0013-7227/04/$15.00/0 Printed in U.S.A.

Endocrinology 145(4):1880 –1888 Copyright © 2004 by The Endocrine Society doi: 10.1210/en.2003-0952

Aromatase-Knockout Mouse Carrying an EstrogenInducible Enhanced Green Fluorescent Protein Gene Facilitates Detection of Estrogen Actions in Vivo KATSUMI TODA, YASUSHI OKADA, MOHAMAD ZUBAIR, KEN-ICHIRO MOROHASHI, TOSHIJI SAIBARA, AND TERUHIKO OKADA Departments of Molecular Genetics (K.T.), Animal Laboratory for Investigation (Y.O.), Gastroenterology and Hepatology (T.S.), and Anatomy and Cell Biology (T.O.), Kochi Medical School, Nankoku, Kochi 783-8505, Japan; Division of Cell Differentiation, National Institute for Basic Biology (M.Z., K.M.), Okazaki, Aichi 444-0085, Japan; and Core Research for Evolutional Science and Technology, Japan Science and Technology Corp. (M.Z., K.M.), Kawaguchi 332-0012, Japan Aromatase is an enzyme that converts androgen to estrogen in the gonads and also at extragonadal sites, including the brain. In this study we developed a transgenic mouse that carries an enhanced green fluorescent protein (EGFP) gene inducible by estrogen through an estrogen response element to facilitate detection of estrogen actions in vivo. The expression of EGFP in aromatase-deficient (Arⴚ/ⴚ) female mice was significantly suppressed at the pituitary gland, ovary, uterus, and gonadal fat pad and was induced by dietary 17␤-estradiol to wild-type (Arⴙ/ⴙ) levels or higher. These results demonstrate that the expression of the EGFP gene is tissue selective and estrogen dependent in vivo. Employing this transgenic mouse, we examined whether estrogen synthesis in the extragonadal sites

E

STROGENS PLAY A central role in the regulation of reproductive, endocrine, as well as skeletal functions (1– 4). They also participate in the developmental control and functional maintenance of the central nervous system (5, 6). Aromatase (CYP19) is an enzyme responsible for the conversion of androgens to estrogens (7). To produce estrogendeficient conditions in vivo for studying the physiological roles of estrogens, aromatase knockout (Ar⫺/⫺) mice were generated by targeted disruption of Cyp19 (8 –10). As expected, Ar⫺/⫺ females are completely infertile. Histology of the reproductive tract demonstrated follicular depletion and hemorrhage formation in the ovaries and diminution of uterine weight (10, 11). Supplementation with 17␤-estradiol (E2) or xenoestrogen apparently supported the development of ovarian follicles and recovered uterine diminution, but did not ameliorate the reproductive failure in Ar⫺/⫺ females (10, 12). As estrogen exerts irreversible organizational actions during ontogeny in addition to the transient activational effects on reproductive functions at various sites within the hypothalamic-pituitary-ovarian axis (13, 14), exogenous supplementation with estrogens at adulthood does not seem sufficient to correct the reproductive failure in Ar⫺/⫺ feAbbreviations: Arⴙ/ⴙ, Wild-type; Ar⫺/⫺, aromatase knockout; E2, 17␤-estradiol; EGFP, enhanced green fluorescent protein; ER, estrogen receptor; ERE, estrogen response element; GAPDH, glyceraldehyde-3phosphate dehydrogenase. Endocrinology is published monthly by The Endocrine Society (http:// www.endo-society.org), the foremost professional society serving the endocrine community.

is necessary for reproduction in female mice. When ovaries of Arⴚ/ⴚ mice were replaced with Arⴙ/ⴙ ovaries, a significant induction of EGFP expression in the pituitary gland and uterus was observed. Histological examinations showed the presence of antral follicles in the replaced ovaries, indicating that the transplants are functional in Arⴚ/ⴚ mice. After crossing with males, three of 10 Arⴚ/ⴚfemales with Arⴙ/ⴙ ovaries became pregnant and fed their pups. Collectively, these observations indicate that estrogen synthesis in the ovary is sufficient for supporting female reproduction, and that infertility of Arⴚ/ⴚ females is primarily due to a defect in estrogen synthesis in the ovary. (Endocrinology 145: 1880 –1888, 2004)

males. In place of exogenous supplementation with estrogens, transplantation of Ar⫹/⫹ ovaries into ovariectomized Ar⫺/⫺ female mice, which is expected to produce a physiological hormonal milieu similar to that occurring in Ar⫹/⫹ mice, might be the method of choice to address the issue of whether the failure in reproduction of Ar⫺/⫺ females is due to a defect in estrogen synthesis in the ovary. Estrogens acts primarily through estrogen receptors (ERs), which are ligand-activated nuclear transcription factors (15, 16). Upon ligand binding, ERs bind to estrogen response elements (EREs) in target gene promoters, leading to transcriptional modulation of the genes. Nevertheless, ligandbound ERs can regulate transcription of genes by other mechanisms (15). In this study we developed a line of transgenic mouse in which an enhanced green fluorescent protein (EGFP) gene, a variant form of the wild-type GFP gene (17), was incorporated into the genome to be expressed in an estrogen-dependent manner via the ERE. As EGFP is an autofluorescent protein, the expression of the protein can easily be visualized without the processing of samples (18). Measurement of the expression levels of the protein in Ar⫺/⫺ mice with or without estrogen allows us to easily evaluate estrogen actions via the ERE in vivo. Employing this animal model, we established that Ar⫺/⫺ females are readily responsive to estrogen derived from the transplanted ovaries of Ar⫹/⫹ mice as well as to dietary estrogen. We then examined the consequences of ovarian transplantation on the reproductive activity of Ar⫺/⫺ females by mating them with fertile males. The findings indicate that the failure in repro-

1880

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2017. at 15:02 For personal use only. No other uses without permission. . All rights reserved.

Toda et al. • ArKO Mice with Estrogen-Inducible EGFP

Endocrinology, April 2004, 145(4):1880 –1888 1881

duction of Ar⫺/⫺ females is primarily due to a defect in estrogen synthesis in the ovary.

