Mammalian sperm nuclear organization: resiliencies and vulnerabilities

The transition proteins

In mammals, hyperacetylated histones are first replaced by transition proteins. This is not the case in all species as for example in mollusks; the histone-protamine exchange does not require an intermediary [41]. Transition proteins (Tnp) are small proteins (between 50 and 140 residues), more basic than histones (but still less basic than protamines) and rich in arginine and lysine. Four Tnp are known in mammals but only Tnp1 and Tnp2 have been well studied. Tnp1 and Tnp2 are encoded by two different single-copy genes composed of 2 exons and an intron. In rodents and humans, the tnp2 gene is part of a cluster along with prm1, prm2, and prm3 genes. This cluster is surrounded by 2 matrix attachment regions (MAR) and involved in the transcriptional regulation of these genes during spermiogenesis [42]. The transcription of these clustered genes and tnp1, located on another chromosome, occurs at the same time in round spermatids. The corresponding mRNAs are stored as ribonucleoproteins for 3 to 7 days until translation. The proteins involved in this storage recognize the 3′UTR regions of the mRNAs. Moreover, these transcripts possess a long polyA tail (about 150 nucleotides) partially cleaved (around 50 nucleotides remaining) before translation. TheTnp mRNAs are then translated subsequently the Tnp proteins are phosphorylated at their C-terminus. This phosphorylation is a prerequisite for binding to the DNA. It is subsequently removed to increase the Tnp-DNA affinity and the chromatin condensation [43]. Tnp1 protein is 54 amino acids long, composed of 20% lysine, 20% arginine, and no cysteine (except in boars, bulls, and rams) in a highly conserved sequence between species. Tnp1 is strongly expressed and evenly distributed in the nucleus of spermatids. In vitro, Tnp1 decreases the melting temperature of DNA [44], destabilizes the nucleosome-DNA interaction and relaxes the chromatin on addition to nucleosome-binding DNA [45]. Tnp1 also increases the topoisomerase I activity [46] and stimulates single-strand break repair [47]. In vivo, tnp1 knock-out in mice did not induce a marked phenotype in sperm nucleus, but was observed to influence fertility [48]. In fact, only 40% of male mice were fertile the litter size was reduced from 7.7 to 1.6 pups/litter when males were mated with females of the same svj129 background. According to the authors, the infertility factor was due to a substantial decrease in sperm motility. In the spermatid nucleus, an abnormal chromatin structure was observed during condensation with the presence of rod-shaped chromatin condensation units in the fine fibrillar chromatin. In fine, the chromatin of epididymal spermatozoa was less condensed than in wild-type (WT) mice. The analysis of protein composition in the spermatid nucleus revealed a normal histone withdrawal but an increased incorporation of Tnp2 and a premature production of the Prm2 precursor protein. Moreover, the processing of the Prm2 precursor by cleavage was delayed and stable intermediary forms of Prm2 were detected in cauda epididymal spermatozoa.

Tnp2 is relatively different from Tnp1 in many aspects. This protein is twice as large as Tnp1, with a 117 to 138 amino acids poorly conserved between species. It is composed of 10% lysine, 10% arginine, 5% cysteine, as well as serine and proline. Tnp2 possesses 2 zinc-finger domains in the N-terminal domain and a highly basic C-terminal domain. Its expression levels vary depending on species. In vitro, Tnp2 increases the melting temperature of the DNA and condenses the nucleosome-binding DNA by oligomerization of close DNA strands [49, 50]. In vivo, Tnp2-null mice were fertile, however a decrease in litter size was observed (from 7.4 to 3.9 pups/litter; [51]. Epididymal spermatozoa presented flagellar defects and an abnormal chromatin structure that was less condensed than in WT mice, resembling that observed in tnp1-null mice. Tnp2 loss was compensated by an increase in Tnp1 expression and maturation defect of the Prm2 precursor (as recorded in tnp1-null mice) was observed. The tnp1/tnp2-null double mutant mice were found to be infertile [52]. These mice showed a great decrease in epididymal sperm counts, motility, viability, and normal morphology. In addition, the chromatin of the few epididymal sperm cells was weakly condensed. Moreover, in vitro fertilization with these spermatozoa revealed poor fertilizing abilities. Thus, this study underlined that Tnp1 and Tnp2 possess some redundant functions, but cannot fully compensate for one another, suggesting some specificity of function. The opposing in vitro properties of Tnp1 and Tnp2 also support these conclusions.

