Expression of cyclooxygenase-1 (COX-1) and COX-2 in human male gametes from normal patients, and those with varicocele and diabetes: a potential molecular marker for diagnosing male infertility disorders
Department of Ecology, University of Calabria, Arcavacata di Rende, Cosenza, Italy
Aquila Saveria, Department Pharmaco-Biology, Faculty of Pharmacy, University of Calabria, Arcavacata di Rende, Cosenza 87036, Italy. T: +39(0)984 496210; F: +39 0984 496203; E:email@example.com; firstname.lastname@example.org
Rising rates of varicocele and diabetes mellitus (DM) pose a significant problem to human fertility. Recent studies have pointed out the impact of cyclooxygenase (COX) in the regulation of testicular function and male fertility. Prominent COX-2 expression has been described recently in the testes of infertile patients, but little is known about the role and identity of COX isoforms in human sperm under certain disease states such as varicocele and DM. We therefore examined the expression profile and ultrastructural localization of COX-1 and COX-2 concomitantly in semen samples from healthy donors, and patients with varicocele and DM. Using Western blotting assay, ‘varicocele’ and ‘diabetic’ sperm showed enhanced COX isoforms expression with respect to the ‘healthy’ sperm. Immunogold labeling revealed human sperm anatomical regions containing COX-1 and COX-2, confirming their increased expression in pathological samples. Our data demonstrate that both COX isoforms are upregulated in the spermatozoa of varicocele and diabetic patients, suggesting the harmful effect of the diseases also at the sperm molecular level, going beyond the abnormal morphology described to date. In conclusion, COX enzymes may possess a biological relevance in the pathogenesis and/or maintenance of male factor infertility associated with varicocele and DM, and may be considered additional molecular markers for the diagnosis of male infertility disorders.
Varicocele and diabetes mellitus (DM) are well recognized causes of male sexual dysfunction and infertility; however, the precise mechanisms by which these pathologies impair male reproductive organs remain uncertain. Varicoceles, defined as abnormally dilated scrotal veins, are present in 15% of the normal male population and in approximately 40% of men presenting with infertility (WHO, 1992; Koksal et al. 2007). Whilst DM, a state of chronic hyperglycemia, is known to cause many systemic complications, male infertility is not widely recognized as one of them. Many efforts have been made in order to find new and suitable semen indicators of varicocele; nonetheless, its management continues to stimulate controversy among reproductive experts. On the contrary, studies addressing the effects of DM on male reproductive function are scarce, and the conflicting nature of the existing data reveals a distinct lack of consensus in the current literature (Bartak, 1979; Glenn et al. 2003; Agbaje et al. 2007). The long-held view that DM has little effect on male reproductive function has been challenged by recent findings showing that this condition influences fertility in numerous previously undetected ways (Mulholland et al. 2011). For instance, it has been demonstrated that there is a significantly higher incidence of primary (16%) and secondary infertility (19.1%) in diabetic patients compared with non-diabetic individuals (La Vignera et al. 2009).
Usually, ejaculates and sperm parameters are affected in a different way in diabetic patients: the most frequently influenced is the sperm motility, then the morphology and/or the volume of the ejaculate, and the least conspicuously is the sperm count (Baccetti et al. 2002; Rama Raju et al. 2011). On the basis of these observations, a negative influence of DM on the quality of the ejaculate seems to be unquestionable.
During its life, sperm undergoes two different physiological conditions. In the male genital tract or upon ejaculation, it remains in a resting state (uncapacitated); in the female genital tract, sperm changes its metabolic state and becomes capacitated. A finely tuned homeostatic system maintains the dynamic balance between energy demand and expenditure in regulating its metabolism. It is also plausible that sperm needs a self-defense mechanism proceeding throughout the male and the female genital tracts (Riccioli et al. 2003).
