Copper/Zinc Superoxide Dismutase Insufficiency Impairs ...

BIOLOGY OF REPRODUCTION (2012) 86(1):16, 1–8 Published online before print 7 September 2011. DOI 10.1095/biolreprod.111.092999

Copper/Zinc Superoxide Dismutase Insufficiency Impairs Progesterone Secretion and Fertility in Female Mice1 Yoshihiro Noda,3 Kuniaki Ota,4 Takuji Shirasawa,5 and Takahiko Shimizu2,3 3Molecular

Gerontology, Tokyo Metropolitan Institute of Gerontology, Tokyo, Japan Department of Obstetrics and Gynecology, School of Medicine, Faculty of Medicine, Toho University, Tokyo, Japan 5Department of Aging Control Medicine, Juntendo University Graduate School of Medicine, Tokyo, Japan 4First

Copper/zinc superoxide dismutase (CuZn-SOD, SOD1) is one of the major antioxidant enzymes, and is localized in the cytoplasm to scavenge superoxide. To investigate the physiological role of SOD1 in the ovaries, we analyzed the fertility of Sod1-deficient female mice. To evaluate their hormonal metabolism, we measured pituitary and ovarian hormone levels in the plasma of the mutant mice. Plasma follicle-stimulating hormone, luteinizing hormone, and estradiol were not altered in the mutant compared to the wild-type females, while the plasma progesterone level was significantly reduced in the mutant females. Furthermore, the mutant mice showed decreased progesterone secretion under the condition of superovulation. In a histochemical analysis, we observed a remarkable reduction in the corpus luteum area in the mutant ovaries without atrophic changes. The mutant mice also displayed enhanced superoxide generation in the region surrounding the corpora lutea, which was associated with increased apoptotic cells and suppressed vasculature. These results suggested that SOD1 deficiency dysregulated luteal formation because of increased superoxide generation in the ovary. In vitro fertilization experiments showed no abnormal fertilization of Sod1-deficient oocytes. In addition, when Sod1-deficient embryos were transferred into the oviducts of wild-type females, mutant embryos developed at a normal rate, indicating that SOD1 deficiency in embryos did not cause miscarriage in the uterus of wild-type females. These results indicated that increased intracellular ROS impaired luteal formation and progesterone production in the mutant females, thus suggesting that SOD1 plays a crucial role in both the luteal function and the maintenance of fertility in female mice. corpus luteum, CuZn-SOD (SOD1), progesterone

INTRODUCTION The corpus luteum (CL) is a temporary endocrine structure in the ovary of mammals. Ovarian follicles generate the CL during the luteal phase of the estrous cycle. The CL also produces estrogen (E2) and progesterone (P4) when stimulated 1Supported by the Program for the Promotion of Basic Research Activities for Innovative Biosciences (to T. Shimizu). 2Correspondence: Takahiko Shimizu, Molecular Gerontology, Tokyo Metropolitan Institute of Gerontology, 35-2 Sakae-cho, Itabashi-ku, Tokyo 173-0015, Japan. FAX: 81 3 3579 4776; e-mail: [email protected]

Received: 22 April 2011. First decision: 12 May 2011. Accepted: 23 August 2011. Ó 2012 by the Society for the Study of Reproduction, Inc. eISSN: 1529-7268 http://www.biolreprod.org ISSN: 0006-3363

1

Article 16

Downloaded from www.biolreprod.org.

by pituitary gonadotropins such as follicle-stimulating hormone (FSH) and luteinizing hormone (LH) to develop the endometrium and to continue embryonic implantation to the endometrium in the uterus. When egg is fertilized in the uterus, P4 plays a crucial role in maintaining the pregnancy after implantation [1, 2]. In the clinic, P4 insufficiency causes spontaneous miscarriage. In these cases, P4 supplementation could potentially maintain the pregnancy. However, there is no evidence to support the routine use of P4 to prevent miscarriage in early to midpregnancy [3]. Oxidative stress caused by the imbalance between reactive oxygen species (ROS) and the biological antioxidant systems can lead to oxidative modification of DNA, lipids, and proteins. There is a delicate balance between ROS and antioxidant enzymes, including several superoxide dismutase (SOD), catalase, and glutathione peroxidases in the ovarian tissues. The antioxidant enzymes neutralize and protect the oocyte and embryo from excessive ROS. Oxidative stress has been reported to play an important role in the normal function of the female reproductive system and in the pathogenesis of female infertility [4]. Simsek et al. [5] also reported that oxidative stress has been implicated as an important cause of recurrent pregnancy loss. In addition, biochemical markers of ROS-induced membrane damage, such as lipid peroxidation products, reach high levels immediately before miscarriage [6]. An oxidant/antioxidant imbalance has also been reported to be associated with miscarriage [7]. Indeed, Suzuki et al. [8] reported that SOD was localized in human ovaries under the normal follicular cycling. However, the pathological effects of intracellular ROS on the ovarian function have not yet been clarified. SOD is one of the major antioxidant enzymes that catalyzes the conversion of superoxide radicals to hydrogen peroxide [9]. SODs consist of three isozymes: copper/zinc SOD (CuZn-SOD, SOD1), which is localized in the cytosol, nucleus, and intermembrane space of mitochondria [10]; manganese SOD (Mn-SOD, SOD2), which occurs in the mitochondrial matrix; and extracellular SOD (EC-SOD, SOD3). We have reported that Sod1-deficient (Sod1/) mice showed drusen formation, which is a typical characteristic of age-related macular degeneration [11], fatty liver [12], and skin thinning [13]. Other groups have reported symptoms of hepatic carcinoma [14], hemolytic anemia [15], and muscle atrophy [16] in Sod1/ mice. Interestingly, several groups have reported that Sod1/ female mice also showed reduced fertility [17, 18]. However, the detailed mechanism responsible for the reduced fertility in Sod1/ mice has not been elucidated. In the present study, we investigated the plasma levels of pituitary and ovarian hormones involved in the reproductive system in Sod1/ female mice. We also histochemically analyzed the organs involved in fertility. Furthermore, using an