The protamines

Transition proteins are replaced in their turns by protamines (Prms). These highly basic small proteins are produced by genes evolutionary derived from an ancestral gene that was also at the origin of the histone H1 gene [53]. Over time, the sequence and the structure of the Prms have much diversified between species [54]. Two Prms, Prm1 and Prm2, have been characterized in mammals. Whereas Prm1 is expressed in all mammals, Prm2 is only expressed in some species such as primates, certain rodents, rabbits, hares, and horses. Although pigs and bulls possess a prm2 gene, it is not functional. Prm1 and prm2 are composed of 2 exons and an intron, as is the case for the tnp genes. As indicated already above, in rats, mice, and humans, prm1 and prm2 form a cluster with tnp2 and a third prm gene, prm3. Prm genes are expressed at the same time in round spermatids and the corresponding mRNAs are stored and inactivated for around 10 days, until the elongating stage, by the same processes than those described for Tnp mRNAs (for reviews see: [43, 55, 56]). It should be noted that prm3 encodes a small cytoplasmic acidic protein, not involved in the process of spermatid chromatin condensation [57]. As Tnps, Prms are phosphorylated immediately after mRNA translation, during the translocation of the proteins into the nucleus. This PTM is necessary for DNA binding, but is removed afterward, increasing the Prm-DNA affinity and the chromatin condensation. Prm1 is translated as a mature protein of about 50 amino acids, composed of an arginine-rich central domain and of cysteine-rich short domains. The N-terminal tail possesses some serine residues, which are phosphorylation sites involved in Prm1 incorporation in the spermatid chromatin. For its part, Prm2 is synthetized as a precursor protein of about a hundred amino acids. Poly-arginine domains are interspersed throughout the mature Prm2 and the content in histidine is higher than in Prm1 (up to 20 and 5% or less, respectively; [58]. As with Prm1, Prm2 is rich in cysteines and is also phosphorylated immediately after its synthesis enabling it to bind to DNA. The DNA-bound Prm2 is progressively matured by successive proteolytic cleavages of its N-terminus, a process that takes several days, increasing chromatin condensation step-by-step. This maturation process eliminates about 40% of the N-terminal domain of Prm2. In mice and humans, six cleavages are necessary to produce a mature protein of about 60 residues long. However, some of the intermediate products can persist in the mature sperm nuclei [59, 60]. Finally, another important difference between Prm1 and Prm2 is the ability of Prm2 to bind zinc. It is hypothesized that the zinc complex will participate in the final condensation of the sperm nucleus, protecting a number of protamine thiol groups from oxidation and thus limiting the formation of intra- and inter protamine disulfide bonds [61]. However, zinc is particularly enriched in the seminal plasma while disulfide-bridging of protamine thiols is a process taking place during epididymal maturation suggesting that the later is likely to be the prominent process driving the final state of compaction of mature spermatozoa (see below).

The final structure of the sperm chromatin

The histone-protamine exchange during the elongating phase of spermiogenesis drastically modifies the structure and the spatial organization of the sperm chromatin. At the end of this process, the sperm chromatin is 6 to 7 times more condensed than in the nucleus of any somatic cells. The high level of sperm DNA condensation almost resembles a crystalline structure, with little room to accommodate even water molecules. The compact structure shields it from mechanical shearing and chemical stressors that may have mutagenic effects. This is particularly important because mature spermatozoa are devoid of any functional DNA repair machinery (see below). Thus, it will be the task of the oocyte after fertilization and prior to the first division of segmentation to repair the paternal DNA in the newly decondensed male pronucleus [62]. Although the processes of sperm nuclear condensation and chromatin reorganization are considered essential, our knowledge of the structural intricacies of the sperm nucleus is scarce. Although it should be pointed out that the sperm nucleus is a highly ordered structure that is well conserved from one sperm cell to another as well as from one individual to another, the organization of the mature sperm nucleus however appears to be species–specific, limiting the ability to translate findings from mouse or bovine to human.

