Human Reproduction Update Advance Access originally published online on June 24, 2004
Human Reproduction Update 2004 10(5):387-399; doi:10.1093/humupd/dmh034
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Regulation of sperm function by reactive oxygen species
University of Bristol, Department of Clinical Sciences South Bristol (Obstetrics & Gynaecology), St Michael's Hospital, Southwell Street, Bristol BS2 8EG, UK
Email: chris.ford{at}bristol.ac.uk
 |
Abstract |
|---|
 TOP Abstract IntroductionReactive oxygen speciesBiological roles of ROSEvidence that ROS modulate...Source of ROS involved...ConclusionsReferences |
|---|
Sperm capacitation can be increased by the addition of reactive oxygen species (ROS) and decreased by antioxidants. Broadly consistent results have been achieved with a wide variety of methods and across different species. Exposure to ROS increases protein tyrosine phosphorylation consequent on an increase in cAMP and activation of tyrosine kinase and inhibition of tyrosine phosphatase. The measurement of ROS production by sperm is complicated by contamination of suspensions by leukocytes, laying many studies open to doubt. In human sperm the observation that extracellular NADPH could support superoxide production detected with the chemiluminescent probe lucigenin and had physiological effects similar to hydrogen peroxide led to the suggestion that they contained NADPH oxidase activity to generate ROS to support capacitation. However, the realization that lucigenin can signal superoxide artefactually, combined with failure to detect superoxide production using spin trapping techniques or to detect NADPH oxidase components in mature sperm, and confirmation of old reports that NADPH solution contains substantial amounts of hydrogen peroxide due to autoxidation, have undermined this hypothesis. Although the presence of significant NADPH oxidase activity in mature human sperm now seems less likely, other observations continue to suggest that they can make ROS in some way. There is stronger evidence that animal sperm can make ROS although these may be mainly of mitochondrial origin.
Key words: capacitation / cyclicAMP / NADPH oxidase / reactive oxygen species / sperm
|
Introduction |
|---|
|
TOPAbstractIntroductionReactive oxygen speciesBiological roles of ROSEvidence that ROS modulate...Source of ROS involved...ConclusionsReferences |
|---|
Sperm are released from the testis with tightly supercoiled DNA which is transcriptionally inert (Sotolongo and Ward, 2000
) but nevertheless they undergo profound changes in their functional capacity as they mature in the epididymis, capacitate in the female reproductive tract and fertilize. These changes must be achieved by modifications of existing proteins rather than changes in gene expression and can be modulated by signals from the sperm's environment or may occur spontaneously. For example, capacitation can be regulated by substances present in semen (Fraser and Adeoya-Osiguwa, 2001), by the female reproductive tract (Smith, 1998) or by progesterone and other substances secreted by the oocyte cumulus complex (Blackmore, 1993) but can also occur spontaneously under suitable conditions in vitro. This article reviews the evidence that reactive oxygen species (ROS) are among the messengers that influence sperm function during their journey from the testis to the oocyte and that they are intrinsic signals produced by sperm themselves. For reasons of space, the effects of reactive nitrogen species (RNS) will only be discussed where they interact with the effects of ROS. |
Reactive oxygen species |
|---|
|
TOPAbstractIntroductionReactive oxygen speciesBiological roles of ROSEvidence that ROS modulate...Source of ROS involved...ConclusionsReferences |
|---|
The term ROS includes reduced forms of oxygen and their reaction products with other molecules. Some but not all are free radicals. Free radicals are highly chemically reactive because they contain unpaired electrons. Diatomic oxygen (O2) contains two unpaired electrons but is relatively inert because these have parallel spins (Halliwell and Gutteridge, 1999
). Successive reduction of oxygen to superoxide and peroxide removes the spin restriction so these molecules are more reactive than O2 even though peroxide is not a free radical. The extremely reactive hydroxyl radical can be formed by reaction between superoxide and hydrogen peroxide or reduction of peroxide catalysed by ferrous ions. Superoxide reacts very avidly with nitric oxide to form peroxinitrite, and hydrogen peroxide can react with chloride to form hypochlorite (Figure 1). These products are among the most biologically important ROS but the term also includes singlet oxygen, ozone and organic peroxyl and alkoxyl radicals (Halliwell and Gutteridge, 1999).
|
|
Biological roles of ROS |
|---|
|
TOPAbstractIntroductionReactive oxygen speciesBiological roles of ROSEvidence that ROS modulate...Source of ROS involved...ConclusionsReferences |
|---|
ROS can react with a wide range of biological molecules, notably unsaturated fatty acids, sulphydryl proteins and nucleic acids and are implicated in a large number of diseases, e.g. arthritis, artherosclerosis and degenerative diseases of ageing (Halliwell and Gutteridge, 1999
). The first report that they could have harmful effects on sperm was published over 60 years ago (Macleod, 1943) and it is now generally accepted that ROS production in sperm suspensions, lipid peroxidation and DNA oxidation are associated with poor sperm function and subfertility (Aitken et al., 1991, 1993; Sukcharoen et al., 1995; Griveau and LeLannou, 1997a; Storey, 1997; Lopes et al., 1998; Whittington et al., 1999; Pasqualotto et al., 2000; Shen and Ong, 2000; Agarwal et al., 2003).
However, ROS also have physiological roles. They are produced by leukocytes as part of the phagocytic process to kill engulfed bacteria (Babior, 1999) but also in smaller amounts by other cell types to act as cell messengers. In the latter function they often act differently from traditional messengers that rely on conformational specificity to bind to sites on their receptor but without producing any covalent modification. By contrast, ROS often change the function of their targets through oxidation, usually of sulphydryl groups, and specificity is achieved through the chemical reactivity of the target. Such effects can be produced by the ROS itself or indirectly through oxidation of intracellular thiols, predominantly glutathione. These mechanisms offer novel possibilities of achieving specificity and cross-talk between different regulatory pathways (Cooper et al., 2002).