ABSTRACT

NODA ET AL. Chronic Gonadotropin (eCG; 7.5 units/mouse; ASKA Pharmaceutical, Tokyo, Japan) followed by human chorionic gonadotropin (hCG; 7.5 units/mouse; ASKA Pharmaceutical) 48 h later. The ova were collected from the oviduct ampulla of mice at 14 h after hCG injection. Essentially, 200-ll drops of Toyoda Yokoyama Hoshi (TYH) medium (Mitsubishi Chemical Medience, Tokyo, Japan) [19] and four 95-ll drops of modified Whitten medium (mWM; Mitsubishi Chemical Medience) [20] were overlaid with mineral oil (M8410; Sigma-Aldrich, St. Louis, MO) in a 35-mm-diameter plastic culture dish (150255; nunc, Roskilde, Denmark) and equilibrated overnight at 378C in an humidified atmosphere of 5% CO2 in air. The subsequent insemination and the embryo culture were also performed under the same conditions. Oviducts were transferred singly beneath the mineral oil in close proximity to the TYH drop. The cumulus-oocyte complexes were released from the ampulla into the mineral oil by rupturing the oviduct with the aid of a needle and then moved into the TYH drops. Spermatozoa of male mice were collected from the cauda epididymis and suspended in TYH medium. After preincubation for 1 h, an aliquot of sperm suspension was inseminated into TYH medium containing oocytes to make the final sperm concentration of 200 sperm/ll. Then, 6–8 h after insemination, pronucleus embryos were counted and washed with mWM. At 24 h after insemination, early two-cell embryos were obtained, washed with mWM, and then transferred to the oviducts (8–10/oviduct) of Jcl:MCH (ICR) recipient females that showed symptoms of sexual excitement that were mated with vasectomized Jcl:MCH (ICR) males on Day 1 for the pseudopregnancy recipients. The number of pups was recorded on the day of birth.

in vitro fertilization and embryo transfer technique, we compared the fertilization and embryogenesis in Sod1/ and Sod1 þ/ þ eggs. We herein discuss the relationship between intracellular ROS due to Sod1 deficiency and ovarian function, especially development of the CL, in female mice. MATERIALS AND METHODS Animals Sod1/ mice were purchased from the Jackson Laboratory (Bar Harbor, ME). The mice were backcrossed with C57BL/6NCrSlc (Japan SLC, Shizuoka, Japan) mice for five or six generations. The genotyping of Sod1/ mice was performed by PCR using genomic DNA isolated from the tail tip as reported previously [12], with slight modifications. The primers for identifying the Sod1 genotype, 5 0 -TGT TCT CCT CTT CCT CAT CTC C-3 0 , 5 0 -ACC CTT TCC AAA TCC TCA GC-3 0 , 5 0 -AGC CCT GGT GCA GGA GTA TT-3 0 , and 5 0 CCT ACC TTG TGT ATT GTC CCC A-3 0 , were used under the following conditions: one cycle of 958C for 10 min, 35 cycles of 968C for 30 sec, 58.58C for 30 sec, and 728C for 60 sec, followed by one cycle of 728C for 7 min. Sod1þ/ þ mice were examined as littermate controls. The estrous cycle was monitored at 0900–1000 h by vaginal smear. After 6–7 h, isolated tissues, and blood were collected at 1600–1700 h as a proestrus period. Jcl:MCH (ICR) mice (8–12 wk of age) were purchased from CLEA Japan Inc. (Tokyo, Japan). The animals were housed under a 12L:12D cycle (light period: 0800–2000) and with ad libitum access to food and water. The mice were maintained and studied according to protocols approved by the Animal Care Committee of the Tokyo Metropolitan Institute of Gerontology.

Determination of Intracellular ROS Generation in Oocytes

Measurement of Plasma Hormone Levels The blood samples were collected at a proestrus period from the intraorbital venous plexus using heparinized capillary tubes under diethyl ether anesthesia at designated time intervals. Blood was collected in ice-cold tubes coated with sodium-EDTA (CAPIJECT; TERUMO, Tokyo, Japan) and centrifuged at 3000 rpm for 10 min at 48C. The plasma was stored at 808C until the hormone assays. FSH, LH, E2, and P4 levels were measured in plasma samples using a rodent ELISA kit according to the manufacturer’s instructions (Endocrine Technologies, Newark, CA). The limits of sensitivity of the assays were 1 ng/ ml for FSH, 0.5 ng/ml for LH, 5 pg/ml for E2, and 0.1 ng/ml for P4.

Statistical Analysis Statistical evaluations were performed using the two-tailed Student t-test for unpaired values. Differences between the data were considered significant for P-values ,0.05. The data are represented as the means plus or minus standard deviations.