The basal unit of the sperm chromatin and its conformation

The nucleoprotamine structure constitutes the basal unit of the sperm chromatin. The Prms and their interaction with the DNA were first studied in salmon and bull [16, 63]. Unfortunately, no detailed crystallographic data are available as the DNA-Prm complex is insoluble. As shown by Raman Spectroscopy, when Prm1 is free in solution, the protein is unfolded, [64]. Prm1 acquires a stable conformation only when it is bound to DNA. Prm1 wraps around the double stranded DNA in one groove of the double helix via electrostatic and hydrogen bond interactions with the DNA backbone. Although, most studies conclude that Prm1 binds to the major groove of the DNA [65, 66], some studies report that it also binds to the minor groove [67]. The interaction of one Prm1 per turn of the DNA helix covers about 11 bp [68] allowing the DNA to curve in a conformation unique to the sperm chromatin. After binding, due to the presence of numerous cysteine residues contained in protamines, intramolecular disulfide bridges are initially formed to stabilize the Prm1-DNA interaction. Subsequently the intermolecular disulfide bridges formed between Prms trigger the adjacent DNA fibers to come closer for a tighter compaction of the sperm chromatin. Atomic force microcopy studies revealed that the addition of bull Prm1 to a free linearized plasmid DNA on a mica surface prompts its condensation into a toroidal structure, called a toroid [69]. Other experiments showed that around 50 kb of DNA can be coiled into a salmon protamine toroid [70]. Toroids were also observed in native human sperm chromatin [71]. The toroid is therefore considered as the first level of organization of the sperm chromatin (see Fig. 1). Similar properties were found for Prm2-DNA interaction with the addition that zinc ions participate in the interaction. Prm2 binds zinc via its cysteine residues with one zinc ion mobilizing 4 cysteine residues. The amount of zinc associated to Prm2 is important since it was reported that there is an equimolar ratio of zinc and Prm2 in human sperm [68]. The zinc-mediated thiol bridges then stabilizes the Prm2 and the sperm chromatin complex [61]. These observations are in agreement with the role associated with zinc in male fertility [61]. It is therefore evident that a fine balance is present between the number of protamine cysteine residues involved in disulfide bridges versus the number of cysteine residues chelated by zinc. This balance is directly linked to the sperm nuclear content in protamines since Prm1 cysteine residues are involved in disulfide bridges while Prm2 cysteine residues are involved in zinc chelation, thus preventing these residues to form disulfides. These observations explain why the sperm nucleus condensation is both sensitive to the redox status and to zinc availability. High levels of zinc will reduce disulfide bond formation leading to a less nuclear compaction while lower concentration of zinc will promote disulfide bridging resulting in a tighter nuclear compaction. Although a tighter sperm nuclear condensation could be perceived as beneficial, it may not be the case since after fertilization a highly condensed paternal DNA will require more time and more energy to be decondensed. Similarly, the redox environment may influence protamine disulfide bridging events consequently sperm nuclear condensation. Thus, we postulate the existence of a fine equilibrium that is critical for optimal sperm nuclear condensation and zinc concentration. The high level of zinc found in the seminal plasma may therefore limit further disulfide bridging events in the sperm nucleus during its journey up to the fertilization site in the female genital tract.

Practically, in high salt conditions and with the use of a reducing agent such as dithiotreitol (DTT) protamines can be extracted from the nucleus of epididymal spermatozoa because of the reduction of the disulfide bridges. This treatment promotes the formation of a halo of DNA loops exiting the sperm nucleus. This halo can be viewed after ethidium bromide staining and its measurement in hamster revealed that average loop length were around 46 kb [72]. This estimate is close to the 50 kb or so of DNA found associated with a toroidal structure of protamines as observed in vitro when using plasmid DNA and salmon protamine [70]. Of note is the fact that with human sperm cells the halo revealed a DNA loop size of about 27 kb [73].

