Estrogen Receptor- Gene Deficiency Enhances Androgen ...

0013-7227/03/$15.00/0 Printed in U.S.A.

Endocrinology 144(1):84 –93 Copyright © 2003 by The Endocrine Society doi: 10.1210/en.2002-220292

Estrogen Receptor-␣ Gene Deficiency Enhances Androgen Biosynthesis in the Mouse Leydig Cell BENSON T. AKINGBEMI, RENSHAN GE, CHERYL S. ROSENFELD, LESLIE G. NEWTON, DIANNE O. HARDY, JAMES F. CATTERALL, DENNIS B. LUBAHN, KENNETH S. KORACH, MATTHEW P. HARDY

AND

Center for Biomedical Research, Population Council (B.T.A., R.G., D.O.H., J.F.C., M.P.H.), New York, New York 10021; Departments of Animal Sciences (C.S.R., L.G.N., D.B.L.), Veterinary Biomedical Sciences (C.S.R.), and Biochemistry and Child Health (D.B.L.), University of Missouri, Columbia, Missouri 65211; and Receptor Biology Section, Laboratory of Reproductive and Developmental Toxicology, National Institute of Environmental Health Sciences, National Institutes of Health (K.S.K.), Research Triangle Park, North Carolina 27709 in ␣ERKO Leydig cells vs. 295 ⴞ 27 pmol/min䡠106 cells in WT cells; P < 0.01). Consistent with steroidogenic enzyme activity, the testis of ␣ERKO mice expressed higher steady state mRNA levels for steroidogenic acute regulatory protein and two enzymes involved in androgen biosynthesis, P45017␣ and 17␤HSD type III, as determined by semiquantitative RT-PCR. Compared with the controls, higher steady state mRNA levels for steroidogenic acute regulatory protein and P45017␣ were also measured in the testis of ICI 182,780-treated mice. In a second set of experiments estrogen administration reduced serum LH and T levels in WT controls, whereas ␣ERKO mice were unaffected. Although exposure of WT and ␣ERKO Leydig cells to estrogen in vitro did not affect androgen biosynthesis, incubation with ICI 182,780 reduced T production by WT, but not ␣ERKO, Leydig cells. These observations indicate that abrogation of the ER␣ gene by targeted deletion or treatment with an antiestrogen increases Leydig cell steroidogenesis in association with elevations in the serum levels of LH, which presumably is the result of estrogen insensitivity at the level of the hypothalamus and/or pituitary gonadotropes. Furthermore, the decrease in T production by WT Leydig cells and not ␣ERKO Leydig cells occasioned by incubation with ICI 182,780 suggests that of the ER subtypes, ER␣ has a regulatory role in Leydig cell steroidogenic function. (Endocrinology 144: 84 –93, 2003)

Leydig cells, which produce the primary male steroid hormone testosterone (T), express the two estrogen receptor (ER) subtypes, ER␣ and ER␤, and have the capacity to convert testosterone to the natural estrogen 17␤-estradiol. Thus, Leydig cells are subject to estrogen action. The development of transgenic mice that are homozygous for targeted deletion of genes encoding the ER subtypes provides an opportunity to examine the role of estrogen in Leydig cell function. In this study androgen biosynthesis was analyzed in Leydig cells from mice that were homozygous for targeted deletion of the ER␣ gene (␣ERKO). T production by ␣ERKO Leydig cells was 2-fold higher than that in wild-type (WT) cells. Serum T levels were accordingly higher in ␣ERKO compared with WT mice (5.1 ⴞ 1.1 vs. 2.2 ⴞ 0.4 ng/ml; P < 0.01) as were serum LH levels (1.31 ⴞ 0.3 vs. 0.45 ⴞ 0.08 ng/ml; P < 0.01). Mice that were treated with the pure antiestrogen ICI 182,780 at 100 ␮g/kg䡠d for 7 d, effectively abrogating ER-mediated activity, also had 2-fold elevations in the serum levels of LH (1.15 ⴞ 0.3 vs. 0.45 ⴞ 0.2 ng/ml) and T (4.3 ⴞ 1.1 vs. 2.2 ⴞ 0.2 ng/ml; P < 0.01). Increased androgen biosynthesis by ␣ERKO Leydig cells was associated with higher steroidogenic enzyme activity, especially of cytochrome P450 17␣-hydroxylase/17–20 lyase (P45017␣) and 17␤-hydroxysteroid dehydrogenase (17␤-HSD), as measured by conversion of radiolabeled steroid substrates to T or its precursors. The largest increases in enzymatic activity were observed for P45017␣ (423 ⴞ 45 pmol/min䡠106 cells

