| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Institute for Biogenesis Research,3 University of Hawaii Medical School, Honolulu, Hawaii 96822
Department of Embryology,4 Institute of Zoology, Warsaw University, Warsaw 02-096, Poland
ABSTRACT
Ejaculated mouse sperm retrieved from the uteri are more susceptible to DNA damage during freeze-drying and freezing without cryoprotection than epididymal sperm. This prompted us to speculate that a factor present in the uterus after mating, either male or female derived, was responsible for increased susceptibility of ejaculated sperm to DNA damage during preservation and that the differences between epididymal and ejaculated mouse sperm in response to stress originated from varying nuclease activity. We first exposed epididymal sperm to the uterine content from females mated to vasectomized males (UCSP), to the uterine content from unmated females in estrus (UC), and to the seminal vesicle fluid (SVF) and examined sperm chromosomes after intracytoplasmic sperm injection (ICSI). We found an increased incidence of chromosome breaks and extremely severe DNA breakage after exposure to UCSP and SVF, respectively, but the chromosomes were normal in sperm exposed to UC. Comet assay results verified that DNA damage after exposure to SVF was present in sperm before fertilization. Next, we examined nuclease activity in sperm and their associated components with a plasmid digestion assay. Nuclease activity was detected in isolated epididymal and ejaculated sperm, as well as in epididymal fluid and seminal plasma, and was much more pronounced in all samples originating from ejaculate. The combined results from the present study imply that there are intrinsic differences between the epididymal and ejaculated mouse sperm preparations in their susceptibility to nuclease-dependent DNA damage that originates from their nuclease activity. This nuclease activity was detected both in the sperm-free fraction of preparations and isolated sperm.
assisted reproductive technology, epididymis, gamete biology, in vitro fertilization, sperm
Nonconventional sperm preservation methods, such as freeze-drying [1–3], freezing without cryoprotection [1, 3, 4], and drying [5, 6], are successful in preserving epididymal mouse sperm. Recently, we attempted to apply these methods for ejaculated mouse sperm, retrieved from the uteri after mating. Surprisingly, when ejaculated spermatozoa were freeze-dried or frozen without cryoprotection in the same manner as epididymal sperm and then injected into the oocytes by intracytoplasmic sperm injection (ICSI), the proportion of fetuses at 15 days postconception was much lower when compared to our past results with epididymal sperm [1, 3], and the chromosome analysis of the zygotes revealed severe DNA damage (the present study). This suggested that the protocols developed for preservation of epididymal sperm were not applicable for ejaculated sperm and that the latter were more prone to DNA damage resulting from stress during preservation. We later established a modified protocol that enabled the efficient preservation of ejaculated mouse sperm retrieved from the uteri [7], but the overall efficiency remained lower than with epididymal sperm. In the present study, we explore the basis of the increased susceptibility of ejaculated sperm to DNA damage.
The major difference between isolated epididymal sperm and ejaculated sperm is that the former are collected from the cauda epididymides in the form of a dense sperm mass (sperm plus a minute amount of epididymal fluid), while the latter are released from the uteri as a suspension (ejaculate plus other uterine components). We therefore speculated that some factor or factors present in the uterus after mating (either male or female derived) were responsible for the increased susceptibility of ejaculated sperm to DNA damage during preservation. Also, the variable behavior of epididymal and ejaculated mouse sperm in response to stress could originate from differences in endogenous nuclease activity.
The existence of sperm-specific nucleases in mice was first reported in a study on the internalization of exogenous DNA in sperm-mediated transgenesis [8]. These nucleases were Ca2+-dependent, could be activated by calcium ionophore, and cleaved both exogenous and genomic DNA. Later, it was shown that the presence of EDTA and the absence of Ca2+ and/or Mg2+ from the media used for mouse head isolation could improve chromosome stability [9, 10], further supporting the presence of nucleases in mouse sperm. Endogenous nucleases were detected in hamster, human, and mouse sperm preparations by pulsed-field gel electrophoresis [11, 12]. Our past studies also pointed out the existence of Ca2+- and/or Mg2+-dependent nucleases that were responsible for the induction of chromosome breaks in sperm DNA and that could be inhibited by ion chelators EDTA and EGTA [1, 3, 13–15]. In spite of the relatively high number of reports available on nucleases involved in sperm DNA degradation, they are not well characterized, and not much is known about their source and mode of action.
DNA damage in the male germ line, and in the mature spermatozoa, is associated with defective fertilization, impaired embryonic development, abortion, and childhood disease [16–18]. The danger of fertilization achieved with a spermatozoon carrying DNA abnormalities increases with the use of assisted reproduction technologies (ART). Testing DNA integrity may help select sperm preparations with intact DNA or with the least amount of DNA damage for use in assisted conception. In turn, this may alleviate the financial, social, and emotional problems associated with failed ART attempts. However, sperm DNA testing is not enough. The ultimate safeguard against sperm DNA damage is to understand the basis of this process so that the underlying causative mechanisms can be addressed in a logical manner.
In the present study, we explored the basis of increased sensitivity to DNA damage in ejaculated mouse sperm retrieved from the uteri. We tested which component present in the uteri after mating was responsible for sperm DNA damage by exposing epididymal sperm to the uterine content from females mated to vasectomized males (UCSP), to the uterine content from unmated females in estrus (UC), and to the seminal vesicle fluid (SVF), followed by ICSI and chromosome analysis in the zygote. We also tested for DNA degradation directly in sperm by the comet assay. Finally, we examined nuclease activity in sperm and sperm-free fluid from ejaculate and epididymides by preparing extracts and using them to digest plasmid DNA. We demonstrated that increased sensitivity of ejaculated sperm to DNA damage resulted from enhanced nuclease activity present in the ejaculate.
Mineral oil was purchased from Squibb and Sons (Princeton, NJ); pregnant mares' serum eCG and hCG were obtained from Calbiochem (San Diego, CA). All other chemicals were obtained from Sigma (St. Louis, MO) unless otherwise stated.