Histology RESULTS

For histological morphology, ovarian tissues were dissected at a proestrus period, fixed in a 20% formalin neutral buffer solution (Wako Pure Chemical Industries, Osaka, Japan) overnight, embedded in paraffin, sectioned on a microtome at 4-lm thickness, and stained with hematoxylin and eosin. The ratio of CL to ovarian areas was quantified using Leica QWin V3 software (Leica Microsystems, Wetzlar, Germany). TUNEL staining was performed using the ApoTag peroxidase in situ apoptosis detection kit (Chemicon, Rosemont, IL) according to the manufacturer’s instructions. Sections were stained with antiactive caspase 3 (1:500; Epitomics, Burlingame, CA) and antiCD31 (1:400 Santa Cruz Biotechnology, Santa Cruz, CA) antibodies using a conventional method. Images were obtained using a Pixera Pro600EX camera attached to a VANOX-S microscope (Olympus, Tokyo, Japan). The number of apoptotic cells was counted in CLs and expressed as apoptotic cells per CL using Leica QWin V3 software.

The Progesterone Levels Were Specifically Down-Regulated in CuZn-SOD-Deficient Mice To investigate the levels of pituitary and ovarian hormones, we measured the plasma hormone level by specific ELISAs in Sod1/ female mice. No differences were observed in the plasma levels of FSH, LH, or E2 at proestrus between the Sod1 þ/ þ and Sod1/ mice (Fig. 1, A–C). On the other hand, the plasma P4 level was significantly repressed by Sod1 deficiency at early pregnancy from Day 1.5 to Day 7.5 compared to Sod1 þ/ þ female mice (Fig. 1D). Furthermore, the plasma P4 level was significantly suppressed in Sod1/ mice when superovulation was induced by gonadotropin injection (Fig. 1F). In contrast, the plasma E2 level was not altered by superovulation (Fig. 1E), supporting the fact that no difference was observed between genotypes in the number of ovulated oocytes after gonadotropin injection (Table 1). These results indicated that Sod1 deficiency specifically down-regulated P4 production.

In Vivo Detection of Superoxide by Dihydroethidium Dihydroethidium (DHE) was obtained from Invitrogen (Molecular Probes, Eugene, OR). To identify cell-specific superoxide formation in the ovaries in vivo, mice were given DHE via tail vein injections (0.2 mg/200 ll in PBS). At 1 h after DHE injection, mice were anesthetized and then perfused with 4% paraformaldehyde (Wako Pure Chemical Industries). Frozen ovaries were sectioned at 10-lm thickness using a cryostat (HM500MV; Microm, Walldorf, Germany). Slices were evaluated for ethidium fluorescence (excitation, 488 nm; emission .590 nm) on a confocal laser scanning microscope (Zeiss LSM5 PASCAL; Carl Zeiss, Oberkochen, Germany).

There Was a Decrease in the Corpus Luteum Area in CuZn-SOD-Deficient Ovaries

In Vitro Fertilization and Embryo Transfer

To investigate the pathology of organs involved in sex hormone production, we analyzed their histology. As shown in Figure 2, a small CL and increased stromal cells were

Oocytes were obtained from Sod1-deficient female mice (8–12 wk of age) induced to undergo superovulation by intraperitoneal injection of equine

2

Article 16


Oocytes were collected from the oviduct ampulla isolated from mice 14 h after hCG injection. The collected cumulus-oocytes were immersed in 0.1% hyaluronidase (TYPE I-S H3506; Sigma-Aldrich) to remove cumulus cells for 5 min. Then oocytes were washed with mWM three times. Oocytes were treated with mWM containing 1 lM 5-(and-6)-chloromethyl-2 0 ,7 0 -dichlorodihydrofluorescein diacetate and acetyl ester (CM-H2DCFDA; C6827; Molecular Probes) for 10 min. Intracellular ROS generation was detected using a Zeiss LSM5 PASCAL microscope (Carl Zeiss). Fluorescence images were captured through a 505–530-nm bandpass filter.

SOD1 REGULATES PROGESTERONE SECRETION

distinctively observed in the Sod1/ ovaries. Furthermore, more vacuoles and degenerated cells were observed in the Sod1/ compared to the Sod1 þ/ þ CL (Fig. 2, C and D). The histological analyses of ovaries obtained during proestrus showed that the cells in the CLs of Sod1/ mice had less cytoplasm and dramatically less vascularity within the CLs (Fig. 2, C and D). A quantitative analysis demonstrated that the ratio of CL to ovarian areas was significantly decreased by 62.5% in the Sod1/ mice (15.0 6 1.1%, P ¼ 0.003, n ¼ 3) compared to that of the Sod1 þ/ þ mice (39.9 6 8.1%, n ¼ 3).

abnormal morphology in the brain or pituitary gland of the Sod1/ mice (Fig. 2, E–H). In contrast, Sod1 depletion does not lead to any change in the weight of the uterus and ovaries themselves (Table 2). These results indicate that Sod1 deficiency induces abnormal morphology in the ovaries but does not alter the ratios of organ weight to body weight. CuZn-SOD Deficiency Causes Apoptosis in the Corpus Lutea Apoptosis is a major determinant of the maintenance of tissue mass. To clarify the effects of cell death on Sod1/ ovaries, we investigated the number of apoptotic cells in ovary sections. As shown in Figure 3, TUNEL-positive cells were clearly observed in the CL of Sod1/ ovaries. An active caspase 3 antibody also distinctively stained the cells in the CL of Sod1/ ovaries (Fig. 3F), thus indicating that Sod1 deficiency induced apoptotic cell death in the CL. In contrast, apoptotic cells were rarely observed in the CL of Sod1 þ/ þ ovaries (Fig. 3, C and E). A quantitative analysis revealed that the number of apoptotic cells in CL was significantly increased by 370% in the Sod1/ mice (16.3 6 7.4 cells/CL, P ¼ 0.002, n ¼ 8) compared to those of the Sod1 þ/ þ mice (4.4 6 5.7 cells/ CL, n ¼ 8).

