The Positive Roles for Reactive Oxygen Species in Human Reproduction; Implications for the Therapeutic Application of Antioxidants

ROS include biologically important signaling molecules that are constantly being generated in cells. Their biochemical significance is built upon three fundamental biochemical principles: (1) ROS can oxidize cysteine thiols to generate the corresponding sulfenic acid (SOH) or create disulfide bridges, thereby fundamentally altering protein conformation and function. In the following review of the reproductive process, this property of ROS is exemplified in numerous ways, from the cross-linking of chromatin in the sperm head to the activation of zymogens (e.g., metalloproteases) and the suppression of protein phosphatase activity [ 16 , 17 ]. The latter is an extremely powerful consequence of ROS exposure that results in phosphorylation-dependent signal transduction cascades being maintained in an activated state. A reproductively important example is the phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT) pathway, which is essential for cell proliferation, protein synthesis, cell cycle progression, the suppression of apoptosis, the resumption of meiosis, and implantation [ 18 , 19 ]. Similarly, mitogen-activated protein kinase (MAPK) activity is promoted by ROS and drives several processes central to reproductive fitness, including spermatogenesis and oocyte maturation [ 20 , 21 ]. Thiol oxidation may also affect cellular activity via changes in transcription factors, such as Nrf2/Keap1 or NF-κB, that regulate the responsiveness of reproductive cells and tissues to OS and are critical for the survival of the germ line and critical processes, such as ovulation, implantation, and labour [ 22 , 23 , 24 ]. At higher oxidation states, sulfenic acid residues can generate the corresponding sulfinic acid, or, if the OS is very severe, sulfonic acid; these oxidation products are not readily reversed and are indicative of terminal oxidation and pathology. (2) At high levels of intensity, ROS can, indeed, be very destructive, attacking a variety of critical biomolecules and inducing physiological cell death [ 12 , 25 ]. This property is an essential element in cellular remodelling processes encountered in reproduction, including luteolysis, menstruation, and implantation, as well as the deletion of defective gametes and embryos. Under these circumstances, ROS is being used as a positive mediator of cellular turnover, driving such key processes as apoptosis, ferroptosis, and autophagy [ 12 , 19 , 26 ]. (3) A third fundamental property of ROS is that it can interact with metal centres at the core of many key proteins and control redox-switching activities. Typically, such metal centres are occupied by transition metals such as iron and copper, the redox status of which controls the overall function of the protein. For example, the redox regulation of heme-iron centres in key proteins, such as cytochrome P450s, cytochrome C, nitric oxide synthase, catalase, and a variety of peroxidases, is important for diverse reproductive processes such as sperm hyperactivation, progesterone synthesis by the corpus luteum, and the generation of prostaglandins at parturition. Lipoxygenase activity is also dependent on the oxidation of an iron atom at its active site, triggering such vital biological processes as eicosanoid generation and ferroptosis [ 27 ]. Similarly, the redox cycling of bound copper supports the functionality of proteins such as superoxide dismutase (SOD) and ceruloplasmin that play critical protective roles at all stages of the reproductive process from gametogenesis to foetal development [ 28 , 29 ]. ROS can oxidize cysteine thiols to generate the corresponding sulfenic acid (SOH) or create disulfide bridges, thereby fundamentally altering protein conformation and function. In the following review of the reproductive process, this property of ROS is exemplified in numerous ways, from the cross-linking of chromatin in the sperm head to the activation of zymogens (e.g., metalloproteases) and the suppression of protein phosphatase activity [ 16 , 17 ]. The latter is an extremely powerful consequence of ROS exposure that results in phosphorylation-dependent signal transduction cascades being maintained in an activated state. A reproductively important example is the phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT) pathway, which is essential for cell proliferation, protein synthesis, cell cycle progression, the suppression of apoptosis, the resumption of meiosis, and implantation [ 18 , 19 ]. Similarly, mitogen-activated protein kinase (MAPK) activity is promoted by ROS and drives several processes central to reproductive fitness, including spermatogenesis and oocyte maturation [ 20 , 21 ]. Thiol oxidation may also affect cellular activity via changes in transcription factors, such as Nrf2/Keap1 or NF-κB, that regulate the responsiveness of reproductive cells and tissues to OS and are critical for the survival of the germ line and critical processes, such as ovulation, implantation, and labour [ 22 , 23 , 24 ]. At higher oxidation states, sulfenic acid residues can generate the corresponding sulfinic acid, or, if the OS is very severe, sulfonic acid; these oxidation products are not readily reversed and are indicative of terminal oxidation and pathology. At high levels of intensity, ROS can, indeed, be very destructive, attacking a variety of critical biomolecules and inducing physiological cell death [ 12 , 25 ]. This property is an essential element in cellular remodelling processes encountered in reproduction, including luteolysis, menstruation, and implantation, as well as the deletion of defective gametes and embryos. Under these circumstances, ROS is being used as a positive mediator of cellular turnover, driving such key processes as apoptosis, ferroptosis, and autophagy [ 12 , 19 , 26 ]. A third fundamental property of ROS is that it can interact with metal centres at the core of many key proteins and control redox-switching activities. Typically, such metal centres are occupied by transition metals such as iron and copper, the redox status of which controls the overall function of the protein. For example, the redox regulation of heme-iron centres in key proteins, such as cytochrome P450s, cytochrome C, nitric oxide synthase, catalase, and a variety of peroxidases, is important for diverse reproductive processes such as sperm hyperactivation, progesterone synthesis by the corpus luteum, and the generation of prostaglandins at parturition. Lipoxygenase activity is also dependent on the oxidation of an iron atom at its active site, triggering such vital biological processes as eicosanoid generation and ferroptosis [ 27 ]. Similarly, the redox cycling of bound copper supports the functionality of proteins such as superoxide dismutase (SOD) and ceruloplasmin that play critical protective roles at all stages of the reproductive process from gametogenesis to foetal development [ 28 , 29 ]. In the following sections, the positive contributions of ROS to all stages of the reproductive process, from gamete maturation to childbirth, are examined. The review is focused on human reproduction, although the concepts expressed are often supplemented with data obtained from animal models, particularly laboratory rodents. Furthermore, this review focuses on ROS-mediated events and only gives limited consideration to other forms of redox-induced change, including S-nitrosylation, sulfhydration, glutathionylation, CoAlation, and protein carbonylation, which have been considered in detail elsewhere [ 30 ]. The results of this analysis clearly illustrate the important biological role of ROS in reproductive biology and highlight why caution should be exercised in determining the type and dose of antioxidants deployed in the therapeutic management of human infertility.