T

HE FUNCTIONS OF estrogen and estrogen receptors (ER) in the female have been well characterized, but their role(s) in the male is only beginning to be clarified. However, localization of ERs in the hypothalamus, pituitary, testis, and reproductive tract suggests estrogen regulation of male reproduction (1). In recent times, the possibility that exposure to environmental agents with estrogenic activity might cause reproductive disorders in human populations has generated public concern and scientific interest in ERmediated activity (2, 3). To date, two ER subtypes have been cloned, ER␣ and ER␤, and both are known to have a wide

tissue distribution pattern in rodents and humans (1, 4). Male mice that were deficient in the ER␣ gene (␣ERKO) were infertile, but had higher serum testosterone (T) levels than their wild-type siblings, indicating that ER␣, albeit along with androgen receptors, has a role in mediating steroid feedback on the pituitary (5, 6). It has also been demonstrated that the expression of ER␣ produces the developmental increase in estrogen sensitivity in the prostate gland (7). On the other hand, mice that were homozygous for targeted disruption of the ER␤ gene (␤ERKO) were fertile, with no obvious phenotypes in the reproductive tract (8). The development of transgenic mice lacking ER subtypes thus affords an opportunity to investigate the influence of estrogen and the ER in specific tissues in the presence of intact pathways for other steroid hormones. The hypothalamus-pituitary-gonadal axis is regulated by gonadal steroids, which act at the pituitary level to inhibit gonadotropin secretion and on the hypothalamus to sup-

Abbreviations: DES, Diethylstilbestrol; E2, 17␤-estradiol; ER, estrogen receptor; ERKO, estrogen receptor knockout; HPTE, 1,1,1-trichloro2,2-bis-(p-hydroxyphenyl)ethane; 3␤-HSD, 3␤-hydroxysteroid dehydrogenase; HT, heterozygous; P45017␣, cytochrome P450 17␣-hydroxylase/ 17–20 lyase; P450scc, cholesterol side-chain cleavage enzyme; SER, smooth endoplasmic reticulum; StAR, steroidogenic acute regulatory protein; T, testosterone; WT, wild-type.

