2017, 64 (11), 1055-1061

Original Effects of ovarian hormones on GPR120 mRNA expression in mouse pituitary gonadotrophs

Ryutaro Moriyama, Kaho Ueda and Chikaya Deura

Department of Life Science, Kindai University, Higashiosaka, 577-8502, Japan

Abstract. GPR120 is a G-protein-coupled receptor that is activated by long-chain fatty acids. In our previous study, GPR120 expression was detected in gonadotrophs of the mouse anterior . It is well known that the function of cells is largely under the influence of circulating sex steroids. Thus, in the present study, we investigated the modulatory roles of the ovarian hormones, estrogen (E2) and progesterone (P), on the expression levels of GPR120 mRNA in mouse pituitary glands. GPR120 mRNA expression levels in the pituitary gland were increased after ovariectomy or P treatment, and were decreased after the administration of E2. Simultaneous injection of E2 and P interfered with the action of E2 on GPR120 mRNA expression. The GnRH antagonist, Cetrotide, did not inhibit the increase in GPR120 expression in ovariectomized (OVX) animals. In addition, immunohistochemistry revealed that more than 95.4% of GPR120 immunoreactive cells colocalized with the β (LHβ) in the anterior pituitary gland of intact, ovariectomized (OVX), estradiol-primed OVX (OVX+E2), or progesterone-primed OVX (OVX+P) animals. Furthermore, GPR120 mRNA expression levels were not significantly different in the pituitary gland of females throughout the ovarian cycle. It is suggested that low levels of P may mask the inhibitory effect of estradiol on the synthesis of GPR120 in the estrous stage in intact animals. These results demonstrate that ovarian hormones may directly regulate GPR120 expression in the reproductive cycle at the pituitary level.

Key words: GPR120, Pituitary, Gonadotroph, Progesterone, Estrogen

FATTY ACIDS are not only nutritional components hormones, such as the luteinizing hormone (LH) and but also act as signaling molecules. G-protein-coupled the follicle-stimulating hormone (FSH). However, the receptor 120 (GPR120) is a receptor for long-chain long-chain fatty acid receptor agonist GW9508 did not fatty acids. It has been reported that GPR120 medi- affect LH secretion [5] or the transcription of the - ates adipogenesis and glucagon-like peptide-1 secre- otropin subunit genes [6] in an LβT2 gonadotropic tion during lipid metabolism [1], and is also involved cell line. Nevertheless, fasting induced an increase in in the anti-inflammatory effects of macrophages [2]. GPR120 mRNA expression in mouse gonadotrophs [4] Furthermore, GPR120 is considered a fatty acid sen- and free fatty acids (FFAs) directly suppressed Fshb sor that controls energy balance, since the dysfunc- mRNA expression in the LβT2 gonadotropic cell line tion of GPR120 leads to obesity in both humans and [7], suggesting that GPR120 could serve as a sensor rodents [3]. In our previous study, GPR120 expression for the detection of serum fatty acid levels in pituitary was detected in gonadotrophs of the mouse anterior gonadotrophs. pituitary gland [4]. This result suggested that GPR120 It is well known that the function of anterior pitu- could play a role in regulating gonadal function by itary cells is largely under the influence of circulat- affecting the synthesis and/or secretion of gonadotropic ing sex steroids in both males and females [8]. In

Submitted Mar. 6, 2017; Accepted Jul. 20, 2017 as EJ17-0102 E-mail: [email protected] Released online in J-STAGE as advance publication Aug. 19, 2017 Abbreviations: GPR120, G-protein-coupled receptor 120; OVX, Correspondence to: Ryutaro Moriyama, Laboratory of ovariectomized; E2, estradiol; P, progesterone; FSH, follicular- Environmental Physiology, Department of Life Science, School stimulating hormone; LH, luteinizing hormone; NGS, normal goat of Science and Engineering, Kindai University, 3-4-1 Kowakae, serum; H-P-G axis, hypothalamic-pituitary-gonadal axis; GnRH, Higashiosaka, Osaka 577-8502, Japan. -releasing hormone. ©The Japan Endocrine Society 1056 Moriyama et al. the present study, we hypothesized that the ovar- RNA extraction and cDNA synthesis ian hormones, estrogen and progesterone (P), regu- Total RNA was prepared from pituitary glands using late GPR120 expression levels in pituitary gonado- TRI Reagent (Sigma-Aldrich Co.), and treated with trophs. To test this hypothesis, we studied the effect RNase-free DNase I to remove genomic DNA contam- on GPR120 expression levels when ovarian hormones ination. cDNA was synthesized using a Superscript were depleted through ovariectomy and replaced by II kit with an oligo(dT)12-18 primer (all reagents were estradiol-17β (E2) or P in ovariectomized (OVX) mice. purchased from Life Technologies Co., Carlsbad, CA).

