The lack of estrogen and excess luteinizing hormone are responsible for the female ArKO mouse phenotype

Molecular and Cellular Endocrinology 327 (2010) 56–64

Contents lists available at ScienceDirect

Molecular and Cellular Endocrinology journal homepage: www.elsevier.com/locate/mce

The lack of estrogen and excess luteinizing hormone are responsible for the female ArKO mouse phenotype夽 Seng H. Liew a,b,∗ , Ann E. Drummond a , Margaret E. Jones a,c , Jock K. Findlay a,b a b c

Prince Henry’s Institute of Medical Research, Australia Department of Obstetrics and Gynaecology, Monash University, Clayton 3168, Australia Research Services, The University of Western Australia, Crawley 6009, Australia

a r t i c l e

i n f o

Article history: Received 13 January 2010 Received in revised form 4 May 2010 Accepted 7 May 2010 Keywords: ArKO mice Estrogen Gonadotrophins Testosterone Ovary Follicles

a b s t r a c t It remains to be established as to whether the absence of estrogen (direct) or the elevated levels of gonadotrophins and androgens (indirect) are responsible for the ArKO (aromatase knockout) ovarian phenotype. The aim of this study was to determine the effects of E2 (17␤-estradiol) replacement, acyline (GnRH antagonist) and flutamide (anti-androgen) treatment on the ovarian phenotype of ArKO mice. E2 replacement and acyline treatment but not flutamide treatment, reduced serum gonadotrophin levels of ArKO mice to within normal ranges. E2 replacement improved uterine and ovarian follicular phenotypes and reduced the number of Sertoli-like filled cords by 62%. Acyline treatment reduced the number of hemorrhagic cysts and the number of Sertoli-like filled cords within ArKO ovaries. The data indicate that the absence of estrogen in concert with elevated levels of circulating gonadotrophins, principally LH, is responsible for the abnormal reproductive phenotype of the female ArKO mouse. Crown Copyright © 2010 Published by Elsevier Ireland Ltd. All rights reserved.

1. Introduction Steroids (androgen and estrogen) and gonadotrophins (follicle stimulating hormone (FSH) and luteinizing hormone (LH)), are essential for the maintenance of the cyclical pattern of ovarian folliculogenesis and female fertility. In mice, the roles of steroidogenic and gonadotrophic hormones in reproductive development and function have been studied by specific hormone alterations and the generation of receptor deficient models (Findlay et al., 2001; Drummond et al., 2002; Britt and Findlay, 2002; Abel et al., 2003; Cheng et al., 2002; Couse et al., 2003; Fisher et al., 1998; Krege et al., 1998; Kumar et al., 1997; Risma et al., 1995, 1997; Lubahn et al., 1993; Walters et al., 2007). However, the precise role of each hormone in various ovarian functions has not been fully elucidated, although it is clear that alterations in the actions of these hormones can alter fertility (Abel et al., 2003; Risma et al., 1995, 1997; Nilson et al., 2000; Britt et al., 2000, 2001; Toda et al., 2001).

夽 The financial support of the National Health and Medical Research Council of Australia (Regkeys 241000, 494802, 198705 and 441101) is acknowledged. ∗ Corresponding author at: Prince Henry’s Institute of Medical Research, P.O. Box 5152, Clayton, Victoria 3168, Australia. Tel.: +61 3 9594 3562; fax: +61 3 9594 6125. E-mail address: [email protected] (S.H. Liew).

Estrogen has been known for a long time to influence fertility (Hisaw, 1947). It modulates steroidogenesis (Roberts and Skinner, 1990), promotes granulosa cell proliferation (Robker and Richards, 1998a,b), facilitates the responsiveness of follicles to gonadotrophins (Robker and Richards, 1998a,b; Richards et al., 1979; Roy and Albee, 2000) and maintains ovarian follicular development in general (Drummond and Findlay, 1999). The aromatase knockout mouse model which lacks estrogen, offers important insights into the role of estrogen in ovarian function. Female ArKO mice are infertile (failure to ovulate); they have elevated levels of circulating gonadotrophins and testosterone, severely underdeveloped uteri (Fisher et al., 1998; Britt et al., 2000, 2001), increased adiposity (Jones et al., 2001, 2000), and apparent sex-reversal of the ovarian somatic cells; that is granulosa cells become Sertoli-like cells (Britt et al., 2002, 2004). E2 (17␤-estradiol) replacement in ArKO mice partially restored folliculogenesis, suppressed serum gonadotrophins to within the normal range and increased uterine weight (Britt et al., 2002, 2004) suggesting that the ovarian phenotype in the ArKO mouse is due to the lack of estrogen alone. However, these results do not dismiss the possibility that the ArKO ovarian phenotype is not due entirely to the lack of E but could also involve elevated gonadotrophins or testosterone. We hypothesised that it is the absence of estrogen (direct) and not the elevated levels of gonadotrophins and androgens (indirect) which is responsible for the ArKO ovarian phenotype.