CuZn-SOD-Deficient Mice Have Lower Body Weights, Without Atrophy of the Genital Organs Next, to evaluate the atrophic changes of the organs, we measured the whole body and individual organ weights in the mice. The Sod1/ female mice exhibited significantly lower body weights compared to the Sod1 þ/ þ mice at 22–25 wk of age (P , 0.01; Table 2). The weights of the brain and pituitary gland in the Sod1/ mice were significantly decreased, but there was not a significant change in the body weight ratio (Table 2). Furthermore, the histochemical analyses showed no

TABLE 1. The number of oocytes produced by female Sod1þ/ þ, Sod1þ/, and Sod1/ mice. Sod1 genotype þ/þ þ/ /

No. of females

No. of collected oocytesa

Average no. of collected oocytes per head (6SD)

36 57 46

1197 1870 1761

33.3 6 7.59 32.8 6 9.40 38.3 6 17.05

CuZn-SOD Deficiency Suppresses Vasculature in the Corpus Lutea To investigate vascularity in Sod1/ tissues, we immunostained the endothelial cells in ovary and uterus sections using anti-CD31 antibody. As shown in Figure 4, CD31-positive vessels were clearly observed in the CL and uteri of Sod1 þ/ þ (Fig. 4, A, C, and E). On the other hand, no CD31-positive

a

Superovulated oocytes were collected from oviduct ampulla at 14 h after hCG administration.

3

Article 16


FIG. 1. The plasma levels of reproductive hormones in CuZn-SOD-deficient mice. The plasma levels of follicle-stimulating hormone (FSH) (A), luteinizing hormone (LH) (B), and estradiol (E2) (C) were determined by ELISA during proestrus in Sod1þ/ þ and Sod1/ mice at 23–25 wk of age (n ¼ 7/ group). D) The plasma levels of progesterone (P4) were determined by ELISA on Days 1.5, 5.5, and 7.5 of pregnancy in Sod1þ/þ and Sod1/ mice at 20– 23 wk of age (n ¼ 8/group and time points). The changes in the plasma estradiol (E) and progesterone (F) concentrations by the induction of ovulation by human chorionic gonadotropin (hCG) treatment of equine chronic gonadotropin (eCG)-primed 11-wk-old Sod1þ/ þ and Sod1/ mice (n ¼ 7/group and time points). Day 0: before the eCG administration; eCG 48 h: treated with eCG for 48 h; hCG 24 h and hCG 48 h: treated with eCG for 48 h followed by hCG for 24 h (hCG 24 h) or 48 h (hCG 48 h), respectively. *P , 0.05, **P , 0.01 vs. Sod1þ/ þ.

NODA ET AL.

Enhanced ROS Production in the CuZn-SOD-Deficient Ovaries and Oocytes

The Fertilization and Development of CuZn-SOD-Deficient Oocytes and Embryos Is Not Abnormal Although Sod1 deficiency is known to reduce fertility in female mice, the molecular mechanisms of impaired fertility have not been clarified. We considered various determinants, including maternal or egg factors, that could induce dysfunctional fertility in Sod1/ females. First, to test the capacity for fertilization and development in Sod1-deficient oocytes, we analyzed their morphology and fertilization rate as well as the two-cell development rate using in vitro fertilization techniques. When female mice were superovulated by sequential gonadotropin injections, we obtained three genotypes of oocytes with an equal proportion and with normal morphology, indicating that loss of the Sod1 allele does not cause problems with oocyte formation (87.6%–90.5%; Table 3). During in vitro fertilization studies, the Sod1/ sperm could efficiently fertilize to Sod1/, Sod1 þ/, and Sod1 þ/ þ oocytes (81.9%– 84.7% of oocytes were fertilized; Table 3). Interestingly, Sod1/ eggs effectively developed to the two-cell stage without delayed cleavage (84.4%; Table 3). These results demonstrated that Sod1 deficiency does not cause impaired

FIG. 2. CuZn-SOD-deficient mice showed abnormal morphology in the ovaries. Ovarian tissue sections were stained with hematoxylin and eosin. A, C) Sod1þ/ þ mice during proestrus at 18 wk of age. B, D) Sod1/ mice during proestrus at 14 wk of age. C, D) Higher magnification of the white boxes of A or B, respectively. E, F) Pituitary sections were stained with hematoxylin and eosin. E) Sod1þ/ þ mice at 55 wk of age. F Sod1/ mice at 55 wk of age. G, H) Higher magnification of the white boxes of E or F, respectively. Bars ¼ 100 lm.

vessel was observed in the CL of Sod1/ (Fig. 4, B and D). Interestingly, we detected CD31-positive vessels in the uteri of Sod1/ (Fig. 4F), thus indicating that Sod1 deficiency selectively suppressed vasculature in the CL.