⫺/⫺

Ar

Materials and Methods mice and preparation of a chow diet

Cyp19 was disrupted by homologous recombination as previously described (10). All animals were maintained on a 12-h light, 12-h dark cycle at 22–25 C, and unless otherwise noted, they were given water and a standard rodent chow diet (NMF, Oriental Yeast, Tokyo, Japan) ad libitum. To examine effects of E2 on EGFP expression, mice at 8 wk of age were fed for 30 d a chow diet containing 10 ppm E2, prepared as follows: 10 mg E2 (Sigma-Aldrich Corp., Tokyo, Japan; purity minimum, 98%) was dissolved in 100 ml acetone, which was then impregnated into 1 kg rodent chow. To examine the effects of tamoxifen on EGFP expression, Ar⫹/⫹ mice at 8 wk of age were fed for 30 d a chow diet containing 100 ppm tamoxifen, prepared as follows: 0.1 g tamoxifen citrate (Nacarai Tesuque, Kyoto, Japan) was dissolved in 100 ml 50% ethanol and impregnated into 1 kg rodent chow. The animal experiments were carried out according to the guidelines of institutional animal regulations.

Construction and selection of transgene A plasmid, nestinhsp68EGFP (19), was digested with EcoRV and NdeI to remove an enhancer element of the nestin gene, but not a basal promoter sequence of the heat shock protein 68 gene with a sequence for the protein-coding region of EGFP (BD Biosciences, Clontech, Palo Alto, CA). Synthetic ERE (5⬘-TCAGGTCACACAGTGACCTGA-3⬘) was then inserted by blunt-end ligation into the EcoRV/NdeI-digested nestinhsp68EGFP, resulting in the generation of plasmids containing various numbers of ERE. The estrogen-dependent expression of EGFP of the plasmids was examined by transient expression analysis with an expression vector for rat ER␣, pSV2RcER (20), in human endometrial carcinoma cells (21). A construct containing four copies of ERE was selected for the generation of transgenic mice (Fig. 1A), because it drove minimal constitutive and strong estrogen-dependent reporter gene expression in the in vitro assay.

FIG. 1. Schematic view of the transgene and Southern blot analysis of genomic DNA of the transgenic mice generated. A, The transgene (2 kb) consists of four copies of the synthetic ERE (5⬘-TCAGGTCACACAGTGACCTGA-3⬘; 4XERE) and a minimal heat shock protein 68 gene promoter (HSP68) linked to the EGFP cDNA (EGFP) and simian virus 40 polyadenylation sequence (PA). B, Total genomic Southern blot analysis. Genomic DNA from Arⴙ/ⴙ mice (lanes 1 and 4), Ar⫹/⫺ mice (lanes 2 and 5), or Ar⫺/⫺ mice (lanes 3 and 6) was digested completely with either PstI (lanes 1–3) or HindIII (lanes 4 – 6). A radiolabeled fragment of cDNA coding for EGFP was used as a probe. Note that regardless of the genotype and the generation (not shown), the same hybridization pattern was observed, indicating that the transgene is incorporated at a single locus in the genome.

Generation and selection of EGFP transgenic mice

Fluorescent images were recorded using a cooled digital color chargedcoupled device camera (C4742-95, Hamamatsu Photonics, Hamamatsu, Japan) mounted on the stereomicroscope.

Transgenic mice were generated according to the method described by Hogan et al. (22). A purified DNA fragment obtained by digestion of the selected transgene with SalI was injected into 170 fertilized eggs of a mouse strain, B6C3F1 (23). The manipulated eggs were transferred into foster mothers. Fifteen (nine males and six females) of 74 pups were identified as independent founders that incorporated the transgene with various copy numbers in the genome, as examined by PCR using tail DNA with the following primers: P1-EGFP, 5⬘-GAGCTGGACGGCGACGTAAAC-3⬘; and P2-EGFP, 5⬘-CACCTTGATGCCGTTCTTCTGC-3⬘ (23). The transgenic founder mice were outcrossed with C57BL6/J to yield F1 offspring. Transmission of the transgene was examined by PCR, and the expression of EGFP was determined by observation of tissues from the F1 offspring under a fluorescence stereomicroscope. F1 offspring with the functional EGFP gene were then crossed with Ar⫹/⫺ mice with a C57BL6/J genetic background to yield Ar⫹/⫺ mice with the EGFP gene, which were intercrossed to generate Ar⫺/⫺ mice with the EGFP gene. The EGFP expression levels in Ar⫺/⫺ mice were compared with those in Ar⫹/⫹ mice under a fluorescence stereomicroscope to select lines of founders. These analyses resulted in the selection of line KT2. Mice with the EGFP gene derived from line KT2 and with heterozygosity for the Cyp19 locus were intercrossed more than five generations, during which the transgene was stably transmitted, resulting in the generation of Ar⫹/⫺ mice homozygous for the EGFP gene, which were used to produce Ar⫹/⫹ and Ar⫺/⫺ mice for the EGFP expression studies.

Visualization of EGFP protein Imaging of EGFP fluorescence was made using a fluorescence stereomicroscope (MZ FLIII; Leica, Deerfield, NJ) with a filter set, GFP2, composed of a 480/440-nm excitation filter and a 510-nm barrier filter.

Quantification of EGFP fluorescence To prepare tissue extracts containing EGFP, tissues (⬍0.2 g) from the transgenic mice were homogenized in 2 ml 10 mm Tris-HCl (pH 7.2) containing 1 mm EDTA and 0.2% sodium dodecyl sulfate (wt/wt) using a Polytron PT1200 homogenizer (KINEMATICA, Lucerne, Switzerland). After centrifugation at 2000 ⫻ g for 10 min at 4 C, the supernatants containing EGFP were recovered for fluorometry using a spectrofluorometer (RF-5300PC, Shimadzu Corp., Kyoto, Japan). The filters used were 480 nm for excitation and 510 nm for emission. Purified recombinant EGFP (BD Clontech, Palo Alto, CA) was employed as a standard to calibrate the amounts of EGFP expressed in the tissues of the transgenic mice. Under our experimental conditions, 3.47 ng purified recombinant EGFP produced 1 arbitrary unit of fluorescent intensity. The amounts of EGFP in the extracts were expressed as picograms of EGFP per microgram of total protein of the tissue extracts examined.

Ovarian transplantation Bilateral ovarian transplantation was performed essentially as described previously (24). Donor and recipient mice at 6 –7 wk of age were anesthetized with pentobarbital. The ovarian fat pads were moved outside through a small longitudinal skin incision. The bursa was opened at one end, keeping most of the membrane intact, and the ovary was teased out. After an ovary to be grafted was pushed into the bursa, the ovary, oviduct, and uterus were placed back into the body cavity. The skin incisions were closed, and the mice were given appropriate postoperative care. The mice were analyzed 2 wk after the operation.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2017. at 15:02 For personal use only. No other uses without permission. . All rights reserved.