0303-7207/$ – see front matter. Crown Copyright © 2010 Published by Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.mce.2010.05.003

S.H. Liew et al. / Molecular and Cellular Endocrinology 327 (2010) 56–64

In order to test this, we compared the effects of E2 replacement, acyline (GnRH antagonist to lower gonadotrophins) or flutamide (anti-androgen) treatment on reproductive parameters of the female ArKO mouse compared to wildtype (WT) and placebo treated mice. Ovarian function was evaluated by quantification of antral follicles, hemorrhagic cysts and corpora lutea (Risma et al., 1995). To assess the extent of formation of Sertoli-like cells, a Sertoli cell specific protein, Sry-like HMG box protein 9 (Sox 9) was immunolocalised to the ovarian granulosa/Sertoli-like cells and was semi-quantified. Estrous cycle lengths and serum hormones were also measured. We conclude that the ovarian phenotype of the ArKO female mouse is due to the absence of estrogen and to the elevated circulating levels of gonadotrophins, possibly LH, but not to elevated levels of testosterone. 2. Materials and methods 2.1. Animals WT and ArKO (Fisher et al., 1998) mice on a J129/C57B6 background were maintained under specific pathogen-free (SPF) conditions, on a 12L:12D regimen and fed ad libitum a soy free mouse chow (Glen Forrest Stockfeeders, Western Australia). All animal procedures were approved by a Monash University Animal Ethics Committee and were carried out in accordance with the Australian Code of Practice for the Care and Use of Animals for Scientific Purposes. 2.2. Treatments WT and ArKO mice 16 weeks of age (n = 8 per group) were used after confirming that the ArKO ovarian phenotype was characterised by the presence of containing Sertoli-like cells, elevated levels of FSH, LH and testosterone and an absence of corpora lutea (Risma et al., 1995) as described previously (Fisher et al., 1998; Britt et al., 2000, 2001, 2002).

57

peroxidase activity was blocked by immersion in 3% hydrogen peroxide in MilliQ water for 5 min, after which the slides were washed twice in Tris buffered saline (TBS). The sections were incubated in a blocking solution, 5% normal goat serum in a humidified chamber at room temperature for 30 min and subsequently incubated with rabbit anti-Sox9 antibody (sc-20095; Santa Cruz; 1:100) overnight at 4 ◦ C. After washing the slides three times in TBS, sections were incubated with rabbit anti-goat serum (1:500) for 1 h, followed by staining using a Vectastain ABC kit (Vector Laboratories, Burlingame, CA). Negative control sections received normal rabbit serum. Immature mouse testis was used as a positive control for the Sox 9 antibody. All slides were masked prior to the study and analysed blind. 2.8. Quantification of corpora lutea, antral follicles, hemorrhagic cysts and Sertoli-like filled cords All counting was performed using a 10× objective on an Olympus BX-50 microscope (Tokyo, Japan). Estimates of numbers of antral follicle, hemorrhagic cysts, CL and Sertoli-like filled cords per group were analysed using analySIS Professional Imaging software, version 5.0. Every 10th section was used and follicles with a visible oocyte nucleus, CL and Sertoli-like filled cords were counted only if they appeared in one section but not in the consecutive section. In order to minimize the structures not to be counted more than once, all pictures taken during quantification were spread on computer monitor to validate the quantification. 2.9. Hormone assays Established RIAs were used to measure LH and FSH levels in serum using reagents supplied by the NIADDK (Britt et al., 2000). The lower limits of detection were 1.13 ng/ml (FSH) and 0.10 ng/ml (LH). The intra-assay CVs for FSH and LH assays were 8.0% and 7.3%, respectively, calculated using a pool of normal mouse serum. Serum E2 and testosterone were measured in single assays using commercially available kits and protocols from Diagnostic Systems Laboratories (E2 , DSL-4800 ultra-sensitive; T, DSL-4100; Webster, TX). The lower limits of detection were 5 pg/ml (E2 ) and 45 pg/ml (T). The intra-assay CVs for E2 and T assays were 4.9% and 6.7%, respectively, calculated using a pool of normal mouse serum.

2.3. Estrogen 2.10. Statistical analysis Pellets containing either 0.05 mg of 17␤-estradiol (E2 ) or placebo (Innovative Research of America, Sarasota, FL) were placed subcutaneously for 21 days in 16week-old WT and ArKO mice as previously described (Jones et al., 2000; Britt et al., 2002). 2.4. Acyline The GnRH antagonist, acyline, was obtained from the National Institutes of Health (Bethesda, MD) and freshly prepared in 5% mannitol immediately before use. Subcutaneous (sc) injection of either 1.5 mg/kg/week of acyline (Porter et al., 2006) or placebo (5% mannitol) were administered weekly for 21 days to 16-week-old WT and ArKO mice. 2.5. Flutamide Sc pellets containing either 3 mg of flutamide or placebo (Innovative Research of America, Sarasota, FL) were implanted for 21 days in 16-week-old WT and ArKO mice as described previously (Cheng et al., 2002). 2.6. Tissue collections and processing Vaginal smears collected daily during the treatment period were stained using Diff Quick Stain (Lab Aids, Narrabeen, Australia). Animals were killed at the end of the treatment period. The abdomen was opened and approximately 1 mL of blood was collected from the inferior vena cava. Blood samples were centrifuged (10 min, 3000 rpm) at room temperature and the serum stored at −20 ◦ C for subsequent hormone analysis. Both ovaries and uterine horns from each animal were collected and weighed. One ovary was immersion-fixed in 10% formalin. Fixed tissue was processed through a graded series of alcohols and embedded in paraffin wax. Serial 5 ␮m sections were cut, and every 10th section was stained with haematoxylin and eosin. The intervening sections were used for immunohistochemistry. 2.7. Immunohistochemical localisation of Sox9 Sox 9 immunolocalisation was used to identify and quantify Sertoli-like filled cord structures in the ovary (n = 4 mice/group). Every 10th section of each ovary was stained with Sox9 according to a method modified from Ikeda et al. (2008). Tissue sections (5 ␮m) were deparaffinized in histolene and put through a graded series of descending concentrations of ethanol. Antigen retrieval was performed by microwaving slides for 10 min in 0.01 M citrate buffer (pH 6.0). Endogenous