Human spermatozoa feature three major physiological sources of ROS, including electron leakage from their mitochondria, the activation of specialized calcium-dependent NADPH oxidases such as NOX5, and the activity of L-amino acid oxidases such as IL4I1 [ 12 ] ( Figure 1 ; Step 1). These ROS are then used to drive various aspects of sperm biology, starting with their fundamental architecture. As spermatozoa mature, cysteine groups in an important protective enzyme, glutathione peroxidase 4 (GPX4), become oxidized, leading to the formation of multiple intermolecular disulphide bridges and the formation of a polymer that stabilizes the spiral arrangement of the mitochondria around the axoneme. In this way, an enzyme that started as a powerful antioxidant becomes transformed, via oxidation, to a key structural element in the sperm midpiece, the mitochondrial sheath [ 31 ]. Another redox-mediated change in maturing spermatozoa involves chromatin cross-linking in the sperm head. This change compacts the DNA into a near-crystalline state and is thought to be important for safeguarding the integrity of the male genome during its long journey from the male to the female reproductive tract [ 32 , 33 ]. The cross-linking of sperm chromatin is achieved by replacement of nuclear histones with cysteine-rich protamines during spermatogenesis, followed by ROS-mediated formation of inter- and intra- molecular disulfide bridges, mediated by GPX4, during epididymal transit ( Figure 1 ; Step 2) [ 34 , 35 , 36 , 37 ]. Although all Eutherian mammals use 6–9 cysteines for disulfide cross-linking, the specific placement and subsequent intra- vs inter-molecular bonding patterns vary. Interestingly, variation in packaging efficiency and DNA integrity appears to correlate with sperm competition levels within a species. Higher competition often selects for more robustly packaged and damage-resistant chromatin to ensure successful fertilisation in a competitive environment. In this context, it is interesting to note that human spermatozoa, which have not evolved to cope with high levels of sperm competition, have notoriously poor chromatin cross-linking and, as a result, are far more susceptible to DNA damage than other Eutherian spermatozoa [ 36 ]. Such deficiencies in chromatin packaging correlate with incomplete thiol oxidation during epididymal transit. As a result, spermatozoa from infertile males are often found to possess a higher thiol content (fewer disulfide bonds) compared with normozoospermic, control samples [ 38 , 39 , 40 ]. Another crucial pathway regulated by redox-responsive mechanisms is capacitation, the process by which mammalian spermatozoa undergo a final maturation during their ascent of the female reproductive tract and gain the capacity to fertilize the oocyte. Capacitation is characterised by increases in pH, intracellular calcium, HCO 3 , cyclic AMP (cAMP), and protein phosphorylation [ 41 ]. ROS are positively involved in many of these processes, as indicated in Figure 1 , Step 3. Thus, the increase in tyrosine phosphorylation that characterises the attainment of a capacitated state is a major redox-driven event in human spermatozoa. This enhancement is achieved via the ability of H 2 O 2 to oxidize catalytic cysteine residues (often protected by glutathionylation or CoAlation) at the active site of tyrosine phosphatases to generate sulfenic acids, which can then react with adjacent cysteines to form disulfide bonds or with nearby amides to form sulfenyl–amide linkages [ 30 , 42 ]. Such oxidation reactions effectively silence phosphatase activity, thereby facilitating the dramatic increase of protein phosphorylation during sperm capacitation. Additionally, ROS promotes tyrosine phosphorylation by stimulating generation of the second messenger that drives this activity, cAMP, via the ability of O 2 •− to stimulate soluble adenylyl cyclase activity [ 43 , 44 , 45 ]. While the biochemical mechanisms responsible for redox-regulated cAMP generation in spermatozoa are not fully understood, the adenylyl cyclase within these cells contains cysteine residues that are susceptible to oxidation, forming disulfide bonds or sulfenic acids. These modifications may induce a structural shift in the catalytic domain, increasing the enzyme’s affinity for its substrate, ATP, or its essential cofactors, Mg 2+ or Mn 2+ . In addition, ROS facilitates capacitation by enhancing the rate of cholesterol efflux from the plasma membrane, thereby contributing to an increase in membrane fluidity [ 46 ]. This phenomenon reflects the ability of ROS to directly induce the formation of oxysterols, which are more hydrophilic than the parent sterols and, as a result, move closer to the sperm surface where they are removed by binding proteins such as albumin ( Figure 1 ; Step 4). The disruption of this redox process using hydrophobic antioxidants, such as vitamins E or A, suppresses the tyrosine phosphorylation events associated with sperm capacitation and also interferes with sperm–zona interaction [ 47 ]. Hyperactivated motility is another feature of capacitation that is redox regulated. This type of movement is characterised by a change in the flagellar waveform to deliver a high-amplitude, high-intensity, asymmetrical beat pattern. This vitally important change is dependent on associated changes in tyrosine phosphorylation and is stimulated by ROS via the MAPK/ERK (Extracellular Signal-Regulated Kinase) pathway [ 48 , 49 , 50 ]. Another ROS-regulated factor involved in the control of hyperactivation is the Epidermal Growth Factor Receptor (EGFR), a transmembrane protein found in many cell types that mediates differentiation, cell proliferation, and migration. EGFR activity has been shown to be directly enhanced by ROS through the inhibition of the receptor’s internalization and, indirectly, via the suppression of protein phosphatase activity [ 51 , 52 , 53 ]. In spermatozoa, ROS-dependent EGFR activation leads to actin polymerization, which, in turn, leads to hyperactivated motility by mechanically facilitating flagellar bending and acting as a regulatory scaffold for key signalling molecules such as Catsper [ 54 , 55 , 56 ]. In keeping with this model, studies have shown that light-induced ROS production successfully activates EGFR, leading to hyperactivated motility, contrasting with the suppression of this process by the presence of SOD [ 54 ]. Taken together, these data clearly indicate that the physiological generation of ROS, particularly H 2 O 2 , plays a critical role in the orchestration of human sperm form and function, in the lead up to fertilization. Following successful hyperactivation and capacitation, further redox-regulated changes lead to egg recognition, the acrosome reaction, and, finally, sperm–oocyte fusion [ 41 , 45 , 55 , 56 ]. The ROS-mediated increase in membrane fluidity during capacitation enables egg-recognition complexes to move anteriorly within the plasma membrane to a location where sperm–zona recognition can be achieved [ 57 ]. This cell–cell interaction then activates signal transduction processes in the spermatozoa, leading to the induction of the acrosomal exocytosis and, ultimately, fusion with the vitelline membrane of the egg. The mechanistic underpinnings of these membrane fusion events appear to involve the ability of ROS to induce lipid peroxidation via the lipoxygenase pathway, followed by activation of phospholipase A2. The latter then enzymatically removes the oxidized fatty acid from position 2 ( sn 2) of the phospholipid, generating a lysophospholipid [ 58 , 59 ]. Possessing just one fatty acid tail, these molecules possess detergent-like properties that promote membrane instability, thereby supporting fusion of the plasma and outer acrosomal membranes during acrosomal exocytosis. ROS might also be involved in the IP3-mediated activation of actin severing proteins that promote the acrosome reaction by further destabilizing the plasma membrane [ 54 , 55 ]. The increase in membrane instability might also explain the observed positive role of ROS in the generation of a fusogenic equatorial segment in the spermatozoa, primed for interaction with the oocyte’s plasma membrane [ 60 ]. Finally, ROS are also involved in the mediation of both sperm survival and senescence. In the context of sperm vitality, ROS has been suggested to act directly on AMP kinase (AMPK), activating this enzyme via the oxidation of key cysteines (299 and 304) on the α-subunit and upregulating both the cells’ antioxidant defences, as well as their capacity for ATP generation [ 61 , 62 ]. On the other hand, ROS are also central to the induction of apoptosis in human spermatozoa, facilitating the apoptotic removal of aged, damaged spermatozoa and ensuring that they do not participate in the fertilization process [ 63 ]. In light of these observations, it has been suggested that sperm capacitation and senescence represent a physiological continuum mediated by ROS [ 64 ]. During the early stages of sperm capacitation, ROS is a positive mediator of biological change, promoting signal-transduction pathways, elevating intracellular calcium, and increasing membrane fluidity under the protective control of antioxidants, notably peroxiredoxin-6 (PRDX6) [ 65 ]. However, if capacitated spermatozoa do not find an oocyte, these same changes will ultimately generate senescence and seal their apoptotic fate [ 64 ].

Oocyte recruitment, development, and maturation involve a series of complex molecular signalling processes within the follicular microenvironment, many of which are mediated by ROS [ 66 ]. Oocytes spend much of their lifespan reposing within primordial follicles and arrested at prophase I. Throughout this prolonged period of time, they are protected from OS because the oocyte mitochondria lack Complex 1, the major site of electron leakage and ROS generation in a majority of cell types [ 67 ]. Following follicle selection and growth, a pre-ovulatory surge of gonadotrophins (LH and FSH) triggers ROS production by granulosa cells in the follicle, precipitating germinal vesicle breakdown (GVB) and meiotic resumption ( Figure 2 ; Step 1). As germinal vesicle breakdown occurs, there is a massive redistribution of mitochondria within the oocyte so that they come to aggregate around the meiotic spindle, which is an ideal position to provide the localized ATP required for spindle assembly, chromosome alignment, and polar body extrusion. In association with germinal vesicle breakdown, oocyte mitochondria acquire Complex 1, along with Complex IV and citrate synthase, at levels comparable to skeletal muscle. As the oocyte maturation process progresses, mitochondrial membrane potential increases, ATP production is accelerated, oxygen consumption is enhanced, and the baseline generation of ROS is elevated [ 68 ]. Mitochondria, therefore, appear to be a major physiological source of ROS in maturing oocytes. Within the whole cumulus-oocyte complex (COC), however, the mitochondrial contribution to redox-regulated oocyte maturation is reinforced by NADPH oxidase activity, particularly NOX4. Thus, FSH-induced maturation of COCs involves translocation of two cytosolic components of NOX4, p47phox, and p67phox to the plasma membrane of cumulus cells, where they activate the oxidase, generating ROS [ 69 ]. Simultaneously, ATP production by the cumulus cell mitochondria rises to meet the energetic demands of oocyte maturation, with ROS produced as a by-product ( Figure 2 ; Step 2) [ 70 ]. Under physiological circumstances, the integrated contributions of mitochondria and NOX4 generate sufficient ROS within the COC to activate signalling pathways such as AMPK and EGFR, which are involved in energy sensing, cumulus expansion, and ovulation. Both ROS-induced activation of AMPK and calcium-signalling pathways have been implicated in meiotic resumption from diplotene arrest ( Figure 2 ; Step 3) [ 71 , 72 ]. LH also stimulates ROS production in the ovary, as suggested by ascorbic acid depletion studies [ 73 ], and this transient OS is functionally important, inducing phosphorylation and activation of the EGF receptor, as well as its downstream effector, p42/44 MAPK (Erk2 and Erk1). These ROS-mediated changes are causally involved in normal cumulus expansion and mucification. Thus, H 2 O 2 exposure can fully mimic the