Materials and Methods Real-time PCR GPR120 mRNA expression levels were deter- Animals mined by real-time PCR, using the SYBR Premix Ex Adult female ICR mice that were obtained from Taq II master mix (Takara Bio Inc., Tokyo, Japan) Japan SLC, Inc. (Hamamatsu, Japan) through a local containing SYBR Green I, and run on a 7,500 Real- distributor (Shimizu Laboratory Supplies Co., Ltd., time PCR System (Applied Biosystems, Darmstadt, Kyoto, Japan) were individually housed in a con- Germany). The following conditions were used: trolled environment (12 h light and 12 h dark; lights on denaturation at 95 °C for 30 s and amplification at 0700 h; temperature, 24 ± 2 °C). Animals had free by cycling 40 times at 95 °C for 5 s, 60 °C for 34 s. access to food (Labo-MR stock, Nihon Nosan Kogyo Data were analyzed using the standard curve method Co., Yokohama) and water until the day of sampling. [12], and normalized to the L19 ribosomal protein Female mice (10-12 weeks of age; n = 6 per group) gene, which was used as a reference gene. The for- that had at least two consecutive regular 4-day estrous ward and reverse primers (Nippon EGT, Toyama, cycles were used in this study. The estrous cycle stage Japan) used for mouse GPR120 (NM_181748) were was monitored by daily vaginal smears that were taken 5ʹ-TCGCTGTTCAGGAACGAATG-3ʹ and 5ʹ-CACCA around 1,000 h. Some mice were sacrificed by rapid GAGGCTAGTTAGCTG-3ʹ, respectively, and those decapitation at the four different stages of the estrus used for ribosomal protein L19 were 5ʹ-CC cycle, and their pituitary glands were collected form AAGAAGATTGACCGCCATA-3ʹ and 5ʹ-CAGCTTGT 1,000 h to 1,200 h for RNA extraction. The study was GGATGTGCTCCAT-3ʹ, respectively. approved by the Committee on Animal Experiments of Kindai University for the care and use of experimen- Tissue preparation for immunohistochemistry tal animals. Tissues were fixed and processed as previously described [4,13]. First, mice were anesthetized using a Ovariectomy and hormonal replacement ketamine/xylazine mixture (8.75 mg ketamine and 1.25 Six days before sampling, female mice were bilater- mg xylazine/100 g BW), and then perfused with 0.05 ally OVX. E2 (250 ng dissolved in 0.1 mL of peanut M phosphate-buffered saline (PBS), followed by 4% oil; Sigma-Aldrich Co., St Louis, MO) and/or P (2 mg paraformaldehyde in 0.05 M phosphate buffer (PB). dissolved in 0.1 mL of peanut oil; Sigma-Aldrich Co.) Pituitary glands were collected immediately, incubated were administered subcutaneously to some of the OVX in 4% paraformaldehyde/PB overnight at 4°C, and then mice, at 1,000 h for five consecutive days [9,10]. Three cryoprotected in PB containing 30% sucrose. A cryo- days before sampling, the GnRH antagonist cetrorelix stat was used to prepare frozen 10-μm horizontal sec- (10 µg in 0.2 mL saline/daily, S.C.; Cetrotide, Merck tions that were mounted on MAS-coated glass slides KGaA, Darmstadt, Germany) was administered sub- (Matsunami Glass Ind. Ltd., Osaka, Japan). cutaneously to some of the OVX mice, at 1,000 h for three consecutive days [11]. Six days after ovariec- Dual-immunostaining for GPR120 and luteinizing tomy, the mice were decapitated between 1,000 and hormone beta (LHβ) 1,200 h, and their pituitary glands and uteri were col- For dual-immunostaining, an anti-LHβ rabbit anti- lected for RNA extraction and wet weight measure- body (1:2,000; kindly provided by Dr. A. F. Parlow, ments, respectively. In addition, diestrus female mice National Institute of Diabetes and Digestive and Kidney that underwent a sham ovariectomy operation were Diseases, Torrance, CA) was pre-conjugated with anti- used as controls. rabbit Alexa Fluor 488 Fab fragments for 5 min at room Effects of ovarian hormones on GPR120 1057 temperature using a Zenon Alexa Fluor 488 Rabbit surgery animals (Fig. 1). This increase in GPR120 IgG Labeling Kit (Life Technologies Co.), according mRNA was reduced following treatment with E2. to the manufacturer’s protocol. Tissue sections were incubated with 10% normal goat serum (NGS) for 1 Colocalization of GPR120 and LHβ in the pituitary h at room temperature to block non-specific binding, gland then incubated with the rabbit anti-GPR120 antibody GPR120 protein expression in gonadotrophs of the (1:166, LS-A2004; LifeSpan BioSciences Inc., Seattle, pituitary glands of Sham, OVX, OVX+E2, OVX+P, WA) in blocking solution at 4 °C overnight. After four OVX+E2+P, and OVX+Cetrotide animals was exam- washes in PBS, the sections were incubated at room ined using dual immunohistochemistry. GPR120 temperature for 90 min with Alexa Fluor 594-conju- immunoreactivity was observed in almost all LHβ- gated goat anti-rabbit IgG (1:200), followed by four immunoreactive cells in intact (96.8 ± 1.7%; Fig. 2 washes in PBS. Sections were then incubated with the and Table 1), OVX (98.0 ± 1.1%; Fig. 2 and Table 1), pre-conjugated anti-LHβ rabbit antibody overnight at 4 OVX+E2 (97.2 ± 2.0%; Fig. 2 and Table 1), OVX+P °C, and washed four times in PBS at room tempera- (98.4 ± 0.0%; Fig. 2 and Table 1), and OVX+E2+P ture. Finally, cover slips were placed over the sections (95.4 ± 0.0%; Fig. 2 and Table 1) animals. Furthermore, using VECTASHIELD Mounting Medium (Vector almost all GPR120 immunoreactive cells were also Laboratories Inc., Burlingame, CA), and examined with LHβ immunopositive in Sham (98.3 ± 1.0%; Fig. 2 a BX51 fluorescence microscope (Olympus Co., Tokyo, and Table 1), OVX (97.0 ± 1.8%; Fig. 2 and Table 1), Japan). To test the specificity of the anti-GPR120 OVX+E2 (97.4 ± 1.9%; Fig. 2 and Table 1), OVX+P antibody, an antigen preadsorption sample was pre- (97.8 ± 0.0%; Fig. 2 and Table 1), and OVX+E2+P pared, in which 50 μL of a 6 μg/mL antibody solution (97.6 ± 0.0%; Fig. 2 and Table 1) animals. The number was incubated at 4 °C overnight with 15 μg of a syn- of LHβ immunoreactive cells could not be counted in thetic GPR120 peptide antigen (LS-P2004; LifeSpan OVX+Cetrotide animals, because the immunoreactiv- BioSciences), prior to incubation with the tissue. ity of LHβ was weak in these animals (Fig. 2 and Table 1). In addition, the preincubation of the anti-GPR120 Quantification of GPR120 immunoreactivity antibody with a synthetic GPR120 peptide abolished Colocalization of GPR120 and LHβ was quanti- the immunostaining of GPR120 in frozen pituitary tis- fied unilaterally in five non-consecutive horizontal sec- sue sections of ad-lib fed animals (Fig. 2, Adsorption). tions obtained from each of the four diestrus, OVX and OVX+E2 animals. The percentages of GPR120 immunoreactivity in LHβ immunoreactive cells and LHβ immunoreactivity in GPR120 immunoreactive cells were calculated. To avoid bias while counting, slides were coded and evaluated in a blinded experi- mental setup.

