Hydroxyurea affects in vitro porcine oocyte maturation through increased apoptosis and oxidative stress

Abstract Hydroxyurea (HU) is an FDA-approved drug used to treat a variety of diseases, especially malignancies, but is harmful to fertility. We used porcine oocytes as an experimental model to study the effect of HU during oocyte maturation. Exposure of cumulus–oocyte complexes (COCs) to 20 µM (P<0.01) and 50 µM (P<0.001) HU reduced oocyte maturation. Exposure to 20 µM HU induced approximately 1.5- and 2-fold increases in Caspase-3 (P<0.001) and P53 (P<0.01) gene expression levels in cumulus cells, respectively, increased Caspase-3 (P<0.01) and P53 (P<0.001) protein expression levels in metaphase II (MII) oocytes and increased the percentage of apoptotic cumulus cells (P<0.001). In addition, HU decreased the mitochondrial membrane potential (Δφm) (P<0.01 and P<0.001) and glutathione (GSH) levels (P<0.01 and P<0.001) of both cumulus cells and MII oocytes, while increasing their reactive oxygen species (ROS) levels (P<0.001). Following parthenogenetic activation of embryos derived from MII oocytes, exposure to 20 µM HU significantly reduced total blastocyst cell numbers (P<0.001) and increased apoptosis of blastocyst cells (P<0.001). Moreover, HU exposure reduced the rate of development of two-celled, four- to eight-celled, blastocyst, and hatching stages after parthenogenetic activation (P<0.05). Our findings indicate that exposure to 20 µM HU caused significant oxidative stress and apoptosis of MII oocytes during maturation, which affected their developmental ability. These results provide valuable information for safety assessments of HU.


Introduction
Hydroxyurea (HU), a United States Food and Drug Administration-approved drug [1], is an inhibitor of ribonucleotide reductase commonly used to treat myeloproliferative disorders and sickle cell anemia [2]. In addition, HU is used as an anti-tumor drug to treat various malignancies [3]. However, HU has several adverse effects and should be used with caution in pregnant women and children. HU can cause abnormal embryonic development in mice, rats, and New Zealand white rabbits [4][5][6][7][8][9]. Moreover, studies have shown that use of HU in pregnant women or babies can cause harmful effects [10,11]. In addition, HU can be extremely toxic to preimplantation embryos because it impacts blastocyst formation and development, compromises folliculogenesis, and reduces ovulation [12]. HU inactivates ribonucleotide reductase and inhibits DNA synthesis in proliferating cells, and can increase apoptosis and induce cell cycle changes [11,13,14]. Accordingly, HU exposure induced apoptosis of fetal tissue cells, which resulted in abnormal tissue development in offspring [15]. HU can increase the production of reactive oxygen species (ROS) [8,16]. The carbamoyl nitroso group is an intermediate of HU that can participate in electron transfer, ROS formation, and oxidative stress [17]. As HU compromises folliculogenesis and can elicit apoptosis and oxidative stress, we hypothesized that HU causes apoptosis and oxidative stress during oocyte maturation. Most research on HU has focused on effects during pre-implantation and post-implantation embryo development, while few reports describe the effects of HU during oocyte maturation. To address this need, the current study investigated the effects of HU on apoptosis and oxidative stress during maturation of porcine oocytes.

Reagents
All chemicals and reagents, except those specifically noted, were purchased from Sigma-Aldrich (St. Louis, MO, U.S.A.).
IVM media supplemented with different concentrations of HU (0, 1, 10, 20, or 50 μM) were prepared prior to incubation with oocytes to examine maturation. We analyzed numbers of metaphase II (MII) oocytes, as well as glutathione (GSH) and ROS levels, φm, and apoptosis.
Next, we cultured parthenogenetically activated MII oocytes [matured in IVM medium with 0 or 20 μM HU, followed by in vitro culture (IVC) medium without HU] to test the effect of HU on development of MII oocytes to blastocysts. We analyzed rates of blastocyst formation, total blastocyst cell numbers, and apoptosis of blastocyst cells.
Finally, we cultured parthenogenetically activated MII oocytes (matured in IVM medium without HU) in IVC medium containing 0, 1, 10, 20, or 50 μM HU to test the effect of HU on embryo development.

