**4. Discussion**

The present study shows for the first time the long-term adverse effect of t-BHP induced in vivo oxidative stress on rat spermatozoa and epididymis and the differential expression of antioxidant enzymes that fight against hydroperoxides in epididymis and testis. The reduced motility and high DNA damage observed in rat cauda epididymal spermatozoa collected at 3 weeks, 6 weeks and 9 weeks after the end of t-BHP treatment indicates the long-lasting adverse effects of oxidative stress

on sperm quality. This result was unexpected since after one cycle of spermatogenesis (9 weeks in the rat) [25], new spermatozoa were produced from spermatogonia and these spermatozoa were those present in the epididymis at the time of collection [28]. Indeed, these collected spermatozoa showed significant DNA oxidative and impaired motility (Figure 1). These findings sugges<sup>t</sup> that the balance between ROS and the antioxidant system has been compromised in the testis and or epididymis after the t-BHP-induced oxidative stress.

Based on the time of the collections and the length and stages of rat spermatogenesis, the collected rat spermatozoa were spermatids, spermatocytes or spermatogonia at the time that the rat was exposed to the t-BHP treatment [28]. When spermatozoa leave the testis, they enter into the epididymis to undergo their maturation before ejaculation [21]. Since the treatment has struck both testis and epididymis, the detrimental effects observed in the spermatozoa retrieved form the cauda epididymis could be a consequence of a direct impact of oxidative stress on the germ cells during spermatogenesis or due to the detrimental and persistent effects of high levels of ROS in the epididymal epithelium that impairs the proper epididymal maturation of spermatozoa.

In the epididymis, we expected that the increased expression of antioxidant enzymes would reduce the oxidative damage in spermatozoa because of the transfer of antioxidant enzymes from the epididymis to spermatozoa through the secretion of epididymosomes [23]. However, high 4-HNE levels in caput and cauda epididymis are an indication of developing lipid peroxidation (Figure 2) and sugges<sup>t</sup> that the epididymal epithelium itself was damaged by the oxidative stress and therefore unable to protect spermatozoa during their maturation. The upregulation of PRDX6 and the trend of increase of PRDX1 and catalase expression in epididymis were not sufficient to scavenge the excessive ROS and restore the healthy cellular environment for a normal sperm epididymal maturation. We observed a similar response by the epididymis of rats treated for two weeks with t-BHP, with an upregulation of PRDX1 and PRDX6 but not of catalase [18].

Contrarily to what was observed in the epididymis, there were no significant differences in the levels of antioxidant enzyme expression and lipid peroxidation in the testes of treated rats compared to controls. Furthermore, the histological analysis and the testis weight and spermatid count indicated that there was no disruption of spermatogenesis. These findings sugges<sup>t</sup> that there was no evidence of oxidative stress during spermatogenesis that could damage the collected spermatozoa at the different end points. Testicular spermatozoa have lower levels of DNA damage compared to the ejaculated counterparts [29,30]. Moreover, the fact that testicular spermatozoa from infertile men with obstructive azoospermia have low levels of DNA oxidation that do not interfere with the formation of an embryo by intracytoplasmic sperm injection suggests that the level of DNA oxidation in testicular sperm is not detrimental for male fertility [31]. It is plausible that the testicular spermatozoa are more resistant to oxidative stress compared to epididymal spermatozoa that flow freely in the lumen because the developing spermatozoa are guarded by the Sertoli cells that provide nutrients [32,33] and antioxidant protection through SOD, GSTs, GPXs, and PRDXs [34–36].

During epididymal maturation, the sperm chromatin is further compacted and could be exposed to oxidative stress generated by different conditions. Thus, it is of paramount importance that the epididymal epithelium protect the maturing spermatozoon against oxidative stress. The finding that PRDX6 expression levels are high when lipid peroxidation (measured by 4-HNE levels) are increased in cauda epididymis of treated rats, collected at week 3 and week 6, while these levels return to those of controls at week 9, indicate that PRDX6 is an essential component of the antioxidant response in the epididymis.

In a previous study, we challenged rats with the same t-BHP treatment and found that caudal epididymal spermatozoa collected 24 h after the end of the treatment had increased DNA oxidation, and reduced motility [18]. These findings indicated the negative effect of in vivo oxidative stress exclusively on epididymal maturation. We observed similar damages in mouse lacking PRDX6, a condition that generates an in vivo oxidative stress and is associated with male infertility [16]. The *Prdx6*−/− spermatozoa have low motility and high levels of DNA oxidation and lipid peroxidation. The *Prdx6*−/− spermatozoa also had higher percentages of cytoplasmic droplet retention compared to wild-type cells [16], an indication of abnormal epididymal maturation. During the epididymal transit, spermatozoa shed the residual cytoplasm; thus, an increase in spermatozoa carrying cytoplasmic droplets is an indication of abnormal epididymal maturation [37].

