*4.4. Sperm DNA Modifications*

At the genomic level, the fidelity of mitochondrial ATP production is controlled by the interplay of mitochondrial (mtDNA) and nuclear (nDNA) DNA, which encode the various components necessary for the proper assembly and function of the mitochondrial complexes of OXPHOS. Specifically, the mtDNA encodes a subset of the protein subunits of Complex I, Complex IV, cytochrome b and ATP synthase, while the nDNA encodes the remainder of the enzymatic components. It follows that the integrity of the mitochondrial and nuclear genomes within the developing germ cell are crucial for the subsequent establishment of normal sperm motility profiles and conversely, that ROS can elicit a deleterious impact on this aspect of sperm function via the induction of DNA damage (reviewed in [129]).

Sperm DNA is highly vulnerable to ROS damage owing to the progressive silencing of the germ cells transcriptional machinery during the latter phases of their development and the attendant reduction in their capacity to mount an e ffective DNA repair response. Indeed, the replacement of sperm histones with protamines during spermiogenesis promotes extreme compaction of the nDNA; a phenomenon that protects sperm chromatin against ROS-mediated, and other sources, of damage. However, if the integrity of the protamination process is compromised such that portions of poorly compacted DNA remain, these genomic regions are placed at heightened risk of oxidative attack. Accordingly, recent genome wide analyses have demonstrated that oxidative sperm DNA damage occurs predominantly on specific chromatin regions with lower compaction associated with histones and inter-linker domains attached to the nuclear matrix [130,131]. Thus, ROS such as H2O2, O2•<sup>−</sup> or •OH are all capable of directly damaging DNA integrity by way of base modifications, induction of single- and double-strand DNA brakes, chromatin cross-linking and/or deletions [129]. Additionally, 4HNE has also been shown to promote the formation of DNA adducts such as 8-oxoguanine (8-oxoG), 1,N6-ethenoadenosine and 1,N2-ethenoguanosine in human sperm cells [132]. Accordingly, elevated levels of oxidative DNA damage, and in particular 8-oxoG lesions, are frequently encountered in the spermatozoa of male infertility patients [129,132]; emphasizing the importance of evaluating sperm DNA fragmentation in individuals considering assisted reproduction treatments.

In males with asthenozoospermia, the levels of sperm DNA fragmentation have been found to be significantly higher than that of fertile controls [133]. Bonanno et al. [20] have also reported that asthenozoospermic males had elevated ROS levels in their semen, which was correlated with decreased mtDNA integrity in their spermatozoa. By comparison, nDNA fragmentation was only detected in less than one-fifth of the patients analyzed in this study [20]. This phenomenon may reflect less efficient packaging of mtDNA, rendering it more susceptible to ROS and 4HNE attack [134–136]. Alternatively, in spermatozoa the mtDNA is placed in much closer proximity to the main source of ROS generation than that of the nDNA. Aside from changes in mtDNA integrity, it has also been reported that spermatozoa with low-motility possess elevated levels of mtDNA copy number; suggesting that these cells incorporate more mtDNA content during spermatogenesis [20,137–139]. By contrast, measurement of cell-free mtDNA copy number in the seminal plasma of asthenozoospermic and oligozoospermic males has identified a reciprocal relationship, whereby lower levels of cell-free mtDNA are associated with increased levels of ROS [21]. A testicular origin for these lesions is supported by the presence of mtDNA deletions in the spermatozoa of infertile males [140,141]. Further, increases in both mtDNA copy number and mtDNA deletions have been recorded in the same patient samples, wherein they were strongly associated with the presence of OS [142].

