Evaluation of the antioxidative enzymes in the seminal plasma of infertile men: Contribution to classic semen quality analysis

ABSTRACT

Protein expression/activity of antioxidative defense enzymes (AD) in seminal plasma of fertile men might be used as biomarkers of male fertility status. To test this concept, the present study examined the semen parameters of males among 14 normal idiopathic (normozoospermia) and 84 subnormal (teratozoospermia, oligoteratozoospermia, oligoasthenoteratozoospermia) infertile individuals\. We investigated levels of protein expression/activity of Cu, Zn superoxide dismutase (CuZnSOD), manganese superoxide dismutase (MnSOD), catalase and glutathione peroxidase (GSH-Px), their association with functional sperm parameters, as well as their potential to serve as biomarkers of specific sperm pathologies. Although the activity of CuZnSOD and protein expression of catalase were significantly correlated with several sperm parameters, underlying their potential role in etiology of various sperm abnormalities, investigation of their potential usefulness as a biomarker of semen quality showed that these AD enzymes could not distinguish subtle differences between various sperm pathologies. In contrast, GSH-Px activity was decreased in all groups with sperm pathologies and was a very good indicator of aberrations in functional sperm parameters, explaining up to 94.6% of infertility cases where functional sperm parameters were affected. Therefore, assessment of GSH-Px activity showed the potential to discriminate between infertile males with normal and subnormal semen characteristics and may prove useful in the evaluation of male (in)fertility.

Abbreviations: AD: antioxidative defense; Cu, Zn SOD: copper, zinc superoxide dismutase; GSH-Px: glutathione peroxidase; MnSOD: manganese superoxide dismutase; NS: normospermia; OATS: oligoasthenoteratozoospermia; OTS: oligoteratozoospermia; ROC: receiver operating characteristic; ROS: reactive oxygen species; TS: teratozoospermia; WHO: world health organization

KEYWORDS:

• Sperm parameters
• seminal plasma
• male infertility
• antioxidative defense
• biomarker

Introduction

In recent years there has been a dramatic increase in infertility, a condition that is now estimated to affect 15% of couples worldwide. In approximately half of these cases, infertility is attributable to a male factor (Buzadzic et al. 2015). Diagnosis of male infertility routinely begins with basic semen analysis, which involves the assessment of sperm count, motility and morphology, based on reference values established by the World Health Organization (WHO) (2010). However, data from these routine tests do not provide comprehensive diagnostic information and do not allow us to determine the underlying cause of infertility (Guzick et al. 2001; Sharma et al. 2013). In addition, some abnormalities of spermatozoa occur at the molecular level and are not linked with any morphological manifestation which could be detectable by light microscopy (Aitken et al. 1989). Thus, it is not uncommon that men with normal WHO parameters are infertile or that fertile men have sub-normal sperm characteristics (de Lamirande et al. 1997). This clearly highlights the lack of sensitive analysis of semen for the precise diagnosis of infertility (Aitken et al. 1989; de Lamirande et al. 1997; Leclerc et al. 1997; Suzuki et al. 1997; Guzick et al. 2001; Sharma et al. 2013; Kratz and Piwowar 2017).

Seminal plasma proteins, i.e., semen proteome have emerged as an important biomarker for male infertility (Suzuki et al. 1997; Cui et al. 2018; Intasqui et al. 2018). Among the range of seminal plasma proteins being investigated, much attention has been dedicated to the importance of antioxidative defense (AD) enzymes in sperm function, including catalase, glutathione peroxidase (GSH-Px) and superoxide dismutase (SOD). This is quite reasonable, considering that AD enzymes are crucial in determining the concentration of reactive oxygen species (ROS), that can exert their effect upon sperm function. These effects can be both beneficial (Aitken et al. 1989; de Lamirande et al. 1997; Leclerc et al. 1997; Suzuki et al. 1997; Otasevic et al. 2013) and detrimental (Aitken and Clarkson 1987; Iwasaki and Gagnon 1992; Sukcharoen et al. 1996; Agarwal and Allamaneni 2004), depending upon the concentration of ROS. In line with these earlier findings, many authors, including ourselves, have reported a relationship between the AD enzymes activity in seminal fluid and semen quality (Alkan et al. 1997; Jóźwik et al. 1997; Lewis et al. 1997; Miesel et al. 1997; Sanocka et al. 1997; Macanovic et al. 2015), although some of these findings have been met with controversy. However, to the best of our knowledge, we were the first to report the relationship between the expression of AD proteins in the seminal plasma and functional parameters of spermatozoa (Macanovic et al. 2015). In our previous pilot study of normozoospermic fertile men, we found that the expression of manganese SOD (MnSOD), copper, zinc SOD (CuZnSOD) and catalase proteins in seminal fluid positively correlates with parameters of semen quality. Furthermore, we demonstrated a negative relationship between the activity of GSH-Px in seminal plasma with both morphology and progressive motility of spermatozoa. These results strongly indicated that the expression and/or the activity of AD components in the seminal plasma might be used as potential biomarkers of fertility status of males (Macanovic et al. 2015).