All mice were obtained at 6 wk of age from the following vendors: B6D2F1 (C57BL/6 x DBA/2) and C57BL/6 from the National Cancer Institute (Raleigh, NC) and CD-1 mice from Charles River Laboratories (Wilmington, MA). Epididymal sperm for ICSI and the comet assay were obtained from 12- to 16-wk-old C57BL/6 males, and epididymal sperm for plasmid digestion were obtained from B6D2F1 males. Mature oocytes for ICSI were obtained from 8- to 16-wk-old B6D2F1 females. Outbred CD-1 females and males were used for mating to obtain different uterine contents. The mice were fed ad libitum with a standard diet and maintained in a temperature- and light-controlled room (22°C, 14L:10D), in accordance with the guidelines of the Laboratory Animal Services at the University of Hawaii and the guidelines presented in National Research Council's "Guide for Care and Use of Laboratory Animals" published by Institute for Laboratory Animal Research of the National Academy of Science, Bethesda, Maryland, 1996. The protocol for animal handling and treatment procedures was reviewed and approved by the Animal Care and Use Committee at the University of Hawaii.
Gamete Handling and Embryo Culture Solutions and Media
Oocyte collection and subsequent oocyte manipulation, including microinjection, were performed in Hepes-buffered CZB medium (Hepes-CZB [19]). Buffer ETBS, an EGTA Tris-HCl-buffered solution consisting of 50 mM EGTA, 50 mM NaCl, and 10 mM Tris-HCl buffer, pH 8.2–8.5 [1, 20], was used for sperm freezing without cryoprotection and for suspending sperm prior to ICSI and the comet assay. Culture of sperm-injected oocytes and embryos was done in CZB medium [21]. CZB was maintained in an atmosphere of 5% CO2 in air, and Hepes-CZB and ETBS were maintained in air.
Collection of Different Uterine Contents and SVF
A detailed explanation of all abbreviations is presented in Table 1. The uterine contents were obtained from females mated to normal stud males (EJA), females mated to vasectomized males (UCSP), and unmated females in estrus (UC). Sperm donors and vasectomized males for mating were caged individually. Females (CD-1) were examined in the evening on the day preceding mating, and those in estrus were separated. We chose CD-1 females for pairing because they have large uteri that are convenient to manipulate, and it is easy to recognize estrus on the basis of vagina color and swelling. We preferred to select females for mating on the basis of their vaginal appearance [22] rather than use hormones to synchronize the estrous cycle, because when mating did not take place, nonstimulated females could be immediately reused, whereas those with hormonally induced estrus required at least 10 days of shelf rest. Early in the morning (0700–0800 h) on the next day, females were placed with males. The females were examined for the presence of vaginal plugs as an indication of successful mating. The first examination was done 30 min after pairing and at 30-min intervals thereafter. Positive females were sacrificed 1 h after finding the plugs, usually from 0900 to 1000 h. The same timing regime was maintained to obtain uterine content from unmated females. The female reproductive tracts were excised and freed from residual adipose tissue and adhering blood. Each uterine horn was cut at its cervical end, and the uterine contents (0.2–0.4 ml) were released into the well of an Organ Tissue Culture Dish (Falcon, Bedford, MA). For plasmid digestion, uterine contents were released into 1 ml of 0.9% NaCl. For the ICSI and comet assay, uterine contents were released into an empty dish and quickly transferred to a 0.2-ml tube. To obtain SVF, seminal vesicles were carefully dissected from B6D2F1 males to free them from the adjacent coagulating glands. The secretions were collected by gently squeezing the glands against the wall of the 1.5-ml tube. The tube was then subjected to a brief centrifugation to collect the fluid at the bottom. The fluid was kept at room temperature until it was mixed with sperm.
|
Collection and Preparation of Epididymal and Vas Deferens Sperm
For the ICSI or comet assay, caudae epididymides were removed from one male, and the epididymal fluid was squeezed out and placed in 1 ml of ETBS buffer. Sperm suspended in ETBS buffer was mixed 1:1 with UCSP, UC, or SVF (see Table 1 for an explanation of abbreviations), incubated at room temperature for 10 min, and frozen without cryoprotection. For plasmid digestion, different subregions of the epididymis (caput, corpus, and cauda) and vas deferens were dissected, and sperm were released into 1 ml of 0.9% NaCl.
Freezing Without Cryoprotection
Aliquots of 10-µl sperm suspensions in ETBS±UCSP, UC, or SVF were loaded in 0.25-ml straws (Edwards Innovations, Spring Valley, VA). Each straw was sealed with Critoseal (Oxford Labware, St. Louis, MO) and placed in a plastic holder on the surface of the liquid nitrogen for 10 min before immersion. Immediately before ICSI, a straw was removed from the storage container and thawed at room temperature (
25°C) for 5 min before expressing the contents into a Petri dish.
Females were induced to superovulate with injections of 5 IU of eCG and 5 IU of hCG given 48 h apart. Oviducts were removed 14–15 h after the injection of hCG and placed in PBS in a Petri dish. The cumulus-oocyte complexes were released from the oviducts into 0.1% of bovine testicular hyaluronidase (300 USP units/mg) in Hepes-CZB medium to disperse cumulus cells. The cumulus-free oocytes were washed with Hepes-CZB medium and used immediately for ICSI.
Intracytoplasmic Sperm Injection
ICSI was carried out as described recently by Szczygiel and Yanagimachi [23]. Briefly, a small drop of sperm suspension was mixed thoroughly with an equal volume of Hepes-CZB containing 12% (w/v) polyvinyl pyrrolidone (Mr 360 kDa) immediately before ICSI. ICSI was performed with Eppendorf Micromanipulators (Micromanipulator TransferMan, Eppendorf, Germany) with a Piezo-electric actuator (PMM Controller, model PMAS-CT150; Prime Tech, Tsukuba, Japan). A single spermatozoon was drawn, tail first, into the injection pipette and moved back and forth until the head-midpiece junction (the neck) was at the opening of the injection pipette. The head was separated from the midpiece by applying one or more piezo pulses. After discarding the midpiece and tail, the head was redrawn into the pipette and injected immediately into an oocyte. Injections were done in Hepes-CZB within 1 h after oocyte collection and sperm reconstitution. Sperm were randomly chosen for injections. Sperm-injected oocytes were transferred into CZB medium for culture. The oocytes were examined at 6 h after ICSI to assess their survival and activation. The oocytes with two well-developed pronuclei and the distinct second polar body were recorded activated. They were then taken for chromosome analysis.