effect of LH, bringing about an extensive mucification/expansion of the follicle-enclosed cumulus–oocyte complexes. In addition, oocyte maturation can be effectively inhibited in vitro in both mice and rats by membrane-permeant antioxidants such as BHA (butylated hydroxyanisole) [ 74 , 75 ]. A core component of oocyte maturation occurs when ROS-activated AMPK participates in the activation of phosphodiesterase 3A (PDE3A) activity, thereby facilitating reduced generation and accelerated breakdown of cAMP and cGMP. The decline in cAMP, in particular, leads to dephosphorylation of maturation-promoting factor (MPF) and resumption of meiosis. However, if the metabolic activity of oocytes fails to meet mitochondrial thresholds, as reflected in poor AMPK activation, then ROS-mediated apoptotic pathways are initiated by cytochrome C release, inducing large-scale cell death and a state of follicular atresia ( Figure 2 . Step 3) [ 75 ]. In addition to ROS-mediated activation of AMPK, EGFR activation by H 2 O 2 stimulates the proteolytic calpain-2 pathway, leading to loosening of granulosa cell adhesions, thereby allowing these cells to expand and reorganise as the follicle grows [ 76 ] ( Figure 2 ; Step 4). Ovarian follicles are, therefore, much like spermatozoa in that they rely significantly on redox regulation. ROS-signalling pathways are crucial for the stimulation of follicle growth and the resumption of meiosis. While ROS are potentially toxic, the short half life and limited diffusion capacity of these molecules helps confer spatial precision to these signalling events within the ovarian follicle. However, if ROS generation does overwhelm the follicle’s defensive capacity, then these same metabolites can trigger pathways leading to apoptosis and cell death. Thus, the ultimate fate of each follicle, and the oocyte it encases, critically depends upon its capacity to manage ROS, divining whether it will successfully achieve ovulation or succumb to atresia [ 77 , 78 , 79 , 80 ]. Ovulation is similarly modulated by redox activity. As indicated above, the LH surge that precedes ovulation is associated with a sharp rise in inflammatory precursors in the ovary and an increase in ROS generation ( Figure 3 ; Step 1) [ 81 ]. A causal relationship between acute inflammation, ROS production, and ovulation has been suggested by several lines of evidence [ 82 , 83 , 84 ]. Critical in this regard is the ability of ROS generated by the granulosa cells to promote the activation of proteases (matrix metalloproteinases, MMPs-14 and 16, and plasminogen activators) that actively digest the collagen and connective tissues of the follicular wall (the theca externa), creating a weakened area called the stigma, which ultimately becomes the point of follicular rupture ( Figure 3 ; Step 2). MMP activation involves a cysteine switch whereby ROS oxidise a thiol group in the pro-domain of the inactive zymogen, displacing a zinc ion and inducing enzyme activation. ROS also promotes MMP by inhibiting key phosphatases, allowing activation of the MAPK pathway and facilitating the stimulation of transcription factors such as activator protein-1 (AP-1), which bind to the promoter region of MMP genes [ 16 , 17 , 84 , 85 ]. Such redox-regulated protease activation, in conjunction with ROS-induced apoptosis of granulosa cells, promotes detachment of the cumulus cells from the follicular wall, further contributing to follicular wall breakdown and ovulation of the mature oocyte [ 84 ]. LH also triggers the activation of EGFs (Epidermal Growth Factor-like factors) in granulosa cells via ROS-mediated activation of MMPs [ 86 ]. The EGFs then mediate the action of LH in the ovarian follicle [ 75 ]. The physiological sources of ROS in mural granulosa cells primarily involve NOX4, possibly under the influence of pigment–epithelium-derived factor (PEDF) [ 87 ], as well as electron leakage from the mitochondria [ 88 ]. In addition, ROS may also be generated via a similar leakage of electrons from the cyclooxygenase-2 pathway during the biosynthesis of PGE2 [ 89 ] ( Figure 3 ; Step 3). In concert, these various sources of ROS combine together to generate the redox drive for follicular rupture and ovulation [ 90 ]. Animal studies using Drosophila support the involvement of ROS in ovulation by demonstrating that O 2 •− generating NOX enzymes work in conjunction with SOD3 to generate H 2 O 2, which then acts as a secondary messenger triggering the apoptotic pathways that lead to follicular rupture [ 91 ]. Moreover, in vivo studies in rodents, as well as ex vivo studies in the rabbit, also corroborate the essential role for ROS in follicular rupture and ovulation [ 75 , 82 , 87 , 91 ]. The generation of ovulatory estrogen is another crucial function of the Graafian follicle. According to the two-cell/two-gonadotrophin theory, LH stimulates follicular thecal cells to generate androgens, which are then converted into estrogen by the granulosa cells under the influence of FSH. This is a redox-regulated process instigated by cytochrome P450 side-chain cleavage activity within the mitochondria to actively convert cholesterol to pregnenolone, and, ultimately, androgens, within the thecal layer. These events depend on the movement of electrons from the mitochondrial electron transport chain (ETC) to the P450 enzyme, mediated by adrenodoxin and the NADPH-dependent FAD flavoprotein, adrenodoxin reductase ( Figure 3 ; Step 4). This reliance on electron transfer elevates the risk of ROS production [ 92 ], which is, in turn, quickly controlled by peroxidase, catalase, and other non-enzymatic antioxidants to ensure that an appropriate redox balance is maintained [ 68 , 69 ]. As antral follicles expand, the thecal layer also becomes heavily vascularized to optimize the uptake of substrates (Low- and High-Density Lipoproteins), as well as the subsequent export of estrogen into the general circulation. In this situation, mitochondrial ROS facilitates stabilization of Hypoxia-Inducible Factor 1-alpha (HIF-1α) that then upregulates Vascular Endothelial Growth Factor (VEGF) in thecal cells, promoting the vascular expansion essential for dominant follicle survival.

