The female reproductive tract holds the key to mammalian sperm selection. After mating there is a preselection of spermatozoa based on their motility, and those that successfully overcome this selection become sequestered within the oviductal sperm reservoir until ovulation takes place, then spermatozoa are untied and proceed towards the oocytes. When mating takes place after ovulation, spermatozoa only may undergo some selection processes during the transport through the female reproductive tract and/or during the zona pellucida (ZP) binding/penetration, since there is no sperm attachment to the oviduct. The aim of this work was to investigate in a post-ovulatory mating mice model (mating 6-8 hours after ovulation), the selection mechanisms that operate in nature and to find out if these mechanisms can discriminate the quality of spermatozoal DNA regardless the source of damage; a heat stress process, or a gamma radiation process. Our results indicate that in postovulatory mating there is a preliminary general selection mechanism of spermatozoa with undamaged DNA during the transport through the female reproductive tract and in the ZP binding, but the ability of the ZP to recognize fragmented-DNA spermatozoa is achieved during sperm-ZP penetration, and it depends on the source and type of damage, because only DNA-damaged sperm cells by heat stress but not by radiation were selected. The preliminary selection could be a consequence of selection based on motility since DNA fragmentation and sperm motility have been related . Here, we have demonstrated that higher motile sperm cells that reach the oviduct have lower DNA fragmentation that sperm with lower motility present in the uterus. However, sperm with some level of DNA damage is able to reach the oviduct and to attach to the ZP. Based on these findings the sperm-ZP selection during penetration should be a specific mechanism, because it is only active when sperm-DNA has been fragmented by effect of heat stress and it is not when the damage has been produced by radiation. It has been suggested that stable chromatin structure provides sperm with a rigidity that enables them to penetrate the zona pellucida .
The damage induced in the testes, and the nature and degree of sperm DNA-damage produced by scrotal heat stress or by γ-radiation seems different and may influence the sperm fertilizing capacity. Heat stress decreased testis weight, sperm counts retrieved from vas deferens and motility as described by other authors . In case of radiation, a decrease of testis weight was also observed, a situation that has been reported by other authors ; however, the decrease in number of spermatozoa and sperm motility was less pronounced than with the heat stress. In agreement with previous reports [16, 31, 32], scrotal heat stress produced changes in testicular architecture and DNA-damage of sperm. Histology of testes indicated similar reduction of spermatogenesis in both treated groups; however, in the radiated group it is observed hyperplasia of Leydig cells that might be a consequence of a supraphysiologic hormonal stimulation of the testes originated by the radiation treatment of the male mouse. At higher doses of radiation, a significant loss of primary spermatocytes has been reported . High-temperature in testis is associated with an increase of oxygen free-radical species, which reacts quickly with substrates as lipids, causing a lipid peroxidation that affects spermatozoa membranes . Next to these effects, mitochondrial membrane can be disrupted in spermatozoa and germ cells [35, 36] activating programmed cell death pathways. Gamma radiation tends to cause damage on DNA strands directly (ds or ss breaks) , and more specifically, mutations and alterations of DNA backbone (crosslinking among base pairs, presence of apurinic or apyrimidinic sites) . Alterations in DNA backbone could alter level of chromatin condensation and nuclear volume , parameters related with susceptibility to damage by radiation (Hawkins, 2005). The assay that we have used for quantifying sperm DNA-fragmentation can not determine this second type of mutations produced by the radiation.
Motility is the main factor that allows the sperm to reach the oviduct through the cervix and the utero tubal junction  but it seems that motility is not the unique factor of the sperm selection mechanisms; chemical barriers (e.g., low pH and viscous mucus) and leuckocytic/phagocytotic responses within the female, determine that only a small proportion of the sperm cells deposed in the female tract ever have the opportunity to encounter an oocyte. During some phases of sperm transport through the female reproductive tract, sperms are subjected to physical stresses, are exposed to factors with cell signaling capabilities, and may sustain oxidative damage to their plasma membrane lipids. The initial stages of sperm transport are also mediated by the female tract, because sperm transport is regulated by a combination of intrinsic sperm motility and peristaltic movements of the female reproductive tract. We have found that sperm DNA fragmentation level in spermatozoa recovered from uterus increased compared to sperm DNA fragmentation observed in vas deferens spermatozoa, both in control and in gamma radiated groups. In heat-stressed group, sperm DNA fragmentation in the uterus was also high, but not higher than the value detected for vas deferens spermatozoa. These observations indicate that some damage in sperm DNA is taking place in the uterus. Two mechanisms could be responsible of this effect. On one hand, immune cells present in the uterine mucosa [41–43] such as neutrophils could extrude their nuclear DNA next to associated proteins to form neutrophil extracellular traps (NETs)  that retain spermatozoa with disturbed motility or plasma membrane. During NETs disaggregation, DNAses present in the seminal plasma [45, 46] can act removing connections between neutrophils, attached sperm and proteins, allowing an efficient cleaning of the reproductive tract. The second mechanism could be the presence of nucleases from seminal fluid affecting spermatozoa located in the uterus [46, 47].
