (total oxidant status, TOS) and nitrite/nitrate contents (nitric oxide radical metabolites, NO x ), as well as the total antioxidant status, assessing the ferric reducing ability of plasma (FRAP), the total antioxidant capacity (TAC by ABTS-based assay), the total thiol levels (TTL) and the free radical scavenging activity (FRSA). Moreover, the advanced oxidation protein products (AOPP) were assessed as a marker of oxidative protein damage. Total proteins (TP), protease (PA) and antiprotease (APA) activities were assessed as biomarkers of proteolysis regulators. The results of these assays have been related to either semen quality or sperm kinetics. Finally, having identified an azoospermic individual within the group of animals trained for semen collection, it seemed interesting to compare the oxidative profile and the proteolytic regulatory potential of this individual with the rest of the normospermic group.

Results
Data report. Means (± SD) and 25-75% range of the oxidative profile and protease regulator parameters evaluated in the SP of the nine normospermic donkey stallions are showed in Table 1. In Table 2, there are the mean (± SD) values of the oxidative profile and protease regulator parameters assessed in the SP of the azoospermic stallion. Considering the availability of only one azoospermic stallion, it was not possible to perform any statistical comparison. However, several differences discriminated this azoospermic individual from normospermic stallions concerning the various methodologies used to evaluate the SP oxidative profile. In particular, the azoospermic stallion showed lower FRAP (83.9 ± 28.5 vs. 537.1 ± 184.0), FRSA (5.7 ± 2.4 vs. 22.6 ± 14.4), and TTL (53.4 ± 2.2 vs. 114.4 ± 36.8) values; vice versa, NO x values were higher in azoo-than normo-spermic stallions (621.9 ± 7.7 vs. 216.7 ± 120.8). However, TAC, TOS, and AOPP were not detected in the azoospermic stallion. In addition, the SP of the azoospermic stallion showed a higher pH, a lower TP concentration, APA, and semen volume whereas the PA and osmolarity in the azoospermic stallion seem similar to those of the normospermic stallions.
Correlations between oxidative profile and protease regulator parameters. In Table 3, the correlation coefficients obtained by comparing each oxidative profile and proteolysis regulator parameter to each other are shown. Almost all the parameters were well associated with each other. However, TOS values were not significantly correlated with any of the other parameters analyzed whereas NO x values were negatively correlated with FRAP and TAC and showed tendential associations with the other parameters. FRAP, FRSA, TAC, and TTL were correlated with the other parameters including those related to proteolysis regulators (PA and APA). AOPP was well correlated with FRSA, and TTL as well as with TP content, PA, and APA.
APA and the TP content practically correlated with all the other parameters except TOS and NO x , although for the latter the correlations are close to the cut-off value. PA was significantly correlated with FRSA, TTL, and AOPP as well as with TP content and APA.