ROS have been implicated in many pathways of cell regulation. These topics have been reviewed recently (Finkel, 1998; Hensley et al., 2000; Thannickal and Fanburg, 2000; Droge, 2001; Chiarugi and Cirri, 2003; Finkel, 2003) and only some of the most recent original papers will be cited in the brief survey that follows. ROS can regulate gene expression, notably up-regulation of anti-oxidant proteins in the face of a higher level of oxidative stress. They are involved in the mechanism of ‘oxygen sensors’, e.g. in the carotid body (Jones et al., 2000). They regulate cell adhesion and antibody production by leukocytes. Although not mandatory for apoptosis, they have pro-apoptotic effects. They can enhance the action of epidermal and platelet-derived growth factors, since hydrogen peroxide facilitates autophosphorylation of the receptor, so increasing its affinity for the agonist. Relatively high hydrogen peroxide concentrations (
1 mmol/l) or large oxidative shifts in thiol status can inhibit tyrosine phosphatases and so increase protein tyrosine phosphorylation triggered by tyrosine kinases. A reversible regulatory system is achieved by the cyclic oxidation and reduction of key sulphydryl groups involving a novel sulphenyl-amide (Salmeen et al., 2003; van Montfort et al., 2003). Relatively small concentrations of hydrogen peroxide or thiol status can activate the tyrosine kinase activity of the insulin receptor as well as intracellular kinases such as MAP kinase and protein kinase C. Oxidants can increase intracellular calcium concentration as well as acting through protein phosphorylation.
|
Evidence that ROS modulate sperm function |
|---|
|
TOPAbstractIntroductionReactive oxygen speciesBiological roles of ROSEvidence that ROS modulate...Source of ROS involved...ConclusionsReferences |
|---|
Capacitation
Freshly ejaculated sperm cannot fertilize until they have spent some time in a suitable environment in order to capacitate. Capacitated sperm acquire the ability to exhibit hyperactivated motility and to undergo a physiological acrosome reaction (Yanagimachi, 1994). Capacitation is associated with a number of biochemical events, most notably an increase in protein tyrosine phosphorylation (Visconti and Kopf, 1998; Visconti et al., 1998; Baldi et al., 2000, 2002; Guraya, 2000; Breitbart, 2003). Although it can occur spontaneously under suitable conditions in vitro, in vivo its progress can be regulated by the female reproductive tract (Smith and Yanagimachi, 1989; Hunter et al., 1998).
ROS have been reported to promote sperm capacitation in hamster, mouse, rat, bull, horse and human sperm. Capacitation has been assessed by the chlortetracycline (CTC) staining pattern, by the ability of the sperm to acrosome-react spontaneously or after stimulation with lysophosphatidyl choline, A23187 or progesterone, by the zona-free hamster oocyte test, by the prevalence of hyperactivation and by IVF. ROS have been generated by enzymes using xanthine/xanthine oxidase or glucose oxidase, by adding activated leukocytes, by direct addition of H2O2 or by addition of substances such as fetal cord serum ultrafiltrate that stimulate endogenous ROS production. NAD(P)H has also been used in this way but it remains uncertain whether it stimulates ROS production or acts as a source of H2O2 due to autoxidation (see below). ROS have been removed with the enzymes superoxide dismutase (SOD) and catalase or with various antioxidants. With few exceptions, adding ROS increased capacitation whereas removing them decreased it whatever methods were used. Some of the effects of ROS could be mimicked by other oxidizing agents, notably thiol oxidants. The studies that support these statements are summarized in Table I. They were done in many different laboratories by a variety of techniques and together provide convincing evidence that redox mechanisms promote sperm capacitation in several species.
|
Maturation
Sperm released from the testis are unable to exhibit progressive motility or to capacitate but acquire these abilities during their passage through the epididymis. These processes are referred to as maturation; other maturational changes include the completion of nuclear condensation and changes in the expression and distribution of molecules on the sperm surface (Cooper, 1995; Moore, 1996).
It has been proposed that ROS play a role in the regulation of maturation. Phospholipid hydroperoxide glutathione peroxidase (GPX4) can utilise the thiol groups in nuclear proteins as an alternative reductant to glutathione. Generation of lipid peroxides by ROS might provide a substrate for GPX4 to drive the oxidation of these proteins and to facilitate nuclear condensation (Aitken and Vernet, 1998). ROS and might also be involved in motility initiation by enhancing cAMP synthesis and protein phosphorylation at ejaculation (Aitken, 2000).
Cell signalling pathways
There is a broad measure of agreement that the potentiation of capacitation by ROS is associated with increased protein tyrosine phosphorylation and shares features with the cAMP-dependent capacitation pathway. The main targets are proteins with a mol. wt of 100 kDa. Aitken et al. (1995) identified a protein of 116 kDa whereas Leclerc et al. (1997) identified proteins migrating with mol. wt of 105 and 81 kDa. It is likely that the bands are complex and that two components are fibrous sheath proteins related to protein kinase A anchoring protein and its precursor (Carrera et al., 1996; Ficarro et al., 2003).
Physiological concentrations of ROS have been proposed to enhance sperm capacitation by increasing cAMP synthesis and by inhibiting protein tyrosine phosphatases whilst activating tyrosine kinase (Figure 2). Evidence that they increase intracellular cAMP concentrations includes the observations that addition of a phosphodiesterase inhibitor or of dibutyryl cAMP had a similar potency to stimulating ROS production by addition of NADPH, in promoting human sperm fusion with zona-free hamster oocytes (although there was some additive effect when both cAMP and NADPH were both added) and that increasing cAMP or adding NADPH produced a similar pattern of tyrosine phosphorylation that could be blocked by the protein kinase A inhibitor H89 (Aitken et al., 1998a). Exposure to superoxide increased sperm cAMP concentration (Zhang and Zheng, 1996) and exposure to NADPH produced a larger increase in intracellular cAMP than the phosphodiesterase inhibitor pentoxifylline (Aitken et al., 1998a). In the latter study, neither pentoxifylline nor dibutyryl cAMP had any effect on ROS production. By contrast, Leclerc et al. (1998) argued that increasing intracellular Ca2 + activity during capacitation provoked an increase in cAMP concentration that stimulated superoxide production which in turn induced protein tyrosine phosphorylation. It has been proposed that in maturing rat sperm, superoxide stimulates tyrosine kinase indirectly by raising the cAMP concentration whereas H2O2 directly stimulates tyrosine kinase and inhibits tyrosine phosphatase (Lewis and Aitken, 2001), and this is broadly consistent with conclusions from human sperm (de Lamirande et al., 1998b). However, 50 µmol/l H2O2 increased the cAMP concentration in bull sperm by stimulating adenyl cyclase activity and increased tyrosine phosphorylation in a similar way to capacitating agents such as heparin. ROS activation of adenyl cyclase has been reported in cultured rat embryo thoracic aorta cells (Tan et al., 1995) and in rat adipocytes (Raimondi et al., 2000) although in the latter study inhibition of phosphodiesterases could be an alternative explanation. In sperm, one intriguing possibility is that the bicarbonate-sensitive soluble adenylyl cyclase (sAC) is activated by H2O2 as well as bicarbonate (Rivlin et al., 2004). sAC is expressed much more strongly in male germ cells and sperm than in other tissues and exists as two functional splice variants (Buck et al., 1999; Sinclair et al., 2000; Jaiswal and Conti, 2001). It is essential for fertility in male mice (Esposito et al., 2004). The 469 amino acid variant of rat sAC contains 12 cysteines of which eight are in catalytic domains (Buck et al., 1999) so it is likely to be open to redox regulation.