1882

Endocrinology, April 2004, 145(4):1880 –1888

Evaluation of reproductive ability The estrous cycle of the mice with ovarian transplants was examined from vaginal smears for 20 d, beginning 2 wk after ovarian transplantation. Then, two females were housed with a male known to be fertile to evaluate reproductive ability. To induce ovulation, Ar⫺/⫺ mice with Ar⫹/⫹ ovaries were injected ip with 5 IU pregnant mare’s serum gonadotropin (Serotropin, Teizo, Tokyo, Japan), and after 48 h, 5 IU human chorionic gonadotropin (Sigma-Aldrich Corp., St. Louis, MO) were administrated.

Determination of serum E2 Serum samples of three mice were pooled and used for determination of E2 concentration by RIA (10). Briefly, the pooled sera (1 ml) were extracted twice with 4 ml diethyl ether. The organic phase was further extracted once with 1 ml 0.1 m NaHCO3 and twice with 1 ml H2O. After drying the samples, RIA was performed, employing 0.2 pmol 2,4,6,7,16,17-3H-labeled E2 as tracer (specific activity, 5.7 TBq/mmol; Amersham Pharmacia Biotech, Little Chalfont, UK) and an anti-E2 antibody (CIDtech Research, Inc., Ontario Canada). Bound and free radiolabeled E2 were separated on a Microspin G-25 column (Amersham Pharmacia Biotech, Piscataway, NJ). E2 levels were assayed four times using different pools of serum samples. As the recovery of E2 after extraction was 67.4 ⫾ 8.6% when calculated with [14C]E2 as an internal tracer, the values obtained by the assay were corrected for the recovery. The detection limit was 2 pg/ml. The intra- and interassay coefficients of variation were 9.0% and 12.0%, respectively.

Gonadotropin assays The concentrations of serum FSH and LH were measured, respectively, by using a rat FSH ELISA test and a rodent LH ELISA test obtained from Endocrine Technologies, Inc. (Newark, CA). All samples for each hormone were assayed in one assay. The intraassay coefficient of variation was 5.7% for both FSH and LH.

Histological examination Ovaries from the mice were fixed in a solution of 10% buffered formalin for 24 h, dehydrated in graded ethanol, and embedded in paraffin. Sections (4 ␮m) were cut and stained with hematoxylin-eosin.

Southern and Northern blot analyses Genomic DNA (20 ␮g) from the mouse tail was cleaved to completion with either HindIII or PstI and electrophoresed on an 0.8% agarose gel (25). The DNA fragments were then transferred to a nylon membrane filter, and the filter was hybridized with the radiolabeled cDNA fragment coding for EGFP. Known amounts of the transgene used for the microinjection were used as a standard to calculate the copy numbers integrated in the genome of the transgenic mice. The signals were quantified using a Bioimage Analyzer BAS2000 (Fuji, Tokyo, Japan). Northern blot analyses were performed using 15 ␮g total RNA from the uteri according to a method previously described (25). cDNA fragments for lactoferrin (26) (gift from Dr. C. T. Teng, NIEHS, Research Triangle Park, NC) and those for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) were used as probes.

Statistical analysis Data are expressed as the mean ⫾ sem. The significance of differences for EGFP expression levels and for expression levels of lactoferrin mRNA in the uterus was analyzed using a one-way ANOVA, employing InStat software (GraphPad, Inc., San Diego, CA), and that for serum gonadotropin levels was analyzed using a Kruskal-Wallis test. P ⬍ 0.05 was considered significant.

Results Generation and selection of transgenic mice with EGFP expression in an estrogen-inducible manner

Among the 15 founder lines, the offspring of five lines (three males and two females) did not express functional

Toda et al. • ArKO Mice with Estrogen-Inducible EGFP

EGFP at all, indicating inactivation of the transgene. The remaining 10 founder lines were then screened on the basis of estrogen dependency of EGFP expression by comparing the fluorescent intensities in the uteri and ovaries of Ar⫹/⫹ mice with those in Ar⫺/⫺ mice. Of the two founder lines selected, line KT2 was employed for the present study because it expressed EGFP more strongly than the other line, KT5. Total genomic Southern blot analysis demonstrated that approximately 10 copies of the transgene were incorporated at a single locus in the genome (Fig. 1B). No other hybridization pattern was observed in progeny after repeated crossing, further supporting the idea that the transgene was incorporated at a single locus in the genome. Estrogen-dependent expression of EGFP in female mice in vivo

The expression levels of EGFP in the uterus, ovary, and pituitary gland were apparently higher in Ar⫹/⫹ than in Ar⫺/⫺ mice, as monitored by fluorescence stereomicroscopy (Fig. 2). Quantitative analysis was designed using 2-month-old females to identify tissues where the expression of EGFP is significantly suppressed in Ar⫺/⫺ mice. In this analysis, the fluorescent intensity in various tissues was normalized to the protein content (Fig. 3). High levels of the EGFP expression (⬎30 pg EGFP/␮g total protein) were detected in the adrenal gland, gonadal fat pad, and skeletal muscle in addition to the ovary, uterus, and pituitary gland in Ar⫹/⫹ females. In Ar⫺/⫺ females, the expression was significantly suppressed in the pituitary gland, ovary, uterus, gonadal fat pad, and skeletal muscle. In the adrenal gland, the level was decreased in Ar⫺/⫺ mice, but not significantly. Low levels of EGFP expression (⬍10 pg/␮g total protein) were detected in the cerebellum, thymus, liver, bone marrow, hypothalamus, heart, parotid, hippocampus, spleen, lung, and cortex without genotype dependency, except for bone marrow, where the expression was at a low level, but was significantly affected by estrogen. Finally, there was no detectable activity in the kidney in Ar⫹/⫹ and Ar⫺/⫺ mice, although the kidney has been reported to have large numbers of genes regulated by E2 (27). Estrogen-dependent expression of the EGFP gene was further evaluated by feeding Ar⫺/⫺ mice a chow diet containing 10 ppm E2. Dietary estrogen induced EGFP expression in the pituitary gland, ovary, uterus, gonadal fat pad, and bone marrow of Ar⫺/⫺ mice nearly to the levels in Ar⫹/⫹ mice or higher (Fig. 3). Thus, these results established that the expression of the EGFP gene is regulated by estrogen in those tissues in Ar⫺/⫺ females in vivo. Interestingly, dietary E2 did not affect the expression level in skeletal muscle, but significantly increased it in the adrenal gland. When Ar⫹/⫹ mice were fed a chow diet containing E2, adrenal expression of the EGFP gene was also increased about 10-fold over that in control mice (Fig. 4). In contrast, dietary E2 suppressed EGFP expression in the uterus and gonadal fat pad of Ar⫹/⫹ mice. When Ar⫹/⫹ mice were fed a chow diet containing 100 ppm tamoxifen, a drug known to possess a tissue-selective, agonistic/antagonistic activity of estrogen (28), EGFP expression was significantly suppressed in the pituitary gland and

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2017. at 15:02 For personal use only. No other uses without permission. . All rights reserved.