TABLE 2. Relative organ weights of the mice. Sod1þ/þ Organ weighed Body Brain Pituitary Uterus Ovary a b

Mean (g) 6 SD 23.67 0.49 1.54 3 103 0.075 0.016

6 6 6 6 6

1.38 0.005 0.23 3 103 0.028 0.003

Sod1/ Organ weight/ body weight (%) 6 SD

2.07 6.46 3 103 0.32 0.067

6 6 6 6

0.11 1.07 3 103 0.12 0.013

Mean (g) 6 SD 20.94 0.45 1.14 3 103 0.070 0.013

6 6 6 6 6

1.03a 0.016a 0.26 3 103b 0.025 0.002

Organ weight/ body weight (%) 6 SD 2.18 5.42 3 103 0.33 0.063

6 6 6 6

0.16 0.91 3 103 0.11 0.011

P , 0.01; significantly different from Sod1þ/ þ mice. P , 0.05; significantly different from Sod1þ/ þ mice.

4

Article 16


To confirm whether there was excessive ROS production in the Sod1/ ovaries, we analyzed superoxide generation in an ex vivo study using dihydroethidium (DHE), which is a specific superoxide-detecting fluorescent compound. When DHE was intravenously injected into Sod1/ mice, excess fluorescence was markedly elevated on stromal cells and the marginal region of the CL in Sod1/ ovaries (Fig. 5B). On the other hand, only slight fluorescence was observed in Sod1 þ/ þ ovaries (Fig. 5A). Quantitative analysis showed that the ROS level was increased by 383% (P , 0.01) in Sod1-deficient ovaries compared to the WT mice (Fig. 5C), indicating that Sod1 deficiency induced superoxide production, which might cause malfunction of corpus luteum in the ovary. Interestingly, we did not find any significant fluorescence in the uterus or oviducts of Sod1/ females (data not shown). Next, we analyzed the ROS levels in oocytes isolated from Sod1/ and Sod1 þ/ þ females. There was increased fluorescence observed in Sod1-deficient oocytes compared to wildtype oocytes (Fig. 5, D and E). A quantitative analysis also showed that the ROS level was increased by 116% (P , 0.01) in Sod1-deficient oocytes (Fig. 5F), thus indicating that Sod1 depletion induced ROS production in oocytes, although at a low level. These results suggested that Sod1 insufficiency impaired the redox balance in the ovary and oocytes and was associated with enhanced ROS production.


DISCUSSION

fertilization or two-cell development under normoxic conditions. Next, to evaluate the maternal effect of Sod1/, we separately transferred two-cell embryos of each genotype into wild-type recipient females. After 20 days, we counted the number of newborn mice. The results indicated that Sod1/ and Sod1 þ/ þ pups were born normally, with an indistinguishable birth rate (45.5% vs. 51.4%, respectively; Table 4), demonstrating that Sod1 does not play a crucial role in embryogenesis during pregnancy. Taken together with these results, it can thus be concluded that Sod1-deficient oocytes can be normally fertilized, Sod1deficient eggs can cleave to the two-cell stage, and Sod1deficient embryos can develop until birth in the uterus of wildtype mice, suggesting that maternal factors, including P4, in the Sod1/ females cause the reduced fertility.

Ovarian function is controlled by the hypothalamuspituitary endocrine system. Pituitary hormones, including FSH and LH, strictly regulate the secretion of ovarian hormones such as E2 and P4 in the ovarian cycle. According to clinical data, patients with luteal insufficiency generally exhibit a decrease in the blood P4 level at the midluteal phase [21]. Furthermore, Sugino et al. [22, 23] reported that the serum P4 level positively correlated with the activities of CuZnSOD and Mn-SOD but negatively correlated with lipid peroxidation in the CL during rat pregnancy. In humans, CuZn-SOD expression in the CL is up-regulated during the early to midluteal phase and is gradually down-regulated thereafter [23]. Similar to the rat, lipid peroxide was inversely correlated with the CuZn-SOD expression in humans [23].

TABLE 3. In vitro fertilization of CuZn-SOD-deficient mouse oocytes. Oocyte genotypea Sod1/ Sod1þ/ Sod1þ/þ

Experimental oocytes

No. (%) of oocytes morphologically normalb

No. (%) of fertilized eggsc

No. (%) of 2-cell embryosd

725 2426 1536

656 (90.5) 2165 (89.2) 1346 (87.6)

537 (81.9) 1833 (84.7) 1132 (84.1)

453 (84.4) 1562 (85.2) 1033 (91.3)

a

Sperm genotype is Sod1/. The number of morphologically normal oocytes per total oocytes. c The number of fertilized eggs per morphologically normal oocytes. d The number of two-cell embryos per fertilized eggs. b

5

Article 16


FIG. 3. CuZn-SOD deficiency increased the number of apoptotic cells in the corpus lutea. The apoptotic cells were analyzed by TUNEL (A–D) and active caspase 3 (E, F) staining in the ovaries of Sod1 þ/ þ (A, C, E) or Sod1/ (B, D, F) mice during proestrus at 23–25 wk of age. C, D) Higher magnification of the white boxes shown in A or B, respectively. Bars ¼ 100 lm.

NODA ET AL.