Collection of porcine oocytes and IVM
Approximately 200 porcine ovaries (obtained from the slaughterhouse) were stored in warm 0.9% saline solution containing 1% antibiotic and sent to the laboratory within 3 h. Using sterile syringes, porcine cumulus-oocyte complexes (COCs) were extracted from 3-6 mm diameter follicles and placed in a 50-ml tube. Collected COCs were washed three times by adding Tyrode's lactate-HEPES buffered medium containing 1% antibiotic and 1 g/l polyvinyl alcohol (PVA). Next, 40-60 COCs were added to 450 μl of IVM medium (TCM-199; 0.91 mM sodium pyruvate, 75 mg/ml kanamycin, 0.6 mM l-cysteine, 10 ng/ml epidermal growth factor, 10 IU/ml luteinizing hormone, 10 IU/ml follicle-stimulating hormone, and 10% v/v porcine follicular fluid) per well of non-tissue culture-treated four-well plates (179830, Thermo Scientific, Waltham, MA, U.S.A.) and covered with 500 μl mineral oil. COCs were cultured in IVM medium for 44-46 h at 38.5 • C with 5% CO 2 /95% air. After maturation, we used HEPES buffer containing 1 mg/ml hyaluronidase to remove cumulus cells from COCs and collected cumulus cells and MII oocytes. We selected MII oocytes using a stereo microscope (180-400× magnification) with a heating stage at 38.5 • C. A 200-μm-diameter glass needle was used to collect oocytes with uniform cytoplasm and an extruded first polar body. We analyzed the rate of MII oocyte development at 46 h.

Evaluation of total cell numbers per blastocyst
Total blastocyst cell numbers were determined for 21 control group blastocysts (7 per group, matured in IVM medium followed by IVC medium without HU) and 21 HU-exposed blastocysts (7 per group, matured in IVM medium with 20 μM HU, followed by IVC medium without HU). Blastocysts were fixed with 4% (w/v) paraformaldehyde, washed three times with phosphate-buffered saline mixed with 1g/l PVA (PBS-PVA), and incubated in 10 μg/ml Hoechst 33342 for 5 min at 37 • C. Finally, blastocysts were placed on a glass slide under a glass coverslip, and images were acquired with a digital camera and fluorescence microscope (E179168, Nikon, Tokyo, Japan).
Immunofluorescence staining and real-time reverse transcription polymerase chain reaction MII oocytes were washed three times with PBS-PVA, fixed with 4% (w/v) paraformaldehyde solution, washed, and incubated with 0.2% (v/v) Triton X-100 for 15-20 min. Fixed oocytes were washed and incubated in 1% (w/v) BSA for 1 h at room temperature to block nonspecific binding. Oocytes were incubated with anti-P53 (1:100; Abcam, Cambridge, U.K.) and anti-Caspase-3 (1:100; Abcam) antibodies at 4 • C overnight. The following day, oocytes were washed, incubated with a secondary antibody (1:100; CY3-goat anti-rabbit; Boster Biological Technology, Wuhan, China) at 37 • C for 1-2 h, washed three times, and placed in Hoechst 33342 for 5 min at 37 • C. We used 24 (8 per group, control group) and 24 (8 per group, 20 μM HU group) MII oocytes to examine P53 expression, and 21 (7 per group, control group) and 21 (7 per group, 20 μM HU group) MII oocytes to examine Caspase-3 expression. Immunostained oocytes were placed on glass slides and covered with a glass coverslip. Images were acquired with a digital camera with a fluorescence microscope.
Total mRNA was extracted from 5 × 10 4 digested cumulus cells (from 20 MII oocytes) using a microRNA extraction kit (Qiagen, Dusseldorf, Germany). mRNA was reverse transcribed into cDNA using a reverse transcription kit (Tiangen Biotech, Beijing, China). SYBR green fluorescent dye (Tiangen Biotech), cDNA, ddH 2 O, and primers (Supplementary Table S1) were added to the samples for PCR using a reverse transcription polymerase chain reaction (RT-PCR) instrument (Eppendorf, Hamburg, Germany). RT-PCR cycles included pre-denaturation at 95 • C for 15 min followed by 45 cycles of 95 • C for 10 s (denaturation), 60 • C for 20 s (annealing), and 72 • C for 30 s (extension), followed by melting curve analysis. The β-Actin gene was used for standardization. Three independent experiments were performed, and the 2 − C t [ C t = C t (case) − C t (control)] method was used to calculate relative mRNA expression.