Cyclooxygenases (COX) are key enzymes in the conversion of polyunsaturated fatty acids and arachidonic acid to prostaglandin (PG). There are two COX isoforms, COX-1 and COX-2 (Rouzer & Marnett, 2009), the first of which is constitutively expressed in most tissues. COX-2 is an inducible isoform, but it may be usually expressed in some organs or tissues (Harris et al. 1994; McKanna et al. 1998; Maslinska et al. 1999; Yang & Bleich, 2004). Earlier studies have reported the localization of COX-2 in adult and fetal human reproductive tissues (Kirschenbaum et al. 2000). The overall effect of COX-2 suppression in mice vas deferens leads to severe motility defects in sperm, without altering sperm count or viability (Didolkar et al. 1980; Kern & Maddocks, 1995). Despite intensive investigation, the effects of COX isoforms on male reproductive biology and physiology are largely unknown. COX expression has been indirectly identified in sperm; it has been reported that the enzyme and PGs are involved in the fertilization process and, more specifically, in the acrosome reaction (Joyce et al. 1987). The presence of COX-2 in human testes of men with deranged spermatogenesis has also been demonstrated (Schell et al. 2007). Deeper insights into the effect of COX isoforms in the testicular system are required to improve novel therapeutic approaches in male infertility. To explore the possible role of COX isoforms on sperm physiology and pathology, we have evaluated the expression of COX-1 and COX-2 in human sperm from healthy donors, varicocele and patients with DM by Western blotting and immunoelectron microscopy.
We demonstrated that COX-1 and COX-2 expression was induced in both ‘varicocele’ and ‘diabetic’ sperm with respect to healthy samples. Our data indicate that varicocele and DM may contribute to a male factor of infertility through a mechanism involving COX-1 and COX-2 in human sperm, evidencing their detrimental impact on sperm at the molecular level.
Materials and methods
Percoll (colloidal PVP-coated silica for cell separation), sodium bicarbonate, sodium lactate, sodium pyruvate, dimethyl sulfoxide, Earle’s balanced salt solution and all other chemicals were purchased from Sigma Chemical (Milan, Italy). Acrylamide bisacrylamide was from Labtek Eurobio (Milan, Italy). Triton X-100, Eosin Y was from Farmitalia Carlo Erba (Milan, Italy). ECL Plus Western blotting detection system, HybondTM ECLTM, Hepes Sodium Salt were from Amersham Pharmacia Biotech (Buckinghamshire, UK). Colloidal gold conjugated goat anti-mouse IgG secondary antibody (Ab) was from Sigma Aldrich (Milan, Italy). Rabbit polyclonal anti-human COX-1 Ab, rabbit polyclonal anti-human COX-2, peroxidase-coupled anti-rabbit IgG secondary Ab were from Santa Cruz Biotechnology (Heidelberg, Germany). Colloidal gold conjugated anti-rabbit IgG secondary Ab was from Sigma Chemical.
Semen samples and spermatozoa preparations
Human semen was collected, according to the World Health Organization (WHO)-recommended procedure, by masturbation from healthy volunteer donors of proven fertility. Spermatozoa preparations were performed as previously described (Aquila et al. 2006). Briefly, semen samples with normal parameters of volume, sperm count, motility, morphology and vitality according to the WHO Laboratory Manual (WHO, 2010) were included in this study. Varicocele samples of patients who consulted us for fertility investigation were also placed in our study. Reflux of blood in the pampiniform plexus was determined by palpation employing the Valsalva maneuver. Physical examination is the reference standard to diagnose varicoceles in subfertile men. Additional radiological imaging is not necessary to diagnose subclinical varicocele, because only a varicocele detected by physical examination should be considered potentially significant (Pryor & Howards, 1987). Varicocele samples used in this study were from oligoastenoteratozoospermic patients with diagnosed varicocele of grade III (visible without palpation) on the left testis, and their ejaculates were found to have total sperm count of 13 × 106 sperm cells per ejaculate, percentage of motility PR + NP of 38%, percentage of normally formed features of 20%, and viability percentage of 60%. Controversial results were reported from the few studies existing in the literature on the effects of type 1 and type 2 DM on sperm parameters (Niven et al. 1995; Agbaje et al. 2007; Delfino et al. 2007). In our finding we had the opportunity to examine samples of patients with type 2 DM that have total sperm count of 15 × 106 sperm cells per ejaculate, percentage of motility PR + NP of 40%, percentage of normally formed features of 22%, and viability percentage of 59%. The study has been approved by the local medical-ethical committee, and all participants gave their informed consent.