Toda et al. • ArKO Mice with Estrogen-Inducible EGFP

Endocrinology, April 2004, 145(4):1880 –1888 1883

FIG. 2. Expression of the estrogen-regulated EGFP transgene in vivo in Ar⫹/⫹ or Ar⫺/⫺ mice. Unfixed tissues from Ar⫹/⫹ (A–C and G–I) or Ar⫺/⫺ (D–F) female mice at 8 wk of age were observed under a fluorescence stereomicroscope. The images show EGFP expression in the gonadal tract (A, D, and G), ovary (B, E, and H), and pituitary gland (C, F, and I). Images of tissues from Ar⫹/⫹ mice without the EGFP gene (G–I) are presented as a negative control.

FIG. 3. Tissue distribution of EGFP expression in the mouse. Ar⫹/⫹ mice fed a control chow diet (䡺), Ar⫺/⫺ mice fed a control chow diet (f), and Ar⫺/⫺ mice fed a chow diet containing 10 ppm E2 (o) were used to measure the expression of EGFP. Tissue extracts from the pituitary gland (Pi), adrenal gland (A), ovary (O), uterus (U), gonadal fat pad (F), muscle (M), cerebellum (Ce), thymus (Th), liver (Li), bone marrow (B), hypothalamus (Hy), heart (He), parotid (Pa), hippocampus (Hi), spleen (S), lung (Lu), cortex (Co), and kidney (K) were prepared by homogenization in 10 mM Tris-HCl (pH 7.4) containing 2 mM EDTA and 0.2% sodium dodecyl sulfate. The amounts of EGFP in the extracts were measured by fluorescent spectrophotometery with excitation at 480 nm and emission at 510 nm and were expressed as picograms of EGFP per microgram of total protein in each tissue examined. The analysis was performed with five animals per group. *, P ⬍ 0.05 vs. Ar⫺/⫺ mice fed a control chow diet. Error bars represent the SEM.

uterus, whereas expression in the other tissues, including the adrenal gland, ovary, and gonadal fat pad, was not affected (Fig. 4 and data not shown). These observations further sup-

port the idea that expression of the EGFP gene with the regulatory sequence of ERE is altered in an estrogen-dependent and tissue-selective manner in vivo.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2017. at 15:02 For personal use only. No other uses without permission. . All rights reserved.

1884

Endocrinology, April 2004, 145(4):1880 –1888

Toda et al. • ArKO Mice with Estrogen-Inducible EGFP

4). Nonetheless, the levels were apparently lower than those in intact Ar⫹/⫹ female mice (14.5 ⫾ 3.6 pg/ml; n ⫽ 4). The serum levels of FSH and LH reflected the feedback effects of estrogen actions occurring in the hypothalamic-pituitaryovarian axis. Both levels were significantly higher in intact Ar⫺/⫺ mice than in Ar⫹/⫹ mice (Fig. 5). The concentration of estrogen secreted from the transplanted Ar⫹/⫹ ovaries in Ar⫺/⫺ mice was sufficient to cause substantial reduction of serum FSH and LH levels. These observations demonstrated that although the E2 concentration is very low in Ar⫺/⫺ mice with Ar⫹/⫹ ovaries, the feedback loop in the hypothalamicpituitary-ovarian axis is functional. Induction of EGFP expression by Ar⫹/⫹ ovaries in Ar⫺/⫺ mice was apparent in the uterus and pituitary gland, as examined under fluorescence stereomicroscope (Fig. 6A). The results of quantitative analysis are shown in Fig. 6B. Northern blot analysis was performed to confirm estrogen action in the uterus. The expression of lactoferrin, a well known molecular marker of estrogen action in the uterus (26, 29), was suppressed in Ar⫺/⫺ mice compared with that in Ar⫹/⫹ mice (the level in Ar⫺/⫺ mice was ⬃30% that in Ar⫹/⫹ mice). The expression was elevated by the ovarian transplants, reaching 9.2-fold over the Ar⫺/⫺ level (Fig. 6C). These findings demonstrate that expression of the EGFP gene in Ar⫺/⫺females reflects the in vivo action of estrogen synthesized in and secreted from the transplanted Ar⫹/⫹ ovaries. Pregnancy of Ar⫺/⫺ females with Ar⫹/⫹ ovaries

FIG. 4. Tissue distribution of EGFP expression in Arⴙ/ⴙ females mice fed a diet containing E2 or tamoxifen. Ar⫹/⫹ mice at 8 wk of age were fed a control chow diet (䡺), a chow diet containing 10 ppm E2 (o), or a chow diet containing 100 ppm tamoxifen (1) for 30 d. The amounts of EGFP in the tissue extracts prepared from the pituitary gland (Pi), adrenal gland (A), ovary (O), uterus (U), and gonadal fat pad (F) were measured by fluorescent spectrophotometery with excitation at 480 nm and emission at 510 nm and were expressed as picograms of EGFP per microgram of total protein in each tissue examined. The analysis was performed with four animals per group. *, P ⬍ 0.05 vs. Ar⫹/⫹ mice fed a control chow diet. Error bars represent the SEM.