Data are presented as mean ± SEM. Statistical analysis was performed using SigmaStat statistical software version 2.1 (Jandel Corporation, San Rafael, CA, USA). Data were subjected to a one-way ANOVA, and significance was determined using a Tukey’s post hoc test. Data was log transformed in the case of unequal variance. Significance (P < 0.05) was determined using the Student–Newman–Keuls test.

3. Results 3.1. Pattern of estrous cycles Placebo-treated ArKO mice are acyclic as described previously (Britt et al., 2000) with vaginal smears showing persistent diestrus. The placebo-treated WT mice have normal estrous cycles. Vaginal smears from WT and ArKO mice treated with E2 indicated persistent estrus; acyline- and flutamide-treated ArKO mice did not cycle and were in constant diestrus or early estrus. Acyline-treated WT mice were also in constant diestrus or early estrus, whereas flutamide-treated WT mice were in constant post-estrus. 3.2. Body (Table 1) and reproductive organ (Fig. 1) weights ArKO mice were heavier than their WT counterparts (P = 0.015). E2 treatment had no significant effect on WT body weight but significantly decreased the body weight of ArKO mice (P = 0.028). There were no significant effects on body weight in the acyline and flutamide treatment groups compared to their respective placebo groups. All placebo-treated ArKO mice had decreased ovarian weights compared with placebo-treated WT animals (P = 0.001). E2 and flutamide treatment had no significant effect on ovarian weight in either WT or ArKO animals. Acyline treatment however, decreased ovarian weight in both WT (P < 0.001) and

58

S.H. Liew et al. / Molecular and Cellular Endocrinology 327 (2010) 56–64

Fig. 1. The effect of placebo (Pl), E2 , acyline (Acy) and flutamide (Flu) treatment on mean ± SEM reproductive organ weights (mg) in WT (closed bar) and ArKO (KO) (open bar) female mice (n = 8 animals per group). Different letters denote statistical significance between treatments at P < 0.05.

ArKO (P < 0.001) animals compared to their respective placebo groups. All placebo-treated ArKO mice had reduced uterine weights (P < 0.001) compared with placebo-treated WT mice. E2 treatment increased uterine weight in both WT and ArKO animals (P < 0.05) with ArKO uterine weights returning to levels equivalent to placebo-treated Wt mice. There were no significant effects of either acyline or flutamide on uterine weights relative to their respective placebo groups.

Table 1 The effects of placebo, E2 , acyline and flutamide treatment on body weight in WT and ArKO mice. WT (g)

ArKO (g)

Placebo E2 treated

24.3 ± 0.8 24.3 ± 1.0

31.8 ± 2.4a 27.1 ± 1.7b

Placebo Acyline treated

24.3 ± 1.1 25.6 ± 1.3

28.0 ± 1.2 29.0 ± 1.9

Placebo Flutamide treated

23.7 ± 0.6 24.6 ± 1.0

27.8 ± 1.4 26.8 ± 1.4

Results expressed as mean ± SEM body weight (g) (n = 8 animals per group). Different letters denote statistical significance between treatments at P < 0.05.

3.3. Serum gonadotrophins (Fig. 2) and steroid hormones (Fig. 3) 3.3.1. FSH Placebo-treated ArKO mice had serum FSH levels on average of 8-fold above those of placebo-treated WT mice (P < 0.001). E2 and acyline treatment significantly reduced the levels of serum FSH (P < 0.001) in ArKO mice, whereas, flutamide treatment had no significant effect. Note that ArKO mice treated with E2 had FSH levels comparable with WT mice treated with either E2 or placebo pellets. 3.3.2. LH Placebo-treated ArKO mice had serum LH levels on average of 17-fold above those of placebo-treated WT mice (P < 0.001). E2 and acyline treatment decreased serum LH levels in ArKO mice (P < 0.001) to the same level as placebo-treated WT mice. Flutamide treatment also decreased LH levels but not to levels comparable with placebo-treated WT mice. 3.3.3. E2 Note that serum E2 levels in the placebo-, acyline- and flutamide-treated ArKO mice were either at or below the lower limits of assay detection, and were therefore considered as background. E2 treatment significantly elevated serum E2 levels in both

S.H. Liew et al. / Molecular and Cellular Endocrinology 327 (2010) 56–64

59

Fig. 2. The effect of placebo (Pl), E2 , acyline (Acy) and flutamide (Flu) treatment on mean ± SEM serum gonadotropins (ng/ml) in WT (closed bar) and ArKO (KO) (open bar) female mice (n = 8 animals per group). Different letters denote statistical significance between treatments at P < 0.05. The dotted line represents the lower limit of detection of the assay.