Flow cytometry detection of apoptosis in cumulus cells
Digested cumulus cells were collected using a low-speed centrifuge and washed in a 1.5-ml tube. The PBS supernatant was gently decanted after centrifugation. PBS-PVA cleaning solution was added and the cell slurry was gently mixed by pipetting. Following three washes, a solution containing 5 μl of Annexin V-FITC (Ca 2+ -dependent phospholipid binding protein; Solarbio Life Sciences, Beijing, China) was added to samples, which were mixed at 20 • C for 10 min. Next, 5 μl of propidium iodide (Solarbio Life Sciences) was added to samples for incubation at 20 • C for 5 min. Finally, samples were gently mixed with 500 μl of PBS (all steps in the dark) and analyzed within 1 h by flow cytometry.

TUNEL assay for detection of apoptosis in blastocysts
To measure apoptosis levels in blastocysts, 18 (6 per group) and 18 (6 per group) blastocysts from parthenogenetically activated MII oocytes (matured in IVM medium with 0 or 20 μM HU, followed by culture in IVC medium without HU) were washed three times with PBS-PVA, fixed with 4% (w/v) paraformaldehyde solution, washed, and incubated with 0.2% (v/v) Triton X-100 for 15-20 min. After washing three times with PBS-PVA, fixed blastocysts were incubated with TdT and fluorescein-conjugated dUTPs (In Situ Cell Death Detection Kit; Roche, Mannheim, Germany) in the dark for 30 min at 37 • C. Next, blastocysts were washed three times, placed in Hoechst 33342 for 5 min at 37 • C, and washed again three times with PBS-PVA for 10 min each. Immunostained blastocysts were placed on glass slides and covered with glass coverslips. Images were acquired with a digital camera and fluorescence microscope.

GSH and ROS levels in MII oocytes and cumulus cells
To measure GSH levels, 48 (15-17 per group) and 50 (16-18 per group) MII oocytes from control and 20 μM HU groups, respectively, were incubated with IVM medium containing 10 μM of 4-chloromethyl-6,8-difluoro-7-hydroxycoumarin (CMF2HC, Thermo Fisher Scientific) cell tracer blue dye for 20 min, washed three times, and analyzed by spectroscopy (blue fluorescence, UV filter, 370 nm). To examine ROS levels, 48 (15-17 per group) MII oocytes from control and 20 μM HU groups were washed three times with PBS-PVA and incubated with IVM medium containing 10 μM of 2,7-dichlorodihydrofluorescein diacetate (H2DCFDA; Thermo Fisher Scientific) for 15 min in non-treated four-well plates. Following incubation, plates were washed three times and analyzed using a fluorescence microscope (green fluorescence, UV filters, 490 nm). All images were stored as TIFF files and analyzed by ImageJ software (http://imagej.nih.gov).
Cumulus cells were washed three times for 5 min at 600×g in 1.5-ml microcentrifuge tubes. Next, cells were incubated with IVM medium containing 10 μM CMF2HC or 10 μM H2DCFDA for 20 min and 15 min, respectively, followed by three washes. Cumulus cells incubated with CMF2HC were analyzed by a fluorescence microscope (blue fluorescence, UV filter, 370 nm). Images were stored as TIFF files and analyzed by ImageJ software. Approximately 1-5 × 10 4 cumulus cells incubated with H2DCFDA were placed on ice in the dark for analysis by flow cytometry (green fluorescence, UV filters, 490 nm). Expression of 2 ,7 -dichlorofluorescein is presented as the peak of the median FITC-A signal. For measurement of φm, cumulus cells were washed three times for 5 min at 600×g in 1.5-ml microcentrifuge tubes. Next, cells were incubated with IVM medium containing JC-1 fluorescent probe at 37 • C for 20 min, and then washed three times with JC-1 staining buffer. Cells were placed in 1 ml of JC-1 staining buffer per tube and transferred in 200-μl aliquots to wells of a six-well plate for analysis. Cells were analyzed using a fluorescence microscope with 490-nm (green fluorescence) and 530-nm (red fluorescence) excitation. Images stored as TIFF files were analyzed by ImageJ software.