Processing and treatments of human ejaculated sperm
For each experiment (many times repeated as reported in the Statistical analysis section), three healthy samples, or four varicocele or four DM samples were pooled. In fact, after liquefaction, semen samples were firstly pooled and then subjected to centrifugation (800 g) on a discontinuous Percoll density gradient (80 : 40% v:v; Aquila et al. 2002). The 80% Percoll fraction was examined using an optical microscope equipped with an oil immersion objective lens (× 100) to ensure that a pure sample containing only spermatozoa was obtained. Sperm had a motility of about 40% (grades PR + NP) and a viability of 80% both in normal and pathological samples. An independent observer inspected several fields for each slide. Particularly, the same number for both normal and pathological samples of Percoll-purified sperm was washed with unsupplemented Earle’s medium (uncapacitating medium), and was resuspended in the same medium.
Western blot analysis of sperm proteins
Percoll-purified sperm samples, washed twice with uncapacitating medium, were incubated for 30 min at 37 °C and 5% CO2, and then centrifuged for 5 min at 5000 g. The pellet was resuspended in lysis buffer as previously described (Aquila et al. 2002). An equal amount of protein (80 μg) was boiled for 5 min, separated on a 11% polyacrylamide gel electrophoresis, transferred to nitrocellulose membranes and probed with an appropriate dilution of the indicated primary Abs. The binding of the secondary Ab was revealed with the ECL Plus Western blotting detection system, according to the manufacturer’s instructions.
Transmission electron microscopy (TEM) with immunogold analysis for COX-1 and COX-2
Sperm fixed overnight in 4% paraformaldehyde were washed in phosphate-buffered saline (PBS) to remove excess fixative, dehydrated in graded alcohol, infiltrated in LR white resin, and polymerized in a vacuum oven at 45 °C for 48 h. Ultrathin sections (60 nm) were cut and placed on coated nickel grids for post-embedding immunogold labeling with the rabbit polyclonal Abs to human COX-1 and to human COX-2. Potential non-specific labeling was blocked by incubating the sections in PBS containing 5% normal goat serum, 5% bovine serum albumin (BSA) and 0.1% cold water fish gelatine at room temperature for 1 h. Sections were then incubated overnight at 4 °C with rabbit polyclonal COX-1 and COX-2 Abs, each at a dilution of 1 : 500 in PBS buffer. After, the grids were washed rigorously several times with drops of PBS + 0.1% BSA, and incubated with 10-nm γ-globulin goat anti-rabbit–gold particle complex at a dilution of 1 : 50 for 2 h at room temperature. The sections were then subsequently washed in PBS, later fixed in glutaraldehyde, counterstained in uranyl acetate and lead acetate, and examined under a Zeiss EM 900 TEM. To assess the specificity of the immunolabeling, negative controls were carried out in corresponding sections of sperm that were labeled with colloidal gold conjugated secondary Ab with normal rabbit serum instead of the primary Ab.
The experiments for TEM assay were performed in at least three independent experiments. The experiments for Western blotting analysis were performed in at least six independent experiments. Data, presented as mean ± SEM, were evaluated by the one-way analysis of variance (anova). The differences in mean values were calculated at a significance level of P ≤ 0.05. The Wilcoxson test was used after anova as post-hoc test.
COX-1 and COX-2 are both expressed in normal, varicocele and DM sperm samples
First we investigated the presence of COX-1 and COX-2 in normal, varicocele and DM sperm samples by Western blotting analysis, demonstrating that both isoforms of the enzyme are expressed in normal human sperm. COX-1 and COX-2 were detected at the expected sizes of about 70–72 kDa, as the two enzymes have similar molecular weight, and at the same level as that reported for MCF7, breast cancer cells, used as positive control (Liu & Rose, 1996). Interestingly, DM and varicocele samples showed a strong expression of COX-1 (Fig. 1a) and COX-2 (Fig. 1b). Therefore, the COX content might distinguish healthy men from those with varicocele and DM. The bands were not detected by non-immune rabbit serum (Fig. 1a1,b1), indicating that the evidenced proteins are specific for COX-1 and COX-2, respectively.