WT (35-fold) and ArKO (198-fold) mice (P < 0.001). However, acyline treatment significantly reduced serum E2 levels in WT mice (P < 0.05), whereas no significant effects on E2 levels were observed in acyline-treated ArKO, or flutamide-treated WT and ArKO mice. 3.3.4. T Placebo-treated ArKO mice had serum T levels on average 2-fold above those of placebo-treated WT mice (P < 0.001). E2 and acyline treatment decreased serum T levels in ArKO mice (P < 0.001) to levels measured in placebo-treated WT mice, whereas, flutamide treatment had no significant effect on serum T levels in either genotype. 3.4. Gross ovarian morphology (Fig. 4) WT mice treated with either E2 or flutamide did not cause any obvious changes in gross ovarian morphology compared with placebo-treated WT mice (Fig. 4A, C, and G). Normal growing follicles and CL were present in the ovaries of WT mice. When ArKO mice treated with either acyline or flutamide, there was impaired

follicular development and the presence of hemorrhagic cysts. In agreement with previous studies (Britt et al., 2002, 2004), E2 treatment partially restored follicular development and in some cases, ovulation as evidenced by the presence of CL in the ovaries of E2 treated ArKO mice (Fig. 4D). No CL were observed in acyline-treated WT or ArKO mice ovaries (Fig. 4E and F). 3.5. Immunohistochemical localisation of Sox9 (Fig. 5) Sox9 was immunolocalised to Sertoli-like filled cords within tubule-type structures of the ArKO ovaries (Fig. 5A and D) and to Sertoli cells in mouse testes (Fig. 5B and E) used as a positive control. Sox9 immunostaining was undetectable in WT ovaries (Fig. 5C and F). 3.6. Quantification of antral follicles, CL, hemorrhagic cysts and Sertoli-like filled cords Antral follicles (Fig. 6A). All placebo-treated ArKO mice had decreased numbers of antral follicles (≥4-fold) compared with

60

S.H. Liew et al. / Molecular and Cellular Endocrinology 327 (2010) 56–64

Fig. 3. The effect of placebo (Pl), E2 , acyline (Acy) and flutamide (Flu) treatment on mean ± SEM serum E2 and testosterone (T) (ng/ml) in WT (closed bar) and ArKO (KO) (open bar) female mice (n = 8 animals per group). Different letters denote statistical significance between treatments at P < 0.05. The dotted line represents the lower limits of detection of the assay.

placebo-treated WT mice. E2 , acyline and flutamide treatment had no significant effect on the number of antral follicles in either WT or ArKO mouse ovaries.

and acyline-treated ArKO mice ovaries, respectively, compared to placebo treated ArKO mice. There was no change in the number of Sertoli-like cell cords in flutamide-treated ArKO mice.

3.6.1. CL (Fig. 6B) There were no significant effects on the number of CL in E2 and flutamide-treated WT mice compared with placebo-treated WT mice. Nevertheless, CL were absent from acyline-treated WT ovaries, and ovaries of ArKO mice treated with acyline or flutamide. E2 -treatment restored CL in 23% of ArKO ovaries.

4. Discussion

3.6.2. Hemorrhagic cysts (Fig. 6C) No hemorrhagic cysts were present in WT ovaries regardless of treatment. However, the number of hemorrhagic cysts/ovary in the placebo-treated ArKO mice (Britt et al., 2002), was significantly reduced by 80% and 45% after treatment with E2 and acyline, respectively. There was no effect of flutamide. 3.6.3. Sertoli-like filled cords (Fig. 7) Using Sox9 as a marker, we showed the incidence of Sertoli-like filled cords was significantly reduced by 62% and 45% in E2 -treated

In this study, hormonal manipulation and phenotypic analyses were undertaken to determine in vivo whether the absence of estrogen (direct) or the elevated levels of gonadotrophins or androgens (indirect) were responsible for the ArKO ovarian phenotype. Similar to previous studies (Britt et al., 2002, 2004), E2 replacement in ArKO mice decreased gonadotrophin levels to within the WT normal range, increased uterine weight, and substantially improved ovarian follicular profiles through a decrease in the number of hemorrhagic cysts and Sertoli-like filled-cords. Furthermore, ovulation was restored in some ArKO mice after E2 replacement. Although no CL were found in acyline-treated ArKO ovaries, the ArKO ovarian phenotype was partly improved as shown by a decrease in the number of hemorrhagic cysts and Sertoli-like filled cords. Flutamide treatment however did not alter the ArKO ovarian phenotype. Therefore, we have demonstrated that the pronounced ovarian

S.H. Liew et al. / Molecular and Cellular Endocrinology 327 (2010) 56–64

61

Fig. 4. The effect of placebo (A and B), E2 (C and D), acyline (E and F) and flutamide (G and H) treatment on the ovarian morphology of WT and ArKO female mice. Placebo, E2 and flutamide-treated WT ovaries (A, C, and G) have a normal complement of growing follicles (*) and corpora lutea (Risma et al., 1995) (†). Ovaries of placebo, E2 , acyline and flutamide-treated ArKO mice (B) show impaired follicular development and hemorrhagic cysts (). Follicular growth improved when ArKO mice were treated with E2 (D). In some cases, ovulation was restored in ArKO mice, as evidenced by the presence of CL (†). No CL were observed in acyline-treated WT and ArKO ovaries (E and F).