Statistical analysis
Each experiment was repeated at least three times and data were analyzed by SPSS 20.0 software (IBM, Armonk, NY, U.S.A.). We used Student's t test to analyze comparisons of two groups and ANOVA test to analyze comparisons of more than two groups. P<0.05 was considered statistically significant.
MII oocytes cultured in IVM media with (20 μM) or without HU were subsequently cultured in IVC media to examine post-meiotic development to the blastocyst stage. Percentages of blastocysts were significantly decreased in the 20-μM HU group compared with the control group (39.00 + − 4.04% vs. 59.17 + − 2.78%, P<0.01, Figure 1C). Moreover, total blastocyst cell numbers were significantly decreased in the 20-μM HU group compared with the control group (49.52 + − 9.55 vs. 74.19 + − 13.92, Figure 1A,B, P<0.001). These results indicate that 20 μM HU affected post-meiotic development of MII oocytes.

Apoptosis of MII oocytes, cumulus cells, and parthenogenetically activated blastocysts exposed to 20 μM HU
To determine the effect of HU on apoptosis, P53 and Caspase-3 expression levels were evaluated using immunofluorescence staining. Expression of P53 protein in MII oocytes was increased in the 20-μM HU group compared with the control group (80.84 + − 1.23 vs. 58.16 + − 2.90 pixels per oocyte, P<0.001, Figure 3A,C, respectively). Expression of Caspase-3 protein in MII oocytes was also increased following HU exposure compared with the control group (60.16 + − 1.04 vs. 46.82 + − 2.41 pixels per oocyte, P<0.01, Figure 3A,D, respectively).

GSH and ROS levels in MII oocytes and cumulus cells exposed to 20 μM HU
To understand the mechanism of action by which HU affected porcine oocyte maturation, GSH and ROS levels were measured after in vitro oocyte maturation. ROS levels in MII oocytes were significantly higher in the 20-μM-HU-exposed group versus the control group (10.24 + − 1.19 vs. 5.15 + − 0.59 pixels per oocyte, P<0.001, Figure  4B,E). Conversely, GSH was significantly down-regulated in the 20-μM HU group (26.35 + − 1.95 vs. 35.75 + − 2.04 pixels per oocyte, P<0.001, Figure 4A,C).