The present study extends the original design used in our pilot study (Macanovic et al. 2015). Here we examine infertile males with normal – normozoospermic (NS) and sub-normal – teratozoospermic (TS), oligoteratozoospermic (OTS) and oligoasthenozoospermic (OATS) semen parameters (according to criteria established by the WHO, 2010) to assess whether protein expression/activity of antioxidative enzymes in seminal plasma of infertile men might serve as biomarkers of male semen quality.

Results and discussion

Nearly two decades ago, it has become evident that male fertility status cannot be accurately evaluated by assessing semen parameters alone (Guzick et al. 2001; Ramya et al. 2010; Sutovsky and Lovercamp 2010). Thus, one of the most important goals for reproductive biologists is to identify new biomarkers of male (in)fertility (Garrido et al. 2004; Sharma et al. 2013; Sutovsky et al. 2015; Kratz and Piwowar 2017). Sperm proteome analysis emerged recently as a reliable diagnostic marker of infertility (Cui et al. 2018; Intasqui et al. 2018). In this respect, AD enzymes in seminal plasma have arose as good candidates (Alkan et al. 1997; Jóźwik et al. 1997; Lewis et al. 1997; Miesel et al. 1997; Sanocka et al. 1997). Our recent findings revealed that the expression of antioxidative enzymes in seminal plasma of fertile men might be used as a predictor of fertility status of males (Macanovic et al. 2015). To extend this, the current study we examined the effect of a series of antioxidative enzymes in seminal plasma infertile males with normal and sub-normal semen parameters. In contrast to the strong positive correlation between CuZnSOD and MnSOD protein content and progressive motility, which we observed in our previous study of fertile subjects (Macanovic et al. 2015), protein expression of CuZnSOD was not associated with any sperm parameter in the infertile groups tested in the present study (Table 1). A negative correlation between MnSOD protein content in the seminal plasma and progressive sperm motility (Table 1 and Supplementary Figure 1), was observed in infertile TS individuals, (r = −0.2972, P = 0.0425), while in the OTS was positive (r = 0.8820, P = 0.0201). In OATS group, MnSOD content was positively correlated with sperm morphology (r = 0.6357, P = 0.0049) (Table 1, Supplementary Figure 1). Clearly, the present findings do not support the assessment of these AD enzymes in the evaluation of sperm pathologies as the degree and nature of correlations with sperm functional parameters varies widely among different groups of infertile subjects. The same holds true for GSH-Px protein expression (Table1, Supplementary Figure 1) that was only significantly correlated in the NS group: negative with sperm count (r = −0.9297, P = 0.0002) and motility (r = −0.5907, P = 0.0094) and positive with sperm morphology (r = 0.6956, P = 0.0374). The absence of differences in SODs and GSH-Px protein expressions in pathologic infertile groups related to NS (Figure 1) support the view that protein expression of these enzymes cannot be used in delineation between various sperm pathologies.

Table 1. Correlation between sperm parameters and relative protein expression of AD enzymes in seminal plasma of normozoospermic, teratozoospermic, oligoterarozoospermic and asthenoteratozoospermic samples.

	CuZnSOD	MnSOD	Catalase	GSH-Px
Number of spermatozoa	/	/	NS (r = −0.577, P = 0.0441)	NS (r = −0.9297, P = 0.0002)
Progressive motility of spermatozoa	/	TS (r = −0.2972, P = 0.0425) OTS (r = 0.8820, P = 0.0201)	TS (r = −0.2456, P = 0.0492)	NS (r = −0.5907, P = 0.0094)
Morphology of spermatozoa	/	OATS (r = 0.6357, P = 0.0049)	OTS (r = −0.8019, P = 0.0167) OATS (r = −0.6402, P = 0.0360)	NS (r = 0.6956, P = 0.0374)

CuZnSOD: copper, zinc superoxide dismutase; MnSOD: manganese superoxide dismutase; GSH-Px: glutathione peroxidase.

Figure 1. Protein expression profiles of GSH-Px (A), CAT (B), CuZnSOD (C) and MnSOD (D) in seminal plasma of normozoospermic (NS), teratozoospermic (TS), oligoteratozoospermic (OTS) and oligoasthenoteratozoospermic (OATS) men. *Compared to normospermic samples, *p < 0.05, #Compared to teratozoospermic samples, #p < 0.05, ##p < 0.01, ###p < 0.001.