Statistical analysis Data were analyzed using the one-way ANOVA and the Bonferroni’s post-hoc analysis. Data are expressed as mean ± SEM and were analyzed using JMP 9. A p value < 0.05 was considered significant. Fig. 1 GPR120 mRNA expression in the pituitary glands of diestrus, ovariectomized (OVX), estradiol-primed OVX Results (OVX+E2), progesterone-primed OVX (OVX+P), both E2 and P-primed OVX (OVX+ E2+P), and GnRH GPR120 mRNA levels in the pituitary gland after E2, antagonist, Cetrotide-injected OVX animals. GPR120 P, or Cetrotide injection mRNA expression, analyzed using real-time PCR, was increased in OVX, OVX+P, OVX+ E2+P and GPR120 mRNA expression levels were signifi- OVX+Cetrotide groups, compared to that in diestrus cantly higher in the pituitary glands of OVX, OVX+P, group. All data are represented as mean ± SEM (n = 6). OVX+E2+P and OVX+Cetrotide animals than in sham * indicates p < 0.05. 1058 Moriyama et al.

Fig. 2 Dual immunohistochemistry for GPR120 (red) and luteinizing hormone β (LHβ, green) in the pituitary tissues of sham operated, ovariectomized (OVX), estradiol-primed OVX (OVX+ E2), progesterone-primed OVX (OVX+P), both E2 and P-primed OVX (OVX+ E2+P) and GnRH antagonist, Cetrotide-injected OVX animals. The GPR120 protein colocalized with LHβ in the pituitary glands of diestrus, OVX, OVX+E2, OVX+P, OVX+E2+P and OVX+Cetrotide animals. GPR120 immunoreactivity was eliminated by preincubation of the anti-GPR120 antibody with a synthetic antigen peptide (Adsorption) in the pituitary tissues of day 1 diestrus animals. Scale bar = 50 µm.

OVX

Diestrus OVX E2 P E2+P Cetrotide

GPR120 / LH 96.8 ± 1.7 98.0 ± 1.1 97.2 ± 2.0 98.4± 0.0 95.4± 0.0 ---

LH / GPR120 98.3 ± 1.0 97.0 ± 1.8 97.4 ± 1.9 97.8± 0.0 97.6± 0.0 ---

Table 1 Immunohistochemical colocalization of GPR120 and luteinizing hormone β (LHβ) in the pituitary gland of diestrus, ovariectomized (OVX), estradiol-primed OVX (OVX+ E2), progesterone-primed OVX (OVX+P), both E2 and P-primed OVX (OVX+ E2+P), and GnRH antagonist, Cetrotide-injected OVX animals. The top row indicates the percentages of GPR120 immunoreactivity in LHβ immunoreactive cells. The bottom row indicates the percentages of LHβ immunoreactivity in GPR120 immunoreactive cells. Effects of ovarian hormones on GPR120 1059

GPR120 mRNA levels in the pituitary gland of nor- Further, this change in GPR120 expression would occur mal cycling females in gonadotrophs, because more than 95.4% of GPR120 GPR120 mRNA expression levels were not sig- immunoreactivities were observed in gonadotrophs. nificantly different in the pituitary gland of females E2 reduced the OVX-induced increase in GPR120 throughout the phases of the ovarian cycle (Fig. 3). mRNA expression in the pituitary gland, indicating that estradiol negatively regulates GPR120 mRNA Uterine wet weight expression in gonadotrophs of the anterior pituitary Uterine wet weights were significantly lower in gland. Interestingly, GPR120 mRNA expression lev- OVX (99.2 ± 10.5 mg), OVX+P (120.7 ± 13.1 mg), els in the estrous stage varied, but only slightly. The and OVX+Cetrotide (82.5 ± 30.8 mg) animals than results suggest that GPR120 mRNA expression is regu- in Sham (187.7 ± 19.1 mg), OVX+E2 (250.1 ± 38.8 lated by the presence or absence of estrogen rather than mg), and OVX+E2+P animals (223.4 ± 38.3 mg) (Fig. in an estrogen dose-dependent manner. If that should 4). Furthermore, uterine wet weights of OVX+E2 and be the case, GPR120 mRNA expression levels may OVX+E2+P animals were almost the same as those of be high in the prepubertal and postmenopausal peri- sham operated diestrus female animals. ods when levels of blood estrogen are very low, and this GPR120 activation may support the inactivation of Discussion gonadotropic hormones in these periods. In addition, it has been reported that some growth factors, such as The present study clearly demonstrates that the expres- the insulin-like growth factor I and the transforming sion of GPR120 increased following OVX in female growth factor β1, are upregulated by E2 in the pitu- mice. This indicates that GPR120 expression in the pitu- itary gland [14,15]. Thus, these growth factors may be itary gland is under the effect of a humoral signal, which involved in the regulation of GPR120 expression. is altered following OVX. Indeed, the OVX-induced The OVX-induced increase in GPR120 mRNA increase in GPR120 mRNA expression levels was mark- expression levels tended to be elevated by P. edly reduced by E2, and tended to be elevated by P, Furthermore, simultaneous injection of E2 and P inter- suggesting these hormones directly or indirectly regu- fered with the action of E2 on GPR120 mRNA expres- late GPR120 mRNA expression in the pituitary gland. sion. Thus, P may counteract the inhibitory effect