Following ovulation, the corpus luteum (CL) is formed from the residual follicular remnants and is tasked with producing progesterone to delay the shedding of the endometrium, thereby allowing implantation to occur. Nitric oxide (NO) promotes blood flow and angiogenesis within the newly formed CL. In a similar fashion to neoangiogenesis within the thecal layer during follicle development, low levels of mitochondrial ROS generation also stabilize HIF-1α, leading to the enhanced generation of VEGF, which then drives the massive influx of blood vessels into the CL ( Figure 3 ; Step 5) [ 93 ]. At the same time, ROS is also central to luteal steroidogenesis through its support of cytochrome P450 activity. Whilst ROS and RNS (reactive nitrogen species) are crucial in establishing the corpus luteum, there is a simultaneous shift towards the upregulation of antioxidant protection. Thus, the CL, literally the “yellow body”, acquires its colour from the presence of a powerful antioxidant, β-carotene, designed to limit the amount of oxidative damage incurred by the CL during the luteal phase. The intracellular ROS responsible for creating such stress are just byproducts of the luteal cell’s intense steroidogenic activity and emanate largely from the mitochondrial cholesterol side-chain cleavage activity (particularly Cytochrome P450 Family 11 Subfamily A Member 1; CYP11A1). Nevertheless, the presence of abundant β-carotene, alongside classical antioxidant enzymes such as SOD, catalase, and GPX, ensures that ROS is effectively scavenged during the lifespan of the CL in order to prolong the generation of progesterone. However, in the absence of a conceptus, SOD expression ultimately decreases and ROS accumulates, leading to the initiation of luteolysis [ 94 , 95 ]. During this process, ROS uncouples the LH receptor from adenylyl cyclase and inhibits steroidogenesis by interrupting transmitochondrial cholesterol transport. ROS (particularly lipid peroxides) also activates the cyclooxygenase (COX-2) responsible for regulating the production of prostaglandin F2α (PGF2α) and triggers the induction and expression of a transcription factor, nuclear factor-kappa B (NF-κB), that binds to the promoter region of the PTGS2 gene encoding COX-2, dramatically enhancing PGF2α synthesis [ 95 , 96 ]. The latter then acts on both luteal cells and inflammatory cells, such as macrophages and neutrophils, in the immediate vicinity, prompting them to upregulate ROS production in a self-perpetuating cycle to complete the destruction of the CL in preparation for the initiation of another menstrual cycle. In parallel with luteolysis, ROS is also a mediator of the endometrial regression that characterises menstruation. In this context, the fall in progesterone levels resulting from a lack of human chorionic gonadotrophin (hCG) signalling from the fetus creates a pro-inflammatory state that activates NF-κB and triggers the expression of several genes with a key role in menstruation, including COX-2, MMPs, and cytokines [ 97 , 98 ]. The latter then triggers the infiltration of leukocytes and the generation of more ROS. Oxidative stress, exacerbated by local hypoxia caused by PGF2α-induced vasoconstriction, triggers widespread apoptosis in the upper layer of the endometrium (stratum functionalis), allowing endometrial shedding while minimizing damage to the underlying tissues and preventing development of a chronic inflammatory condition that might damage future fertility [ 97 , 98 , 99 ].

Following ovulation, ROS are centrally involved in the rapid senescence of the unfertilized oocyte, thereby ensuring that an ageing egg is rapidly eliminated and cannot contribute a potentially damaged genome to the next generation [ 100 ]. If fertilization does occur, physiological concentrations of ROS are involved in pronuclear formation, initiation of cleavage, and subsequent cell proliferation [ 101 , 102 ]. However, it is important to note that the needs of the embryo change as embryo development progresses, and, therefore, reasonable to infer that ROS production would also vary. During the cleavage stages of embryonic development, energy production relies on the oxidation of pyruvate via carboxylic acid metabolism, with lactate being produced as a by-product [ 101 , 102 ]. At this stage of embryogenesis, ROS production is relatively low, and the embryo is extremely vulnerable to OS. This offers a mechanistic explanation for the limitations experienced in early embryo culture systems, including the two-cell block in mice and the four-cell block in human embryos [ 103 , 104 ]. Such blocks occur at the moment of zygotic genome activation due to ROS accumulation in the high-oxygen-tension environments employed for embryo culture, impacting effective activation of the embryonic genome and thus hindering cellular division [ 105 , 106 ]. When the embryo develops into a blastocyst, the mitochondria increase their contribution to overall metabolism, and ROS generation is accelerated due to electron leakage from the ETC ( Figure 4 ; Step 1). The ROS generated at this time influence patterns of gene expression, generating proteins that support embryo growth and differentiation [ 107 , 108 ]. Central to this developmental response to OS is the ability of ROS to activate signal transduction pathways such as MAPK, which is centrally involved in the process of lineage determination [ 109 , 110 , 111 ]. Through their ability to cross-link cysteines on the redox-sensing protein KEAP1, ROS are also able to stabilise the transcription factor NRF2, which regulates genes involved in maintaining redox homeostasis [ 112 ]. Other transcription factors important for development, including hypoxia-inducible factors and proteins in the FoxO subfamily, are also responsive to ROS and, again, play a key role in ensuring that the developing embryo does not suffer from OS [ 113 ]. The physiological generation of ROS in the early embryo largely involves NOX, complexes I-III of the mitochondrial ETC, and xanthine oxidase, while superoxide dismutase, catalase, glutathione peroxidase, and peroxiredoxins are the major defences against oxidative damage [ 113 , 114 , 115 , 116 ]. In addition to their role in embryonic cell division and differentiation, ROS are also a powerful mediators of cell death, ensuring that damaged cells do not contribute to the developing embryo and that defective embryos do not contribute to future generations. The use of ROS in the apoptotic deletion