Catalase content was negatively correlated with sperm count (r = −0.577, P = 0.0441) (Table 1, Supplementary Figure 1) in normozoospermic infertile patients, contrary to the positive correlation we identified between catalase protein content and both sperm motility and morphology in normozoospermic fertile men (Macanovic et al. 2015). This negative correlation persisted across all infertile groups, although the parameters showing correlation varied. In TS subjects, catalase expression correlated negatively with motility (r = −0.2456, P = 0.0492), but in OTS and OATS, catalase expression correlated negatively with sperm morphology (r = −0.8019, P = 0.0167; r = −0.6402, P = 0.0360, respectively). These negative correlations between sperm parameters and catalase protein content in all infertile groups (Supplementary Figure 1) might be a key feature of all infertile groups. This data, along with previously shown positive association between catalase protein content and sperm parameters in fertile men (Macanovic et al. 2015) suggest that direction of correlation (positive or negative) of catalase protein expression with sperm parameters could make some distinction between fertile and infertile subjects.

In addition, when sperm function parameters were correlated with the activity of CuZnSOD, a certain regularity was observed for the infertile groups with a sub-normal sperm count (OTS and OATS) (Table 2, Figure 2). Namely, there was a strong negative correlation with both sperm count (r = −0.7051, P = 0.0228) and morphology (r = −0.5796, P = 0.0490) in OTS subjects, and positive relationship between CuZnSOD activity and sperm count (r = 0.7289, P = 0.0168) and morphology (r = 0.6778, P = 0.0313) in OATS subjects (Figure 2). In teratozoospermic samples, CuZnSOD activity positively correlated with motility of spermatozoa (r = 0.5638, P = 0.0487). Collectively, these results demonstrated an absence of correlation between CuZnSOD activity and semen quality in both fertile (Macanovic et al. 2015) and infertile normozoospermic subjects, indicating that the nature and type of CuZnSOD correlations with sperm quality parameters could discriminate between various sperm pathologies. Although the activity of CuZnSOD in the seminal plasma correlates with some semen parameters in various groups of infertile men, this had little diagnostic value with respect to the infertile normozoospermic group. Moreover, investigation of the potential usefulness of catalase and CuZnSOD as a biomarker of semen quality demonstrated these AD enzymes could not distinguish subtle differences between various sperm pathologies (data not shown).

Table 2. Correlation between sperm parameters and enzyme activity of AD enzymes in seminal plasma of normozoospermic, teratozoospermic, oligoterarozoospermic and asthenoteratozoospermic samples.

	CuZnSOD	GSH-Px
Number of spermatozoa	OTS (r = −0.7051, P = 0.0228) OATS (r = 0.7289, P = 0.0168)	NS (r = −0.7671, P = 0.0096)
Progressive motility of spermatozoa	TS (r = 0.5638, P = 0.0487)	/
Morphology of spermatozoa	OTS (r = −0.5796, P = 0.0490) OATS (r = 0.6778, P = 0.0313)	NS (r = −0.5866, P = 0.0474) OTS (r = 0.8146, P = 0.0075)

CuZnSOD: copper, zinc superoxide dismutase; GSH-Px: glutathione peroxidase.

Figure 2. Correlation analysis between seminal plasma CuZnSOD and GSH-Px activity and functional sperm parameters; sperm concentration (circle), progressive motility (square) and morphology (triangle) in infertile normozoospermic (NS), teratozoospermic (TS), oligoteratozoospermic (OTS) and oligoasthenoteratozoospermic (OATS) men.

In comparison, the relationship between the activity of the GSH-Px and semen quality parameters in normozoospermic infertile subjects (Table 2, Figure 2) did not differ from those obtained previously in fertile normozoospermic subjects (Macanovic et al. 2015). Namely, GSH-Px activity negatively correlated with sperm number (r = −0.7671, P = 0.0096) and morphology (r = −0.5866, P = 0.0474) in NS infertile group. However, in pathological infertile subjects, the only correlation between GSH-Px activity and sperm parameters (Figure 2) is observed in the group with a low sperm count (OTS), where GSH-Px activity was positively correlated with sperm morphology (r = 0.8146, P = 0.0075). This is in accordance with previous findings described by Foresta et al. (2002) and Crisol et al. (2012). Nevertheless, the activity of GSH-Px was significantly decreased in all infertile groups with sperm pathologies related to NS (Figure 3A), indicating that GSH-Px activity in seminal plasma could make difference between infertile males with normal and subnormal semen characteristics (TS, OTS, and OATS). The receiver operating characteristic (ROC) curve analysis suggested that low levels of GSH-Px activity could be a very good indicator of aberrations in functional sperm parameters. GSH-Px activity values ≤1.918 nM NADPH min⁻¹ mg⁻¹ protein (average for control subjects: 3.177 nM NADPH min⁻¹ mg⁻¹ protein) may be indicative of infertility caused by subnormal functional sperm parameters. All controls were ≤1.918 nM NADPH min⁻¹ mg⁻¹ and 78/84 of the TS, OTS, and OATS were ≥1.918 nM NADPH min⁻¹ mg⁻¹ affording good sensitivity and specificity.