Fertilized oocytes were transferred after 6 to 8 h of culture into CZB containing vinblastine at 0.006 µg/ml to inhibit syngamy. Between 19 and 21 h after ICSI, oocytes were treated with 1% pronase (1000 tyrosine units/mg; Kaken Pharmaceuticals, Tokyo, Japan) for 5 min at room temperature to soften the zonae pellucidae. Then, the oocytes were treated with hypotonic solution (1:1 mixture of 1% sodium citrate and 30% fetal bovine serum) for 5 min at 37°C or 10 min at 25°C. Chromosomes were spread on clean glass slides by the gradual fixation/air-drying method [24]. The preparations were stained with 2% Giemsa (Merck, Darmstadt, Germany) in PBS (pH 6.8) for 10 min for conventional chromosome analysis. The chromosomes of a spermatozoon were considered normal when an oocyte contained 40 normal metaphase chromosomes. It was not always possible to distinguish between chromosomes of paternal and maternal origin. However, because oocyte chromosomes rarely show structural aberrations at first cleavage metaphase after parthenogenetic activation (M. Ward, unpublished results), any abnormal chromosomes within fertilized oocytes were believed to be of sperm origin.
DNA replication analysis was done as described earlier [25] with modifications. Fertilized oocytes were incubated in CZB with 10 µM 5-bromo-2-deoxyuridine (BrdU) for 30 min. Following incubation, the oocytes were fixed 2.5% paraformaldehyde, 0.5 M NaOH, and Dulbecco PBS, pH 7.3 (Gibco, Grand Island, NY), at room temperature for 15 min. Fixed oocytes were washed in 10% fetal bovine serum (FBS), 0.2% Triton X-100 (TX-100), and Dulbecco PBS, pH 7.3, and blocked in the same solution for 30 min in 37°C. The oocytes were then washed in 2% FBS, 0.1% TX-100, and Dulbecco PBS, pH 7.3; incubated in drops of anti-BrdU antibody conjugated with Alexa Fluor 488 (Molecular Probes, Eugene, OR); diluted 1:19 in 2% FBS, 0.1% TX-100, and Dulbecco PBS, pH 7.3, for 1 h in 37°C; and washed again in 2% FBS, 0.1% TX-100, and Dulbecco PBS, pH 7.3. The oocytes were placed on the microscope slides coated with polylysine (1 mg/ml). The preparations were covered with VectaShield mounting media (Vector Laboratories, Burlingame, CA) and cover glasses and examined by fluorescence.
Chromatin fragmentation in sperm was assessed by the Trevigen Comet Assay kit (Trevigen, Gaithersburg, MD). Two microliters of sperm suspension was collected from sperm samples prepared for ICSI and diluted 1:250 in Ca2+- and Mg2+-free PBS. One to two microliters of sperm suspension (depending on the sperm concentration) was added to 75 µl of low-melting-point agarose (38°C) and pipetted on a slide. Agarose was allowed to solidify at 4°C for 20 min. The slides were immersed in a lysis buffer containing 40 mM dithiothreitol (DTT) for 1 h at room temperature and protected from light. Next, proteinase K was added to the lysis solution to a final concentration of 10 mg/ml, and the additional lysis was performed at 37°C for 2.5 h. After lysis, the slides were washed through three changes of deionized water at 20-min intervals to remove salt and detergent from the microgels. The slides were placed in a horizontal electrophoresis unit, equilibrated for 20 min in TBE buffer (89 mM Tris base, 89 mM boric acid, and 2 mM EDTA, pH 8.0), and electrophoresed (1 V for each centimeter between electrodes) for 20 min. After electrophoresis, the slides were rinsed with water, air dried, and stored protected from light until analysis. To analyze DNA fragmentation, Sybr Green diluted 1:104 in TE buffer (10 mM Tris-Cl and 1 mM EDTA, pH 8.0) was placed onto the dried agarose, and the slide was viewed under the fluorescent microscope (Nikon Eclipse E600; Nikon, Kobe, Japan) at 200x magnification. One hundred DNA tails were photographed and analyzed per slide, and each experiment was repeated three times. The length of each tail was measured from the center of the comet head to the end of the tail by Image J software [26].
Air-Dried Giemsa-Stained Sperm Preparations for Assessment of Sperm Chromatin Remodeling
Sperm chromatin remodeling after ICSI was examined as described in our recent study [27]. Briefly, fertilized oocytes were washed in Dulbecco PBS, pH 7.3, and air dried on microscope slides starting from 0 time (reflecting the approximate time of sperm-oocyte fusion) and at 30-min intervals for up to 4.5 h (10 time points). The fertilization was achieved within 5 min per oocyte group. The 0-min time point was defined as the end of injection of one group of oocytes. The preparations were fixed in absolute ethanol and glacial acetic acid (3:1) for 10 min and stained in 2% Giemsa in buffered saline solution, pH 6.8, for 10 min. The slides were examined with a light microscope having 1000x magnification. More than 10 (range, 11–19) oocytes were examined at each time point in each group. The following sequential remodeling stages were differentiated, as described before [27]: unchanged sperm head, chromatin decondensation, chromatin recondensation, beginning of pronuclei formation, and developed pronuclei. Their frequency at specific time points was noted. In the chromatin decondensation and recondensation groups, both partially and fully decondensed or recondensed sperm, respectively, were included. The developed pronuclei group contained early and fully developed pronuclei.