of defective cells is, therefore, a quality-control measure ensuring that the maternal investment in pregnancy will ultimately be productive. Implantation involves synchronised interactions between the trophoblast, derived from the trophectoderm of the blastocyst, and the epithelial lining of the uterus. In humans, the process of implantation involves three basic phases: apposition of the activated and hatched blastocyst to the endometrial lining, adhesion, and finally invasion of the decidualised endometrial lining. The decidual cells regulate invasion of the blastocyst through immune responses and paracrine signalling, as well as other molecular signalling pathways. Mounting evidence suggests that physiological concentrations of ROS contribute to the cellular and molecular events underlying the implantation process [ 117 , 118 ]. Once formed, the blastocyst must hatch out of the zona pellucida before implantation can occur. The vigorous pulsing activity of the blastocyst at the time of hatching necessitates a sudden increase in mitochondrial ATP production, which, in turn, leads to a spike in local O 2 •− production as a consequence of electron leakage from the mitochondria, with possible additional input from NADPH and xanthine oxidase activities [ 119 , 120 ] ( Figure 4 ; Step 1). Thomas et al. demonstrated the pivotal role of ROS, specifically O 2 •− , plays in blastocyst hatching in the mouse [ 119 ]. Their studies on murine embryos involved comparison of blastocysts in pre-, peri-, and post-hatching stages. While blastocysts in both pre- and post-hatching stages recorded low levels of O 2 •− and high SOD activity, blastocysts in the peri-hatching stage showed the opposite. The addition of SOD to peri-hatching blastocysts in vitro decreased hatching, while direct exposure of such embryos to O 2 •− enhanced this process [ 119 ]. ROS generated by this oxidative burst activated a family of enzymes called ADAMs (A Disintegrin and Metalloproteinases), particularly ADAM17, which cleaved pro-HB-EGF (Heparin-binding epidermal growth factor-like growth factor) to release the biologically active molecule, HB-EGF ( Figure 4 ; Step 2). The latter then promoted the rapid proliferation of trophectoderm cells and the accumulation of fluid within the blastocoel. The resultant increase in internal hydrostatic pressure physically stretches the ZP from the inside, forcing this structure to fracture. HB-EGF signalling also triggers the blastocyst to produce and secrete lysins or proteases that chemically digest and weaken the ZP, making it easier for the embryo to escape. ROS also silence protein tyrosine phosphatases, which would otherwise suppress HB-EGF signalling by dephosphorylating the cognate receptor (EGFR) [ 121 , 122 , 123 , 124 , 125 ]. A hatched blastocyst must then engage in a series of interactive events with the endometrium to allow for successful implantation, such as apposition, adhesion/attachment, and invasion. Hormone synthesis, inflammatory-like events, uterine secretions, and gene expression all have a part to play in orchestrating these critical changes, allowing for interactions between the endometrial lining and the blastocyst [ 125 ]. During apposition, the uterine wall secretes lysophosphatidic acid (LPA), the production of which involves the oxidation of low-density lipoprotein (LDL) by ROS [ 126 ]. LPA, in turn, triggers a downstream signalling pathway involving G protein-coupled receptors and phospholipase C, leading to the release of Ca 2+ and activation of protein kinase C (PKC). This process culminates in the shedding of embryonic HB-EGF into the uterine environment, signalling the blastocyst’s presence [ 127 ]. LPA also drives COX-2-derived prostaglandin E 2 (PGE 2 ) production in the luminal epithelium and the stroma at the site of adhesion. In addition, LPA increases nitric oxide synthase activity, leading to the production of NO at the implantation site [ 127 , 128 ]. The combination of PGE 2 and NO plays a key role in angiogenesis, leading to the evolution of a highly vascularised nidus capable of supporting implantation. Decidualisation of the endometrial cells is another important facet of implantation, characterized by the morphological remodelling and differentiation of maternal stromal cells during the secretory phase of the menstrual cycle. Decidualisation is mediated by oestrogen, progesterone, transcription factors, cytokines, and other complex signal-transduction pathways. This process sees the elongated, fibroblast-like cells of the uterine stroma transforming into rounded, epithelioid-like cells. Interestingly, decidualisation not only involves a morphological transformation of existing stromal cells but is also associated with the influx of inflammatory cells into the uterus prior to blastocyst adhesion [ 128 , 129 ]. The initiation of decidualization is a biphasic process involving ROS [ 130 ]. While progesterone and cAMP are the primary drivers, they use ROS to activate certain genes. Thus, decidualization triggers the activation of a free radical-generating NADPH oxidase (NOX4), producing a burst of O 2 •− and H 2 O 2 inside the cell, which is needed to activate the transcription factor C/EBPβ (CCAAT/enhancer-binding protein beta) ( Figure 4 ; Step 3). This factor then binds to the promoters of key decidual markers like prolactin and insulin-like growth factor-binding protein-1 ( Figure 4 ; Step 4) [ 130 , 131 ]. Significantly, antioxidants or NOX inhibitors can block the generation of ROS during this initiating phase of decidualization, reducing the expression of decidual markers. In addition to this positive role for ROS in initiating the decidualisation process, the formation of these cells is associated with the upregulation of antioxidant enzymes, like SOD2, catalase, glutathione peroxidase, and glucocorticoid-inducible kinase-1 [ 132 , 133 ]. This sudden elaboration of defensive enzymes designed to protect against oxidative damage does not impair the ability of ROS to drive positive physiological changes during implantation but rather helps prepare the uterus for the looming oxidative burden associated with pregnancy. The final step of implantation is the invasion of the endometrial lining by the blastocyst ( Figure 4 ; Step 5). To accomplish this step, the trophoblast layer differentiates into different subtypes: villous cytotrophoblasts (vCTBs) and syncytiotrophoblasts (STs). The invasive behaviour and functional characteristics of cytotrophoblasts