Figure 3. Activity of GSH-Px (A) and CuZnSOD (B) in seminal plasma of normozoospermic (NS), teratozoospermic (TS), oligoteratozoospermic (OTS) and oligoasthenoteratozoospermic (OATS) men. *Compared to normozoospermic samples, **p < 0.01, ***p < 0.001; #Compared to teratozoospermic samples, ##p < 0.01.

Clearly, data from the present study confirm and significantly extended our previous findings showing that the evaluation of protein expression and/or activity of AD enzymes in the seminal plasma might be important in the evaluation of male (in)fertility status. Significant correlations between catalase and CuZnSOD protein expression with functional sperm parameters were shown. However, the expression of these AD enzymes could not distinguish differences between various sperm pathologies. The resultant data clearly showed that the evaluation of enzyme activity of GSH-Px in seminal plasma has the potential to discriminate the difference between infertile males with normal and subnormal semen characteristics. Therefore, assessment of GSH-Px activity may be useful along with classic semen analysis in the evaluation of male (in)fertility. Additional translational studies are required to validate these findings.

Materials and methods

Samples

This study was performed according to Declaration of Helsinki and approved by the Institutional Ethics Board of the Clinic for Gynecology and Obstetrics ‘Narodni Front’ as well as by the Institute for Biological Research, Belgrade University. Study inclusion criteria were: no history of chronic illness, e.g., cancer, diabetes, autoimmune diseases; no use of drugs, dietary supplements, chronic medication no alcohol abuse (<three drinks per week). Subject with leukocytospermia, as well as subjects exposed to pesticides, heavy metals, organic solvents and welding (patients declared that they were not exposed to above stated environmental pollutants) were excluded from the study. Ninety-eight males attending the infertility clinic for male factor infertility (subjects that did not obtain pregnancy after two years of unprotected sexual intercourse; the female factors were excluded) took part in the study after providing written informed consent. After 3–5 days of ejaculatory abstinence, semen samples were collected. For evaluation, all semen samples were allowed to completely liquefy (room temperature, mean time = 20 min), and then volume, pH, sperm count, motility and morphology were evaluated according to criteria established by the World Health Organization (WHO) (2010) and semen samples were divided into four groups: normozoospermia (n = 14), teratozoospermia (n = 54), oligoteratozoospermia (n = 20) and oligoasthenoteratozoospermia (n = 10). In the NS group, sperm number was 55.6 ± 12.92, motility 51.73 ± 6.34 and normal morphology 5.11 ± 0.33. In TS, all sperm parameters were decreased related to NS – sperm number (38.97 ± 18.31; P < 0.001), motility (42.24 ± 9.81; P < 0.001) and normal morphology (2.22 ± 0.66; P < 0.001). OTS had significantly lower number of spermatozoa (9.8 ± 2.61) related to both NS (P < 0.005) and TS (P < 0.01), while motility (42.2 ± 9.17) and normal morphology (1.44 ± 0.52) was lower than in NS (P < 0.01 and P < 0.001), respectively. OATS had the lowest sperm parameters, number (6.88 ± 3.33), motility (24.44 ± 8.81) and normal morphology (1.0 ± 0.5). Mean participant parameters (age and semen characteristics) are shown in Table 3. To obtain seminal plasma, samples were centrifuged at 3000 g for 30 min to obtain total elimination of the cellular components. Supernatants were collected, aliquoted and kept at −80ºC for later assessment of antioxidative defense activity and protein expression.

Table 3. The mean participant’s age and semen parameters.