It is well established that estrogen is synthesized not only in the ovary, but also in other tissue sites, including the brain (30). Thus, we examined whether estrogens synthesized only in the ovary are sufficient to support reproduction using Ar⫺/⫺ females whose ovaries were replaced with Ar⫹/⫹ ovaries. Firstly, the estrous cycle of the mice with the transplants was examined from vaginal smears. Ar⫹/⫹ females receiving Ar⫹/⫹ ovaries showed a regular estrous cycle of 5to 6-d intervals when examined 2 wk after the operation. This interval is consistent with that observed in intact Ar⫹/⫹ females (31). Examination of vaginal smears demonstrated

In male mice, high levels of expression were observed in the testis, pituitary, and adrenal gland, and low levels were found in the muscle, hypothalamus, heart, parotid, and cerebellum. Nevertheless, expression of the EGFP gene in neither tissue revealed significant genotype dependency, except for the gonadal fat pad, where the expression level was 30-fold higher in Ar⫺/⫺ mice than in Ar⫹/⫹ mice. EGFP induction in Ar⫺/⫺ mice with Ar⫹/⫹ ovaries

We next assessed the responsiveness of EGFP gene expression to estrogen derived from Ar⫹/⫹ ovaries bilaterally transplanted into the ovarian bursa of ovariectomized Ar⫺/⫺ mice. The serum level of E2 in intact Ar⫺/ mice was less than the detection limit, as described previously (8, 10). The level in Ar⫺/⫺ mice with Ar⫹/⫹ ovaries was 5.4 ⫾ 0.8 pg/ml (n ⫽ 4), which was nearly the same as that in ovariectomized Ar⫹/⫹ mice receiving Ar⫹/⫹ ovaries (6.4 ⫾ 0.4 pg/ml; n ⫽

FIG. 5. Serum levels of FSH and LH. Concentrations of FSH (A) and LH (B) were measured by ELISA using serum collected from intact Arⴙ/ⴙ mice (lane 1; n ⫽ 8), Ar⫺/⫺ mice (lane 2; n ⫽ 7), Ar⫹/⫹ mice with transplanted Ar⫹/⫹ ovaries (lane 3; n ⫽ 11), and Ar⫺/⫺ mice with transplanted Ar⫹/⫹ ovaries (lane 4; n ⫽ 11). *, P ⬍ 0.05 vs. the mice of all other groups. Error bars represent the SEM.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2017. at 15:02 For personal use only. No other uses without permission. . All rights reserved.

Toda et al. • ArKO Mice with Estrogen-Inducible EGFP

Endocrinology, April 2004, 145(4):1880 –1888 1885

FIG. 6. Expression of EGFP and lactoferrin mRNA in Ar⫺/⫺ mice with transplanted Ar⫹/⫹ ovaries. A, EGFP expression in the uterus (a and c) and pituitary gland (b and d) of intact Ar⫺/⫺ mice (a and b) or Ar⫺/⫺ mice with Ar⫹/⫹ ovaries (c and d). The images were obtained by fluorescence stereomicroscopy. B, Amounts of EGFP in the pituitary gland (Pi), uterus (U), gonadal fat pad (F), adrenal gland (A), and bone marrow (B) of intact Ar⫺/⫺ mice (䡺) or Ar⫺/⫺ mice with Ar⫹/⫹ ovaries (f) were quantified by fluorescent spectrophotometery and expressed as picograms of EGFP per microgram of total protein. *, P ⬍ 0.05 vs. intact Ar⫺/⫺ mice. Error bars represent the SEM. C, Uterine expression of lactoferrin mRNA in intact mice or mice with ovarian transplants. The expression of the lactoferrin gene was analyzed by Northern blot hybridization using 20 ␮g total RNAs from the uteri of Arⴙ/ⴙ mice (lane 1), intact Ar⫺/⫺ mice (lane 2), sham-operated Ar⫺/⫺ mice (lane 3), or Ar⫺/⫺ mice with ovarian transplants of Ar⫹/⫹ mice (lane 4; upper panel). The same membrane filter was reprobed with a radiolabeled cDNA fragment coding for GAPDH to examine loading variations between the samples (middle panel). The signals of lactoferrin mRNA were quantified using a radioactive image analyzer (BAS 2000) and normalized relative to GAPDH mRNA levels to calculate the relative intensity. The data are presented as the fold induction over the relative intensity in intact Ar⫹/⫹ mice (bottom panel). *, P ⬍ 0.05 vs. intact Ar⫺/⫺ mice.

FIG. 7. Histological analysis of transplanted ovaries. The ovaries of Ar⫹/⫹ mice were transplanted into the ovariectomized Ar⫹/⫹ mice (A) or into ovariectomized Ar⫺/⫺ mice (B). The ovaries of Ar⫺/⫺ mice were transplanted into ovariectomized Ar⫺/⫺ mice (C). The ovaries were collected and processed for histological analysis 2 months after the operation. The sections were stained with hematoxylin-eosin. Bar, 500 ␮m.

that Ar⫺/⫺ mice with Ar⫹/⫹ ovaries showed an image for the estrous stage, but the cyclicity was irregular, with a mean interval of 6 d. For instance, one mouse stayed at the estrous stage for 4 d, and the other mouse repeated the metestrous and diestrous stages, then proceeded through the cycle. These observations indicate that the regulatory system controlling the estrous cycle is not fully recovered by transplantation of Ar⫹/⫹ ovaries in Ar⫺/⫺ females. Histology of transplanted Ar⫹/⫹ ovaries in Ar⫹/⫹ mice demonstrated the presence of follicles at various maturational stages, including primary to large antral follicles (Fig. 7A). We observed similar morphology in Ar⫹/⫹ ovaries transplanted into ovariectomized Ar⫺/⫺ mice (Fig. 7B). In contrast, Ar⫺/⫺ ovaries transplanted into ovariectomized Ar⫺/⫺ mice showed depletion of follicles, the presence of many atretic follicles, and formation of a large hemorrhage (Fig. 7C), which is a typical in ovaries with estrogen insufficiency (10, 11). These results indicate that although the cyclicity of the estrous cycle is not regular, folliculogenesis

seems to proceed normally in Ar⫹/⫹ ovaries transplanted into Ar⫺/⫺ mice. The recovery in reproductive ability was further confirmed by treatment of Ar⫺/⫺ females with Ar⫹/⫹ ovaries with pregnant mare’s serum gonadotropin, followed by injection of human chorionic gonadotropin, which resulted in the induction of ovulation from transplanted Ar⫹/⫹ ovaries (not shown). When Ar⫺/⫺ females with transplanted Ar⫹/⫹ ovaries were mated with fertile Ar⫹/⫹ males in a continuous manner, three of 10 Ar⫺/⫺ female mice with Ar⫹/⫹ ovaries became pregnant. Two of them delivered pups (three pups in one case and six pups in the other; Table 1) and successfully weaned them. Thus, these results together with the results showing EGFP expression indicate that a failure in reproduction of Ar⫺/⫺ females is primarily due to a defect in estrogen synthesis in the ovary, whereas the other control mechanisms, including those regulating the regularity of the estrous cycle, also need to be corrected for full recovery.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2017. at 15:02 For personal use only. No other uses without permission. . All rights reserved.