|
ROS-dependent tyrosine phosphorylation was blocked by H89 whereas this inhibitor had no effect on tyrosine phosphorylation induced with the tyrosine phosphatase inhibitor vanadate (Rivlin et al., 2004
). Although these results must be interpreted with caution, since vanadate is a very non-specific inhibitor, they suggest that activation of the kinase pathway is an important component of the effect of ROS on protein phosphorylation. The cAMP-dependent increase in protein tyrosine phosphorylation may be catalysed at least in part by the tyrosine kinase c-yes (Leclerc and Goupil, 2002). An extracellular signal-regulated kinase (ERK) pathway has been implicated in the regulation of capacitation upstream of tyrosine phosphorylation (de Lamirande and Gagnon, 2002). Double phosphorylation of the threonine-glutamine-tyrosine motif characteristic of ERK1/2 activation is regulated by nitric oxide (Thundathil et al., 2003). Stimulation of this event by fetal cord serum ultrafitrate was blocked by the nitric oxide synthase inhibitor NG-nitro-L-argenine methyl ester (L-NAME) but not SOD or catalase. However, superoxide did influence phosphorylation of the Thr-Glu-Tyr motif in proteins of lighter mol. wt (16–33 kDa) and regulated phosphorylation of some insoluble ERK1/2 substrates in parallel with the ERK pathway (de Lamirande and Gagnon, 2002). It is also likely that superoxide may react with nitric oxide to form peroxinitrite, which can induce tyrosine nitration of sperm proteins as well as tyrosine phosphorylation (Herrero et al., 2001).
It is possible that the effects of ROS are transduced through changes in the redox state of protein sulphydryl groups. Addition of reduced sulphydryl compounds can block capacitation and antagonize the effects of ROS. Addition of compounds that react with sulphydryl groups promoted capacitation; this effect was concomitant with increased superoxide production and was blocked by SOD, although SOD did not affect changes in protein tyrosine phosphorylation (de Lamirande and Gagnon, 1998). By contrast, sulphydryl reducing agents impeded capacitation and decreased superoxide production and protein tyrosine phosphorylation (Aitken et al., 1996b; de Lamirande and Gagnon, 1998). Capacitation was associated with rapid (10–15 min) changes in the sulphydryl content of 
10 proteins; five exhibiting an increase and five a decrease. The changes were reversed after 30–120 min. The changes were prevented by SOD and catalase but were similar whether induced by fetal cord serum ultrafiltrate, follicular fluid ultrafiltrate or progesterone (de Lamirande and Gagnon, 2003). The increasing availability of proteomic technologies may assist the identification of target proteins.
Conclusions on the regulation of sperm function by ROS
There is convincing evidence that ROS can modulate sperm capacitation in vitro. Capacitation judged by a variety of endpoints can be accelerated by addition of ROS in different ways and retarded by trapping ROS in various ways. Similar results can be achieved with sperm from different species. However, many questions remain unanswered about the mechanism of action. What are the roles of the different ROS family members? What are the ROS receptors in the sperm cell? Do ROS activate sperm adenylyl cyclase and, if so, how? Other questions concern the source of the ROS: are they generated by the sperm themselves and, if so, how is their production controlled; or are they derived from the environment? The remainder of this review will be concerned with the latter problem.
|
Source of ROS involved in regulation of sperm function |
|---|
|
TOPAbstractIntroductionReactive oxygen speciesBiological roles of ROSEvidence that ROS modulate...Source of ROS involved...ConclusionsReferences |
|---|
Pathways of ROS production in cells
In cells, ROS can be produced by intracellular oxidases and peroxidases, e.g. xanthine oxidase, or as an alternate product by other enzymes such as cytochrome p450 and nitric oxide synthase; but perhaps the main source in most cells is leakage of electrons from the electron transport chain (Chance et al., 1979; Halliwell and Gutteridge, 1999). ROS produced by these means are usually regarded as toxic by-products of other metabolic processes and are implicated in the aetiology of disease and ageing (Raha and Robinson, 2000) although mitochondrial ROS production can be controlled and is involved in cell regulation including the initiation of apoptosis (Brookes et al., 2002).
Some cells have other mechanisms to produce ROS for physiological purposes. Phagocytic leukocytes contain a NADPH oxidase which secretes superoxide into the phagocytic vesicle during the ‘killing reaction’. The ‘core’ of this enzyme, termed gp91phox, contains a flavoprotein and a cytochrome b558 component and is activated by combining with regulatory subunits (Segal and Abo, 1993; Babior et al., 1997; Babior, 1999). It has become clear that gp91phox is a member of a NOX enzyme family widely distributed in other cell types and involved in generating ROS for regulatory and other purposes (Lambeth et al., 2000). There is current controversy about whether sperm contain NADPH oxidase activity of this sort. Another potential source of ROS production is reduction of oxygen by plasma membrane redox systems (PMRS) designed to transport reducing power across the cell plasma membrane (de Grey, 2003)
Here, I review the evidence that mammalian sperm produce ROS and contain NADPH oxidase activity and discuss some of the technical problems that complicate the investigation of these questions. A sceptical view has been adopted to encourage readers to consider the quality of the evidence critically before accepting that sperm produce ROS by a given mechanism.