Ultrastructural COX-1 and COX-2 expression in ‘healthy’ controls
Immunoelectron microscopy demonstrated that both isoforms of COX were expressed in normal human sperm. Spermatozoa from healthy donors showed a weak but clearly identifiable immunoreaction for both COX-1 (Fig. 2) and COX-2 (Fig. 3). The electron-dense gold particles localized to the entire tail, from the middle piece to the end piece, with a faint head immunoreaction. In the sperm head, gold particles marking COX-1 and COX-2 were mainly present on the apical region of the acrosome and in the nucleus, while no appreciable labeling was detected over the post-acrosomal area and in the neck region. The density of gold particles appeared similar in the midpiece compared with the principal piece. In the midpiece of the sperm tail, label for COX-1 and COX-2 was found in the axoneme, in the swollen space between the mitochondria and only occasionally in association with the outer mitochondrial membrane. There was also some labeling between the ribs of the fibrous sheet both in the middle and the principal piece of the tail. All corresponding sections treated with BSA/PBS instead of primary antibodies, which served as negative controls, were free of labeling (Fig. S1).
Ultrastructural COX-1 and COX-2 expression in ‘varicocele’ sperm
A significant increase of COX-1 (Fig. 4) and COX-2 (Fig. 5) expression was observed in ‘varicocele’ sperm with respect to the healthy samples. The two COX isoforms exhibited a similar distribution pattern in the sperm cells; however, COX-2 appears to be the most abundant isoform. In the head, immunoreactions for COX-1 and COX-2 localized both at the plasma membrane and in the nucleus. Heavily labeled regions appeared to correspond to the connecting piece (neck), especially the basal plate and the segmented columns, and the mitochondria-rich midpiece of the sperm tail. In both longitudinal and transverse sections, label was found along the entire length of the middle piece and often near the edges of the mitochondrion, suggesting that both COX-1 and COX-2 were predominantly present on the outer mitochondrial membrane. COX-1 and COX-2 staining was also associated with the axoneme and the outer dense fibers. In the principal piece, gold particles detecting COX-1 and COX-2 were distributed between the elements of the axoneme and the fibrous sheet. Negative controls were almost completely free of gold-sphere labeling (Fig. S1).
Ultrastructural COX-1 and COX-2 expression in sperm from diabetic men
When compared with the healthy controls, sperm from diabetic donors displayed an intense immunoreaction for both COX-1 (Fig. 6) and COX-2 (Fig. 7). In the sperm head, COX-1 was sparsely present throughout the nucleus and along the plasma membrane, while COX-2 was found exclusively in the nucleus. The sperm neck showed staining for both COX-1 and COX-2 next to the basal plate and the segmented columns. Patches of labeling were found along the middle and the principal piece of the tail. In both these regions, immunolabeling for COX-1 and COX-2 was found to be associated with the axoneme, the outer dense fibers and the fibrous sheet. In the midpiece region, gold particles marking both the isoforms were also found around the mitochondria and in association with the outer mitochondrial membrane. Negative controls were completely devoid of label (Fig. S1).
Hormonal imbalances, altered sperm morphology and sperm functional deficits have been demonstrated in varicocele and DM; however, the negative impact of these disorders on male fertility still remains poorly understood. It has been recently proposed that COX and PGs may have particular relevance in most male fertility disorders and may play a key role in the regulation of sperm metabolism. Two isoforms of COX have been described in mammalian cells, referred to as COX-1 and COX-2 (Cryer & Dubois, 1998; Botting, 2006). In terms of their molecular biology, COX-1 and COX-2 are of similar molecular weight, approximately 70 and/or 72 kDa, having 65% amino-acid sequence homology and near-identical catalytic sites. Generally, while COX-1 is ubiquitous, COX-2 operates as an inducible enzyme, although some tissues, such as brain, kidney and reproductive organs, do express it constitutively (O’Neill & Ford-Hutchinson, 1993; Maslinska et al. 1999; Adegboyega & Ololade, 2004). It is known that COX-2 levels are markedly increased in inflamed tissues; however, it is also induced by factors such as cytokines, steroid hormones and mitogenic stimuli (Seibert & Masferrer, 1994). With respect to reproductive biology, induction and/or presence of COX-2 has been reported in the male reproductive tract; however, its localization and expression levels significantly differ between the species. In rat testis, COX-2 has been found to localize in germ and