phenotype observed in the adult ArKO mouse is due to the lack of E2 action and to the elevated levels of gonadotrophins, possibly LH, but not to elevated testosterone. 4.1. Estrogen replacement WT and ArKO mice treated with E2 were maintained in a state of persistent estrus and had increased uterine weights due to constant exposure of plasma E2 released from the pellet into the circulation. In the mouse, estrogens negatively regulate gonadotrophin release (Kolibianakis et al., 2005) and thus it is not surprising that E2 reduced the serum levels of FSH and LH to within the normal range in ArKO mice. Folliculogenesis and subsequent ovulation was significantly improved in ArKO mice treated with E2 , as indicated by improved follicular growth and the appearance of CL in some ArKO ovaries. In addition, the number of hemorrhagic cysts decreased in E2 -treated ArKO ovaries, possibly due to the decrease

in serum LH levels (Nilson et al., 2000) and the presence of E2 (see below). Sox9 expression is strongly upregulated in Sertoli cells soon after the expression of SRY begins; whereas it is downregulated in the ovary (Sekido and Lovell-Badge, 2008). However, the ovaries of female ArKO mice contain cords filled with Sertoli-like cells and express Sox9 mRNA similar to those observed in WT testis (Britt et al., 2002). Hence, we extended this finding by quantifying the Sertoli-like filled cords in the ArKO ovaries using the Sertoli cell specific marker Sox9. We found that the number of Sertoli-like filled cords in ArKO ovaries was significantly decreased by E2 treatment. This suggests that estrogen acts as a suppressor of Sox9 expression either via an indirect effect on regulatory elements involved in Sox9 action and/or directly via the presence of an estrogen responsive element in the Sox9 gene (Liew et al., 2009). The exact mechanism of estrogen action on the Sertoli-like cells thought to be formed by transdifferentiation of granulosa cells (Dupont et al., 2000) and

62

S.H. Liew et al. / Molecular and Cellular Endocrinology 327 (2010) 56–64

Fig. 5. Positive immunohistochemical stain for (brown) for Sox9 on Sertoli-like cell within ArKO ovary ((A) ×100 and (D) ×40), WT testis ((B) ×100 and (E) ×40). Staining is absent within the WT ovary ((C) ×100 and (F) ×40).

Sox9 expression is still unclear, although the fact that granulosa and interstitial cells express estrogen receptors (Schomberg et al., 1999) points to a direct action of estrogen on the gonadal somatic cells. This is supported by observations in the ER␣␤KO mouse, where Sertoli-like cells were also observed in the ovaries (Dupont et al., 2000). A role of the oocyte in these actions of estrogen on the phenotype of gonadal somatic cells has not been reported. 4.2. Acyline treatment The effectiveness of acyline treatment in reducing serum FSH and LH levels in female ArKO mice was observed as early as day 4 of treatment (data not shown). In addition, acyline-treated WT and ArKO mice were maintained in a state of persistent early estrus, consistent with the lack of ovulation and an absence of CLs. The uterine weights of the WT and ArKO mice remained unchanged after acyline treatment, whereas, the ovarian weights were decreased when compared to their control counterparts. Although serum levels of E2 were decreased by acyline treatment, the levels were still sufficient to maintain uterine weight. The decrease in gonadotrophic hormones and absence of the ovulatory surge, is the most likely reason for the absence of CL in the ovaries. Similar to E2 replacement, acyline treatment also decreased the number of hemorrhagic cysts and the appearance of Sertoli-like filled cords. This could be due to a reduction in the levels of one or both gonadotrophins driving ovarian cells (see below) or an increase in apoptotic activity. However, we did not observe any increase in apoptotic activity by terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assay in E2 -treated and acyline treated ArKO mice compared to WT (data not shown). The development of hemorrhagic cysts from follicles in ArKO mice ovaries has previously been attributed to the elevated levels of LH, since these cystic structures are also observed in the ovaries of mice overexpressing LH (Risma et al., 1995; Nilson et al., 2000) and estrogen receptor-alpha knockout (ERKO) mice which also have elevated LH (Schomberg et al., 1999). Transgenic human FSH mice (Tg-FSH) do not have elevated levels of LH (McTavish et al., 2007). A recent study (Lee et al., 2009) also provides evidence supporting the view that exogenous gonadotrophin hyperstimula-