One of the best-described large-scale mtDNA deletions is the specific 4977 bp deletion (mtDNA4977), which occurs between nucleotides 8470 and 13447 and eliminates seven genes encoding four subunits of Complex I (ND3, ND4, ND4L, partial ND5), one subunit of Complex IV (COIII) and two subunits of ATP-synthase (ATP6 and partial ATP8); all of which are crucial for OXPHOS (depicted in Figure 3). Accordingly, mtDNA<sup>4977</sup> has been linked to a spectrum of disorders, including heart disease, different forms of cancer and mitochondrial diseases [143–146]. Notably, this mutation has also been reported to accumulate in different human tissues as a consequence of natural aging [147]. If the mtDNA<sup>4977</sup> deletion occurs during spermatogenesis and/or spermiogenesis, the mature spermatozoa are endowed with dysfunctional mitochondria, which can act as a source of ROS. Interestingly, mtDNA<sup>4977</sup> has been recorded in the spermatozoa of infertile males with asthenozoospermia and oligoasthenozoospermia [148,149]. An alternative large-scale mtDNA deletion comprising 7436 bp (mtDNA7436) has also been reported in low-motility human spermatozoa [150]. In addition to the genes excised by mtDNA4977, mtDNA<sup>7436</sup> also eliminates genes encoding subunit 6 of Complex I (ND6) and cytochrome b from the mtDNA genome (see Figure 3). Although positive correlations between mtDNA deletions and motility have not been universally established [151,152], Kumar et al. [19] have shown that mutations in sperm mtDNA (giving rise to nucleotide changes in subunits of: ATP6, ATP8, ND2, ND3, ND4 and ND5) do exist in males with oligoasthenozoospermia and that these defects are associated with elevated ROS levels in their spermatozoa. In any case, the crucial role of mitochondria in sperm bioenergetics, as well as the potential for defective mitochondria to generate excessive ROS, motivates a better understanding of the factors responsible for perturbation of mtDNA in developing germ cells.

**Figure 3.** Mitochondrial DNA and location of the mtDNA 4977 bp and 7436 bp deletions. The mtDNA<sup>4977</sup> deletion includes genes encoding two subunits of ATP-synthase (ATP6 and ATP8), cytochrome oxidase III (COIII), NADH dehydrogenase subunit 3 (ND3), ND4, ND4 subunit L (ND4L), and ND5, whereas mtDNA<sup>7436</sup> includes ATP6, COIII, ND3, ND4, ND4L, ND5, ND6 and the entire cytochrome b (Cyt b). As a consequence, spermatozoa harboring mtDNA4977 and mtDNA7436 deletions lack several essential OXPHOS genes, fail to assemble functional ETC complexes, and experience compromised energy production. Abbreviations: CI, ETC Complex I; CIII, ETC Complex III; CIV, ETC Complex IV.

### **5. Antioxidant Systems in the Male Reproductive Tract**

The male reproductive tract produces a wide range of antioxidant scavengers capable of defending spermatozoa from ROS attack. Given that mature spermatozoa are translationally inactive and carry with them minimal endogenous antioxidant defenses, they are highly dependent on these exogenous sources of enzymes during spermatogenesis and their residence in the male reproductive tract. In semen, the most abundant antioxidant enzymes are those belonging to the glutathione peroxidase (GPX) and peroxiredoxin (PRDX) families (reviewed in [153]). The GPX4 isoform is predominantly synthesized in the testis and in mature spermatozoa is abundantly localized in the midpiece. Transgenic mice lacking GPX4 display impaired sperm quality, including deficits in sperm motility and structural abnormalities in the midpiece of the flagellum [154]. An alternative GPX isoform, GPX5, is highly expressed in the caput epididymis and has been implicated in the protection of sperm DNA based on the demonstration that the spermatozoa of *Gpx5*-null mice display lower levels of DNA compaction and higher levels of 8-oxoG than their wildtype counterparts [155]. In addition to GPXs, PRDXs are also highly expressed in the caput and cauda epididymis and have been documented within seminal plasma and in virtually all domains of the human spermatozoon [156]. PRDXs are a highly-conserved family of thiol-dependent peroxidases that regulate antioxidant defense systems by virtue of their ability to reduce H2O2, peroxynitrite (ONOO−) and hydroperoxides (ROOH); themselves becoming inactivated upon oxidization [156,157]. PRDX1 and PRDX6 have been shown to be abundantly expressed in rat epididymal spermatozoa and to become highly oxidized after the induction of OS with substrates such as tert-butyl hydroperoxide (tert-BHP) [158]. Due to its catalase and calcium-independent phospholipase A2 (Ca<sup>2</sup>+-iPLA2) activities, PRDX6 has also been implicated in the prevention of LPO and the repair of oxidized membranes [159]. Accordingly, the inhibition of PRDX6 Ca2<sup>+</sup>-iPLA2 activity in human spermatozoa has recently been shown to promote extensive oxidative damage (including high levels of LPO, DNA oxidation and reduced ΔΨ), which contributes to a phenotype of reduced sperm motility [160]. Aside from GPXs and PRDXs, alternative enzymatic antioxidants such as superoxide dismutase (SOD) and catalase (CAT) collaborate to protect spermatozoa held within male reproductive tract. Such enzymatic defenses can be supplemented by dietary derived non-enzymatic antioxidants, including vitamins (C, E and B), carotenoids, glutathione, inositol, carnitines, cysteines, hyaluronan, serum albumin, and zinc (reviewed in [161]).