These observations support that oxidative stress modulates the ovarian function and plasma hormone level. In addition, patients with unexplained recurrent spontaneous miscarriage (URSM) showed a highly significant decrease in their serum levels of SOD and catalase compared to controls [24]. El-Far et al. [24] also reported that higher serum TNF-a and LH were detected in URSM patients compared to normal individuals, suggesting that SOD, catalase, TNF-a, and LH may play a major role in the pathogenesis of URSM. Furthermore, Lim et al. [25] reported age-related increases in oxidative lipids and proteins, as well as DNA, and decreases in the expression of antioxidant genes in aged mouse ovaries, suggesting that ovarian aging might impair the intracellular redox balance in mice.

In the present study, cytoplasmic CuZn-SOD-deficiency led to decreased luteinization (Fig. 2, A and B) and a decreased serum P4 level (Fig. 1D) that were associated with increased ROS production in the ovaries (Fig. 5, A and B). Furthermore, we also observed increased apoptotic cell death in the CL (Fig. 3D) and decreased vasculature (Fig. 4D) of the mutant mice, which resulted in a specific impairment of P4 secretion. Taken together, these data indicate that CuZn-SOD in the ovary plays a pivotal role in the maintenance of CL function and P4 secretion. Ho et al. [17] first reported that SOD1 deficiency led to reduced fertility in female mice. However, female Sod1/ mice exhibited a normal estrous cycle and comparable numbers of ova compared with the Sod1 þ/ þ and Sod1 þ/ females. Although they reported that embryos were aborted at Days 10– 14 after fertilization, they did not measure the pituitary and ovarian hormones in the Sod1/ females [17]. Matzuk et al. [18] also reported that Sod1-deficient females showed subfertility, with primary follicles and reduced CL. They indicated that reduced FSH and LH caused subfertility in Sod1/ females on a C57BL/6/129SvEv genetic background. On the other hand, we failed to detect any difference in the blood FSH and LH levels between Sod1/ and Sod1 þ/ þ females (Fig. 1, A and B). Genetic or environmental factors may modulate the production of pituitary hormones, including FSH and LH. Wong et al. [26] reported reduced fertility in copper chaperone for superoxide dismutase (CCS)-deficient

TABLE 4. The development of CuZn-SOD-deficient embryos after transfer into wild-type recipient mice. Oocyte genoype

No. of recipients

No. of embryos transferred

No. (%) of neonatesa

Sod1/ b Sod1þ/ b Sod1þ/þ c

63 61 8

1327 1391 214

604 (45.5) 564 (40.5) 110 (51.4)

a b c

The number of neonates per transferred oocytes. Sperm genotype is Sod1/. Sperm genotype is Sod1þ/ þ.

6

Article 16


FIG. 4. CuZn-SOD deficiency decreased blood vessels in the corpus lutea. The blood vessels were stained with anti-CD31 antibody in the ovaries (A–D) and uteri (E, F) of Sod1þ/ þ (A, C, E) or Sod1/ (B, D, F) mice during proestrus at 14–18 wk of age. C, D) Higher magnification of the white boxes shown in A or B, respectively. Bars ¼ 100 lm.


VEGF signaling in the midluteal phase resulted in a decrease in progesterone secretion and endothelial cell death [30, 32]. Angiopoietins have also been reported to regulate the number of blood vessels in collaboration with VEGF via angiogenesis and blood vessel stabilization in the CL during early pregnancy [33]. Recently, Groleau et al. [34] reported that SOD1 expression is essential to protect against excessive ROS generation in endothelial progenitor cells and to preserve their angiogenic properties. Furthermore, increased oxidative stress can alter the placental vasculature, leading to early miscarriage [35]. In this context, we investigated the number and morphology of blood vessels and showed decreased blood vessels in the CL of Sod1/ females (Fig. 4), suggesting that suppressed vascularity resulted in decreased luteinization and P4 secretion, leading to infertility in the Sod1/ females. Further studies of the expression of VEGF and angiopoietins might clarify the role of angiogenesis and vessel function in the ovaries of Sod1/ females. In conclusion, we herein presented evidence strongly suggesting that increased oxidative stress in the ovary could cause luteal insufficiency leading to miscarriage in mice. In the clinic, older pregnant patients show a higher rate of miscarriage. These patients suffer from an unexplained etiology, making it impossible to choose appropriate therapies. Although the etiology of miscarriage remains unclear, intracellular oxidative stress may be one of the causative factors. Since several antioxidants without adverse effects on pregnancy should suppress intracellular ROS production, it may be beneficial to apply these antioxidants for older patients or patients with recurrent pregnancy loss. It therefore remains necessary to investigate the efficacy of antioxidant supplementation to prevent pregnancy loss.

mice. The CCS mutant mice showed loss of SOD1 activity due to Cu depletion in the enzyme. In their histological analysis of the ovaries, they observed fewer mature follicles and CL. The CL dysplasia in the CCS-deficient females was recapitulated in the morphology of the CL in Sod1/ females. Recently, Kimura et al. [27] reported that the development of Sod1-deficient embryos was arrested at the two-cell stage under conventional culture conditions (20% O2) but not at 1% O2. Since oxygen promotes the generation of harmful ROS under the 20% O2 condition, SOD1 deficiency likely causes increased ROS production, leading to cell cycle arrest of embryos in vitro. Although the oxygen concentration is maintained at 2%–8% in the uterus and oviduct [28], both Sod1/ and Sod1 þ/ embryos were aborted in the uteri of Sod1/ females. In the present study, we demonstrated that Sod1/ embryos developed normally in the uteri of wild-type females using in vitro fertilization and embryo transfer techniques (Tables 3 and 4), indicating that Sod1/ embryos develop without abnormalities under the physiological condition in vivo. These results strongly suggest that the reduced P4 level caused the subfertility in the Sod1/ females. However, we cannot rule out the possibility that redox imbalance causes the abortion in the uterus of Sod1/ females. Further analyses will be needed to clarify the pathological role of ROS during embryogenesis in mice. Angiogenesis plays an important role in the formation and development of the CL as well as maintenance of luteal function [29]. Vascular endothelial growth factor (VEGF) is required for the maintenance of vascular function and structure during the normal luteal phase [30]. Fraser et al. [31] also reported that VEGF is the major regulator of luteal angiogenesis in the primate ovary. In addition, inhibition of ovarian 7