1886

Endocrinology, April 2004, 145(4):1880 –1888

Toda et al. • ArKO Mice with Estrogen-Inducible EGFP

TABLE 1. Fertility of ArKO mice with ovarian transplants Genotype

Recipient ⫹/⫹ ⫺/⫺ ⫺/⫺

Donor ⫹/⫹ ⫹/⫹ ⫺/⫺

No. of mice examined

No. of mice showed estrus

Regularity (interval, days)

No. of mice that got pregnant

Average litter size

7 10 4

6 8 0

Regular (5– 6) Irregular (4 –9) No estrous cycle

6 3 0

4.5 ⫾ 0.4 4.5 0

Discussion

In the present study a new line of transgenic mouse was developed, in which transcription of the EGFP gene is regulated by estrogen via the ERE and not via other responsive elements. Because EGFP can be visualized without processing the samples, the expression can be directly monitored in vivo (18). Analyses of the expression patterns of the reporter gene in Ar⫺/⫺ females demonstrated that the EGFP gene is expressed in a tissue-selective and estrogen-dependent manner, and that Ar⫺/⫺ females are responsive to the physiological concentrations of estrogen derived from transplanted Ar⫹/⫹ ovaries. Generally, it is well accepted that estrogen exerts its effects not only through the action of ERs, but also through other transcriptional regulatory factors in a variety of tissues in addition to nongenomic actions (15, 32). Thus, the present system does not necessarily reflect the full nature of the range of estrogen-dependent gene expression patterns in vivo. Nonetheless, expression of the EGFP gene was regulated by estrogen in the pituitary gland, ovary, uterus, and gonadal fat pad of Ar⫺/⫺ females, because the expressions were suppressed in the tissues of Ar⫺/⫺ mice compared with those in Ar⫹/⫹ mice and were increased significantly by dietary E2 in Ar⫺/⫺ mice. Dietary E2 caused attenuation of EGFP expression in the uterus and gonadal fat pad of Ar⫹/⫹ females, whereas it increased the expression in Ar⫺/⫺ mice. We observed that uterine expression of EGFP in Ar⫺/⫺ mice was increased approximately 6-fold over the control levels by sc injection of diethylstilbestrol at 2 ␮g/kg, but the degree of the increase was dose-dependently reduced at higher concentrations of diethylstilbestrol (3-fold increase at 20 ␮g/kg and 2-fold increase at 200 ␮g/kg; Toda, K., unpublished observations). Thus, attenuation of EGFP expression seems to occur depending on the amount of estrogen in a tissue-selective manner in vivo. EGFP expression in the adrenal glands of both Ar⫹/⫹ and Ar⫺/⫺ females is markedly increased by dietary E2. Histological study of the adrenal gland of Ar⫹/⫹ mice revealed that EGFP is exclusively localized to the adrenal cortex (Toda, K., unpublished observations). The presence of ERs in the region has been documented (33–35). Therefore, EGFP expression in the adrenal gland seems to be estrogen dependent in the present transgenic animal model, although basal expression levels are not statistically different between Ar⫹/⫹ and Ar⫺/⫺ mice. We failed to detect an estrogendependent expression of EGFP in other target tissues of estrogen, including the hypothalamus, implying that suboptimal conditions of E2 supplementation might be employed for those tissues. Two transgenic murine models containing estrogenregulatable reporter genes were recently generated indepen-

dently. One model contains a luciferase gene as a reporter gene, whose transcription was designed to be regulated by a thymidine kinase promoter in an estrogen-dependent manner (36), and the other contains the ␤-galactosidase gene as a reporter gene (37). Using ovariectomized transgenic mice, estrogen-dependent expressions of the reporter genes were examined and were found in the liver, lung, spleen, brain, and thymus (36) or in the liver, kidney, hypothalamus, and thyroid (37). The differences in the expression patterns of the reporter genes in vivo among transgenic mice might be attributable to different experimental conditions, including the structures of the transgenes (38) and the strains of mouse employed. Both copy numbers of the transgene in the genome and the site to which the gene is integrated modify its expression pattern in vivo. The stability of the reporter gene products in vivo and the manner of detecting the activities of the products are other factors that affect the regulatory manner of transgene expression. In the present study the transgene was introduced in Ar⫺/⫺ mice. There is therefore absolutely no estrogen present at the baseline. This may not have been the case in pervious studies. Despite the fact that ovariectomized females were used, these mice probably had the possibility of producing at least small amounts of estrogens through aromatization of adrenal androgens. Genotype-independent expression of the EGFP gene in the tissues examined in male mice was unexpected, because estrogen action is definitely required for lipid and glucose metabolism (39 – 41) as well as for reproduction in male mice (42). As suggested by Weihua et al. (43), 5␣-androstane3␤,17␤-diol, a possible endogenous metabolite of dihydrotestosterone, might be produced and functional as an estrogenic steroid to drive transcription of the EGFP gene in vivo in male mice under physiological conditions where serum testosterone and estrogen levels are elevated and lowered, respectively. In addition, genotype-independent expression of the EGFP gene in male mice might represent ligand-independent activities of ERs as previously described (15). We showed that Ar⫹/⫹ ovaries failed to drive a regular estrous cycle in Ar⫺/⫺ mice when transplanted at 6 –7 wk of age. This observation might represent permanent alterations in the hypothalamic-pituitary-ovarian axis due to a lack of estrogen actions during development of Ar⫺/⫺ females. Actions of estrogen in the central nervous system at the earlier stage of female life are proposed to be necessary to establish reproductive ability (44, 45). Ar⫺/⫺ mice with Ar⫹/⫹ ovaries displayed partial restoration of reproductive function despite the irregularity of the estrous cycle, indicating that the reproductive failure of Ar⫺/⫺ females is primarily due to a defect in estrogen synthesis in the ovary. Nonetheless, the pregnancy rate of Ar⫺/⫺ mice with Ar⫹/⫹ ovaries is apparently lower than that of Ar⫹/⫹ mice with Ar⫹/⫹ ovaries. The

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2017. at 15:02 For personal use only. No other uses without permission. . All rights reserved.