Highlighting the physiological importance of these collective antioxidant defenses, patients with poor sperm motility parameters commonly have an attendant deficit in their levels of seminal plasma antioxidants. By way of example, studies of asthenozoospermic individuals with idiopathic or varicocele-related background have shown these lesions are commonly accompanied by relatively low levels of PRDX6 and PRDX1 enzymes within their seminal plasma and spermatozoa, respectively [162]. Similarly, the levels of lactoferrin (LTF; an iron-binding glycoprotein secreted by the mammalian epididymis [163,164] that possesses antioxidant properties and is capable of binding receptors on the sperm head and midpiece [165]) are reduced in spermatozoa from asthenozoospermic males [166]. Whilst others have reported the opposing trend; i.e., higher levels of LTF in spermatozoa from males with asthenozoospermia [18], such discrepancies may reflect the alternative defects that give rise to this pathology and/or the integrity of sperm membrane domains wherein LTF receptors are localized. Building on this body of evidence, the levels of vitamin C, vitamin E and the reduced form of glutathione (GSH; an endogenous antioxidant that can function synergistically with vitamin C [167]), have each been found to be diminished in the semen of asthenozoospermic males compared to normozoospermic controls [168,169]. In contrast to this general trend, data from meta-analyses have failed to document significant changes in the levels of alternative seminal plasma antioxidants such as SOD between males with different forms of infertility (including oligoasthenozoospermia) and that of healthy controls [170], suggesting that not all antioxidants are of equivalent importance in terms of maintaining sperm function. One possible explanation is that the attenuation of antioxidant capacity in infertile individuals may be due, at least in part, to redox driven modifications of specific antioxidant enzymes. Illustrative of this, PRDXs, which serve as primary antioxidants in ejaculated human spermatozoa, are especially vulnerable to ROS-induced modifications such as S-nitrosylation, tyrosine nitration, and S-glutathionylation (reviewed in [171]). Oxidative damage to specific antioxidants may therefore be among the first steps in the cascade of events that contribute to OS-mediated male infertility.

Taking into account the potential of antioxidants to ameliorate OS related pathologies, it is perhaps not surprising that different cocktails of enzymatic and non-enzymatic antioxidants have found therapeutic application during assisted reproductive interventions [172]. Among the most popular of these are vitamin C, zinc and L-carnitine [173–175], although it must be acknowledged the outcomes of these trials are far from consistent. In this context, zinc supplementation has been reported to improve sperm motility by way of reducing OS, apoptosis and DNA fragmentation; but notably, these results were only achieved in the presence of vitamins C and E [174]. Similarly, in a study by Garolla et al. [175], L-carnitine was shown to improve sperm motility but only in samples wherein normal levels of GPX were maintained. This caveat could explain the opposing data obtained by Sigman et al. [176], who reported that L-carnitine had no effect on sperm motility. In view of these dichotomous results, there is a clear imperative to continue basic research to advance our understanding of the interplay of ROS and sperm biology in order to inform the development of effective therapies to combat OS-mediated lesions.