Leydig cells, while in mice it is restricted to interstitial cells of Leydig (Lazarus et al. 2002; Neeraja et al. 2003). Discordant data are reported about the presence of COX-2 in human testes. For instance, it has been shown that COX-2 is abundantly expressed in testicular biopsies of men suffering from different forms of impaired spermatogenesis, while it is absent in the testes of men with normal spermatogenesis (Frungieri et al. 2002, 2007). It has been also demonstrated that the extent and intensity of COX-1 and COX-2 expression in testicular cancer cells is higher than in normal tissues (Yoshimuraa et al. 2003). COX-2 expression has been recently shown in the testes of diabetic rats (Kushwaha & Jena, 2012), while no data have been reported for varicocele. Herein, we used Western blot and immunogold electron microscopy to define the expression pattern and the subcellular localization of COX-1 and COX-2 in ‘healthy’ and ‘diseased’ human sperm. To the best of our knowledge, the present study is the first in which immunocytochemistry analysis was applied to analyze COX-1 and COX-2 expression in semen samples from healthy volunteer donors of proven fertility, varicocele and DM patients. So far, just the ‘COX’ was reported in the homogenates of bovine spermatozoa (Shalev et al. 1994). Our results demonstrate the presence of COX-1 and COX-2 in healthy human sperm and a concomitant increase in the amounts of the enzymes in the pathological samples. COXs analyzed by Western blotting were readily detectable in the proteic lysates of human ejaculated spermatozoa at the same apparent molecular weight (70–72 kDa) as the stained band of MCF7 used as positive control cells (Liu & Rose, 1996). The immunocytochemical data corroborate the findings of Western blot analysis, demonstrating an augmented expression of COXs in both varicocele and diabetes samples. In ‘healthy’ sperm, we found rare gold particles that prevalently decorate the apical region of the acrosome region and the entire tail, from the middle piece to the end piece. A similar pattern of COX expression (without a clear distinction between the two isoforms) was detected by immunofluorescence microscopy in bovine spermatozoa, where the protein was found to be localized to both in the sperm head and in the midpiece (Shalev et al. 1994). ‘Varicocele’ sperms showed a strong increase of the COXs both in the head (sperm membrane and nucleus) and in the tail, specifically on the outer mitochondrial membrane. An augmented expression of COX-1 and COX-2 has been also observed, with some differences in the distribution and content, in the sperm from diabetic patients. Intriguingly, COX-2 was mainly present in the nucleus rather than on the sperm membrane as demonstrated for COX-1. Additionally, COX-2 protein levels are often higher than those observed for COX-1. It is generally recognized that COX-1 is constitutively expressed in most tissues, while COX-2 may be induced by a variety of stimuli. It has been demonstrated that under basal conditions, inducible COX-2 dominates over COX-1 in pancreatic islet cells (Sorli et al. 1998). Because sperm and beta-pancreatic cells (Aquila et al. 2005a) have many common characteristics, it may be possible that under normal conditions they also possess the same COX expression pattern. In the pathological samples, much of the increase in COX-2 expression occurs in the midpiece, suggesting an altered oxidative metabolism. On the basis of these observations it may be supposed that the COX isoforms are involved in different sperm activities according to their specific cellular localization, and therefore in acrosome reaction, energy metabolism and sperm motility. The physiological presence of COX in the above-mentioned cellular compartments suggests a functional importance for their eicosanoid products in human sperm. PGs play key roles in the male reproductive tract, and high concentrations of PGs are reported to exist in seminal fluid (Templeton et al. 1978). In addition, human spermatozoa have been shown to synthesize PGs (Roy & Ratnam, 1992). PGs are involved in all stages of sperm maturation, from spermatogenesis to the acrosome reaction (Aitken & Kelly, 1985; Aitken et al. 1986; Hayashi et al. 1988; Shimizu et al. 1998). These autacoids have been found to be associated with the acrosome of male germ cells rather than with the other parts of the sperm, thus suggesting that PGs and COX might have a major impact on the acrosome reaction (Roy & Ratnam, 1992). PGs can alter the biochemical activity of the membrane in somatic cells (Hall & Behrman, 1982), thus their production in ‘healthy’ sperm or in the surrounding environment might enhance fertilizing capacity probably by augmenting the acrosome reaction (Vesin & Harbon, 1974; Rogers & Bentwood, 1980). Under disease