Toda et al. • ArKO Mice with Estrogen-Inducible EGFP

low pregnancy rate of Ar⫺/⫺ females with Ar⫹/⫹ ovaries might be attributable not only to the irregularity of the estrous cycle, but also to low levels of sexual behaviors of the mice, as the behaviors are dependent on estrogens in both males and females. Furthermore, defects in other processes required for pregnancy, including implantation, are conceivable. Because Ar⫹/⫹ ovaries do not survive well in Ar⫺/⫺ females when transplanted (Toda, K., unpublished observations), we are unable to argue that estrogens synthesized in extragonadal sites can support ovarian function. A recent study suggested that estrogen synthesized at extragonadal sites plays locally important roles in a paracrine/autocrine fashion (46). The experimental system described in the present study would open the way to study how estrogen secreted into the circulation from the ovaries affects hypothalamic-pituitary functions in female mice. Ar⫺/⫺ mice carrying the estrogen-inducible EGFP gene could be applicable in many research fields in addition to physiological studies of Ar⫺/⫺ mice. For instance, the expression levels of EGFP in the pituitary gland, ovary, uterus, gonadal fat pad, and adrenal gland can be used as an indicator of estrogen action in vivo, which will help us to develop novel medical drugs with tissue-selective antagonist/agonist activity of estrogen (28, 47, 48). Actually, we showed that tamoxifen suppressed EGFP expression in the pituitary gland and uterus without affecting expression in the adrenal gland, ovary, and gonadal fat pad of Ar⫹/⫹ females. Apparently, we need further analyses to clarify whether the effects of tamoxifen are due to its antagonistic activity or to the agonistic activity of estrogen actions in those tissues, because of the observation that E2 suppressed EGFP expression in the uterus and gonadal fat pad of Ar⫹/⫹ mice. Recently, Ar⫺/⫺ mice have been shown to be a suitable animal model for detecting xenoestrogens such as bisphenol A using histochemistry and Northern blot analyses (12). Ar⫺/⫺ mice with the estrogen-inducible EGFP gene seem to be more suited for the detection of xenoestrogens than Ar⫺/⫺ mice, as the presence or absence of estrogenic chemicals could be readily determined by fluorescence stereomicroscopy without processing of the samples. Acknowledgments We thank Dr. H. Okano (Keio University School of Medicine) for providing us with the plasmid nestinhsp68EGFP, and Dr. C. T. Teng (NIEHS, Research Triangle Park, NC) for providing us with cDNA for lactoferrin. Received July 28, 2003. Accepted December 5, 2003. Address all correspondence and requests for reprints to: Dr. Katsumi Toda, Department of Molecular Genetics, Kochi Medical School, Nankoku, Kochi 783-8505, Japan. E-mail: [email protected]. This work was conducted as part of research projects by the Japan Food Industrial Center.

References 1. Findlay JK, Britt K, Kerr JB, O’Donnell L, Jones ME, Drummond AE, Simpson ER 2001 The road to ovulation: the role of oestrogens. Reprod Fertil Dev 13:543–547 2. Hess RA, Bunick D, Bahr J 2001 Oestrogen, its receptors and function in the male reproductive tract. Mol Cell Endocrinol 178:29 –38 3. O’Donnell L, Robertson KM, Jones ME, Simpson ER 2001 Estrogen and spermatogenesis. Endocr Rev 22:289 –318

Endocrinology, April 2004, 145(4):1880 –1888 1887

4. Manolagas SC, Kousteni S, Jilka RL 2002 Sex steroids and bone. Recent Prog Horm Res 57:385– 409 5. McEwen B 2002 Estrogen actions throughout the brain. Recent Prog Horm Res 57:357–384 6. Wise PM, Dubal DB, Wilson ME, Rau SW, Bottner M, Rosewell KL 2001 Estradiol is a protective factor in the adult and aging brain: understanding of mechanisms derived from in vivo and in vitro studies. Brain Res Rev 37: 313–319 7. Simpson ER, Mahendroo MS, Means GD, Kilgore MW, Hinshelwood MM, Graham-Lorence S, Amarneh B, Ito Y, Fisher CR, Michael D, Mendelson CR, Bulun SE 1994 Aromatase cytochrome P450, the enzyme responsible for estrogen biosynthesis. Endocr Rev 15:342–355 8. Fisher CR, Graves KH, Parlow AF, Simpson ER 1998 Characterization of mice deficient in aromatase (ArKO) because of targeted disruption of the cyp19 gene. Proc Natl Acad Sci USA 95:6965– 6870 9. Honda S, Harada N, Ito S, Takagi Y, Maeda S 1998 Disruption of sexual behavior in male aromatase-deficient mice lacking exons 1 and 2 of the cyp19 gene. Biochem Biophys Res Commun 252:445– 449 10. 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 17␤-oestradiol. J Endocrinol 170:99 –111 11. Britt KL, Drummond AE, Cox VA, Dyson M, Wreford NG, Jones ME Simpson ER, Findlay JK 2000 An age-related ovarian phenotype in mice with targeted disruption of the Cyp 19 (aromatase) gene. Endocrinology 141:2614 – 2623 12. Toda K, Miyaura C, Okada T, Shizuta Y 2002 Dietary bisphenol A prevents ovarian degeneration and bone loss in female mice lacking the aromatase gene (Cyp19). Eur J Biochem 269:2214 –2222 13. Hsueh AJW, Billig H, Tsafriri A 1994 Ovarian follicle atresia: a hormonally controlled apoptotic process. Endocr Rev 15:707–724 14. Tobet SA 2002 Genes controlling hypothalamic development and sexual differentiation. Eur J Neurosci 16:373–376 15. Nilsson S, Makela S, Treuter E, Tujague M, Thomsen J, Andersson G, Enmark E, Pettersson K, Warner M, Gustafsson JA 2001 Mechanisms of estrogen action. Physiol Rev 81:1535–1565 16. Couse JF, Korach KS 1999 Estrogen receptor null mice: What have we learned and where will they lead us? Endocr Rev 20:358 – 417 17. Cormack BP, Valdivia RH, Falkow S 1996 FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173:33–38 18. Hadjantonakis AK, Nagy A 2001 The color of mice: in the light of GFP-variant reporters. Histochem Cell Biol 115:49 –58 19. Kawaguchi A, Miyata T, Sawamoto K, Takashita N, Murayama A, Akamatsu W, Ogawa M, Okabe M, Tano Y, Goldman SA, Okano H 2001 Nestin-EGFP transgenic mice: visualization of the self-renewal and multipotency of CNS stem cells. Mol Cell Neurosci 17:259 –273 20. Koike S, Sakai M, Muramatsu M 1987 Molecular cloning and characterization of rat estrogen receptor cDNA. Nucleic Acids Res 25:2499 –2513 21. Tada A, Sasaki H, Nakamura J, Yoshihama M, Terashima Y 1993 Aromatase activity and the effect of estradiol and testosterone on DNA synthesis in endometrial carcinoma cell lines. J Steroid Biochem Mol Biol 44:661– 666 22. Hogan B, Beddington R, Costantini F, Lacy E 1994 Manipulating the mouse embryo: a laboratory manual. 2nd ed. Cold Spring Harbor, NY: Cold Harbor Laboratory Press 23. Zubair M, Watanabe E, Fukada M, Noda M 2002 Genetic labeling of specific axonal pathways in the mouse central nervous system. Eur J Neurosci 5: 807– 814 24. Cunliffe-Beamer TL 1983 Biomethodology and surgical techniques. In: Foster HL, Small JD, Fox JG, eds. The mouse in biomedical research. New York: Academic Press; 401– 437 25. Sambrook J, Fritsch EF, Maniatis T 1989 Molecular cloning: a laboratory manual. 2nd ed. Cold Spring Harbor, NY: Cold Spring Harbor Laboratory Press 26. Liu YH, Teng CT 1991 Characterization of estrogen-responsive mouse lactoferrin promoter. J Biol Chem 266:21880 –21885 27. Jelinsky SA, Harris HA, Brown EL, Flanagan K, Zhang X, Tunkey C, Lai K, Lane MV, Simcoe DK, Evans MJ 2003 Global transcription profiling of estrogen activity: estrogen receptor ␣ regulates gene expression in the kidney. Endocrinology 144:701–710 28. McDonnell DP, Connor CE, Wijayaratne A, Chang CY, Norris JD 2002 Definition of the molecular and cellular mechanisms underlying the tissueselective agonist/antagonist activities of selective estrogen receptor modulators. Recent Prog Horm Res 57:295–316 29. Newbold RR, Hanson RB, Jefferson WN 1997 Ontogeny of lactoferrin in the developing mouse uterus: a marker of early hormone response. Biol Reprod 56:1147–1157 30. Lephart ED, Lund TD, Horvath TL 2001 Brain androgen and progesterone metabolizing enzymes: biosynthesis, distribution and function. Brain Res Rev 37:25–37 31. Rugh R 1990 The mouse: its reproduction and development. New York: Oxford University Press; 38 – 43