tion induces follicular cysts in theca-specific estrogen receptor-␣ (Esr1) knockout (thEsr1KO) mouse ovaries. However, this phenotype was prevented by GnRH antagonist treatment in ␣ERKO mice, which lowered LH levels, reduced the incidence of hemorrhagic cysts and restored ovarian morphology to that of WT (Couse et al., 1999). Interestingly, the reduced serum FSH and LH levels in ArKO mice treated with acyline did not completely restore a WT phenotype in the ArKO ovary as hemorrhagic cysts were still present. These data therefore suggest that the development of hemorrhagic cysts in ArKO mice may not solely be due to the elevated levels of LH, but also to the absence of estrogen. In support of this, it should be noted that phytoestrogens can reduce the abundance of hemorrhagic cysts in ArKO mice, without decreasing LH levels (Britt et al., 2005), and Tg-FSH mice have normal circulating LH and E2 levels and no hemorrhagic cysts (McTavish et al., 2007). Nevertheless, the presence of hemorrhagic cysts of varying sizes may reflect the sensitivity of follicles to the altered hormonal milieu in the absence of estrogen and could be follicle specific. The mechanisms underlying the development and size of hemorrhagic cysts within the ovary remain unclear. 4.3. Flutamide treatment Androgen receptors (AR) are present in the oocyte, granulosa cells and theca cells of rodent ovaries (Hirai et al., 1994; Tetsuka and Hillier, 1996; Tetsuka et al., 1995). Female AR knockout mice are subfertile (Walters et al., 2009) highlighting the important role of androgen in maintaining female fertility. In addition, Cheng et al. (2002) have shown that arrested follicular maturation correlated with a high level of expression of AR in ER␤ knockout (BERKO) ovaries and that these effects can be reversed with the antiandrogen flutamide. However, in the present study there were no obvious differences in the ovarian phenotype between flutamidetreated WT and ArKO mice, and placebo-treated WT and ArKO mice, respectively. Interestingly, flutamide-treated ArKO mice still contained hemorrhagic cysts and Sertoli-like filled cords similar to the placebo-treated ArKO mice, suggesting that hemorrhagic cysts and the formation of Sertoli-like filled cords formation in the ArKO ovary does not involve the AR signaling pathway.

S.H. Liew et al. / Molecular and Cellular Endocrinology 327 (2010) 56–64

63

Fig. 7. Mean ± SEM of Sertoli-like filled cords of ArKO (KO) mice treated with either placebo (Pl), E2 , acyline (Acy) or flutamide (Flu). Different letters denote statistical significance between treatments at P < 0.05.

Acknowledgements We would like to thank Prof. Evan Simpson for the supply of animals, Dr. Mai Sarraj for her assistance with the Sox 9 immunostaining procedure and Susan Hayward from Monash Institute of Medical Research for serum E2 , T, FSH and LH measurements. References

Fig. 6. Mean ± SEM of (A) antral follicles and (B) CL of WT (closed bar) and ArKO (KO) (open bar) mice treated with either placebo (Pl), E2 , acyline (Acy) or flutamide (Flu). Mean ± SEM of (C) hemorrhagic cysts of ArKO mice treated with either placebo (Pl), E2 , acyline (Acy) or flutamide (Flu). Different letters denote statistical significance between treatments at P < 0.05.

In summary, the ovarian phenotype of the ArKO female mouse is due to the absence of estrogen and the elevated circulating levels of LH. We conclude that this study provides evidence for a role of estrogen in maintaining the female phenotype of the eutherian ovary. The robust ArKO mouse model somewhat simulate pathologies of human genetic aberrations of estrogen action, due to single or double mutations in various exons of the CYP19 gene encoding aromatase (Morishima et al., 1995; Mullis et al., 1997; Conte et al., 1994; Ito et al., 1993). However, the application of this knowledge to human fertility and infertility will require further research. The ArKO mouse model therefore offers an opportunity to identify estrogen responsive genes in the ovary that may be responsible for the actions of estrogen.

Abel, M.H., Huhtaniemi, I., Pakarinen, P., Kumar, T.R., Charlton, H.M., 2003. Agerelated uterine and ovarian hypertrophy in FSH receptor knockout and FSHbeta subunit knockout mice. Reproduction 125, 165–173. Britt, K.L., Findlay, J.K., 2002. Estrogen actions in the ovary revisited. J. Endocrinol. 175, 269–276. Britt, K.L., Drummond, A.E., Cox, V.A., Dyson, M., Wreford, N.G., Jones, M.E., Simpson, E.R., Findlay, J.K., 2000. An age-related ovarian phenotype in mice with targeted disruption of the Cyp 19 (aromatase) gene. Endocrinology 141, 2614–2623. Britt, K.L., Drummond, A.E., Dyson, M., Wreford, N.G., Jones, M.E., Simpson, E.R., Findlay, J.K., 2001. The ovarian phenotype of the aromatase knockout (ArKO) mouse. J. Steroid Biochem. Mol. Biol. 79, 181–185. Britt, K.L., Kerr, J., O’Donnell, L., Jones, M.E., Drummond, A.E., Davis, S.R., Simpson, E.R., Findlay, J.K., 2002. Estrogen regulates development of the somatic cell phenotype in the eutherian ovary. FASEB J. 16, 1389–1397. Britt, K.L., Stanton, P.G., Misso, M., Simpson, E.R., Findlay, J.K., 2004. The effects of estrogen on the expression of genes underlying the differentiation of somatic cells in the murine gonad. Endocrinology 145, 3950–3960. Britt, K.L., Simpson, E.R., Findlay, J.K., 2005. Effects of phytoestrogens on the ovarian and pituitary phenotypes of estrogen-deficient female aromatase knockout mice. Menopause 12, 174–185. Cheng, G., Weihua, Z., Mäkinen, S., Mäkelä, S., Saji, S., Warner, M., Gustafsson, J.A., Hovatta, O., 2002. A role for the androgen receptor in follicular atresia of estrogen receptor beta knockout mouse ovary. Biol. Reprod. 66, 77–84. Conte, F.A., Grumbach, M.M., Ito, Y., Fisher, C.R., Simpson, E.R., 1994. A syndrome of female pseudohermaphrodism, hypergonadotropic hypogonadism, and multicystic ovaries associated with missense mutations in the gene encoding aromatase (P450arom). J. Clin. Endocrinol. Metab. 78, 1287–1292. Couse, J.F., Bunch, D.O., Lindzey, J., Schomberg, D.W., Korach, K.S., 1999. Prevention of the polycystic ovarian phenotype and characterization of ovulatory capacity in the estrogen receptor-alpha knockout mouse. Endocrinology 140, 5855–5865. Couse, J.F., Yates, M.M., Walker, V.R., Korach, K.S., 2003. Characterization of the hypothalamic-pituitary–gonadal axis in estrogen receptor (ER) Null mice reveals hypergonadism and endocrine sex reversal in females lacking ERalpha but not ERbeta. Mol. Endocrinol. 17, 1039–1053. Drummond, A.E., Findlay, J.K., 1999. The role of estrogen in folliculogenesis. Mol. Cell. Endocrinol. 151, 57–64. Drummond, A.E., Britt, K.L., Dyson, M., Jones, M.E., Kerr, J.B., O’Donnell, L., Simpson, E.R., Findlay, J.K., 2002. Ovarian steroid receptors and their role in ovarian function. Mol. Cell. Endocrinol. 191, 27–33. Dupont, S., Krust, A., Gansmuller, A., Dierich, A., Chambon, P., Mark, M., 2000. Effect of single and compound knockouts of estrogen receptors alpha (ERalpha) and beta (ERbeta) on mouse reproductive phenotypes. Development 127, 4277–4291. Findlay, J.K., Britt, K., Kerr, J.B., O’Donnell, L., Jones, M.E., Drummond, A.E., Simpson, E.R., 2001. The road to ovulation: the role of oestrogens. Reprod. Fertil. Dev. 13, 543–547.