Discussion
Our results show that HU decreased the maturation rate of MII oocytes and significantly decreased the parthenogenetic activation rate of blastocysts. In addition, HU significantly decreased total blastocyst cell numbers. These findings indicate that HU can decrease the maturation and developmental ability of porcine oocytes. We also investigated the effects of HU on the two-celled to hatching stage of porcine embryonic development and found that HU significantly reduced development during the two-celled to hatching stage. Our experimental results are consistent with previous studies [12,21] showing that HU inhibits early embryonic development.
The P53 pathway plays a central role in embryonic stress response and teratogenesis [22]. While activation of the P53 pathway in aneuploid cells reduces cell proliferation [23], P53 activation leads to cell cycle arrest and apoptosis during embryonic kidney development and in mouse embryonic stem cells [24,25]. During the organogenesis stage of embryonic development, HU significantly increased levels of P53 and the P53-dependent protein Caspase-3 [26]. In the current study, we observed a similar increase in both P53 and Caspase-3 protein levels following HU exposure of MII oocytes. The observed effects in oocytes demonstrate HU-induced toxicity in post-implantation embryos. Past work showed that HU induced high levels of P53 expression during IVM of oocytes, which led to cell developmental arrest and disrupted extrusion of the second polar body [21]. Therefore, HU exposure results in reduced maturity of oocytes, affecting their IVM. During maturation, signaling between cumulus cells and oocytes elicits interactions of cell-secreted factors [27,28]. Proliferation and differentiation of cumulus cells are essential for the development of oocytes [29][30][31][32][33]. The maturation of cumulus cells greatly influences the maturation of porcine oocytes. Previous work showed that reduced apoptosis of porcine cumulus cells leads to decreased expression of Caspase-3 and related apoptotic proteins, as well as a reduced BCL2/BAX ratio [34]. Additionally, increased expression levels of apoptotic genes, such as p38, have been observed during porcine maturation [35,36]. We observed increased levels of P53 and Caspase-3 mRNA following HU exposure of cumulus cells. In parallel, flow cytometry and TUNEL assay analyses showed increased cumulus and blastocyst cell apoptosis because of HU exposure. Our results confirm that apoptosis of cumulus cells and oocytes is synchronized, but the mechanisms of HU-mediated apoptosis of cumulus cells and oocytes remain undetermined and need further study. We hypothesized that HU blocks oocyte maturation by increasing apoptosis of both oocytes and cumulus cells, and by directly inhibiting oocyte maturation.
During embryonic development, HU has been shown to induce nitric oxide (NO) through regulation of production and binding to fetal hemoglobin. Increased NO may result in oxidative stress and damage to the fetus. During oxidative stress, ROS levels increase and GSH levels decrease [25,37]. Several teratogens affect developing embryos by increasing their oxidative stress through increased ROS levels, especially during early organogenesis. Oxidative stress, in turn, leads to severe embryo damage [38]. GSH is important in mouse embryo organogenesis, and GSH depletion significantly impacts oxidative stress and drug teratogenicity [39]. We examined GSH and ROS levels in immature oocytes and found that ROS levels increased following HU exposure, while GSH levels were decreased. These findings, which are consistent with previous studies of post-implantation embryos, suggest that the toxicological effects of HU extend to the maturation process. During embryonic development, cumulus expansion [31,40,41], apoptosis [41], and cell cycle regulation [42] are correlated with ROS levels in oocytes [43]. Paracrine factors regulate intracellular ROS levels during IVM of porcine oocytes [44]. In addition, regulation of cumulus cell lipid metabolism can increase GSH levels and decrease ROS levels during IVM, which results in improved developmental competence of somatic cell nuclear transfer embryos [45]. As GSH and ROS levels appear to be critical for IVM of oocytes, we also analyzed GSH and ROS levels of cumulus cells. Cumulus cells exposed to HU showed decreased GSH levels and increased ROS levels, which suggests that HU exposure led to oxidative stress in these cells. This oxidative stress may result in diminished signaling between cumulus cells and oocytes, and inhibition of oocyte maturation. The main endogenous source of ROS is mitochondria, where φm is a key regulator of mitochondrial respiration. Depolarization of φm can lead to excessive ROS production. Our results indicate that the oxidative stress of porcine oocytes and cumulus cells are synchronized, but the mechanism of HU-mediated oxidative stress in oocytes and cumulus cells is unclear and needs further study. Our current study investigated the role of mitochondria in HU toxicity and found that HU increased mitochondrial depolarization. Our findings indicate that HU may have an adverse effect on electron transfer in mitochondria that results in increased ROS.
The current study shows that the toxic effect of HU on porcine oocyte maturation in vitro is modulated through increased apoptosis of oocytes and cumulus cells, and increased oxidative stress. These effects hinder polar body extrusion in oocytes, which negatively influences their maturation and embryo development. The current study helps with risk assessment associated with HU exposure during oocyte maturation.

Data Availability
All data generated or used during the study are available from the corresponding author by request. All data generated or used during the study appear in the submitted article.