The problem of leukocyte contamination
Sperm preparations usually contain some contaminating leukocytes. Activated leukocytes are prolific sources of ROS (Babior et al., 1997) and leukocytes are the predominant source of ROS in crude suspensions of human sperm (Aitken and West, 1990; Aitken et al., 1992b, 1994a; Kessopoulou et al., 1992; Whittington and Ford, 1999) and the ROS they produce can explain the long-recognized toxicity of oxygen towards human sperm (Macleod, 1943; Whittington and Ford, 1998; Whittington et al., 1999). As discussed in greater detail below, human sperm themselves make relatively small amounts of ROS, if any. Consequently sensitive detection methods have to be used and these can detect ROS production from very few leukocytes, e.g. electron paramagnetic resonance spectroscopy with the spin trap DEPMPO can detect superoxide produced by as few as 2000 activated leukocytes/ml. Experiments with sperm often use sperm concentrations of 20 x 106/ml. Therefore the presence of one activated leukocyte per 20 000 sperm would produce a detectable amount of ROS. Luminescent detection methods are more sensitive still, implying that a yet lower level of leukocyte contamination might be significant (Ford, 2003). Fortunately leukocytes can be detected very sensitively and specifically by immunocytochemistry (Kessopoulou et al., 1992) or by challenging the suspension with the chemotactic peptide N-formylmethionyl-leucyl-phenylalanine (NFMLP) whilst measuring ROS production with the luminol–horseradish peroxidase (HRP) system. Phagocytes have NFMLP receptors and respond with a large increase in ROS production whereas sperm do not (Krausz et al., 1992). Leukocytes can be removed from sperm suspensions by using magnetic beads coated with antibodies against the common leukocyte antigen CD45 often combined with anti-CD15 for greater effectiveness against neutrophils and monocytes (Aitken et al., 1996a; Richer and Ford, 2001). Unless leukocytes are removed in this way and their absence is confirmed by an NFMLP challenge test or by immunocytochemistry then there is a risk that any ROS detected are produced by leukocytes and claims that they are derived from sperm have to be treated with caution.
Detection of ROS production by sperm
The first indication that human sperm might make ROS was the demonstration by Macleod (1943) that catalase could protect sperm against the harmful effects of oxygen. In retrospect this was probably due to the presence of leukocytes in the sperm suspension. Other early work demonstrated that bull, boar and ram sperm could produce H2O2 through an oxidase acting on L-aromatic amino acids (Tosic and Walton, 1950).
Holland and Storey (1981) demonstrated that rabbit sperm produced H2O2 by three routes, a cytoplasmic system that required low mol. wt co-factors and rotenone-sensitive and -insensitive mitochondrial pathways. This is likely to be due to the dismutation of superoxide since a later paper demonstrated that rabbit sperm produce superoxide much of which is degraded by SOD (Holland et al., 1982). Mouse sperm also produced superoxide much of which was converted to H2O2 by SOD (Alvarez and Storey, 1984). These studies used acetylated cytochrome C to detect superoxide and acetylated cytochrome C oxidation by cytochrome C peroxidase to detect H2O2. These are reliable methods but although likely to be less than for human ejaculated sperm the impact of leukocyte contamination on these results is unclear. Similar techniques were used to demonstrate the production of superoxide and H2O2 by human sperm (Alvarez et al., 1987) but this is more likely to reflect contamination by leukocytes.
Aitken and Clarkson (1987) used the chemiluminescent probe, luminol, to demonstrate ROS production by human sperm. ROS production was associated with defective sperm function. It was greater in less dense fractions from Percoll gradients and was increased by centrifugation (Aitken and Clarkson, 1988; Aitken et al., 1989a,b). Similar results were reported by Gagnon's group (Iwasaki and Gagnon, 1992). It became clear that much of the ROS detected in these experiments was produced by leukocytes as discussed above. Nevertheless a number of papers detected ROS production in sperm suspensions where no leukocytes could be detected or that had been depleted of leukocytes with Dynabeads (Aitken and West, 1990; Aitken et al., 1994b, 1996a; Whittington and Ford, 1999). ROS production by these fractions could be stimulated by phorbol esters but not by NfMLP (Krausz et al., 1992, 1994). Sperm in such fractions were characterized by a high cytoplasmic volume associated with a high content of cytoplasmic enzymes including creatine kinase and glucose-6-phosphate dehydrogenase (Aitken et al., 1994c; Gomez et al., 1996; Gil-Guzman et al., 2001). It was proposed that the presence of excess glucose 6-phosphate dehydrogenase allowed the production of greater amounts of NADPH that drove superoxide production by an NADPH oxidase-like enzyme located in the sperm plasma membrane (Aitken et al., 1994c; Gomez et al., 1996). In these experiments, ROS was detected with a peroxidase-enhanced luminol assay that detects mainly extracellular H2O2, which may derive from the dismutation of superoxide by SOD (Aitken et al., 1992a).
Aitken et al. (1995) demonstrated that externally added NADPH could promote sperm capacitation in a similar way to hydrogen peroxide, as judged by the zona-free hamster oocyte test and tyrosine phosphorylation. Superoxide production measured by lucigenin chemiluminesence could be supported by externally added NADPH or NADH and was inhibited by flavoprotein inhibitors in a similar way to the leukocyte NADPH oxidase. Chemiluminescence was not affected by mitochondrial or diaphorase inhibitors. NADPH oxidase activity was localized in a membrane fraction. It was also possible to detect superoxide production in response to high NADPH concentrations by SOD-sensitive cytochrome C reduction, although the effect of inhibitors was not tested in this system (Aitken et al., 1997; Griveau and LeLannou, 1997b). It was proposed that superoxide production by the NADPH oxidase supported a novel cAMP-dependent pathway of tyrosine phosphorylation required for capacitation (Aitken et al., 1998a). Addition of increasing concentrations of NADPH to human sperm suspensions at first mimicked the pro-capacitation effects of H2O2 but as the concentration exceeded 5–10 mmol/l caused lipid peroxidation and DNA oxidation similar to the pathological effects of higher concentrations of H2O2 (Aitken et al., 1998b; Twigg et al., 1998a,b). In these experiments leukocytes were removed with Dynabeads and their absence confirmed by a NFMLP challenge. These observations are consistent with the presence of NADPH oxidase-like activity in the human sperm plasma membrane. Further support for this idea came from the demonstration that the gene for the NADPH oxidase family member NOX5 is expressed in human primary spermatocytes (Banfi et al., 2001) although the presence of the enzyme has never been confirmed in mature sperm.