Correlations between oxidative and protease parameters vs. semen and sperm kinematic parameters. Table 4 shows the correlation coefficients between OS parameters and protease/proteases inhibitor activities vs. seminal and sperm kinematic parameters. Among the seminal parameters, the osmolarity and pH were not associated with any of the parameters analyzed, the seminal volume was only significantly associated with TOS and NOx values whereas the sperm concentration and production were significantly correlated with almost all the OS parameters, TP content, and APA but not related to TOS and PA. The sperm kinematic parameters (Tot mot, Prog, VCL, VSL, and VAP) were significantly correlated to FRAP, TAC, NO x whereas AOPP was only correlated with total motility. www.nature.com/scientificreports/ www.nature.com/scientificreports/ The azoospermic stallion showed, together with a lower seminal volume and TP and APA, a lower antioxidant status when measured with FRAP, TTL, and TAC assays as well as higher NO x and AOPP values in comparison with the normospermic stallions. It seems important to point out that the azoospermic stallion was affected by a form of non-obstructive azoospermia; this condition can be discriminated from obstructive azoospermia based on the alkaline phosphatase (AP) content in SP 27 . In our study, the AP content in azoospermic stallion did not differ from that of the normospermic stallions (193.2 ± 77.7 vs. 153.3 ± 90.4 µkat/L in the azoo-and normospermic stallions, respectively) (data not shown). Therefore, the differences in the oxidative profile and in the proteolysis regulators discriminating the azoospermic individual from the normospermic stallions cannot be addressed to the lack of the testicular and epididymal fluid contribution to the SP. Recently, the SP antioxidant status, measured by the activities of four antioxidant enzymes (superoxide dismutase, catalase, glutathione peroxidase, glutathione reductase), was related to sperm functionality in horse and donkey stallions 28 . In this study, SP antioxidant capacity was significantly higher in donkey than in horse stallions, and the activities of superoxide dismutase and catalase in donkeys positively correlated with some sperm kinematic parameters, as progressive motility, linearity, and straightness; however, no correlation was found in horse stallions 28 . The higher enzymatic activities in donkey SP might be related to the higher functional activity of the accessory glands in donkey than in horse stallions 29 . Moreover, analyzing blood serum and SP in azoo-and normo-spermic men, Gulum and colleagues 30 found higher levels of oxidative parameters and lower levels of antioxidant parameters in the azoospermic group. In the present paper, the SP antioxidant status was assessed by different methods whose analytical results are the expression of the cumulative actions of all the antioxidants present in this biological matrix, thus providing an integrated parameter rather than the sum of the single measurable substances. Among these assays, FRAP and TAC positively correlated with most of the analyzed parameters of Table 3. Correlation matrix between oxidative profile and proteolysis regulator parameters in the seminal plasma of donkey stallions (n = 9). FRAP: ferric reducing ability of plasma; TAC: total antioxidant capacity; FRSA: free radical scavenging activity; TTL: total thiol levels; TOS: total oxidant status; NOx: nitric oxide metabolites; AOPP: advanced oxidation protein products; PA: protease activity; APA: antiprotease activity; TP: total protein. *, **, ***: indicate a statistic significance for P < 0.05, P < 0.01 and P < 0.001, respectively; † : indicate a tendency for P < 0.10.  Table 4. Correlation matrix between oxidative profile and proteolysis regulator parameters in the seminal plasma, semen parameters and sperm kinetics in donkey stallions (n = 9). FRAP: ferric reducing ability of plasma; TAC: total antioxidant capacity; FRSA: free radical scavenging activity; TTL: total thiol levels; TOS: total oxidant status; NOx: nitric oxide metabolites; AOPP: advanced oxidation protein products; PA: protease activity; APA: antiprotease activity; TP: total protein; Osm: osmolarity; Vol: volume; Conc: sperm concentration; Prod: sperm production; Tot mot:total sperm motility; Prog: progressive sperm motility; VCL: curvilinear velocity; VSL: straight-line velocity; VAP: average path velocity. *, **, ***: indicate a statistic significance for P < 0.05; P < 0.01 and P < 0.001, respectively; † : indicate a tendency for P < 0.10.  www.nature.com/scientificreports/ sperm functionality. FRAP reflects the effect of low molecular weight antioxidants, mainly measuring uric acid, α-tocopherol, bilirubin, and ascorbic acid levels, rather than antioxidants containing sulfhydryl groups in their structure, as glutathione (GSH) and albumin. In contrast, TAC assay also measures albumin, whose proteoforms correlated with stallion sperm motility parameters 31 . Moreover, the uric acid, whose concentration is higher in high fertility stallions 32 , affects the ABTS-based method less than the FRAP assay 33,34 . All these assays are widely used in fertility diagnostics in men [35][36][37][38] ; however, recently, FRAP and TAC assays have been applied to bovine SP and sperm lysates and positively correlated with sperm motility parameters 39 . Alternative methodological approaches for evaluating SP antioxidant activity are FRSA and TTL. They positively correlated with sperm concentration and production; however, they did not correlate with sperm kinematic parameters. Lower FRSA and TTL levels were found in the azoospermic stallion in agreement with previous studies in human sperm 40 showing a lower content of thiol groups in infertile than control patients. TTL assay is a little-used method for the assessment of semen quality in livestock species. Notwithstanding in our study TTL only correlated with sperm concentration and production, we believe it represents a valid tool for sperm quality assessment based on the pivotal role played by thiols in protecting spermatozoa from oxidative damage 41 . This is supported by a previous study in stallions showing that low TTL and ABTS-based assay values in SP were associated with a massive occurrence of sperm DNA breaks 42 .