Chromatin defects occurring during spermatogenesis

There are numerous ways in which the sperm nuclear structure could be compromised frequently resulting in local or global abnormal sperm nuclear condensation. When local, it is often associated with nuclear vacuoles usually visible at high magnification (above x6000) depending on their size and number. It is now well accepted that these sperm head vacuoles are indeed nuclear in nature and relate to local impairment in nuclear condensation [113, 114]. Local defects in protamination are likely the cause, however, even with the phenotype characterized, the mechanisms involved are not well understood and requires further research. To avoid the selection of spermatozoa with vacuoles for IntraCytoplasmic Sperm Injection (ICSI) procedure, some clinicians now recommend the use of differential interference contrast microscopy (also called Normarski contrast) that allows the observation of live sperm at high magnification. The use of the Motile Sperm Organelle Morphology Examination (MSOME) technique allows the detection of sperm head vacuoles that otherwise would go unnoticed at the regular magnification (x300) used in routine for the selection of sperm for ICSI. However, the use of this protocol in improving ICSI outcome is not yet confirmed due to the paucity of comparative studies [115–121]. Indeed only a few studies comparing the results of IMSI (Intracytoplasmic morphologically selected sperm injection) with regular ICSI have been reported with the blastocyst implantation rate, miscarriages and birth rates as the end points. The sole indication of the presence of chromatin defects that prevails to date is when clinicians face recurrent implantation failures while performing ICSI [122]. The drawback of such sperm selection protocols is the time they require and the exposure of sperm to bright light and media that may lead to oxidative DNA damage.

In addition to these local alterations of the sperm nuclear condensation, there are instances of global sperm nuclear decondensation. These may have several causes such as defective protamination and excessive DNA fragmentation. The latter may be due to unrepaired meiotic strand breaks, mechanical shearing during the cytodifferentiation step at spermiogenesis, or/and the result of oxidative damage essentially in response to environmental conditions (chemical and physical stressors). To better understand the histone to protamine exchange process and the importance of an optimal sperm chromatin organization and compaction on male fertility, different mutant mice were generated. Mice deleted for the two poly(ADP-ribose) polymerases (Parp1 and Parp2), essential proteins involved during the replacement of histone by Prm, are infertile and show increased amount of histones in spermatocytes along with reduced sperm chromatin condensation [123, 124]. Mice knockout for Brdt (Bromodomain-containing protein) which recognizes hyperacetylated histones allowing their replacement by transition proteins (Tpn) are also infertile. They present a severe reduction both in sperm counts and sperm motility associated with high percentage of sperm morphological abnormalities including misshaped heads and failure of nuclear condensation [125]. A similar phenotype was observed in the chromodomain helicase DNA binding protein 5 (Chd5) mutant mice. This was not a surprise since Chd5, similarly to Brdt, orchestrates histone replacement, histone H4 hyperacetylation, histone variant expression and nucleosome eviction [126]. It is worth noting that CHD5 was found highly expressed in the human testis during spermiogenesis and that low CHD5 expression was associated with some human infertile situations [126]. Another knockout model, the rnf8 ^−/− mouse, associates sperm chromatin compaction defect with male infertility. Rnf8 is a ubiquitin E3 ligase that ubiquitinates in particular histones H2A and H2B. This PTM appears to promote histone H4K16 acetylation, a critical modification for the replacement of histones by protamines during spermiogenesis. As a consequence, rounded germ cell nuclei of Rnf8-deficient male mice contained histones but no Prm leading to severely impaired sperm chromatin compaction. In addition, Tnp1 and Tnp2 levels were dramatically reduced and a mild elevation of germ cell apoptosis was recorded in both testes and epididymis. rnf8 ^−/− sperm also displayed abnormal rounded heads and retained residual cytoplasm [127]. Likewise, Tnp1- and Tnp2-null double mutant mice present with fertility problems associated with abnormal sperm chromatin condensation due to incomplete protamination, a decrease in sperm count as well as a decrease in sperm motility [128]. With respect to the mammalian protamines Prm1 and Prm2, we have seen above that the incorporation of these two proteins into the sperm chromatin is strictly regulated, resulting in a species-specific, tightly controlled Prm1/Prm2 ratio. In human sperm the ratio of PRM1/PRM2 is approximately equal (1:1) in fertile men whereas it was found to be displaced in some infertile situations [129–131]. Patients with altered P1/P2 ratio were shown to be more likely to display decreased sperm concentration, decreased motility, a higher frequency of abnormal sperm morphology and an elevated level of DNA damage. In mice, the knockout of Prm1, Prm2 or both is associated with infertility and non-functional spermatozoa showing abnormal morphology, reduced sperm counts and motility as well as increased DNA fragmentation [57, 132–134]. Recently, Prm1-deficient female chimeric mice carrying Prm1-deficient oocytes were generated. These mice successfully produced Prm1(+/−) male mice. Via in vitro fertilization (IVF) healthy Prm1(+/−) offspring were produced demonstrating that spermatozoa lacking Prm1 can fertilize and produce viable embryos. However a detailed survey of the offspring showed midification in expression profiles [135].