However, experiments using the alternative luminescent probe 2-methyl-6-(p-methoxyphenyl)-3,7-dihydroimidazo [1,2a]pyrazin-3-one (MCLA) failed to demonstrate an increase in superoxide production in response to NADPH, although stimulation of superoxide production was observed in response to biological fluid ultrafiltrates (De Lamirande et al., 1998a). The discrepancy might be explained by the properties of the lucigenin detection system. The detection of superoxide by lucigenin requires its reduction to a lucigenin cation radical. This reaction can be catalysed by many intracellular dehydrogenases. The lucigenin cation radical can react with superoxide to form an unstable dioxetane intermediate that decomposes to produce two molecules of N-methylacridone, one of which is in an excited state and emits a photon upon its return to the ground state. Unfortunately the lucigenin cation radical can also react with molecular oxygen to form superoxide and regenerate lucigenin. The superoxide can react with other lucigenin cation radicals to produce light emission. Thus lucigenin can both produce and signal superoxide in the absence of any physiological production (Figure 3a) (Liochev and Fridovitch, 1997, 1998). Richer and Ford (2001) demonstrated that such redox cycling occurred in human sperm since NADPH consumption was increased by the addition of lucigenin (Figure 3b). They were unable to detect ROS production by human sperm using EPR spectroscopy and the spin trap DEPMPO or with the sensitive H2O2 probe ‘Amplex red’ and they were unable to detect consistently SOD-sensitive cytochome C reduction. In samples where cytochome C reduction was detected it was not sensitive to diphenylene iodonium, an effective although non-specific inhibitor of NADPH oxidase (Hancock and Jones, 1987). They could not detect the cytochrome b558 component of NADPH oxidase. Likewise, Armstrong et al. (2002) failed to detect ROS production by human sperm using spin trapping with DEPMPO, MCLA chemiluminesence or nitroblue tetrazolium reduction, although all techniques readily detected ROS production by leukocytes and they too were unable to detect cytochrome b558 in sperm. Superoxide production by rat sperm has been measured by EPR spin trapping (Kumar et al., 1991) and there is one report that it has been measured by this method in human sperm (Zhang and Zheng, 1996) but it is uncertain if leukocytes were removed effectively from the sperm suspensions. For comparison, most investigators who have demonstrated positive effects of H2O2 on sperm have used concentrations of 10 µmol/l (see above) and such amounts should be easy to detect.
|
If we accept the results that purified sperm suspensions do not make superoxide in response to added NADPH, how can we explain that addition of NAD(P)H causes effects similar to the addition of H2O2? The answer may lie in the fact that traces of H2O2 are always present in NAD(P)H solutions due to autoxidation. This process is faster in acidic media (Halliwell, 1978
; Halliwell and Gutteridge, 1999) but occurs at a substantial rate in Biggers–Whitten–Whittingham (BWW) medium at pH 7.4. The generation of H2O2 in 0–20 mmol/l solutions of NADPH in BWW buffer is shown in Figure 4. In 20 mmol/l NADPH the H2O2 concentration exceeded 80 µmol/l. These amounts are nearly sufficient to account for the oxidative effects of NADPH on human sperm based on comparing effects of NADPH and H2O2 on sperm motility and DNA oxidation (Aitken et al., 1998b) (Figure 5). However, the system is quite complex: the data in Figure 4 show the total accumulation of H2O2 measured by trapping it by reaction with the detection reagent Amplex Red; H2O2 undergoes spontaneous dismutation and the rate of this increases in solutions whose pH is >7, therefore at any given time the concentration of H2O2 may be considerably less than shown in Figure 4. A further factor is that on the one hand H2O2 will be actively metabolized by sperm whereas on the other hand any peroxidases present in sperm will increase the rate of H2O2 formation from NADPH. Therefore it will be difficult to prove conclusively that the spontaneous oxidation of NADPH to form H2O2 can explain the pro-oxidative effects of NADPH on sperm. However, this hypothesis appears to account for the published observations and removes the need to postulate NADPH oxidase activity to account for the biological effects of high extracellular NADPH concentrations on sperm.
|
|
Aitken et al. (2003)
now envisage that human sperm contain multiple forms of redox activity that include a plasma membrane redox system (PMRS) (de Grey, 2003) capable of reducing the tetrazolium compound 2-(4-iodophenyl)-3–4-(nitrophenyl)-5–2,4-disulphophenyl)-2H-tetrazolium (WST-1) in the presence of the intermediate electron acceptor phenazine methosulphate (PMS) or extracellular NADH, a diaphorase-like enzyme catalysing the reduction of WST-1 by NADPH, a different enzyme capable of eliciting lucigenin chemiluminesence in the presence of NADPH and finally a calcium-sensitive system tentatively identified as NOX5 that can generate a luminescent signal from luminol–HRP.