84

Akingbemi et al. • Leydig Cell Steroidogenesis in ER␣ Knockout Mice

press GnRH pulses (9, 10). ER␣ and aromatase (the enzyme that catalyzes the synthesis of endogenous estrogen from T) are coexpressed in gonadotropes and in many hypothalamic nuclei (11, 12). The hypothalamus also contains ER␤ mRNA and protein in rodents (13, 14). Analysis of the promoter regions of gonadotropin genes in the pituitary has shown that some of the feedback effects of steroid hormones are mediated through ER interactions with response elements and other transcription factors (15). Therefore, disruption of ER gene function is expected to affect reproductive activity through modulation of pituitary gonadotropin secretion. The primary male steroid hormone, T, is mainly produced by Leydig cells under stimulation by LH. Cholesterol is the substrate used in androgen biosynthesis in a series of reactions catalyzed by four enzymes: cholesterol side-chain cleavage enzyme (P450scc), 3␤-hydroxysteroid dehydrogenase (3␤-HSD), cytochrome P450 17␣-hydroxylase/17–20 lyase (P45017␣), and 17␤-hydroxysteroid dehydrogenase (17␤-HSD). Leydig cells have ER, and estrogen inhibits Leydig cell development and function because exposure of Leydig cells to estrogen affects the expression of the genes encoding cytochrome P450 enzymes (16, 17). The higher serum T levels observed in ␣ERKO mice are coupled to elevated LH concentrations, and chronic LH stimulation is known to stimulate the proliferation of developing Leydig cells and to increase steroidogenesis in mature Leydig cells (18). However, it is not clear whether enhanced androgen biosynthesis in ␣ERKO mice is due to a higher number of Leydig cells or can be accounted for at least in part by enhanced steroidogenic capacity in individual Leydig cells. This study was thus designed to measure Leydig cell steroidogenesis in the ER␣deficient mouse Leydig cell and to test whether Leydig cells are sensitive to estrogen action. Materials and Methods Chemicals Trypsin inhibitor, EDTA, HEPES, BSA, bovine lipoprotein, sodium bicarbonate (NaHCO3), DMEM nutrient mixture [Ham’s F-12 (DMEM/F-12; 1:1 mixture without phenol red)], heparin, diethylstilbestrol (DES), 17␤estradiol (E2), 5-pregnen-3␤-ol-20-one (pregnenolone), dibutyryl cAMP, 22(R)-hydroxycholesterol, albumin, Percoll, etiocholan-3␤-ol-17-one, nicotinamide adenine dinucleotide, nitroblue tetrazolium, and gentamicin were purchased from Sigma (St. Louis, MO). Dulbecco’s PBS, medium 199, and 10⫻ Hanks’ balanced salt solution were obtained from Life Technologies, Inc. (Grand Island, NY), and collagenase, dispase, and deoxyribonculease were purchased from Roche Molecular Biochemicals (Mannheim, Germany). 25-[26,27-3H]Hydroxycholesterol, [7-N-3H]pregnenolone, [1,2-N3 H]17␣-hydroxyprogesterone, and [1␤, 2␤-N-3H] androst-4-ene-3, 17dione were purchased from NEN Life Science (Boston, MA). Estradiol benzoate [1,3,5-(10)-estratien-3,17␤-diol-3-benzoate] was purchased from Steraloids (Newport, RI), and the antiestrogen ICI 182,780 (7␣-[9-(4,4,5,5pentafluoropentylsulfinyl)estra-1,3,5-(10)-triene-3,17␤-diol]) was a gift from Zeneca Pharmaceuticals (Cheshire, UK). 1,1,1-Trichloro-2,2-bis-(phydroxyphenyl)ethane (HPTE) was synthesized by Dr. W. R. Kelce (Upjohn-Pharmacia, Kalamazoo, MI), and ovine LH was provided by the National Hormone and Pituitary Program (NIDDK, Bethesda, MD). The antibody to mouse P45017␣ was donated by Dr. Buck Hales (University of Illinois at Chicago, Chicago, IL).

Animals The ER␣ gene was disrupted in embryonic stem cells as previously described (19). Genotyping of mice was performed by PCR of DNA isolated from tail snips and amplified using four oligonucleotide primers

Endocrinology, January 2003, 144(1):84 –93 85

placed strategically around the neomycin insertion (two within and two flanking) in the ER␣ gene. The products were resolved on 2% agarose gels, and band sizes were used to distinguish among three genotypes: mice homozygous for targeted deletion of the ER␣ gene (␣ERKO, ER␣⫺/⫺), and their heterozygous (HT; ER␣⫹/⫺) and wild-type (WT; ER␣⫹/⫹) siblings. Experiments were conducted with C57BL6J/129 mice that were sexually mature, i.e. not younger than 10 wk, but no more than 13 wk of age, because testicular atrophy is observed in ␣ERKO mice after 60 – 80 d of age (5). Animals were fed Purina chow (Ralston Purina, St. Louis, MO) ad libitum. Although this diet contains soybean meal, all animals were exposed to the same levels of phytoestrogen. Drinking water was also freely provided. Animals were maintained on a 12-h dark, 12-h light cycle and were killed by using a protocol approved by the institutional animal care and use committee of Rockefeller University (Protocol 91200-R2). In preliminary experiments serum LH and T concentrations and androgen production by Leydig cells were measured in all three genotypes (WT, HT, and ␣ERKO), and these measurements were similar in WT and HT mice. Therefore, only WT and ␣ERKO mice were analyzed in this study.