64

S.H. Liew et al. / Molecular and Cellular Endocrinology 327 (2010) 56–64

Fisher, C.R., Graves, K.H., Parlow, A.F., Simpson, E.R., 1998. Characterization of mice deficient in aromatase (ArKO) because of targeted disruption of the cyp19 gene. Proc. Natl. Acad. Sci. U.S.A. 95, 6965–6970. Hirai, M., Hirata, S., Osada, T., Hagihara, K., Kato, J., 1994. Androgen receptor mRNA in the rat ovary and uterus. J. Steroid Biochem. Mol. Biol. 49, 1–7. Hisaw, F.L., 1947. Development of the Graafian follicle and ovulation. Physiol. Rev. 27, 95–119. Ikeda, Y., Tanaka, H., Esaki, M., 2008. Effects of gestational diethylstilbestrol treatment on male and female gonads during early embryonic development. Endocrinology 149 (8), 3970–3979. Ito, Y., Fisher, C.R., Conte, F.A., Grumbach, M.M., Simpson, E.R., 1993. Molecular basis of aromatase deficiency in an adult female with sexual infantilism and polycystic ovaries. Proc. Natl. Acad. Sci. U.S.A. 90, 11673–11677. Jones, M.E., Thorburn, A.W., Britt, K.L., Hewitt, K.N., Wreford, N.G., Proietto, J., Oz, O.K., Leury, B.J., Robertson, K.M., Yao, S., Simpson, E.R., 2000. Aromatase-deficient (ArKO) mice have a phenotype of increased adiposity. Proc. Natl. Acad. Sci. U.S.A. 97, 12735–12740. Jones, M.E., Thorburn, A.W., Britt, K.L., Hewitt, K.N., Misso, M.L., Wreford, N.G., Proietto, J., Oz, O.K., Leury, B.J., Robertson, K.M., Yao, S., Simpson, E.R., 2001. Aromatase-deficient (ArKO) mice accumulate excess adipose tissue. J. Steroid Biochem. Mol. Biol. 79, 3–9. Kolibianakis, E.M., Papanikolaou, E.G., Fatemi, H.M., Devroey, P., 2005. Estrogen and folliculogenesis: is one necessary for the other? Curr. Opin. Obstet. Gynecol. 17, 249–253. Krege, J.H., Hodgin, J.B., Couse, J.F., Enmark, E., Warner, M., Mahler, J.F., Sar, M., Korach, K.S., Gustafsson, J.A., Smithies, O., 1998. Generation and reproductive phenotypes of mice lacking estrogen receptor beta. Proc. Natl. Acad. Sci. U.S.A. 95, 15677–15682. Kumar, T.R., Wang, Y., Lu, N., Matzuk, M.M., 1997. Follicle stimulating hormone is required for ovarian follicle maturation but not male fertility. Nat. Genet. 15, 201–204. Lee, S., Kang, D.W., Hudgins-Spivey, S., Krust, A., Lee, E.Y., Koo, Y., Cheon, Y., Gye, M.C., Chambon, P., Ko, C., 2009. Theca-specific estrogen receptor-alpha knockout mice lose fertility prematurely. Endocrinology 150, 3855–3862. Liew, S.H., Sarraj, M.A., Drummond, A.E., Findlay, J.K., 2009. Estrogen dependent gene expression in the mouse ovary. In: Proceedings of the 52nd Endocrine Society of Australia Annual Scientific Meeting, ESA. abstr. 377, Adelaide, Australia. Lubahn, D.B., Moyer, J.S., Golding, T.S., Couse, J.F., Korach, K.S., Smithies, O., 1993. Alteration of reproductive function but not prenatal sexual development after insertional disruption of the mouse estrogen receptor gene. Proc. Natl. Acad. Sci. U.S.A. 90, 11162–11166. McTavish, K.J., Jimenez, M., Walters, K.A., Spaliviero, J., Groome, N.P., Themmen, A.P., Visser, J.A., Handelsman, D.J., Allan, C.M., 2007. Rising follicle-stimulating hormone levels with age accelerate female reproductive failure. Endocrinology 148, 4432–4439. Morishima, A., Grumbach, M.M., Simpson, E.R., Fisher, C., Qin, K., 1995. Aromatase deficiency in male and female siblings caused by a novel mutation and the physiological role of estrogens. J. Clin. Endocrinol. Metab. 80, 3689–3698. Mullis, P.E., Yoshimura, N., Kuhlmann, B., Lippuner, K., Jaeger, P., Harada, H., 1997. Aromatase deficiency in a female who is compound heterozygote for two new point mutations in the P450arom gene: impact of estrogens on hypergonadotropic hypogonadism, multicystic ovaries, and bone densitometry in childhood. J. Clin. Endocrinol. Metab. 82, 1739–1745.