Recent work on animal sperm has followed a similar line to that the human but with the added complication that there is a more significant contribution from the mitochondria. Epididymal sperm from rodents generated lucigenin chemiluminesence but the spontaneous rate was lower in sperm from guinea-pigs and hamsters compared to mice and rats. In a given species, there was no difference in spontaneous chemiluminesence between different regions of the epididymis, but addition of extracellular NADPH caused a big increase in chemiluminesence with the greatest effect in caput sperm. Among rat immature germ cells the highest NADPH-dependent lucigenin chemiluminesence was seen in pachytene spermatocytes. By contrast, spontaneous H2O2 production measured with luminol–HRP was highest in cauda sperm and did not vary between rat germ cells at different stages of development (Fisher and Aitken, 1997; Lewis and Aitken, 2001). It was postulated that the lucigenin activity reflected the ability of the sperm to generate peroxides to act as hydrogen acceptors during nuclear condensation, whereas the ability to produce H2O2 reflected acquisition of the ability to capacitate (Aitken and Vernet, 1998) and these ideas were elaborated into a model incorporating ROS in motility activation and capacitation (Aitken, 2000). The generation of lucigenin-dependent chemiluminesence could be divided into a mitochondrial component due to electron leakage at complexes 1 and 2 of the respiratory chain and a contribution from a putative plasma membrane NADPH oxidase that was inhibited by a cytosolic component (Vernet et al., 2001). We have since confirmed that rat sperm produce hydrogen peroxide using the Amplex Red assay and have shown that this is increased by substrates such as succinate or glycerol-1-phosphate that feed into complex 2 of the mitochondrial electron transport chain (M.Adlam and W.C.L.Ford, unpublished observations). Mouse sperm from the cauda epididymis produced a chemiluminescent signal from luminol–HRP that increased slowly from the start of the incubation to reach a plateau value proportional to the sperm concentration. This signal was inhibited by catalase, SOD and diphenylene iodonium and by omitting bicarbonate from the medium. It was not seen with caput sperm (Ecroyd et al., 2003). A similar activity in rat sperm has been ascribed to formation of peroxinitrite. Activity was blocked by inhibitors of plasma membrane redox systems and of nitric oxide synthase and by scavengers of peroxinitrite. These inhibitors also suppressed the ability of the sperm to undergo the acrosome reaction, which suggests that the redox activity is physiologically important (Aitken et al., 2004). Recently Aitken's group have demonstrated that the NAPDH oxidase activity reported in caput epididymal rat sperm could be accounted for by a cytochrome p450 reductase located in epithelial cells contaminating the sperm suspensions (Baker et al., 2004). This undermines most of the evidence supporting NADPH oxidase activity in rodent sperm and focuses attention on the mitochondria as the main source of ROS production by sperm from these species.
Equine sperm produced H2O2, measured with an ‘Amplex Red’ assay; activity was increased by permiabilizing the cells by freezing and thawing or by adding NADPH. Activity was stimulated by A23187 but not phorbol esters or NFMLP and was greater in sperm from the low density region of a polyvinylpyrrolidone-coated silica gradient (Ball et al., 2001). Bovine sperm produced increased amounts of H2O2 (measured with a p-hydroxyphenylacetic acid–HRP system) after capacitation by addition of heparin, although superoxide appeared to be the species involved in cell regulation (O'Flaherty et al., 2003). Further work is required to elucidate the source of ROS production in sperm from these species.
|
Conclusions |
|---|
|
TOPAbstractIntroductionReactive oxygen speciesBiological roles of ROSEvidence that ROS modulate...Source of ROS involved...ConclusionsReferences |
|---|
There is robust evidence that ROS can promote sperm capacitation. One mechanism is by increasing protein tyrosine phosphorylation with a pattern similar to that produced by cAMP, though this cannot explain all the effects of ROS. ROS may stimulate the sperm sAC but further evidence is needed to confirm this. It remains unclear which ROS are involved but superoxide and peroxide may each have their own specific effects and the role of other ROS, notably peroxinitrite, are only beginning to be investigated. It is also unclear how the promotion of capacitation by ROS fits into the pattern of events in natural fertilization or IVF.
By contrast, the question of whether sperm can produce ROS is steeped in controversy. Evidence for NADPH oxidase activity looks increasingly uncertain. Lucigenin is an unreliable probe for superoxide and NADPH-dependent superoxide production has never been convincingly demonstrated by other methods. The pro-oxidative effects of NADPH may well be explained by the spontaneous reduction of oxygen to H2O2 in NADPH solutions. Although NOX gene expression has been demonstrated in human spermatocytes there is no immunological evidence of NOX protein components in mature sperm and at least two groups have recorded their failure to detect the cytochrome b558 component of gp91phox. It now appears that NADPH-dependent lucigenin chemiluminesence in rodent epididymal sperm is produced by a cytochrome p450 reductase present in contaminating epididymal epithelial cells (Baker et al., 2004). The mechanism in human sperm remains to be elucidated but it is clear that the plasma membrane of these sperm contains multiple redox enzyme activities.
There is stronger evidence that animal sperm make ROS. The possibility of leukocyte contamination always has to be borne in mind but is less likely be a serious problem than for human sperm. Much of this activity may be accounted for by the mitochondria, at least in rodents. Human sperm have relatively few mitochondria and these appear to be less metabolically active than in most animal species (Ford and Rees, 1990) but it is surprising that no superoxide production from this source has been reported.
Some evidence that suspensions of human sperm produce ROS remains. Superoxide was detected with the probe MCLA after stimulation with biological fluids such as fetal cord serum ultrafiltrate but not NADPH (De Lamirande and Gagnon, 1995; De Lamirande et al., 1998a). However, although these experiments used sperm from healthy donors after purification on a 95% Percoll gradient, no further steps were taken to remove leukocytes or to test for their presence. Luminol–HRP chemiluminesence was generated by some leukocyte free sperm fractions (Aitken et al., 1994b). We also have to account for why removing ROS from sperm suspensions with SOD and/or catalase promotes capacitation (see Table I).
Much of the confusion that surrounds ROS production by sperm arises from the assays that have been used. Chemiluminescent assays can be extremely sensitive but are difficult to calibrate in molar units. This makes it difficult to compare results between different laboratories or to judge the physiological or pathological implications of the data. Assays that allow ROS production to be measured in molar units are to be preferred. As well as carrying routine controls with SOD and catalase, researchers should confirm that the probe they choose to use does not perturb the system in other ways, e.g. by increasing substrate utilization. Above all, they must ensure that their sperm preparations are free from leukocytes.