Fig. 3 GPR120 mRNA expression, analyzed using real-time Fig. 4 Mean uterine wet weights (mg) of diestrus, PCR, in the pituitary glands of normal cycling female ovariectomized (OVX), estradiol-primed OVX (OVX+ animals. D1, day 1 diestrus; D2, day 2 diestrus; P, E2), progesterone-primed OVX (OVX+P), both E2 and proestrus; E, estrus. All data are represented as mean ± P-primed OVX (OVX+ E2+P), and GnRH antagonist, SEM (n = 6). Cetrotide-injected OVX animals. Uterine wet weights were lower in the OVX, OVX+P, and OVX+Cetrotide groups compared to those in diestrus animals. * indicates p < 0.05. 1060 Moriyama et al. of estradiol on the synthesis of GPR120 in the pitu- mRNA expression. itary gland. The results suggest that the activation of Uterine weight is classically used as a bioassay for GPR120 expression in response to P, which is increased estrogen levels [19, 20]. In the present study, the uter- in the luteal phase and during pregnancy, may contrib- ine wet weights of OVX+E2 and OVX+E2+P animals ute to the inactivation of gonadotropic hormones dur- were almost equal to those of sham operated diestrus ing these phases. Indeed, FFAs directly suppress Fshb animals, indicating that the blood estrogen levels of mRNA levels in the LβT2 gonadotropic cell line [7], OVX+E2 and OVX+E2+P animals were almost the and the plasma levels of FFA in female rodents are same as those for diestrus animals in this study. commonly elevated during pregnancy [16,17]. In summary, the current study clearly demon- The , which is also regulated by E2 strates that the expression of GPR120 in the pituitary and P, may regulate GPR120 mRNA expression in gland, and particularly in gonadotrophs, is augmented the pituitary gland indirectly through GnRH secre- by ovariectomy or P, and is down regulated by E2. tion in response to E2 and P injection, because cells Furthermore, P can interfere with the activity of E2 in in the hypothalamus also express E2 and P receptors GPR120 expression. These results suggest that ovarian [18]. Therefore, we decided to investigate whether hormones may directly regulate GPR120 expression at E2 directly regulates GPR120 mRNA expression in the pituitary level. Additional studies are needed to the pituitary gland by using the LβT2 gonadotropic further elucidate the physiological function of GPR120 cell line in vitro. However, the data were inconsis- in the pituitary gland. tent and the results of the study were inconclusive (data not shown). Thus, we conducted another exper- Acknowledgements iment that involved GnRH antagonist, Cetrotide, treatment to examine the direct effects of E2 and P We are grateful to Mses. Reina Hara, Mari Toyonaga, on GPR120 mRNA expression in the pituitary gland. and Yuki Sogo for their technical assistance. This Cetrotide did not reduce the increase in GPR120 work was supported in part by a Japan Society for the mRNA expression levels in OVX animals. This sug- Promotion of Science KAKENHI Grant (20780202, gests that E2 and P act not on the hypothalamus but 26450452 awarded) and a Kindai University Grant directly on the pituitary gland to regulate GPR120 (GS03, RK-058) to R.M.