PRDX6 is a unique antioxidant enzyme as it is the only antioxidant enzyme known to date, with calcium-independent phospholipase A2 (iPLA2) [38] and lysophosphatidylcholine acyl transferase activities (LCAT) [39]. Both PRDX6 iPLA2 and LCTAT activities are essential to remove and replace the oxidized phospholipids with newly synthesized phospholipids [39,40]. The epididymis increases PRDX6 in response to the oxidative damage caused by ROS to try to repair oxidized membrane lipids. A recovery of lipid peroxidation to the control level was observed in epididymis and spermatozoa at week 9, suggesting that PRDX6 repaired the damaged lipid membranes in this organ. This tendency was found in cauda epididymis but not in caput epididymis. We previously indicated that PRDX6 plays a crucial role in protecting both the epithelium and the spermatozoa in the cauda epididymis segmen<sup>t</sup> specifically [18]. Noteworthily, the higher levels of 4-HNE found in caput compared to cauda epididymis at week 9 sugges<sup>t</sup> a di fferential capacity of antioxidant response in the di fferent parts of the epididymis.

Although some repair of oxidative damage such as lipid peroxidation was observed in the caput and cauda epididymis at 6 and 9 weeks (Figure 2) and in spermatozoa at 9 weeks (Figure 1), epididymal spermatozoa had significant DNA oxidation at all time points. There is a possibility that the oxidative damage sperm DNA is a consequence of the impact of the treatment on the testis. Treated rats had similar sperm production than controls; their testes did not show long-lasting oxidative damage as the epididymis and were morphologically similar to control testis. Thus, it is less likely that the sperm DNA oxidation is due to problems during spermatogenesis. However, we cannot exclude the possibility that some of the damage observed in the sperm DNA may occur during the formation of spermatozoa in the testis. The oxidative stress generated by t-BHP altered the expression of miRNAs involved in the antioxidant response and spermatogenesis in mouse testis [41]. Although we did not see significant changes in spermatogenesis and the antioxidant response appears to be intact in rat testis, there is a possibility for the disruption of molecular mechanisms driven by miRNAs or epigenetic changes that can be associated with the permanent sperm DNA damage observed in this study. Further studies are required to rule out these possibilities.

While we presented evidence that the antioxidant response of the rat epididymis against oxidative stress is altered and may explain the poor quality of spermatozoa observed in this study, multiple factors could contribute to the persistent DNA oxidation observed in spermatozoa from t-BHP-treated rats. Sperm chromatin is a highly organized structure that di ffers from that of the somatic cells. Protamines replace histones during spermatogenesis, allowing the chromatin to tightly compact [42,43]. During epididymal transit, protamines become thiol oxidized and make disulfide bridges among them, thus, making the sperm chromatin more compacted [43]. Low mature protamine to protamine precursor ratio has been found in infertile patients and is correlated with high DNA damage, suggesting that chromatin compaction is critical for the protection of sperm DNA [43,44]. In our study, the persistent DNA damage observed in rat spermatozoa could be due to alterations of the sperm chromatin structure that interfered with normal sperm chromatin compaction, thus making sperm DNA more susceptible to the oxidative stress seen after 9 weeks of the end of treatment. Permanent oxidative stress in the male reproductive system as the one observed in *Prdx6*−/− male mice leads to changes in the sperm chromatin with increased DNA oxidation and lower protamination (amounts of protamines) and DNA compaction compared to the wild-types controls [16]. Exposure of male rats to the chemotherapeutic agen<sup>t</sup> cyclophosphamide, known to produce ROS as part of the mechanism of action, decreased the level of protamination and subsequently increased DNA damage of rat spermatozoa [45].

Similar long-lasting damages as those found in this study were observed in testicular cancer survivors who underwent chemotherapy with cisplatin and bleomycin, two drugs that generate high levels of ROS in cells exposed to them [46,47]. Cancer patients treated with polychemotherapy, including ROS-generating compounds, have high levels of lipid peroxides in blood, indicating the establishment of oxidative stress due to the treatment [48]. Spermatozoa from testicular cancer survivors displayed high DNA damage and low DNA compaction up to two and one years, respectively, after the end of chemotherapy [49,50]. With this significant clinical relevance, the present study provides insight into the understanding of the long-term e ffect of oxidative stress, a condition often seen in male infertility [2,3].

The molecular mechanism behind the long-term, lasting oxidative stress observed in this study is ye<sup>t</sup> to be determined. A potential candidate is the dysregulation of mitochondria due to high levels of 4-HNE. We hypothesize that the high oxidative stress due to t-BHP promoted significant 4-HNE levels that impaired mitochondrial proteins leading to dysregulation of this organelle. We observed that the inhibition of PRDX6 iPLA2 activity by MJ33 increased the levels of 4-HNE and impaired the sperm mitochondrial membrane potential, leading to the generation of oxidative stress and the oxidation of the DNA in human spermatozoa [15]. 4-HNE is capable of inducing mutations of the mitochondrial DNA and form adducts with mitochondrial proteins that lead to mitochondrial dysfunction [51]. Further studies are needed to elucidate the molecular mechanisms behind the alterations in spermatozoa and epididymis observed in the present study.