Chromatin defects due to chromosomal abnormalities

Sperm aneuploidy originates from segregation errors during the meiotic divisions of spermatogenesis, though the exact causes of sperm aneuploidy are unknown. Perrin et al. [136] found that men with structural chromosomal abnormalities (reciprocal translocations, Robertsonian translocations and pericentric inversions) had higher rates of DNA fragmentation than men without abnormal karyotypes. DNA fragmentation was significantly correlated with the percentage of aneuploidy in chromosomes X, Y, 13, 18 and 21 [137]. Muriel et al. [138] and Enciso et al. [139], have confirmed these findings and shown that sperm aneuploidy may lead to an increase in sperm DNA damage. This is consistent with the fact that sperm DNA integrity partly depends on the structural organization of the sperm nucleus and that chromosome alterations may lead to local architecture modifications [140]. To date, it is not that clear whether DNA damage and aneuploidy in sperm are causally linked or associated by a common mechanism of damage, or if it is just the consequence of a disrupted nuclear architecture [141]. It should be noted that recent findings report that chromosome structural aberrations (disomy, translocation) found in the offspring can be also caused by the oocyte attempt to repair DNA alterations accumulated on sperm DNA after spermatogenesis [142].

Post-testicular chromatin defects

Leaving the testes, spermatozoa are not completely mature both at the structural and functional levels. Amongst the numerous modifications that occur on spermatozoa while going down the epididymal tubule up to the storage compartment one concerns the completion of the sperm nucleus compaction. We and others have demonstrated that during the epididymal journey, the sperm nuclear protamines rich in cysteine thiol-containing residues are modified by oxidative events resulting in the formation of intra and inter-protamine disulfide bridges. This nuclear process requires the activity of an enzyme located in the sperm nuclear compartment, the snGpx4 protein (sperm nucleus glutathione peroxidase 4, [143]) that uses hydrogen peroxide (H₂O₂) to make disulfide bonds, thus working as a disulfide isomerase. There is now a large body of evidence showing that a finely tuned level of oxidation is required to further condense the sperm nucleus. Absence of sngpx4 induces a delay in post-testicular sperm nuclear compaction with abnormal chromatin condensation and sperm head fragility when cells are eventually challenged by mild reducing conditions, resulting in a giant head phenotype [143, 144]. We have shown earlier that in mouse an epididymal luminal scavenger, the glutathione peroxidase 5 protein (Gpx5) is critical in this post-testicular disulfide bridging process since it contributes to fixing the optimal H₂O₂ concentration in the epididymal fluid [145]. Consequently it also determines the optimal level of disulfide bridging on the sperm nucleus [146]. When Gpx5 is absent (in gpx5 ^−/− animals) it results in DNA oxidative damage, mainly to the sperm nucleus, a cell compartment that is difficult to protect even though the epididymis of gpx5-deficient animals does it best to limit the increase of luminal oxidative attacks on the sperm plasma membrane [145]. Sperm DNA oxidation was associated with reproductive failures including an increase in miscarriages, an increase in abnormal embryonic development and an increase in perinatal mortality. When both sngpx4 and gpx5 were knocked out we recorded cumulative effects on sperm cells that show both nuclear susceptibility to decondensation and nuclear oxidation [144]. Therefore, when the testis is not at the origin of the defect, the sperm nuclear condensation can also be challenged in its post-testicular life especially during epididymal transit where oxidative processes are at work to complete it [146, 147]. Any disruption of the oxidative balance in or around sperm cells in their post-testicular life may thus affect the level of nuclear condensation. Oxidative stress, but also reductive stress, may end-up having the same effects on the sperm nucleus state of condensation. Excessive oxidation while it may serve nuclear condensation at first will lead to DNA fragmentation, resulting ultimately in decondensation [148]. Reductive stress will destroy the disulfide bridges sticking the nuclear protamines together and therefore will result in decondensation of the sperm nucleus [148].