Osm
Since reactive oxygen species (ROS) and reactive nitrogen species (RNS) are highly unstable and reactive, an approach for evaluating the effects of oxidative and nitrosative stress on biological matrices consists in assessing more stable and reliable derived-ROS and -RNS compounds by simple spectrophotometric assays that can be automated on clinical auto-analyzers allowing rapid and inexpensive data collections. Nitric oxide (NO) is a free radical widely involved in many biological processes, including reproduction. It is very difficult to detect and, hence, it is indirectly evaluated through its metabolites 43 . The role of NO in sperm physiology is still unclear although it is well-known that it is involved in sperm motility, capacitation, and fertilization in mammals 44,45 . A recent study demonstrated that elevated NO metabolites were associated with sperm dysfunction in horse stallions 46 . Our results showed higher NO x levels in the azoospermic individual than in the normospermic stallions and significant negative correlations with all sperm function parameters. These findings are in agreement with studies in humans, where NO metabolite levels, analyzed as nitrite (NO 2 − ) concentration, were statistically higher in infertile men than in controls; besides, the NO 2 − concentration negatively correlated with sperm motility 47 .
TOS assay evaluates the total oxidant substances present in biological fluids 48 . In our study, this test did not correlate with sperm function but it correlated with ejaculate volume. Based on these findings, TOS does not seem a valid predictor of sperm functionality in donkeys. In contrast, in human semen, higher TOS values were associated with sperm motility impairment and lower total antioxidant status, assessed by ABTS-based test; in this study, TOS and TAC values were negatively correlated 49 . Also, Verit et al. 50 , comparing fertile and idiopathic sub-fertile men, found TOS negatively correlated with sperm concentration, motility, and morphology.
AOPP assay evaluates the oxidative protein modifications by detecting dityrosine-containing protein crosslinked products. This analysis was established to assess proteins, mostly aggregates of albumin, in blood plasma 51 ; however, it was efficiently applied in other organic matrices as SP 37 . In our study, the AOPP levels were positively correlated with some descriptive parameters of sperm function, as sperm concentration, production, and total motility. Moreover, AOPP levels were not detectable in the SP of the azoospermic stallion. This finding may depend on the lower total protein content found in the azoospermic individual with respect to the normospermic samples, ponying the detection below its analytical limit. The consequences of protein oxidation on reproductive function in mammals are still unclear. In human SP, AOPP levels were negatively related to fertility, with higher concentrations in azoo-and terato-in comparison to normo-spermic individuals 37 . Similarly, when protein oxidation level of SP was measured as carbonyl content, its concentration was higher in infertile men 52 . On the contrary, in stallion spermatozoa, carbonyl content was positively related to motility and viability, with a higher level of oxidized sperm proteins in fertile animals during the breeding season 53 . This occurrence, however, was not associated with sperm DNA damage 53 and could be part of that paradoxical effect between free radical production and sperm function found in stallion sperm by Aitken and colleagues 54 . On the other hand, SP protein oxidation has been related to several defects in sperm function of sub-fertile stallions during the non-breeding season 53 .