Article 16


FIG. 5. Enhanced ROS production in the ovaries and oocytes of CuZn-SOD-deficient mice. Superoxide was detected using dihydroethidium in ovary sections from 30- to 38-wk-old female mice on Day 8.5 of pregnancy (A, B). C) The quantified ROS level (relative fluorescence intensity) in the ovaries of Sod1þ/ þ and Sod1/ mice, expressed as the means 6 SD (n ¼ 5). The production of intracellular ROS was detected with 2 0 ,7 0 -dichlorofluorescin diacetate fluorescence staining in the oocytes of Sod1þ/ þ (left) or Sod1/ (right) mice (D, E). Bar ¼ 200 lm. F) The quantified intracellular ROS level (relative fluorescence intensity) in oocytes of Sod1þ/þ and Sod1/ mice, expressed as the means 6 SD (n ¼ 10). Bar ¼ 100 lm.

NODA ET AL.

ACKNOWLEDGMENT We thank Drs. Hiroaki Ohta and Ken Ishitani (Tokyo Women’s Medical University) for helpful discussions and Ms. Eiko Moriizumi, Ms. Midori Ogawara, Dr. Yasutaka Ikeda, Dr. Kazuma Murakami, Mr. Syuichi Shibuya, Dr. Tetsuro Horie, Ms. Chizuru Tsuda (Tokyo Metropolitan Institute of Gerontology), and Ms. Izumi Aoyama and Mr. Masahito Fujikawa (Labotech Co., Inc.) for technical assistance.

18. 19. 20.

REFERENCES 21. 22.

23. 24.

25. 26.

27.

28. 29. 30.

31. 32.

33.

34.

35.

8

Article 16


1. Findlay JK. Angiogenesis in reproductive tissues. J Endocrinol 1986; 111:357–366. 2. Sugino N, Okuda K. Species-related differences in the mechanism of apoptosis during structural luteolysis. J Reprod Dev 2007; 53:977–986. 3. Haas DM, Ramsey PS. Progestogen for preventing miscarriage. Cochrane Database Syst Rev 2008; 16. CD003511. 4. Agarwal A, Allamaneni SS, Said TM. Chemiluminescence technique for measuring reactive oxygen species. Reprod Biomed Online 2004; 9:466– 468. 5. Simsek M, Naziroglu M, Simsek H, Cay M, Aksakal M, Kumru S. Blood plasma levels of lipoperoxides, glutathione peroxidase, beta carotene, vitamin A and E in women with habitual abortion. Cell Biochem Funct 1998; 16:227–231. 6. Sane AS, Chokshi SA, Mishra VV, Barad DP, Shah VC, Nagpal S. Serum lipoperoxides in induced and spontaneous abortions. Gynecol Obstet Invest 1991; 31:172–175. 7. Lagod L, Paszkowski T, Sikorski R, Rola R. The antioxidant-prooxidant balance in pregnancy complicated by spontaneous abortion. Ginekol Pol 2001; 72:1073–1078. 8. Suzuki T, Sugino N, Fukaya T, Sugiyama S, Uda T, Takaya R, Yajima A, Sasano H. Superoxide dismutase in normal cycling human ovaries: immunohistochemical localization and characterization. Fertil Steril 1999; 72:720–726. 9. Wallace DC, Melov S. Radicals r’aging. Nat Genet 1998; 19:105–106. 10. Okado-Matsumoto A, Fridovich I. Subcellular distribution of superoxide dismutases (SOD) in rat liver: Cu, Zn-SOD in mitochondria. J Biol Chem 2001; 276:38388–38393. 11. Imamura Y, Noda S, Hashizume K, Shinoda K, Yamaguchi M, Uchiyama S, Shimizu T, Mizushima Y, Shirasawa T, Tsubota K. Drusen, choroidal neovascularization, and retinal pigment epithelium dysfunction in SOD1deficient mice: a model of age-related macular degeneration. Proc Natl Acad Sci U S A 2006; 103:11282–11287. 12. Uchiyama S, Shimizu T, Shirasawa T. CuZn-SOD deficiency causes ApoB degradation and induces hepatic lipid accumulation by impaired lipoprotein secretion in mice. J Biol Chem 2006; 281:31713–31719. 13. Murakami K, Inagaki J, Saito M, Ikeda Y, Tsuda C, Noda Y, Kawakami S, Shirasawa T, Shimizu T. Skin atrophy in cytoplasmic SOD-deficient mice and its complete recovery using a vitamin C derivative. Biochem Biophys Res Commun 2009; 382:457–461. 14. Elchuri S, Oberley TD, Qi W, Eisenstein RS, Jackson Roberts L, Van Remmen H, Epstein CJ, Huang TT. CuZnSOD deficiency leads to persistent and widespread oxidative damage and hepatocarcinogenesis later in life. Oncogene 2005; 24:367–380. 15. Iuchi Y, Okada F, Onuma K, Onoda T, Asao H, Kobayashi M, Fujii J. Elevated oxidative stress in erythrocytes due to a SOD1 deficiency causes anaemia and triggers autoantibody production. Biochem J 2007; 402:219– 227. 16. Muller FL, Song W, Liu Y, Chaudhuri A, Pieke-Dahl S, Strong R, Huang TT, Epstein CJ, Roberts LJ II, Csete M, Faulkner JA, Van Remmen H. Absence of CuZn superoxide dismutase leads to elevated oxidative stress and acceleration of age-dependent skeletal muscle atrophy. Free Radic Biol Med 2006; 40:1993–2004. 17. Ho YS, Gargano M, Cao J, Bronson RT, Heimler I, Hutz RJ. Reduced