Germ cell apoptosis and DNA fragmentation during spermatogenesis

Apoptosis is a major feature of male germ cell development. Apoptosis is used as a mechanism for the removal of damaged germ cells from seminiferous tubules so that they do not continue to differentiate into spermatozoa. During spermiogenesis, Sertoli cells are responsible for the induction of apoptosis in 50 to 60% of all germ cells that enter meiosis. Germ cells marked with apoptotic markers of the Fas type are eliminated by the Sertoli cell via a phagocytic process [149, 150]. However, this mechanism may not always operate efficiently and a variable percentage of apoptotic germ cells enter the process of sperm remodeling and appear later in the ejaculate. They present DNA strands-breaks associated to the apoptotic process (=abortive apoptosis). Apoptosis during spermatogenesis has been suggested to play a role in the etiology of spontaneous male infertility in light of the excessively high numbers of apoptotic germ cells observed in the testes of some infertile males [151]. In addition, apoptotic markers including caspase activation and phosphatidylserine exteriorization have been detected in mature spermatozoa from infertile males [152]. Besides the apoptotic process there is another way by which DNA fragmentation may occur during spermatogenesis. During spermiogenesis it was shown that chromatin packaging requires endogenous nuclease and ligase activities to create and ligate nicks that facilitate the protamination step. McPherson and Longo [153] hypothesized that the presence of high level of DNA nicks in ejaculated sperm may be indicative of impaired spermiogenesis [153]. These nicks are thought to provide relief of torsional stress to support chromatin arrangement during the displacement of histones by the protamines [154]. If not repaired completely in a timely manner they may affect sperm chromatin packaging and render spermatozoa more susceptible to post-testicular damage. As an illustration, mice deficient for poly(ADP-ribose) show a high level of unrepaired DNA nicks that are associated with male infertility [31].

Sperm DNA fragmentation induced by Reactive Oxygen Species (ROS)

ROS are natural products of cellular metabolism. In physiological concentrations, sperm cells require ROS at different moments of their life. During epidiymal maturation ROS (especially H₂O₂) participates in the processes of sperm maturation (disulfide bridges on sperm proteins). ROS are also required for successful oocyte fertilization acting as second messengers in the capacitation processes including hyperactivated motility and acrosomal exocytosis. However, when ROS generation exceeds ROS recycling it contributes to a large proportion of instances of male infertily [155, 156]. ROS target the polyunsaturated fatty acid (PUFA) rich sperm plasma membrane altering membrane fluidity and mitochondria functions resulting in impaired mobility and decreased fusogenic capacity with the oocyte. ROS, especially hydrogen peroxide (H₂O₂) may also reach the sperm nucleus leading to oxidative DNA damage that may lead to mutagenic effects which may be transmitted. There are many common situations that may lead to sperm exposure to ROS whether it is secondary to aging, environmental factors (exposures to toxicants, drugs, UV, ionizing radiations, heat…), pathological situations (infection, inflammation, metabolic disorders….), lifestyle (unbalanced diet, smoking, alcohol addiction…) [157–162]. In addition, sperm exposure to ROS may happen during assisted reproductive technologies (ART) procedures for example during sperm cryopreservation, exposure to culture media or sperm handling during selection processes, especially for ICSI (intracytoplasmic Sperm Injection) [163]. As we have seen above, because ROS and especially H₂O₂ are such important contributors to the completion of the sperm nucleus compaction during epididymal transit and beyond, it is no surprise to find them involved in nearly all situations of defective spermatozoa.