Among the constituents of mammalian SP, a complex of proteins and enzymes with protease and protease inhibitor activities, namely proteolysis regulators, plays a role in the reproductive biology that is still poorly known. In humans, proteolysis regulators affect sperm maturation, activation, and storage as well as semen coagulation and liquefaction. SP protease inhibitors are further involved in sperm protection from proteolytic enzymes released from damaged/dead spermatozoa 26 . Proteolysis regulators seem to be involved in fertilization, allowing the spermatozoon to penetrate both the zona pellucida and oocyte plasma membrane 55 . For an overall assessment of these activities in donkey SP, we applied two simple and reproducible spectrophotometric methods, previously used to assess the proteolysis regulator activities in fish 56 , rather than proteomics approaches already experienced in mammals 26 . In our study, both protease and antiprotease activities correlated with many of the other analyzed SP parameters. Moreover, antiprotease activity strongly correlated with sperm concentration and motility (P < 0.001). This fits with results obtained in some fish species, showing a strong correlation between antiprotease activity and sperm concentration 57 , and in humans, where protease inhibitors are involved in sperm capacitation and motility and are associated with semen fertility 26,58 . All this suggests that the assessment of the activity of proteolysis regulators represents a useful tool for obtaining more information on seminal quality.
In conclusions, the analysis of the oxidative profile and the activity of proteolysis regulators in donkey seminal plasma showed that most of the results obtained by the different methods used were well correlated with each other. However, only some of them, as FRAP, TAC, and NO x , were significantly correlated with seminal characteristics and sperm kinetics. The comparison carried out between different methods was aimed at evaluating www.nature.com/scientificreports/ the oxidative profile of an organic matrix such as seminal plasma and did not intend to define selection criteria among assays, considering that each test interrogated different groups of substances. Instead, the correspondence between each assay and sperm function allowed to identify effective tools for assessing semen quality in the donkey species. Preliminary discriminant variations emerged between an azoospermic individual and the normospermic stallions in the antioxidant activity, based on FRAP, TAC, and TTL assays as well as in the nitric oxide metabolites, based on the nitrite/nitrate (NO x ) levels. Also, the anti-protease activity seems to differ between the azoospermic individual and the normospermic stallions. These differences should be reanalyzed following an increase in the number of cases and, if confirmed, the physiological mechanism underlining them should be investigated.

Materials and methods
Reagents. All reagents and media were purchased from Merck (Milan, Italy) unless otherwise stated and cell culture tested. Experimental design. Stallions previously trained for semen collection were subjected to repeated twice weekly semen collections to empty their epididymal reservoir and to grossly assess their semen quality (data not shown). Subsequently, after 4 d from the last preliminary collections, a single ejaculate for each stallion was collected and analyzed in this study. However, in the azoospermic stallion, the semen collection was repeated after one month and both ejaculates were analyzed. SP of each stallion was analyzed for evaluating its oxidative profile by using several methodologies as well as its proteolysis regulator potential. The experimental blinding was assured through an independent assignment of tasks.

Animals.
Semen collection and preliminary evaluations. Semen was collected from March to June, as previously described 59 . Semen was filtered by a sterile gauze; the volume was measured by a cylinder and sperm concentration was estimated by a Makler chamber. Sperm production was obtained by sperm concentration × semen volume. A homogeneous part of the semen was centrifuged (2000g for 10 min) to separate the SP from cells and debris. SP osmolarity was measured using an osmometer (Digital Osmometer, Roebling, Berlin, DE) and pH was measured a Jenway 3020 pH meter (Jenway, London, UK). SP aliquots of each stallion were stored at −80 °C until analysis.  www.nature.com/scientificreports/ samples were mixed with 200 μl of acetic acid-sodium acetate buffer (pH 5.8) and measured at 660 nm with the microplate reader. Thereafter, 20 μL of acetic acid-sodium acetate buffer (pH = 3.6) containing 2 mM H 2 O 2 and 10 mM ABTS were added to the mixtures. After 5 min of incubation at room temperature, the ODs were read again at 660 nm. The assay was calibrated with ascorbic acid (AA) and the results are presented as AA equivalents (μM).