References 1. Hirasawa A, Tsumaya K, Awaji T, Katsuma S, Adachi T, release. 152: 3905-3916. et al. (2005) Free fatty acids regulate gut incretin gluca- 6. Moriyama R, Yamazaki T, Kato T, Kato Y (2016) Long- gon-like peptide-1 secretion through GPR120. Nat Med chain unsaturated fatty acids reduce the transcriptional 11: 90-94. activity of the rat follicle-stimulating hormone beta- 2. Oh DY, Talukdar S, Bae EJ, Imamura T, Morinaga H, subunit gene. J Reprod Dev 62: 195-199. et al. (2010) GPR120 is an omega-3 fatty acid recep- 7. Sharma S, Morinaga H, Hwang V, Fan W, Fernandez tor mediating potent anti-inflammatory and insulin-sen- MO, et al. (2013) Free fatty acids induce Lhb mRNA sitizing effects. Cell 142: 687-698. but suppress Fshb mRNA in pituitary LbetaT2 gonado- 3. Ichimura A, Hirasawa A, Poulain-Godefroy O, tropes and diet-induced obesity reduces FSH levels in Bonnefond A, Hara T, et al. (2012) Dysfunction of lipid male mice and disrupts the proestrous LH/FSH surge in sensor GPR120 leads to obesity in both mouse and female mice. Endocrinology 154: 2188-2199. human. Nature 483: 350-354. 8. Genazzani AR, Lucchesi A, Stomati M, Catarsi S, 4. Moriyama R, Deura C, Imoto S, Nose K, Fukushima N Genazzani AD, et al. (1997) Effects of sex steroid hor- (2015) Expression of the long-chain fatty acid receptor mones on the neuroendocrine system. Eur J Contracept GPR120 in the gonadotropes of the mouse anterior pitu- Reprod Health Care 2: 63-69. itary gland. Histochem Cell Biol 143: 21-27. 9. Matsumura S, Takeuchi S, Takahashi S (2004) Effect 5. Garrel G, Simon V, Denoyelle C, Cruciani-Guglielmacci of estrogen on melanocortin-3 receptor mRNA expres- C, Migrenne S, et al. (2011) Unsaturated fatty acids sion in mouse pituitary glands in vivo and in vitro. stimulate LH secretion via novel PKCepsilon and -theta Neuroendocrinology 80: 143-151. in gonadotrope cells and inhibit GnRH-induced LH 10. Chung Y, Kim J, Im E, Kim H, Yang H (2015) Effects of ovarian hormones on GPR120 1061

Progesterone and 17β-estradiol regulate expression of sion. J Biol Chem 289: 16164-16175. nesfatin-1/NUCB2 in mouse pituitary gland. Peptides 16. Cripps AW, Williams VJ (1975) The effect of pregnancy 63: 4-9. and lactation on food intake, gastrointestinal anatomy 11. Bittner AK, Horsthemke B, Winterhager E, Grümmer R and the absorptive capacity of the small intestine in the (2011) Hormone-induced delayed ovulation affects early albino rat. Br J Nutr 33: 17-32. embryonic development. Fertil Steril 95: 2390-2394. 17. Zhang L, Sugiyama T, Murabayashi N, Umekawa T, 12. Bustin SA, Benes V, Garson JA, Hellemans J, Huggett Ma N, et al. (2011) The inflammatory changes of adi- J, et al. (2009) The MIQE guidelines: minimum infor- pose tissue in late pregnant mice. J Mol Endocrinol 47: mation for publication of quantitative real-time PCR 157-165. experiments. Clin Chem 55: 611-622. 18. Couse JF, Lindzey J, Grandien K, Gustafsson JA, 13. Moriyama R, Reyes BAS, Tsukamura H, Maeda K. Korach KS (1997) Tissue distribution and quantita- (2003) Glucoprivation-induced Fos expression in the tive analysis of estrogen receptor-alpha (ERalpha) and hypothalamus and medulla oblongata in female rats. J estrogen receptor-beta (ERbeta) messenger ribonucleic Reprod Dev 49: 151-157. acid in the wild-type and ERalpha-knockout mouse. 14. Faure MO, Nicol L, Fabre S, Fontaine J, Mohoric N, et Endocrinology 138: 4613-4621. al. (2005) BMP-4 inhibits follicle-stimulating hormone 19. Evans SJ, Varny FR, Koch CF (1941) The mouse secretion in ewe pituitary. J Endocrinol 186: 109-121. uterine weight method for the assay of estrogens. 15. Choi SG, Wang Q, Jia J, Pincas H, Turgeon JL, et al. Endocrinology 28: 747-752. (2014) Growth differentiation factor 9 (GDF9) forms an 20. Rubin BL, Dorfman AS, Black L, Dorfman RI (1951) incoherent feed-forward loop modulating follicle-stim- Bioassay of estrogens using the mouse uterine response. ulating hormone beta-subunit (FSHbeta) gene expres- Endocrinology 49: 429-439.