Oxidative alterations of cellular components are a very common problem of aerobic cells which are usually well–equipped in enzymatic and non-enzymatic primary and secondary antioxidants to deal with it. Oxidative alterations can affect the nuclear compartment of any cells, especially because reactive oxygen species such as H₂O₂ can freely pass through plasma membranes. When this happens and when ROS reaches the nucleus it may be at the origin of modified bases, abasic sites, chromatin protein cross-linking and DNA strand breaks (both single and double) depending on the intensity of the oxidative attack [148]. In any somatic cell as well as in an oocyte, the base excision repair (BER) pathway will replace these oxidized bases by non-oxidized bases correcting the alterations. Mature sperm cells cannot do that as they have been shown to lack the necessary equipment [62]. Only the first player of the Base Excision Repair (BER) pathway, the 8-oxoguanine DNA glycosylase 1 (Ogg1/OGG1) was shown to work in rodent and human spermatozoa [62]. OGG1 activity marks the oxidized base to be removed [62]. Sperm cells will then rely on the oocyte BER equipment that will correct the paternal genome after fertilization upon the decondensation step of the male pronucleus and before the first division of segmentation. The consequence of sperm DNA damage with respect to normal embryo development is therefore the result of an equilibrium between the extent of nuclear oxidation and the DNA repair capacity of the oocyte [164, 165]. It has been shown that the zygote responds to sperm DNA damage through a non-apoptotic mechanism that acts by slowing paternal DNA replication and ultimately leads to the arrest of embryonic development [164, 166–168]. After induction of DNA oxidative damage on Rhesus sperm using xanthine and xanthine oxidase, Burruel et al. [169] have shown that ICSI-produced embryos present severe fragmentation, multi-nucleation, and early cell arrest essentially around the four-cell stage. Because of this inability to repair its DNA the mature spermatozoa is very sensitive to DNA oxidative alterations. Paradoxically, mature spermatozoa are prone to suffer oxidative attacks because they harbor a peculiar plasma membrane rich in polyunsaturated fatty acids (PUFA) that is highly susceptible to ROS. When oxidized in situations with an excess of ROS or weakness in the antioxidant protective activities, these PUFA will amplify the generation of ROS in a vicious oxidative stress circle [170]. In addition, even-though they are most sensitive to oxidative stress we have seen above that mature spermatozoa are physiologically exposed to ROS. We, and others have demonstrated that part of their post-testicular (ie. epididymal) maturation step utilizes a finely tuned concentration of H₂O₂ to complete the condensation of the sperm nucleus via disulfide bridging events on the thiol-containing protamines. How relevant sperm DNA oxidation is with respect to male infertility is difficult to say at this stage in the absence of clinical trials in which the level of sperm DNA oxidation is correlated with reproductive success. However, there are recent reports suggesting that it is a major concern since it was shown that over 60% of male entering ART programs present medium to high levels of sperm DNA oxidative alterations [171]. The oxidized base adduct, 8-hydroxy-2′-deoxyguanosine (8-OHdG) has been used in studies to demonstrate that oxidative DNA damage is significantly elevated in the spermatozoa of patients attending infertility clinics [172]. Robust clinical trials are necessary to correlate reproductive success with sperm DNA oxidation.

A common belief is to think that sperm nuclear fragmentation is always associated with sperm nuclear oxidation. Although this is true when the oxidative stress around sperm cells is high, in many situations a mild oxidative stress will not lead to sperm DNA fragmentation. Therefore, one cannot simply say that there is no sperm DNA oxidative damage by assessing the level of sperm DNA fragmentation. This is clearly demonstrated in transgenic animal models having medium sperm DNA oxidative alterations that are not associated with DNA fragmentation [145]. In these models the level of sperm DNA oxidation is sufficient to lead to reproductive failures when transgenic males are crossed with WT females [145]. The reproductive problems recorded ranged from increased miscarriages, increased abnormal development and increased perinatal mortality, all classical issues in reproductive defects both in natural and artificial reproduction.