Preparation of Extracts and Plasmid Digestion
Extracts for plasmid digestion were prepared from ctEPIsp, csEPIsp, caEPIsp, VDsp, EJAsp, caEPIsp*, EJAsp*, caEPI-F1–5, EJA-F1–5, UCSP, and UC (see Table 1 for explanation of abbreviations). In the initial step, the contents of different subregions of epididymis, vas deferens, or uteri were released into Center Well Petri Dish containing 1 ml of 0.9% NaCl and mixed. The obtained initial suspensions were treated as described in detail below. All centrifugations were done with the following conditions: 1020 x g, 10 min, 4°C. The solution used for all washings, dilutions, and extraction was sterile 0.9% NaCl (subsequently called NaCl). The extracting solution was 1% TX-100 in 0.9% NaCl (subsequently called TX-NaCl). All experiments were repeated at least three times.
Experiment 1. In this experiment, the extracts were prepared from caEPIsp and EJAsp. To prepare caEPIsp extract, the initial caEPI suspension was transferred into a 1.5-ml centrifuge tube, mixed by pipetting, and centrifuged. The supernatant was discarded; the sperm pellet was resuspended in 0.5 ml of NaCl, and the sperm concentration was measured. A portion of the suspension containing 4 x 106 sperm was transferred to another tube and centrifuged. The sperm pellet was resuspended in 100 µl of TX-NaCl, vigorously mixed, and centrifuged once more. The supernatant of this last spin was collected as the caEPIsp extract. The extract from EJAsp was prepared in the same way, starting from the initial suspension EJA, obtained after releasing the content of the uterus from a female mated to a normal stud male.
Experiment 2. In this experiment, extracts were prepared from caEPI-F1–5 and EJA-F1–5 and from caEPIsp* and EJAsp*. To prepare extracts from caEPIsp* and caEPI-F1–5, the initial caEPI suspension was transferred into a 1.5-ml centrifuge tube, mixed by pipetting, and centrifuged. A portion of the supernatant (45 µl) from the first spin was collected for preparation of caEPI-F1. The sperm pellet was resuspended in 0.5 ml of NaCl, and the sperm concentration was measured. A portion of the suspension containing 4 x 106 sperm was transferred to another tube, NaCl was added to make up a 1-ml total suspension volume, and the sample was centrifuged. A portion of the supernatant (45 µl) from the second spin was collected for preparation of caEPI-F2. The same steps (pellet resuspension in 1 ml of NaCl, spinning, and collection of 45 µl of the supernatant) were repeated to collect samples for preparation of caEPI-F3–5. The sperm pellet after the fifth spin was resuspended in 100 µl of TX-NaCl, vigorously mixed, and centrifuged once more. The supernatant of this last spin was collected as caEPIsp* extract. To make extracts from caEPI-F1–5, TX-100 was added to each sample to give a final concentration of 1%, and the suspensions were mixed vigorously. Extracts from EJAsp* and EJA-F1–5 were prepared in the same way as from caEPIsp* and caEPI-F1–5, starting from the initial suspension EJA, obtained after releasing the content of the uterus from a female mated to a normal stud male.
Experiment 3. In this experiment, extracts were prepared from EJAsp, EJA-F1, UCSP, and UC. Respective initial suspensions were centrifuged. Ten-microliter samples were collected from each supernatant and diluted 100x with NaCl. This dilution was necessary to avoid the complete degradation that was seen before with EJA-F1. Next, TX-100 was added to a 1% final concentration, and the suspensions were mixed vigorously. These suspensions were used as EJA-F, UCSP, and UC extracts. The sperm pellet obtained after centrifugation of the initial EJA suspension was resuspended in 0.5 ml of NaCl, and the sperm concentration was measured. A sample containing 4 x 106 (or less if 4 x 106 were not available) sperm was collected and centrifuged. The sperm pellet was resuspended in 100 µl of TX-NaCl, vigorously mixed, and centrifuged. Supernatant was collected as EJAsp extract.
Experiment 4. In this experiment, extracts were prepared from ctEPIsp, csEPIsp, caEPIsp, and VDsp. To prepare extracts, respective initial suspensions were transferred into 1.5-ml centrifuge tubes, mixed by pipetting, and centrifuged. Supernatants were discarded; sperm pellets were resuspended in 0.5 ml of NaCl, and the sperm concentration was measured. A portion of each suspension containing 4 x 106 sperm was transferred to another tube and centrifuged. The sperm pellets were resuspended in 100 µl of TX-NaCl, vigorously mixed, and centrifuged once more. Supernatants of this last spin were collected as extracts. If less than 4 x 106 sperm were obtained from the initial suspension, proportionately less TX-NaCl was added to the sperm pellet.
Extracts prepared as described for experiments 1, 2, 3, and 4 were mixed with supercoiled (sc) or linearized (ln) PUC19 plasmids and incubated for 1 h at 37°C or used immediately after mixing. Ten microliters of each extract-plasmid mixture was loaded into a well of 0.7% agarose gel, and the samples were electrophoresed for 20 min at 100 mV. The gels were stained with ethidium bromide and analyzed under UV light. Plasmid alone and extract alone were electrophoresed as controls.
The experiments were designed to assess the extent and origin of chromosome damage resulting from sperm exposure to various uterine content and to test for the differences in nuclease activity between ejaculated and epididymal sperm. Three sets of experiments were performed: 1) chromosome analysis after ICSI with sperm exposed to various components of uterine contents; 2) comet assay on sperm exposed to various components of uterine contents; and 3) plasmid digestion with extracts prepared from epididymal and ejaculated sperm and their associated components. In chromosome analysis experiments, the same sperm sample was used for ICSI in the tested (exposed) and the control (nonexposed) groups, both fresh and frozen. The comet assay was performed on the same sperm samples that were used for ICSI. In the plasmid digestion experiments, the sperm number was adjusted in compared samples. All experiments (ICSI, comet, and plasmid digestion) were repeated at least three times.
Chi-square, Likelihood Ratio, Fisher Exact Probability, and Student t-tests were used for analyzing all responses. Lack of statistical significance was reported when all tests gave P > 0.05. The presence of statistical significance was noted when at least one of the three tests showed P 
0.01 or P 0.05. The computations were done by KyPlot version 2.0 beta 13 software (copyright 1997–2000, Koichi Yoshioka, available online at http://www.woundedmoon.org/win32/kyplot.html).