Free radical scavenging activity (FRSA). FRSA was analyzed by the 2,2-Di(4-tert-octylphenyl) -1-picrylhydrazyl (DPPH) reduction assay, based on the reduction of DPPH· to 1,1-diphenyl -2-picryl hydrazine, as described by Blois 62 with minor modifications 63 . In brief, 25 µL of the SP samples were mixed with 475 µL of 10 mM PBS, pH 7.4, and 500 µL of a 0.1 mM DPPH diluted in methanol. The mixture was kept for 30 min in darkness at room temperature, before absorbance reading at 520 nm against the blank, using a spectrophotometer (SmartSpec 3000 UV/Vis, Bio-Rad, Segrate, Italy). The absorbance of each sample was related to DPPH in PBS solution. The decrease of DPPH bleeding was calculated as follows: % of inhibition = [1 − (As/A0)] × 100; where As is the absorbance of the sample and A0 is the absorbance of the DPPH solution.
Total thiol levels (TTL). TTL (sulfhydryl group, -SH) were measured as indicated by Hu 64 65 . In brief, 100 µL of deproteinized SP samples were treated with 100 µL of 0.8% vanadium (III) chloride in 1 M HCl to reduce nitrate to nitrite. Thereafter, 100 μL of Griess reagent (0.2% sulfanilamide and 0.1% N-1-(naphthyl)ethylenediamine) were added and the mixtures were incubated for 30 min at 30 °C. Finally, the ODs were read at 540 nm in the microplate reader. The assay was calibrated with sodium nitrate (NaNO 3 ) and the results were reported as NaNO 3 equivalents (μM).
Advanced oxidant products (AOPP). AOPP was analyzed as Hanasand et al. 66 . Briefly, 200 µl SP samples were added to 800 µl 0.2 M citric acid. Different concentrations of chloramine-T (3.125-100 µM) in citric acid (calibration standards) were added to 10 µL of 1.16 M potassium iodide. After 2 min, ODs were read at 340 nm in the spectrophotometer and the AOPP concentrations were expressed as chloramine-T equivalents (μM).
Protease activity (PA). PA of samples was determined using the azocasein hydrolysis assay 67 , with minor changes. Briefly, an equal volume of each SP sample was added to 0.7% azocasein in 100 mM ammonium bicarbonate, pH 7.8. Tubes were placed on a shaker at 30 °C for approximately 19 h. The reaction was stopped by adding trichloroacetic acid (4.5% final concentration). After centrifugation (10,000 g, 15 min), 100 µl of each supernatant in duplicate was added to 100 µl of 0.5 M NaOH and ODs read at 450 nm with the microplate reader. Results are expressed as a percentage of activity in relation to a positive control that was the bovine trypsin.
Anti-protease activity (APA). APA was determined by the capacity of the sample to inhibit trypsin activity 68 . In brief, 20 µL of the SP samples were incubated with 5 mg mL -1 trypsin solution for 10 min at 22 °C. Subsequently, 200 µL of 0.1 M phosphate buffer solution, pH 7.0, and 250 µL 2% azocasein were added and incubated for 10 min at 22 °C. The reaction was stopped by the addition of 500 µL of 10% trichloroacetic acid and further incubated for 30 min at 22 °C. After centrifugation (8000g, 10 min), 100 µL 1 N NaOH was added to 100 µL of supernatants and the ODs were read at 450 nm with the microplate reader. In the positive control, buffer replaced the SP sample whereas, in the negative control, buffer replaced both the SP sample and the trypsin. The percentage of inhibition of trypsin activity by each SP sample was calculated by comparing it to the positive control.
Total protein (TP) concentration. TP concentrations were measured by the Biuret method using a total protein reagent (T1949). The assay was calibrated with bovine serum albumin (BSA) and the TP concentration is presented as BSA equivalents (mg mL -1 ).
Statistical analysis. Data from the nine normospermic and one azoospermic stallions entered into a datasheet, as follows: the experimental date, the sampling number, the duplicate number (1-2), the ten SP parameters (i.e., FRAP, TAC, FRSA, TTL, TOS, NO x , and AOPP assays as well as TP, PA, and APA), the five semen characteristics (i.e., semen pH, osmolarity and volume as well as sperm concentration and production) and the