Finally we should remember that only relatively small numbers of sperm reach the isthmal region of the Fallopian tube where capacitation begins in vivo, and that this environment is rich in antioxidant enzymes (Lapointe et al., 1998; El Mouatassim et al., 2000; Kaneko et al., 2001).Given that the effect of ROS on sperm capacitation can be blocked by SOD and/or catalase, they must be presented extracellularly to exert their effects. If sperm produce only small amounts of ROS, could they generate sufficient to achieve the 10–100 µmol/l concentrations apparently required to promote capacitation? Alternative sources should be considered; one possibility might be that they are generated by the cumulus mass surrounding the oocyte. The observation that cumulus of bovine oocytes enhances fertilization in an oxygen-dependent manner (Tanghe et al., 2003) is consistent with this idea and it seems logical that the oocyte should greet the approaching sperm with a battery of messengers to signal them to complete the process of capacitation and prepare to fertilize.
|
Acknowledgements |
|---|
I am grateful to Dr J.T.Hancock, School of Biosciences, University of the West of England at Bristol for reading the manuscript and making helpful suggestions for its improvement.
|
References |
|---|
|
TOPAbstractIntroductionReactive oxygen speciesBiological roles of ROSEvidence that ROS modulate...Source of ROS involved...ConclusionsReferences |
|---|
Agarwal A, Saleh RA and Bedaiwy MA (2003) Role of reactive oxygen species in the pathophysiology of human reproduction. Fertil Steril 79, 829–843.[CrossRef][ISI][Medline]
Aitken RJ (2000) Possible redox regulation of sperm motility activation. J Androl 21, 491–496.[ISI][Medline]
Aitken RJ and Clarkson JS (1987) Cellular basis of defective sperm function and its association with the genesis of reactive oxygen species by human spermatozoa. J Reprod Fertility 81, 459–469.
Aitken RJ and Clarkson JS (1988) Significance of reactive oxygen species and antioxidants in defining the efficacy of sperm preparation techniques. J Androl 9, 367–376.
Aitken RJ and Vernet P (1998) Maturation of redox regulatory mechanisms in the epididymis. J Reprod Fertil, (Suppl. S3), 109–118.
Aitken RJ and West KM (1990) Analysis of the relationship between reactive oxygen species production and leukocyte infiltration in fractions of human semen separated on percoll gradients. Int J Androl 13, 433–451.[ISI][Medline]
Aitken RJ, Clarkson JS and Fishel S (1989a) Generation of reactive oxygen species, lipid peroxidation and human sperm function. Biol Reprod 41, 183–197.[Abstract]
Aitken RJ, Clarkson JS, Hargreave TB, Irvine DS and Wu FCW (1989b) Analysis of the relationship between defective sperm function and the generation of reactive oxygen species in cases of oligozoospermia. J Androl 10, 214–220.
Aitken RJ, Irvine DS and Wu FC (1991) Prospective analysis of sperm-oocyte fusion and reactive oxygen species generation as criteria for the diagnosis of infertility. Am J Obstet Gynecol 164, 542–551.[ISI][Medline]
Aitken RJ, Buckingham DW and West KM (1992a) Reactive oxygen species and human spermatozoa—analysis of the cellular mechanisms involved in luminol-dependent and lucigenin-dependent chemiluminescence. J Cellular Physiol 151, 466–477.[CrossRef][ISI][Medline]
Aitken RJ, Buckingham D, West K, Wu FC, Zikopoulos K and Richardson DW (1992b) Differential contribution of leukocytes and spermatozoa to the generation of reactive oxygen species in the ejaculates of oligozoospermic patients and fertile donors. J Reprod Fertil 94, 451–462.[Abstract]
Aitken RJ, Harkiss D and Buckingham D (1993) Relationship between iron-catalyzed lipid-peroxidation potential and human sperm function. J Reprod Fertil 98, 257–265.[Abstract]
Aitken RJ, West K and Buckingham D (1994a) Leukocytic infiltration into the human ejaculate and its association with semen quality, oxidative stress, and sperm function. J Androl 15, 343–352.
Aitken RJ, Krausz C and Buckingham D (1994b) Relationships between biochemical markers for residual sperm cytoplasm, reactive oxygen species generation, and the presence of leukocytes and precursor germ cells in human sperm suspensions. Mol Reprod Dev 39, 268–279.[CrossRef][ISI][Medline]
Aitken RJ, Krausz C and Buckingham D (1994c) Relationships between biochemical markers for residual sperm cytoplasm, reactive oxygen species generation, and the presence of leukocytes and precursor germ-cells in human sperm suspensions. Mol Reprod Dev 39, 268–279.[CrossRef][ISI][Medline]
Aitken RJ, Paterson M, Fisher H, Buckingham DW and Vanduin M (1995) Redox regulation of tyrosine phosphorylation in human spermatozoa and its role in the control of human sperm function. J Cell Sci 108, 2017–2025.[Abstract]
Aitken RJ, Buckingham DW, West K and Brindle J (1996a) On the use of paramagnetic beads and ferrofluids to assess and eliminate the leukocytic contribution to oxygen radical generation by human sperm suspensions. Am J ReprodImmunol 35, 541–551.
Aitken RJ, Buckingham DW, Harkiss D, Paterson M, Fisher H and Irvine DS (1996b) The extragenomic action of progesterone on human spermatozoa is influenced by redox regulated changes in tyrosine phosphorylation during capacitation. Mol Cell Endocrinol 117, 83–93.[CrossRef][ISI][Medline]
Aitken RJ, Fisher HM, Fulton N, Gomez E, Knox W, Lewis B and Irvine S (1997) Reactive oxygen species generation by human spermatozoa is induced by exogenous NADPH and inhibited by the flavoprotein inhibitors diphenylene iodonium and quinacrine. Mol Reprod Dev 47, 468–482.[CrossRef][ISI][Medline]
Aitken RJ, Harkiss D, Knox W, Paterson M and Irvine D (1998a) A novel signal transduction cascade in capacitating human spermatozoa characterised by a redox-regulated, cAMP-mediated induction of tyrosine phosphorylation. J Cell Sci 111, 645–656.[Abstract]
Aitken RJ, Gordon E, Harkiss D, Twigg JP, Milne P, Jennings Z and Irvine D (1998b) Relative impact of oxidative stress on the functional competence and genomic integrity of human spermatozoa. Biol Reprod 59, 1037–1046.