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2017. at 15:02 For personal use only. No other uses without permission. . All rights reserved.

1888

Endocrinology, April 2004, 145(4):1880 –1888

32. Kelly, MJ, Levin, ER 2001 Rapid actions of plasma membrane estrogen receptors. Trends Endocrinol Metab 12:152–156 33. Hirst JJ, West NB, Brenner RM, Novy MJ 1992 Steroid hormone receptors in the adrenal glands of fetal and adult rhesus monkeys. J Clin Endocrinol Metab 75:308 –314 34. Saunders PT, Maguire SM, Gaughan J, Millar MR 1997 Expression of oestrogen receptor ␤ (ER␤) in multiple rat tissues visualised by immunohistochemistry. J Endocrinol 154:13–16 35. Wu Z, Maric C, Roesch DM, Zheng W, Verbalis JG, Sandberg K 2003 Estrogen regulates adrenal angiotensin AT1 receptors by modulating AT1 receptor translation. Endocrinology 144:3251–3261 36. Ciana P, Di Luccio G, Belcredito S, Pollio G, Vegeto E, Tiveron LC, Maggi A 2001 Engineering of a mouse for the in vivo profiling of estrogen receptor activity. Mol Endocrinol 15:1104 –1113 37. Nagel SC, Hagelbarger JL, McDonell DP 2001 Development of an ER action indicator mouse for the study of estrogens, selective ER modulators (SERMs) and xenobiotics. Endocrinology 142:4721– 4728 38. Hall JM, McDonnell DP, Korach KS 2002 Allosteric regulation of estrogen receptor structure, function, and coactivator recruitment by different estrogen response elements. Mol Endocrinol 16:469 – 486 39. Nemoto Y, Toda K, Ono M, Fujikawa-Adachi K, Saibara T, Onishi S, Enzan H, Okada T, Shizuta Y 2000 Altered expression of fatty acid-metabolizing enzymes in aromatase-deficient mice. J Clin Invest 105:1819 –1825 40. Jones ME, Thorburn AW, Britt KL, Hewitt KN, Wreford NG, Proietto J, Oz OK, Leury BJ, Robertson KM, Yao S, Simpson ER 2000 Aromatase-deficient

Toda et al. • ArKO Mice with Estrogen-Inducible EGFP

41.

42.

43. 44.

45.

46. 47. 48.

(ArKO) mice have a phenotype of increased adiposity. Proc Natl Acad Sci USA 97:12735– 41270 Takeda K, Toda K, Saibara T, Nakagawa M, Saika K, Onishi S, Sugiura T, Shizuta Y 2003 Progressive development of insulin resistant phenotype in male mice lacking aromatase gene (Cyp19). J Endocrinol 176:237–246 Toda K, Okada T, Takeda, K, Akira S, Saibara T, Shiraishi M, Onishi S, Shizuta Y 2001 Oestrogen at neonatal stage is critical for reproductive ability of male mice as revealed by supplementation with 17␤-oestradiol to aromatase (CYP19) gene knockout mice. J Endocrinol 168:455– 463 Weihua Z, Andersson S, Cheng G, Simpson ER, Warner M, Gustafsson JA 2003 Update on estrogen signaling. FEBS Lett 546:17–24 Toran-Allerand CD 1984 On the genesis of sexual differentiation of the general nervous system: morphogenetic consequences of steroidal exposure and possible role of alpha-fetoprotein. Prog Brain Res 61:63–98 Bakker J, Honda S, Harada N, Balthazart J 2002 The aromatase knock-out mouse provides new evidence that estradiol is required during development in the female for the expression of sociosexual behaviors in adulthood. J Neurosci 22:9104 –9112 Simpson ER 2002 Aromatization of androgens in women: current concepts and findings. Fertil Steril 77:6 –10 Dhingra K 2001 Selective estrogen receptor modulation: the search for an ideal hormonal therapy for breast cancer. Cancer Invest 19:649 – 659 Park WC, Jordan VC 2002 Selective estrogen receptor modulators (SERMS) and their roles in breast cancer prevention. Trends Mol Med 8:82– 88

Endocrinology is published monthly by The Endocrine Society (http://www.endo-society.org), the foremost professional society serving the endocrine community.

The Endocrine Society. Downloaded from press.endocrine.org by [${individualUser.displayName}] on 14 January 2017. at 15:02 For personal use only. No other uses without permission. . All rights reserved.