Protocols Developed for Freeze-Drying and Freezing Without Cryoprotection of Epididymal Sperm Are Less Successful with Ejaculated Sperm
Nonconventional sperm preservation methods, such as freeze-drying or freezing without cryoprotection, are successful in preservation of epididymal mouse sperm [1–4]. When ejaculated spermatozoa retrieved from the uteri were preserved in exactly the same way and then injected into the oocytes by ICSI, the proportion of fetuses at 15 days postconception from two-cell embryos transferred remained below 10% (8% [12 of 158]; 2% [4 of 175]; in freezing without cryoprotection and freeze-drying, respectively). This was much lower when compared with the results obtained earlier with epididymal sperm of the same strain (up to 70%) [1]. When we performed the chromosome analysis of the zygotes produced with sperm from the ejaculated sperm samples that yielded a low proportion of fetuses, we found severe DNA damage in the paternal complements. Only 24% (20 of 85) and 43% (26 of 60) of ejaculated sperm preserved by freeze-drying and freezing without cryoprotection, respectively, were chromosomally normal. These results demonstrated that the protocols for freeze-drying and unprotected freezing developed for epididymal sperm [1] were not successful with ejaculated sperm and that ejaculated sperm were more susceptible to DNA damage during preservation.
Sperm Exposure to UCSP Results in Increased Incidence of Paternal Chromosome Breaks in the Zygote
When epididymal sperm were exposed to UCSP (see Table 1 for explanation of abbreviations) and then injected into the oocytes by ICSI, 50% of the resulting zygotes were chromosomally normal (Table 2). This was significantly less than in the control (83% vs. 57%, P < 0.001). When sperm exposed to UCSP were frozen without cryoprotection, only 16% of sperm chromosome complements were normal. There were no differences between nonexposed fresh and frozen sperm in the proportion of normal chromosomes after ICSI, indicating that freezing per se does not impair chromosome integrity (83% vs. 85%, P > 0.05). The severity of chromosome damage, reflected by the number of chromosome breaks per oocyte examined (aberrations per spermatozoon ratio), was slight when fresh sperm exposed to UCSP were injected and significant when the sperm were exposed to UCSP and subsequently frozen without cryoprotection (Table 2 and Fig. 1, C and D). Sperm exposure to UCSP also resulted in an increased incidence of oocytes arrested at the pronuclei stage (14% vs. 3%, P < 0.01), and more arrested oocytes were noted after freezing when compared with fresh sperm (25% vs. 14%, P < 0.05; Table 2).
|
|
Sperm Exposure to SVF Followed by Freezing Results in a High Incidence of Pronuclei Stage Oocyte Arrest and Extremely Severe Chromosome Breaks in the Zygote
To test which component of UCSP caused paternal chromosome breaks, we exposed epididymal sperm to either UC or SVF. After exposure, the sperm were frozen without cryoprotection to enhance any possible effect. Sperm exposed to UC were chromosomally normal (95%) and did not cause oocyte arrest at the pronuclei stage (Table 3 and Fig. 1F). On the contrary, more than 60% of sperm exposed to SVF resulted in oocyte arrest. All oocytes except for one (98% total) that overcame the arrest contained broken paternal chromosomes. The damage was extremely severe, with the aberrations per spermatozoon ratio reaching 8.4 (Table 3 and Fig. 1E). This was the most severe DNA damage that we have ever seen when conducting studies that examine sperm chromosome integrity after various treatments [1, 3, 13, 14, 28, 29].
|
The Comet Assay Demonstrates a High Incidence of DNA Damage in Sperm Exposed to SVF and Frozen Without Cryoprotection
A significant proportion of the oocytes injected with sperm exposed to SVF and subsequently frozen were arrested at the pronuclei stage, and the accumulation of sperm chromosome complements available for analysis was slow. Therefore, we performed comet assays on sperm prior to injection. Comet assays were done for sperm exposed to SVF and for sperm exposed to UC and subsequently frozen, as well as for their appropriate controls (nonexposed sperm were manipulated in the same way). The same sperm samples were also used for ICSI. Four types of comet tail expressing different levels of DNA damage were differentiated (Fig. 2). The results are shown as the analysis of the frequency of comets with different tail lengths and with different tail types (Fig. 3). Both tail length and tail type reflect the severity of DNA damage. The distribution pattern of different tail lengths is shown in Figure 3. The difference in mean comet tail length between sperm exposed to SVF and control was highly significant (mean ± SD; 278.53 ± 37.79 vs. 159.64 ± 57.48; P < 0.0001). Sperm exposed to SVF had mostly tail types 3 (66%) and 4 (11%), while in controls, types 1 (78%) and 2 (10%) dominated. No differences in mean tail length were noted between sperm exposed to UC and control (mean ± SD; 167.90 ± 59.19 vs. 159.47 ± 53.45; P > 0.05). In both groups, the prevailing tail length was 100–200 µm, and the prevailing tail type was 1 (Fig. 3). When sperm exposed to SVF were compared with sperm exposed to UC, the difference was also highly significant (P < 0.0001).
|
|
Oocytes Injected with Sperm Exposed to SVF and Frozen Without Cryoprotection Maintain DNA Replication
We assessed DNA replication to test whether pronuclei stage oocyte arrest and high incidence of DNA damage after ICSI with sperm exposed to SVF were associated with the disturbances in DNA synthesis. Two experiments were performed. The oocytes injected with sperm were exposed to SVF, and the controls were fixed and subjected to BrdU staining 24 h after injection. Among the oocytes injected with sperm exposed to SVF, 12% (3 of 25) remained at metaphase, 64% (16 of 25) arrested at interphase of the first embryo division, and 24% (6 of 25) reached the two-cell stage. In the control group, 88% (7 of 8) of the oocytes reached the two-cell stage, and one oocyte (12% [1 of 8]) remained at interphase of the first embryo division. The presence of BrdU staining was noted in all tested and control oocytes, but the intensity of staining seemed weaker in arrested oocytes when compared with those that cleaved. Moreover, in 4 of 6 two-cell embryos obtained with sperm exposed to SVF, few small nuclei in addition to the primary cell nucleus were observed in blastomeres, perhaps because of damaged DNA, which may have prevented the correct separation of chromosomes.