Aitken RJ, Ryan AL, Curry BJ and Baker MA (2003) Multiple forms of redox activity in populations of human spermatozoa. Mol Hum Reprod 9, 645–661.
Aitken RJ, Ryan AL, Baker MA and McLaughlin EA (2004) Redox activity associated with the maturation and capacitation of mammalian spermatozoa. Free Radical Biol Med 36, 994–1010.[CrossRef][ISI][Medline]
Alvarez JG and Storey BT (1984) Lipid peroxidation and the reactions of superoxide and hydrogen peroxide in mouse spermatozoa. Biol Reprod 30, 833–841.[Abstract]
Alvarez JG, Touchstone JC, Blasco L and Storey BT (1987) Spontaneous lipid peroxidation and production of hydrogen peroxide and superoxide in human spermatozoa: superoxide dismutase as major enzyme protectant against oxygen toxicity. J Androl 8, 338–348.
Armstrong JS, Bivalacqua TJ, Chamulitrat W, Sikka S and Hellstrom WJ (2002) A comparison of the NADPH oxidase in human sperm and white blood cells. Int J Androl, 25223–25229.
Babior BM (1999) NADPH oxidase: an update. Blood 93, 1464–1476.
Babior BM, El Benna J, Chanock SJ and Smith RM (1997) The NADPH oxidase of leukocytes: The respirtory burst oxidase. In Scandalios JG (ed) Oxidative stress and the molecular biology of antioxidant defenses. Cold Spring Harbor Laboratory Press, New York, pp 737–784.
Baker MA, Krutskikh A, Curry BJ, McLaughlin EA and Aitken RJ (2004) Identification of cytochrome p450-reductase as the enzyme responsible for NADPH-dependent lucigenin and tetrazolium salt reduction in rat epididymal sperm preparations. Biol Reprod, in press.
Baldi E, Luconi M, Bonaccorsi L, Muratori M and Forti G (2000) Intracellular events and signaling pathways involved in sperm acquisition of fertilizing capacity and acrosome reaction. Front Biosci 5, E110–E123.[ISI][Medline]
Baldi E, Luconi M, Bonaccorsi L and Forti G (2002) Signal transduction pathways in human spermatozoa. J Reprod Immunol 53, 121–131.[CrossRef][ISI][Medline]
Ball BA, Vo AT and Baumber J (2001) Generation of reactive oxygen species by equine spermatozoa. Am J Vet Res 62, 508–515.[CrossRef][ISI][Medline]
Banfi B, Molnar G, Maturana A, Steger K, Hegedus B, Demaurex N and Krause KH (2001) A Ca2+-activated NADPH oxidase in testis, spleen, and lymph nodes. J Biological Chemistry 276, 37594–37601.
Baumber J, Sabeur K, Vo A and Ball BA (2003) Reactive oxygen species promote tyrosine phosphorylation and capacitation in equine spermatozoa. Theriogenology 60, 1239–1247.[CrossRef][ISI][Medline]
Bize I, Santander G, Cabello P, Driscoll D and Sharpe C (1991) Hydrogen-peroxide is involved in hamster sperm capacitation in vitro. Biol Reprod 44, 398–403.[Abstract]
Blackmore PF (1993) Rapid non-genomic actions of progesterone stimulate ca2+ influx and the acrosome reaction in human sperm. Cell Signalling 5, 531–538.[CrossRef][ISI][Medline]
Breitbart H (2003) Signaling pathways in sperm capacitation and acrosome reaction. Cell Mol Biol 49, 321–327.[ISI][Medline]
Brener E, Rubinstein S, Cohen G, Shternall K, Rivlin J and Breitbart H (2003) Remodelling of the actin cytoskeleton during mammalian sperm capacitation and acrosome reaction. Biol Reprod 68, 837–845.
Brookes PS, Levonen AL, Shiva S, Sarti P and Darley-Usmar VM (2002) Mitichondria: Regulators of signal transduction by reactive oxygen and nitrogen species. Free Rad Biol Med 33, 755–764.[CrossRef][ISI][Medline]
Buck J, Sinclair ML, Schapal L, Cann MJ and Levin LR (1999) Cytosolic adenylyl cyclase defines a unique signalling molecule in mammals. Proc Natl Acad Sci USA 96, 79–84.
Carrera A, Moos J, Ning XP, Gerton GL, Tesarik J, Kopf GS and Moss SB (1996) Regulation of protein-tyrosine phosphorylation in human sperm by a calcium/calmodulin-dependent mechanism—identification of a kinase anchor proteins as major substrates for tyrosine phosphorylation. Dev Biol (Orlando) 180, 284–296.
Chance B, Sies H and Boveris A (1979) Hyperoxide metabolism in mammalian organs. Physiol Rev 59, 527–605.
Chiarugi C and Cirri C (2003) Redox regulation of protein tyrosine phosphatases during receptor tyrosine kinase activation. Trends Biochem Sci 28, 509–514.[CrossRef][ISI][Medline]
Cohen N, Lubart R, Rubinstein S and Breitbart H (1998) Light irradiation of mouse spermatozoa: Stimulation of in vitro fertilization and calcium signals. Photochem Photobiol 68, 407–413.[CrossRef][ISI][Medline]
Cooper C, Patel RP, Brookes PS and Darley-Usmar VM (2002) Nanotransducers in cellular redox signalling: Modification of thiols by reactive oxygen and nitrogen species. Trends Biochem Sci 27, 489–492.[CrossRef][ISI][Medline]
Cooper TG (1995) Role of the epididymis in mediating changes in the male gamete during maturation. Adv Exp Med Biol 377, 87–101.[Medline]
de Grey ADNJ (2003) A hypothesis for the minimal overall structure of the mammalian plasma membrane redox system. Protoplasma 221, 3–9.[CrossRef][ISI][Medline]
de Lamirande E and Gagnon C (1993a) A positive role for the superoxide anion in triggering hyperactivation and capacitation of human spermatozoa. Int J Androl 16, 21–25.[ISI][Medline]
de Lamirande E and Gagnon C (1993b) Human sperm hyperactivation and capac