have often been compared to those observed in malignant cells, especially when considering the expression of Tubulointerstitial nephritis antigen-like 1 (TINAGL1 or lipocalin7), a structural matrix protein that interacts with integrins on the trophoblast surface and endometrial epithelium, enabling adhesion [ 134 ]. This adhesion event then triggers intracellular signalling cascades, including the FAK (Focal Adhesion Kinase) and MAPK/ERK pathways, which are essential for trophoblast outgrowth and promoted by the local generation of ROS, through the latter’s ability to inhibit tyrosine phosphatases [ 135 ]. Cytotrophoblasts, at the time of invasion, also exhibit a downregulation of E-cadherin, a negative regulator of trophoblast invasion. By analogy with cancer cell invasion, the generation of ROS is thought to trigger this change in E-cadherin expression through the upregulation of transcription factors such as Snail (SNAI1) and Slug (SNAI2), which are transcriptional repressors of the CDH1 gene (which encodes E-cadherin), as well as stimulation of the MAPK/ERK kinase pathway [ 136 , 137 , 138 ]. ROS can also stabilize Hypoxia-Inducible Factor 1-alpha (HIF-1α), which further promotes the expression of E-cadherin repressors to facilitate trophoblast invasion [ 139 , 140 ]. During the invasive stage of embryo implantation, the syncytiotrophoblast expresses endothelial nitric oxide synthase to increase the production of NO, a vasodilator, which is of critical importance in establishing the placentation process [ 140 ]. Trophoblast invasion is also facilitated by a range of MMPs (MMP-1, MMP-2, MMP-3 MMP-9, MMP-11, and MMP-14) that have been identified in human placentae [ 141 , 142 ]. ROS is critical for the activation and production of MMPs via pathways that involve the upregulation of MAPK and PI3K/Akt [ 143 , 144 , 145 ]. These enzymes are central to the implantation process, promoting degradation of the ECM and allowing the trophoblast to break through the uterine lining.

The fertility journey culminates in parturition, a complex process featuring hormonal fluctuation, inflammatory reactions, and rapid tissue remodelling. ROS are physiologically involved in several aspects of parturition, including the myometrial contractions that initiate labour and the ripening of the cervix, largely through activation of the MAPK pathway. P38 MAPK proteins, modulated by ROS, initiate the release of pro-inflammatory cytokines and prostaglandins from the uterine lining [ 146 ]. Within the same tissues, oxytocin activates NF-κB-mediated inflammatory-signalling pathways, leading to stimulation of ROS generation via NADPH oxidases (primarily NOX 1 and NOX4), the sensitisation of oxytocin receptors, and the onset of myometrial contractions, leading to labour [ 147 , 148 ]. Further evidence of the crucial role of ROS during parturition can be seen in the ability of non-enzymatic antioxidants to reduce the risk of preterm birth [ 148 ]. In this context, N-acetylcysteine was found to decrease COX-2 and prevent the activation of NF-κB by scavenging local ROS generation and inhibiting some of the key downstream-signalling pathways involved in the onset of labour [ 149 ].

Given that ROS play such an important role in the biology of reproduction, it would be reasonable to ask whether the indiscriminate administration of antioxidants, albeit with the best of therapeutic intentions, might impede rather than promote this process. The oversupplementation of healthy individuals who are not suffering from OS with antioxidants runs the risk of impairing the myriad cellular-signalling pathways that rely on physiological levels of ROS, generating a damaging state of reductive stress [ 150 ]. In vitro scavenging of ROS suppresses tyrosine phosphorylation events associated with sperm capacitation, inhibiting hyperactivated motility, disrupting acrosomal exocytosis, and, in animal models, reducing fertilization rates [ 49 , 151 , 152 , 153 , 154 ]. Similarly, with oocyte maturation, while low doses of antioxidants can facilitate this process, high doses delay or completely suppress the ability of mammalian oocytes to undergo meiotic maturation in vitro and ovulate in vivo [ 155 , 156 , 157 , 158 , 159 ]. High doses of antioxidants have also been found to have a negative impact on both cleavage and blastocyst development rates in vitro, while compounds with antioxidant properties such as sanguinarine and EGCG have also been found to suppress blastocyst implantation and post-implantation embryonic development [ 160 , 161 ]. Later in pregnancy, coenzyme Q10 administration has been found to increase OS in rats [ 162 ]. Furthermore, because ROS plays a crucial role in the fetal brain-sparing response (prioritization of oxygen and nutrient delivery to the brain, heart, and adrenal glands at the expense of other organs during pregnancy), there are also concerns that excessive antioxidant use by pregnant women could weaken fetal defences against acute hypoxia, increasing the risk of hypoxic–ischaemic encephalopathy [ 163 ]. Towards the end of pregnancy, antioxidants have been used in an attempt to address complications such as pre-eclampsia, pre-term labour, fetal death, fetal growth restriction, and stillbirth, but no positive outcomes have been recorded. Indeed, occasionally detrimental impacts of such treatment have even surfaced, including decreases in human chorionic gonadotrophin generation, fetal growth restriction, low birthweight, gestational hypertension, and others [ 164 , 165 , 166 , 167 ]. In terms of mechanisms, over-supplementation with antioxidants can drive mitochondrial ROS generation by enhancing the reduced status of key electron donors to the ETC. This leads to increases in the NADH:NAD + and FADH 2 :FAD ratios to the point that the ETC cannot cope; electrons leak directly, or following reverse electron transport to Complex I, and are swept up by oxygen to generate O 2 •− , which then rapidly dismutates to H 2 O 2 under the influence of SOD. Similarly, increases in the NADPH:NADP + ratio can favour the activation of NADPH oxidases such as NOX 4 and NOX5, generating excess ROS and impairing cell function [ 168 ]. High doses of powerful reductants like vitamin C can also enhance OS by promoting Fenton chemistry, whereby the reduced form of transition metals, such as iron and copper, can promote the generation of free radicals. High levels of certain antioxidants can also interfere with cellular homeostasis by