Sperm Chromatin Remodeling Is Not Affected after Sperm Exposure to SVF and Subsequent Freezing
We have recently shown that paternal chromatin remodeling differs when sperm are subjected to various manipulations [27]. In the present study, we evaluated whether sperm exposure to seminal vesicle and subsequent freezing affected paternal chromatin remodeling. Previously reported differences in paternal chromatin remodeling were mostly in its synchrony. Each remodeling stage was defined as synchronous when 
80% of the oocytes with the chromatin in this stage were accumulated in at least one time point. There were no differences in the number of synchronous remodeling stages between sperm exposed to SVF and their nonexposed controls; both had three stages of five that were synchronous (Fig. 4). Unchanged sperm head, decondensation, and formed pronuclei stages were synchronous in the SVF group (Fig. 4A), while unchanged sperm head, recondensation, and formed pronuclei were synchronized in the control (Fig. 4B). Thus, the differences between examined groups were in the dynamics of decondensation and recondensation. These differences, however, were not statistically significant (P = 0.13 and P = 0.09, for decondensation at 1 h and recondensation at 2 h, respectively). Overall, the pattern of chromatin remodeling in sperm exposed to SVF was similar to that of control sperm.
|
Extracts Prepared from Ejaculated and Epididymal Sperm and Associated Components Digest Plasmid DNA Differently
To test for nuclease activity epididymal and ejaculated sperm, we prepared sperm extracts from the respective sperm samples and associated components and tested their ability to degrade plasmid DNA. Plasmid degradation induced by extracts prepared from ejaculated sperm (EJAsp) was much stronger when compared with cauda epididymal sperm (caEPIsp). Cauda epididymal sperm resulted in weak degradation of supercoiled plasmid after 1 h of incubation (Fig. 5, lines 3 and 5) and no degradation of the linear form (Fig. 5, lines 4 and 6). Ejaculated sperm completely degraded both supercoiled and linear plasmids, even without prior incubation (Fig. 5, lines 7–10).
|
To test for nuclease activity in the components associated with cauda epididymal sperm (caEPI-F) and ejaculated sperm (EJA-F), plasmids were incubated with extracts (F1–F5) prepared from supernatants from five consecutive washes, to which the diluted contents of the cauda epididymides or ejaculate obtained from the uteri were subjected. When the extracts prepared from cauda epididymal sperm were compared, the highest degradation of supercoiled plasmid was noted in F1 (Fig. 6A, line 7). The degradation decreased in F2 and F3 (Fig. 6A, lines 9 and 11), was barely visible in F4 (Fig. 6A, line 13), and was not present in F5 (Fig. 6A, line 15). The linear form of plasmid, more resistant to digestion, was affected only in F1 (Fig. 6A, line 8). Much stronger degradation was obtained with extracts derived from supernatants obtained after washing ejaculated sperm obtained from the uteri (Fig. 6B). Both forms of plasmid were completely digested in F1 and F2 (Fig. 6B, lines 7–10), and supercoiled plasmid was also completely digested in F3 (Fig. 6B, line 11). In F4, the degradation of supercoiled plasmid was partial but seemed stronger than in EPI-F4 (Fig. 6, A and B, line 13). In F5, no degradation was noted (Fig. 6B, lines 15 and 16). The results also confirmed the presence of nuclease activity in both cauda epididymal and ejaculated sperm. The extract prepared from ejaculated sperm degraded plasmid much more prominently (Fig. 6B, lines 3–6) than the extract prepared from cauda epididymal sperm (Fig. 6A, lines 3–6). Importantly, this degrading activity was present in sperm that were subjected to five centrifugations (caEPIsp* and EJAsp*).
|
All ejaculates in the present study were obtained from the uteri after mating. Thus, it was possible that the strong nuclease activity observed in sperm-free components of ejaculate (Fig. 6B, F1–F3) originated from the uterus rather than the ejaculate per se. Also, because the nuclease activity was noted in extracts prepared from both sperm-only and sperm-free fractions of the ejaculate, it was not clear whether this activity was sperm-specific. To clarify this, we prepared extracts from the UCSP and the UC and compared their degrading activity with that of ejaculated sperm (EJAsp) and sperm-free supernatant from the first ejaculate wash (EJA-F1). EJA, UCSP, and UC were obtained from a single female each and were processed in exactly the same way. Similar degradation patterns were noted for EJA-F1 (Fig. 7, lines 5–8) and UCSP (Fig. 7, lines 9–12), while no degradation was observed when plasmids were exposed to UC (Fig. 7, lines 13–16). The plasmid digestion after incubation with EJAsp shown in Figure 7 (lines 1–4) was weaker than that shown in Figure 5 (lines 7–10) and in Figure 6B (lines 3–6). This was because of the lower sperm number (0.9 x 106) that was used for making the extract in the experiments from which a photograph of gel is shown.
|
To test for the differences in nuclease activity in different regions of the male reproductive tract, we prepared extracts from sperm obtained from the caput (ctEPIsp), corpus (csEPIsp), and cauda (caEPIsp) epididymes and from the vas deferens (VDsp) (Fig. 8). Prior to performing this experiment, we tested how many spermatozoa could be obtained from different subregions of the male reproductive tract. The following sperm numbers (mean ± SD, n = 10 sexually mature B6D2F1 males) were obtained: 2.1 x 106 ± 0.9433, 2.1 x 106 ± 0.8306, and 26.2 x 106 ± 11.3207 from the caput, corpus, and cauda epididymides, respectively; 13.5 x 106 ± 6.0207 from the vas deferens; and 3.9 x 106 ± 1.8681 from a single ejaculate retrieved from the uterus. Usually, in our hands, the number of sperm obtained from cauda epididymides is lower (10 x 106 to 20 x 106) [30]. The higher number obtained in the present study was due to the more advanced age of the examined males.