fertility in female mice lacking copper-zinc superoxide dismutase. J Biol Chem 1998; 273:7765–7769. Matzuk MM, Dionne L, Guo Q, Kumar TR, Lebovitz RM. Ovarian function in superoxide dismutase 1 and 2 knockout mice. Endocrinology 1998; 139:4008–4011. Toyoda Y, Yokoyama M, Hoshi T. Studies on the fertilization of mouse eggs in vitro fertilization of eggs by fresh epididymal sperm [in Japanese]. Jpn J Anim Reprod 1972; 16:147–151. Whitten WK. Nutrient requirements for the culture of preimplantation mouse embryos in vitro. In: Raspa G (ed.), Intrinsic and Extrinsic Factors in Early Mammalian Development. Advances in the Bioscience 1971; 6:129–141. Hinney B, Henze C, Kuhn W, Wuttke W. The corpus luteum insufficiency: a multifactorial disease. J Clin Endocrinol Metab 1996; 81:565–570. Sugino N, Hirosawa-Takamori M, Zhong L, Telleria CM, Shiota K, Gibori G. Hormonal regulation of copper-zinc superoxide dismutase and manganese superoxide dismutase messenger ribonucleic acid in the rat corpus luteum: induction by prolactin and placental lactogens. Biol Reprod 1998; 59:599–605. Sugino N, Takiguchi S, Kashida S, Karube A, Nakamura Y, Kato H. Superoxide dismutase expression in the human corpus luteum during the menstrual cycle and in early pregnancy. Mol Hum Reprod 2000; 6:19–25. El-Far M, El-Sayed IH, El-Motwally AE, Hashem IA, Bakry N. Serum levels of TNF-alpha and antioxidant enzymes and placental TNF-alpha expression in unexplained recurrent spontaneous miscarriage. J Physiol Biochem 2009; 65:175–181. Lim J, Luderer U. Oxidative damage increases and antioxidant gene expression decreases with aging in the mouse ovary. Biol Reprod 2011; 84:775–782. Wong PC, Waggoner D, Subramaniam JR, Tessarollo L, Bartnikas TB, Culotta VC, Price DL, Rothstein J, Gitlin JD. Copper chaperone for superoxide dismutase is essential to activate mammalian Cu/Zn superoxide dismutase. Proc Natl Acad Sci U S A 2000; 97:2886–2891. Kimura N, Tsunoda S, Iuchi Y, Abe H, Totsukawa K, Fujii J. Intrinsic oxidative stress causes either 2-cell arrest or cell death depending on developmental stage of the embryos from SOD1-deficient mice. Mol Hum Reprod 2010; 16:441–451. Fischer B, Bavister BD. Oxygen tension in the oviduct and uterus of rhesus monkeys, hamsters and rabbits. J Reprod Fertil 1993; 99:673–679. Sugino N, Matsuoka A, Taniguchi K, Tamura H. Angiogenesis in the human corpus luteum. Reprod Med Biol 2008; 7:91–103. Fraser HM, Wilson H, Wulff C, Rudge JS, Wiegand SJ. Administration of vascular endothelial growth factor Trap during the ‘‘post-angiogenic’’ period of the luteal phase causes rapid functional luteolysis and selective endothelial cell death in the marmoset. Reproduction 2006; 132:589–600. Fraser HM, Duncan WC. Vascular morphogenesis in the primate ovary. Angiogenesis 2005; 8:101–116. Fraser HM, Wilson H, Morris KD, Swanston I, Wiegand SJ. Vascular endothelial growth factor Trap suppresses ovarian function at all stages of the luteal phase in the macaque. J Clin Endocrinol Metab 2005; 90:5811– 5818. Kashida S, Sugino N, Takiguchi S, Karube A, Takayama H, Yamagata Y, Nakamura Y, Kato H. Regulation and role of vascular endothelial growth factor in the corpus luteum during mid-pregnancy in rats. Biol Reprod 2001; 64:317–323. Groleau J, Dussault S, Haddad P, Turgeon J, Menard C, Chan JS, Rivard A. Essential role of copper-zinc superoxide dismutase for ischemiainduced neovascularization via modulation of bone marrow-derived endothelial progenitor cells. Arterioscler Thromb Vasc Biol 2010; 30:2173–2181. Todorova K, Mazneikova V, Ivanov S, Genova M. The frequency of mild and severe fetal malformations in diabetic women with high values of glycosilated hemoglobin in early pregnancy. Akush Ginekol 2005; 44:3– 10.