states, COX upregulation might induce a progressive PG imbalance that in turn might contribute to an increased oxidative stress with a consequent damage of mitochondrial and nuclear sperm DNA. For instance, it has been proven that diabetic men have an increased sperm nuclear and mtDNA damage (Agbaje et al. 2007), and also patients with varicocele show high levels of DNA damage, reactive oxygen species production and apoptosis (Dada et al. 2010). It has been reported that COX-1 appears to be upregulated in different carcinomas (Hwang et al. 1998), and both COX isoforms are induced in testicular cancer. The extent and intensity of immunoreactive COX-1 and COX-2 in these cells were greater than in normal testis tissues (Yoshimuraa et al. 2003). Therefore, the increased proportion of COX expression in varicocele and diabetes may be due, at least in part, to an alteration of the metabolic state of the sperm cells, as occurs in other diseases. Besides, because COX and PGs are players present in different types of immune response and both in varicocele and diabetes encompass inflammatory diseases, the increased expression of COX might also reflect a protective role against inflammation (Li et al. 2009). In the normal testes, sperms release suppressor factors to inhibit immune reactions, generating an immunosuppressive environment that promotes sperm transport through the female reproductive tract. These data, considering that sperms have the capability to synthesize PGs, raise the possibility that PGs could participate in the formation of the immunoprotective barrier to ensure fertilization. The possibility that in humans PGs might play a self-protective role for the spermatozoon cannot be excluded, and awaits experimental information. From our previous studies it appears that sperm is able to modulate its own metabolism independently by the systemic regulation (Aquila et al. 2005b). Recently, a putative role for COX-2-derived PGJ2 in the lipid metabolism has been suggested in human sperm cells (Aquila et al. 2006). In fact, we have previously reported that PGJ2 primarily acts through a peroxisome proliferator-activated receptor (PPAR)-gamma-dependent pathway, addressing a lipolytic effect on human sperm metabolism. During capacitation, energy request increases, and capacitated sperms show enhanced metabolism and overall energy expenditure. In this scene, COX enzymes might cooperate with other factors providing additional metabolic fuel to sustain the capacitation process. Nearly all reproductive tissues have been shown to produce PGs (LeMaire et al. 1975; Warnes et al. 1978; Gerozissis & Dray, 1981; Harrison et al. 1987). The concentrations of PGs in semen and menstrual fluid probably represent the highest found in the human body. The effect of PGs in seminal plasma has been claimed to influence sperm motility and sperm migration, but the evidence is controversial. Because even though under physiological conditions seminal plasma rarely reaches the uterus, it seems unlikely that seminal PGs may affect the acrosome reaction and fertilization in vivo. However, it is certainly possible that the PGs present in the female genital tract may influence the fertilizing capacity of spermatozoa (Schuetz & Dubin, 1981; Evans et al. 1983). These findings demonstrate that COXs and PGs have an important role in sperm physiology.
Altogether, these experimental observations lead us to hypothesize that the PGs may act using two different mechanisms: through an autocrine short loop involving PPAR-gamma, which provides the recruitment of energy substrates according to sperm metabolic needs; and/or in a paracrine fashion, actuating a self-protective mechanism against inflammation and oxidative stress.
In conclusion, the current finding showed that the COX isoforms might have a biological relevance in the pathogenesis and/or maintenance of infertility states. A potential therapeutic value of COXs as targets in future strategies designed for the treatment of fertility disorders remains to be further investigated.
Our special thanks to Dr Vincenzo Cunsolo (Biogemina Italia Srl, Catania – Italy) for the technical and scientific assistance. We would like also to thank Perrotta Enrico for the excellent technical and scientific assistance, and Sturino Domenico for the English language review of the manuscript. This work was supported by MIUR Ex 60% -2011.
Perrotta I.: contributions to concept/design/acquisition of data; Santoro M.: contributions to concept/design/acquisition of data; Guido C.: acquisition of data; Avena P.: acquisition of data; Tripepi S.: contributions to concept; De Amicis F.: data analysis/interpretation; Gervasi M.C.: data analysis/interpretation; Aquila S.: design, drafting of the manuscript, critical revision of the manuscript and approval of the article.