disrupting the intricate redox-signalling processes highlighted above or by interfering with critical homeostatic mechanisms such as apoptosis or protein folding. Some antioxidants, such as polyphenols, may also have chemical structures (large planar hydrophobic molecules) that, at high concentrations, can insinuate themselves into membranes, disrupting cellular activity by promoting electron leakage and ROS generation from the mitochondria or disrupting receptor activation and signal transduction at the plasma membrane. They can also intercalate into the DNA, distorting the DNA backbone, disrupting chromatin compaction, and inducing DNA fragmentation [ 169 , 170 , 171 ]. At high doses, instead of protecting the genome from OS, such molecules can induce DNA strand breaks and/or inhibit repair enzymes like topoisomerase [ 172 ]. So, while there is abundant enthusiasm for the use of antioxidant supplements to treat a range of reproductive pathologies, we are still a long way short of this goal. First of all, we lack a simple, clinically validated test to determine which patients are suffering from OS and require antioxidant treatment. The same lack of a diagnostic test means that we do not know when to cease antioxidant treatment, so that the risk of over-supplementation and the instigation of reductive stress can be avoided [ 173 , 174 ]. The specific antioxidants that should be used, in terms of their bioavailability, mechanism of action, and site of action, have still not been optimized for different clinical conditions, and the quality of clinical trials in this area has been generally poor [ 175 ]. The indiscriminate use of poorly selected antioxidants to treat reproductive disorders, without regard for the redox status of the patient, the source, and biochemical nature of the stress, or the danger of over-supplementation, has led to widespread concern that this treatment strategy is ineffective. According to the WHO, “There are insufficient data to recommend the use of supplemental antioxidant therapies for the treatment of men with abnormal semen parameters and/or male infertility” [ 176 ]. Similarly, antioxidants have yielded disappointing results when used to manage complications of pregnancy [ 164 ], endometriosis [ 177 ], polycystic ovarian syndrome (PCOS) [ 178 ], and female infertility [ 179 ]. Although there are signs of definite promise for these reagents, much more fundamental work needs to be done on the diagnosis and characterization of OS before antioxidant supplementation can be adopted as an effective, robust, therapeutic option.

Oxidative stress (OS) is a well-recognized pathological phenomenon that plays a central role in the aetiology of human infertility. OS can occur as a result of a variety of factors, including: pathological conditions such as varicocele, polycystic ovary syndrome, endometriosis, fibroids, and inflammation [ 1 , 2 , 3 ], exposure to environmental stressors such as air pollutants, pesticides, and radiofrequency electromagnetic radiation [ 4 , 5 , 6 , 7 ], and the impact of a wide variety of lifestyle factors, including diet, obesity, and smoking [ 8 , 9 , 10 , 11 , 12 ]. Fundamentally, OS occurs because of an imbalance between reactive oxygen species (ROS) production and the body’s ability to neutralise those ROS through the action of enzymatic and non-enzymatic antioxidants [ 13 ]. This imbalance can lead to cellular damage because their unstable and reactive nature allows ROS to abduct electrons from nearby molecules, including proteins, nucleic acids, and carbohydrates, leading to lipid peroxidation, protein degradation, and the formation of advanced glycation products, respectively. In addition, ROS are instrumental in the activation of cellular defence mechanisms, such as apoptosis/ferroptosis, which can lead to the large-scale depletion of reproductive cells, including the male and female germ lines [ 12 ]. Within the microcosm of reproductive health, OS is recognized as having a key role in disrupting such fundamental biological processes such as fertilization, oocyte maturation, ovulation, decidualisation, embryo development, blastocyst implantation, pregnancy, and parturition [ 13 , 14 , 15 ]. Given all these negative impacts of ROS on the reproductive process, it is reasonable to question why evolution would have encouraged the generation of these potentially toxic metabolites in cells that are so critically important for procreation and transgenerational carriage of the genome. This review addresses the physiological roles of ROS in the reproductive biology of mammals, with particular reference to human fertility and the ability of men and women to generate normal, healthy offspring.

This review clearly emphasises that ROS is physiologically important at all stages of the reproductive process. While OS may well contribute towards many of the pathologies affecting human fertility, antioxidants should be utilised with care, with the aim of ensuring that an appropriate redox balance is maintained and that critical biological processes can continue unimpaired. This is especially important as there is now growing evidence for reductive stress negatively impacting the reproductive process from gametogenesis to parturition [ 173 ]. To rationalize the use of antioxidants to treat reproductive deficiencies in vivo and in vitro, we need to consider much more carefully the dose and specific structure of the antioxidants used in clinical practice. Too often, antioxidant administration is not calibrated with the level of OS being experienced by the patient or the cells being targeted. Indeed, antioxidants are frequently administered in vivo and in vitro without any diagnostic assessment of OS [ 175 ]. In addition, too little consideration has been given to the source and type of OS when selecting antioxidants in terms of their physicochemical properties (charge, size, hydrophobicity, half-life, bioavailabilty) or mode of antioxidant action (one electron-, two electron-, or hydrogen atom-donating) in optimizing their biological action. With the introduction of new ART culture media formulations enriched with antioxidants, and an abundance of articles highlighting the negative impacts of ROS on ART outcomes, there is a risk that antioxidants will be used in doses that compromise the physiological role of ROS in driving the reproductive process and inadvertently create a state of reductive stress, which can be just as damaging as its oxidative counterpart.