Our results agree with previous reports indicating that the passage thought the utero-tubal junction in vivo, is one of the most important selective points that are involved in a plausible selection mechanism in which motility plays a main role . In addition, we have found that the transit form the uterus to the oviduct is critical to reduce sperm with fragmented-DNA. Spermatozoa from the three groups analyzed arriving at the oviduct showed drastic reduction in DNA-damage. This would indicate that there is a selection mechanism in the female reproductive tract based on sperm motility but that indirectly, ensures that spermatozoa with the most intact DNA arrive to the oviduct probably a consequence of a relationship between motility and DNA-fragmentation. We have analyzed the DNA-fragmentation by COMET of high motile sperm selected in vitro, taking the faster population of spermatozoa that are able to cross thought drops of media and we have found a correlation between high motility and low level of DNA-fragmentation (data non published). Other authors have also suggested this relation between motility and DNA-fragmentation by ex vivo or in vitro approaches [7, 48]. However, we have found that in vivo, fertilization with radiated sperm produced a higher number of resorptions, suggesting that in radiated sperm, several DNA-damages that are carried by motile sperm are selected neither in the female tract nor in the ZP penetration. Collectively, these results suggest that there is a relationship between motility and the sperm with damage on DNA strands (ds or ss breaks), but not with mutations and/or alteration of DNA backbone (crosslinking among base pairs, presence of apurinic or apyrimidinic sites).
To analyze if the ZP binding and/or penetration plays a role in the selection of sperm with fragmented-DNA, we have examined the DNA quality of sperm cells attached to the ZP and the preimplantation development of oocytes fertilized in vitro or by ICSI. Sperm cells from treated mice that are attached to the ZP exhibits lower levels of DNA fragmentation that the non-attached sperm, indicating that either ZP performs a positive selection towards unfragmented DNA sperm or that motile sperm (with lower level of fragmented-DNA) are more capable to attach to the ZP. This agrees with previous findings demonstrating that a high number of spermatozoa with normal chromatin were attached to ZP . Samples taken from the fertilization drop showed that there is an increase in sperm DNA fragmentation levels compared to the levels of fragmentation observed in the vas deferens of the control and radiated groups (data not shown). This fact could be explained by the presence of endonucleases delivered from dead sperm cells during IVF [49–51]. Some reports have suggested that membrane-altered spermatozoa liberate some factors to the media, including endonucleases, that could fragment DNA of spermatozoa with unaltered membranes . Also, an increase in ROS production during IVF could take place, escalating damage on alive spermatozoa . In addition, IVF experiments showed that sperm form heat stressed mice that are able to fertilize are also able to produce blastocysts, but when the spermatozoa attached to the ZP are delivered by ICSI, blastocysts development is reduced, indicating that the ability of the ZP to select fragmented-DNA spermatozoa is achieved during sperm-ZP penetration. In the case of sperm from radiated mice, blastocyst development is reduced in comparison to control both after IVF or ICSI, indicating that the ZP is unable to select sperm with DNA damage induced by radiation. These findings agree with our in vivo results since many of the blastocysts produced, apparently normal, could implant, but failed to carry on a normal foetal development and were reabsorbed. On the other hand, in vivo results with heat stressed mice indicate that resulting blastocysts after mating are as competent as control blastocysts to develop into live foetuses. Our results of IVF from heat stress males differ from other study where a block of embryonic development was detected using C57BL/6 inbred mice . The differences may be due to the strain of mice, because we have used CD1 outbreed mice, and it is well established that the genetic background of outbreed and inbred strains of mice influence sperm-assessment parameters, resistance and susceptibility to DNA fragmentation, in vitro fertilization rate, and in vitro embryo development rate; moreover, sperm from the C57BL/6 backgrounds is particularly sensitive . For this reason, results from some specific inbred strains should be considered with precaution.