Parameters (Average ± S.E.M)	NS	TS	OTS	OATS
Age of donors (years)	40.07 ± 5.44	37.56 ± 6.06	39.2 ± 8.53	36.11 ± 5.15
Height (cm)	184.5 ± 5.91	182.2 ± 6.48	184.4 ± 5.77	182.5 ± 13.43
Weight (kg)	90.10 ± 11.13	87.14 ± 12.06	89. ± 10.28	94.17 ± 17.6
BMI (kg/m^2)	26.48 ± 3.1	26.17 ± 2.54	26.34 ± 1.78	27.99 ± 1.89
Volume of semen (ml)	3.64 ± 1.7	3.55 ± 1.93	4.54 ± 2.74	3.87 ± 1.77
Abstinence time (day)	3.48 ± 0.09	4.17 ± 0.53	3.34 ± 0.78	2.99 ± 0.87
Number of spermatozoa (10^6/ml)	55.6 ± 12.92	38.97 ± 18.31	9.8 ± 2.61	6.88 ± 3.33
Motility (%)	51.73 ± 6.34	42.24 ± 9.81	42.2 ± 9.17	24.44 ± 8.81
Normal morphology (%)	5.11 ± 0.33	2.22 ± 0.66	1.44 ± 0.52	1.0 ± 0.5
Number of leukocytes	Absent (-)	Absent (-)	Absent (-)	Absent (-)
Number of erythrocytes	2.11 ± 0.44	1.45 ± 1.12	1.78 ± 0.87	2.47 ± 1.69

Western blotting

Western blots were done as reported earlier (Petrovic et al. 2005). After polymerization, gels were loaded with 10 µg aliquots of seminal plasma protein samples for each Western blot analysis. The primary antibodies: a rabbit polyclonal antibodies against: CuZnSOD (ab13498; 0.2 g l⁻¹), MnSOD (ab13533; 1:5000), catalase (ab1877; 1:1000), GSH-Px (ab16798; 1:2000) and β-actin (ab8225; 1:1000) were incubated with membranes overnight at 4ºC. The appropriate horseradish peroxidase-conjugated secondary antibody – goat anti-rabbit (ab6721; 1:6000) was incubated for 2 h at room temperature. All antibodies were purchased from Abcam, Cambridge, UK. Immunoreactive bands were visualized using enhanced chemiluminescence and Hyperfilm MP (Amersham). Densitometric quantitative analysis of immunoreactive bands was conducted using ImageJ software version 1.50i (National Institutes of Health, USA). Total band density was calculated as the sum of pixel intensities within a band (one pixel ¼ 0.007744 mm²). It was obtained by plotting pixel density and measuring area integral using “Gels” option in Image J. We averaged the ratio of dots per band for the target protein and β-actin (the gel loading control) from three independent experiments and expressed changes in protein expression as a percentage of normozoospermic infertile group, which was standardized as 100%. Data were then statistically analyzed. Representative blots are shown in Supplementary Figure 1.

AD enzymes activity

Activity of SOD was assayed using the method described by Misra and Fridovich (1972), but at 26ºC and expressed in units mg⁻¹ of protein. Following inhibition with 4 mM KCN, the total specific SOD activity and MnSOD activity were measured, and then CuZnSOD activity was calculated. SOD units were defined as the amount of the enzyme inhibiting epinephrine oxidation by 50% under the appropriate reaction conditions. Activity of catalase was determined by the method of Beutler (1982) and expressed in μmol H₂O₂ min⁻¹ mg⁻¹ protein. Activity of GSH-Px was determined using the method of Paglia and Valentine (1967) and expressed in nmol NADPH min⁻¹ mg⁻¹ protein;

Additional assays and statistical analysis

Protein content was assessed by the method of Lowry (Lowry et al. 1951). Normality of data distribution was tested by the Shapiro-Wilk normality test. To evaluate the relationship between the activity/protein expression of AD enzymes and semen parameters, Pearson’s (for normally distributed data) and Spearman’s (for data that are not normally distributed) correlation analysis were used. Analysis of variance was done by one-way ANOVA and Kruskal–Wallis test. Differences between groups were tested by Tukey and Dunn’s post-hoc tests, where appropriate, according to the normality of data distribution (GraphPad Prism version 6.01 software). The results are given as means ± standard error of the mean (SEM) and the statistical significance level was set at 0.05. The ROC curves were constructed using Med Calc (Version 16.8, MedCalc Software, Belgium), and cutoff values were determined that could be used to delineate infertile normospermic, from infertile men with various sperm pathologies.

Authors’ contributions

Research design, writing paper, data interpretation: VO, AK, BK; Performed experiments, statistical analysis AK, BM; Data interpretation and revised manuscript: AJ, AS, EG.

Acknowledgments

We thank the International Science Editing for language editing and polishing the paper.

Disclosure statement

No potential conflict of interest was reported by the authors.

Supplementary material

Supplemental data for this article can be accessed here.

Additional information

Funding

This work was supported by the Ministry of Education, Science and Technological Development of the Republic of Serbia, Grant no. 173054.