Nilson, J.H., Abbud, R.A., Keri, R.A., Quirk, C.C., 2000. Chronic hypersecretion of luteinizing hormone in transgenic mice disrupts both ovarian and pituitary function, with some effects modified by the genetic background. Recent Prog. Horm. Res. 55 (discussion 89–91). Porter, K.L., Shetty, G., Meistrich, M.L., 2006. Testicular edema is associated with spermatogonial arrest in irradiated rats. Endocrinology 147, 1297–1305. Richards, J.S., Uilenbroek, J.T., Jonassen, J.A., 1979. Follicular growth in the rat: a reevaluation of the roles of FSH and LH. Adv. Exp. Med. Biol. 112, 11–26. Risma, K.A., Clay, C.M., Nett, T.M., Wagner, T., Yun, J., Nilson, J.H., 1995. Targeted overexpression of luteinizing hormone in transgenic mice leads to infertility, polycystic ovaries, and ovarian tumors. Proc. Natl. Acad. Sci. U.S.A. 92, 1322–1326. Risma, K.A., Hirshfield, A.N., Nilson, J.H., 1997. Elevated luteinizing hormone in prepubertal transgenic mice causes hyperandrogenemia, precocious puberty, and substantial ovarian pathology. Endocrinology 138, 3540–3547. Roberts, A.J., Skinner, M.K., 1990. Estrogen regulation of thecal cell steroidogenesis and differentiation: thecal cell-granulosa cell interactions. Endocrinology 127, 2918–2929. Robker, R.L., Richards, J.S., 1998a. Hormone-induced proliferation and differentiation of granulosa cells: a coordinated balance of the cell cycle regulators cyclin D2 and p27Kip1. Mol. Endocrinol. 12, 924–940. Robker, R.L., Richards, J.S., 1998b. Hormonal control of the cell cycle in ovarian cells: proliferation versus differentiation. Biol. Reprod. 59, 476–482. Roy, S.K., Albee, L., 2000. Requirement for follicle-stimulating hormone action in the formation of primordial follicles during perinatal ovarian development in the hamster. Endocrinology 141, 4449–4456. Schomberg, D.W., Couse, J.F., Mukherjee, A., Lubahn, D.B., Sar, M., Mayo, K.E., Korach, K.S., 1999. Targeted disruption of the estrogen receptor-alpha gene in female mice: characterization of ovarian responses and phenotype in the adult. Endocrinology 140, 2733–2744. Sekido, R., Lovell-Badge, R., 2008. Sex determination involves synergistic action of SRY and SF1 on a specific Sox9 enhancer. Nature 453, 930–934. Tetsuka, M., Hillier, S.G., 1996. Androgen receptor gene expression in rat granulosa cells: the role of follicle-stimulating hormone and steroid hormones. Endocrinology 137, 4392–4397. Tetsuka, M., Whitelaw, P.F., Bremner, W.J., Millar, M.R., Smyth, C.D., Hillier, S.G., 1995. Developmental regulation of androgen receptor in rat ovary. J. Endocrinol. 145, 535–543. Toda, K., Takeda, K., Okada, T., Akira, S., Saibara, T., Kaname, T., Yamamura, K., Onishi, S., Shizuta, Y., 2001. Targeted disruption of the aromatase P450 gene (Cyp19) in mice and their ovarian and uterine responses to 17beta-oestradiol. J. Endocrinol. 170, 99–111. Walters, K.A., Allan, C.M., Jimenez, M., Lim, P.R., Davey, R.A., Zajac, J.D., Illingworth, P., Handelsman, D.J., 2007. Female mice haploinsufficient for an inactivated androgen receptor (AR) exhibit age-dependent defects that resemble the AR null phenotype of dysfunctional late follicle development, ovulation, and fertility. Endocrinology 148, 3674–3684. Walters, K.A., McTavish, K.J., Seneviratne, M.G., Jimenez, M., McMahon, A.C., Allan, C.M., Salamonsen, L.A., Handelsman, D.J., 2009. Subfertile female androgen receptor knockout mice exhibit defects in neuroendocrine signaling, intraovarian function, and uterine development but not uterine function. Endocrinology 150, 3274–3282.