|
The extract prepared from sperm obtained from caput epididymides resulted in a slight degradation of the supercoiled plasmid when freshly mixed and stronger after 1 h of incubation (Fig. 8, lines 1 and 3). The extracts prepared from corpus and cauda epididymal sperm resulted in a weak degradation of the supercoiled plasmid only after 1 h of incubation (Fig. 8, lines 7 and 11). The degradation was slightly more pronounced with extract obtained from cauda sperm, evidenced as a weak smear (Fig. 8, line 11). The extract prepared from vas deferens sperm yielded stronger plasmid degradation when compared with all segments of the epididymis. The supercoiled plasmid was slightly degraded when freshly mixed (Fig. 8, line 13) and completely digested after 1 h of incubation (Fig. 8, line 15). Nuclease activity in sperm prepared from the vas deferens was nevertheless weaker than that of ejaculated sperm (compare Fig. 8, lines 13–16, and Fig. 5, lines 7–10). None of the extracts prepared from epididymal or vas deferens sperm was capable of digesting the linear form of plasmid (Fig. 8, lines 2, 4, 6, 8, 10, 12, 14, and 16).
In the present study, we demonstrated that a factor present in the uteri after mating, but of male origin, evokes a response in epididymal sperm that increases their susceptibility to DNA damage. We also provided evidence for the existence of endogenous nucleases in isolated ejaculated and epididymal sperm and documented the variance in nuclease activity between sperm from different segments of the male reproductive tract.
To clarify whether the factor responsible for the increase in DNA damage observed in the sperm exposed to the uterine content was of male or female origin, we exposed epididymal sperm to the content of uteri from an unmated female or to seminal vesicle fluid (SVF) and froze them without cryoprotection to enhance the possible effect. We tested SVF because it is known to contribute to the bulk of seminal plasma in rodents [31]. By two different approaches, paternal chromosome analysis in the zygote and the comet assay, we demonstrated significant DNA degradation after sperm exposure to SVF and no DNA damage after exposure to female uterine secretions. Thus, the factor responsible for DNA damage in ejaculated sperm originated from male seminal plasma components—more specifically, from the seminal vesicle.
The DNA damage after exposure to SVF and freezing was extremely severe, as judged by the aberrations per sperm ratio. We also observed a high incidence of pronuclei stage arrest. Previously, it was shown that DNA synthesis can be initiated with severely degraded DNA when chromatin is degraded within its extended loops [32], but complete DNA synthesis inhibition takes place when degradation occurs at the bases of the loop domain attachment regions, which are Topoisomerase II Beta (TOP2B) sites of action [33]. Here, DNA replication took place in all oocytes injected with sperm exposed to SVF. Oocyte arrest at the pronuclei stage can be caused by one of the two checkpoints detecting DNA damage: Intra S Phase and G2/M [34]. Because we observed weaker BrdU staining in single-cell arrested embryos than in the oocytes that cleaved, it is more likely that the former checkpoint was activated.
Sperm exposure to SVF did not significantly alter chromatin remodeling after fertilization. This was rather remarkable considering that sperm DNA was severely damaged. The only difference between sperm exposed to SVF and their chromosomally normal controls was the more asynchronous recondensation observed in the former. The same phenomenon was previously noted in sperm with damaged DNA resulting from treatment with TX-100+DTT [14, 27]. Prolonged recondensation might reflect some problems with chromatin packaging and achieving proper nucleosome structure [35, 36] by damaged sperm DNA.
The mechanism by which SVF induced DNA damage in sperm is not clear. It is known that seminal plasma from a variety of species contains nucleases [31]. It was also reported that SVF contained nucleases that were Ca2+- and Mg2+-dependent and could be inhibited by EDTA. In our study, sperm exposed to SVF were suspended in ETBS buffer (devoid of divalent ions and containing ion chelators), and yet the degradation was still noted. Thus, nucleases different from those described before could be involved. It is also possible that SVF contains some factor that is able to activate sperm-specific nucleases or that the rich ion composition of seminal vesicle secretions provided a substrate that interfered with the EGTA present in the sperm handling medium and prevented the inhibition of sperm-specific nucleases by this chelating agent.
One of the controversies relating to the nucleases responsible for degrading sperm DNA is whether they originate from the sperm itself (i.e., are sperm-specific). In past studies [1, 3, 13–15], we tested for the presence of sperm nucleases indirectly, by subjecting sperm to various DNA-damaging treatments and examining paternal chromosomes in the zygotes after ICSI. Thus, we could not exclude the possibility that DNA degradation took place in the oocytes after fertilization and that oocyte nucleases were involved. In the present study, we confirmed the presence of DNA damage in sperm prior to fertilization. DNA degradation could be seen both during chromosome analysis of the zygote and during comet analysis of sperm prior to injection.
Even if the degradation of sperm DNA took place before fertilization, it was still not clear whether the nucleases responsible for digestion originated from the sperm per se or from the surrounding fluid. To define the source of DNA-damaging nucleases, we tested the nuclease activity in epididymal and ejaculated sperm and their associated fluids. We demonstrated that nuclease activity was present both in sperm isolated by repeated centrifugation and in the surrounding fluids. The same was observed with other methods of sperm separation, such as Sephadex column separation or Percoll gradient centrifugation (our unpublished results). Regardless of the examined fraction, the nuclease activity was much stronger in extracts prepared from the ejaculate when compared with the epididymal sample. The presence of nuclease activity in the sperm-free fraction of ejaculate was further confirmed by the analysis of plasmid degradation with extracts prepared from the uterine content from females mated to normal males from which sperm were removed by centrifugation and from the UCSP that did not contain sperm to start with. The lack of nuclease activity in the uterine content from unmated females indicated that the nucleases present in ejaculate retrieved from the uteri originated from male components. However, it cannot be excluded that the process of mating induced the secretion of uterine fluids containing nucleases that would not otherwise be found in the female tract.
The activity of nucleases in the surrounding fluid decreased with consecutive washes but was undoubtedly present in the supernatants from at least the first three. One explanation would be that fluid-specific nucleases were attached to sperm and required repeated washing to shed. Another explanation could be that these were sperm-specific internal nucleases that leaked out in the course of subsequent centrifugations. Mouse sperm are very sensitive to several types of mechanical stress [37–40], and even simple pipetting or mild centrifugation can result in membrane damage and subsequent chromosome breaks [13, 14]. We cannot at this point clarify if the nucleases present in the first three washes were sperm-specific, fluid-specific, or both. However, because the nuclease activity was observed in the sperm subjected to five washes while no activity was noted in the supernatant from the fifth wash (and almost none in the fourth wash), it is very likely that some nucleases were retained within the sperm. On the basis of the results obtained in our plasmid digestion experiments, we hypothesize that the nucleases observed in sperm samples can be grouped into three categories: 1) nucleases not specific to sperm and originating from epididymal or ejaculate components; 2) nucleases specific to sperm that are released only after strong detergent extraction; and 3) nucleases that are either sperm-specific but easily released with centrifugation or fluid-specific but with a high affinity to sperm membrane and requiring centrifugation to release. It is hard to comment which group has the highest activity, but it is clear from all digestion experiments that the nucleases originating from ejaculate (regardless of category) are much more effective than those from epididymides.
The finding that there are different types of nucleases involved in sperm DNA damage is in agreement with previous reports. In our past studies, the nucleases that caused DNA degradation in epididymal sperm exposed to detergent and DTT [14] or to exogenous DNA [13] seemed to require Ca2+ and/or Mg2+ because their activity was inhibited by EDTA/EGTA. However, this inhibition was not complete, suggesting the involvement of some other nucleases with different requirements. In this study, ETBS buffer (containing EGTA), which was found to be effective in preventing DNA damage in freeze-dried and frozen epididymal sperm [1, 3], did not maintain DNA integrity in epididymal sperm after exposure to uterine or seminal vesicle contents. This, once again, suggested that different nucleases were involved. Recently, TOP2B was found in sperm and implied to be involved in sperm DNA degradation. It was proposed that it worked in conjunction with some Ca2+- and Mg2+-dependent external nuclease originating from epididymal fluid, required DTT for activation, and was reversible by EDTA [41]. This study was later expanded, and the results from in vivo experiments have shown that when TOP2B was induced to cleave sperm DNA before fertilization, the paternal DNA was further degraded by some putative nuclease in the pronuclei [33]. This nuclease could be brought there by sperm but could also originate from the oocyte. Thus, at least three different possible nucleases involved in sperm DNA degradation emerge from these studies: TOP2B, a nuclease from epididymal fluid, and a nuclease acting in the oocyte, all with specific requirements for activity. In our plasmid digestion experiments, epididymal and ejaculate samples were released directly into saline, and no further addition of ions took place. The nucleases were active without Ca2+ and Mg2+ and did not require any other activating factors.
The presence of nucleases in sperm that are capable of degrading its DNA may seem in contradiction to what is conventionally thought of as the major role of sperm in fertilization—to deliver paternal DNA in its pristine and untouched form. We have previously suggested that nuclease activity in sperm could be part of a mechanism that sperm use to prevent the transmission of potentially damaged DNA to the embryo when they encounter a stressful environment [15]. The results of the present study support this hypothesis and demonstrate that nuclease activity might also be present under normal, physiological conditions. Nuclease activity was detected in sperm from different subregions of the epididymis and vas deferens, and it seems that, overall, it increased during sperm maturation. It was weakest in the corpus, slightly stronger in the cauda, moderate in the vas deferens, and strongest in ejaculated sperm. The exception to this pattern was the enhanced nuclease activity observed in the sperm obtained from the caput epididymis. However, it is possible that this activity was a remnant of nucleases acting in the testis and regulating cell death during spermatogenesis [42]. The highest nuclease activity observed in ejaculated sperm suggests positive selection toward maintaining this function.
In the present study, we demonstrated that mouse ejaculates contained high nuclease activity both in the sperm and in the seminal plasma. This activity was much higher than that observed in epididymal sperm and epididymal fluid and might account for the increased susceptibility of ejaculated sperm to DNA damage during stress. The nucleases did not require Ca2+ and/or Mg2+ or any other form of activation, and their activity was also observed before fertilization. Thus, it is likely that these nucleases were different from those described in the past [8, 13, 14, 33, 41]. The recent progress in the sperm nuclease field not only shows interesting new information but also clearly points out how much is still unknown. It seems that there are many sperm-associated nucleases and that they exhibit different modes of action. It is hoped that future studies will help define them more thoroughly.
ACKNOWLEDGMENTS
The authors thank Raymond Machi for help with plasmid digestion experiments.
FOOTNOTES
1This material is based on work supported by NIH HD048446 and HD048845 grants to M.A.W. 
Correspondence: 2Monika A. Ward, Institute for Biogenesis Research, John A. Burns School of Medicine, University of Hawaii, 1960 East-West Rd., Honolulu, HI 96822. FAX: 808 956 7316; e-mail: mward{at}hawaii.edu
Received: 20 April 2007.
First decision: 13 May 2007.
Accepted: 22 June 2007.
REFERENCES
This article has been cited by other articles:
|
|
 |
|
  W.-M. Cheng, L. An, Z.-H. Wu, Y.-B. Zhu, J.-H. Liu, H.-M. Gao, X.-H. Li, S.-J. Zheng, D.-B. Chen, and J.-H. Tian Effects of disulfide bond reducing agents on sperm chromatin structural integrity and developmental competence of in vitro matured oocytes after intracytoplasmic sperm injection in pigs Reproduction, April 1, 2009; 137(4): 633 - 643. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Ehmcke and S. Schlatt Animal models for fertility preservation in the male Reproduction, December 1, 2008; 136(6): 717 - 723. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Yamauchi, B. Doe, A. Ajduk, and M. A Ward Genomic DNA Damage in Mouse Transgenesis Biol Reprod, November 1, 2007; 77(5): 803 - 812. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