Serum hormones The serum concentrations of LH and T were measured in WT and ␣ERKO mice. To determine whether elevations in serum LH and T concentrations in ␣ERKO mice are associated with a decrease in ERmediated activity, we measured the serum levels of these hormones in the absence of ER activity. Although lacking in estrogenic activity mediated via ER␣, ␣ERKO mice nevertheless retain aromatase and ER␤mediated activity. However, the pure antiestrogen ICI 182,780, which binds ER␣ and ER␤ and has no partial ER agonist activity within the rodent testis (20), can be used to achieve functional inactivation of the ER. This compound is known to down-regulate the ER and cause abrogation of ER-mediated effects (21). Therefore, in another set of experiments, mice in a C57BL/6J background (Charles River Laboratories, Inc., Wilmington, MA) received ICI 182,780 by daily sc injection at 100 ␮g/kg䡠d for a total period of 7 d. This dose was selected after pilot experiments showed that this treatment regimen increased serum T levels in mice. Control animals received only the oil vehicle. Within 24 h of the last administration of the antiestrogen, animals were killed. Blood was collected by cardiac puncture, and serum was stored at ⫺20 C until analyzed for LH and T concentrations. The serum concentrations of LH were measured using [125I]rat LH (Covance Laboratories, Inc., Vienna, VA) and materials obtained from the National Hormone and Pituitary Program, i.e. rat antibody NIDDK anti-rLH-S11 and LH reference standards (NIDDK rLF-RP-3). The lower limit of detection for this assay is 0.12 ng/ml, and LH values are expressed in relation to the RP-3 standards. The intra- and interassay coefficients of variation were 5% and 10%, respectively. Serum T levels, and T production by Leydig cells, were measured with a tritium-based RIA as described previously, with a 7– 8% interassay variation (22).

Leydig cell T production A major objective of this study was to characterize steroidogenic capacity in individual Leydig cells with and without ER␣ and determine whether increased T production per Leydig cell is contributory to the higher levels of T seen in the ␣ERKO mouse. Purified Leydig cells were prepared as previously described (23), and this procedure has been validated and used routinely for mouse Leydig cells (24). Briefly, testes were subjected to collagenase digestion. Before loading onto a 55% continuous Percoll density gradient, the cell suspension was subjected to centrifugal elutriation to remove germ cell and sperm contaminants. The fraction of Leydig cells remaining in the Elutriator chamber at a flow rate greater than 16 ml/min and a rotor speed of 2000 rpm was collected. Leydig cells were harvested from the Percoll gradient at 1.070 and greater bands. The cells were counted using a hemocytometer, and purity was determined by histochemical staining for 3␤-HSD using 0.4 mm etiocholan-3␤-ol-17-one as the enzyme substrate (25). Leydig cell preparations were typically 95–97% intensely stained. The amounts of T produced by WT and ␣ERKO Leydig cells were measured after incubation in DMEM/F-12 culture medium containing 0.1% BSA and 0.5 mg/ml bovine lipoprotein and buffered with 14 mm NaHCO3. Incubations were conducted with a density of 0.05– 0.1 ⫻ 106

86

Endocrinology, January 2003, 144(1):84 –93


cells in microcentrifuge tubes and at a temperature of 34 C for 3 h without (basal) or with a maximally stimulating dose of ovine LH (100 ng/ml). To determine whether ER␣ deficiency affects LH signaling via the classical cAMP-dependent phosphate kinase A pathway, Leydig cells were incubated with dibutyryl cAMP (100 ␮m). The latter is an activator of the cAMP-dependent phosphate kinase A pathway of LH signaling that bypasses LH receptors in Leydig cells. Steroidogenic capacity was investigated further by incubation of Leydig cells with steroid substrates at high doses. These T precursors have the ability to diffuse readily into Leydig cells, and the amount of T produced is a measure of enzyme capacity: 50 ␮m 22(R)-hydroxycholesterol (P450scc) or 20 ␮m pregnenolone (3␤-HSD), progesterone (P45017␣), and androstenedione (17␤-HSD) (26, 27). All measurements of T production in vitro were conducted 4 times, using purified Leydig cell fractions pooled from at least 12 males/genotype. Aliquots of the spent medium were assayed by RIA to measure T production, and values were normalized to 106 cells.

drostenedione, and T, whereas C17–20 cleavage activity was indicated by the conversion of 17␣-hydroxyprogesterone to androstenedione and T. The activity of 17␤-HSD was assayed by measuring the conversion of androstenedione to T. The activity of P450scc was determined by measuring the conversion of side-chain-labeled 25-[26,27-3H]hydroxycholesterol to radioactive 4-hydroxy-4-methylpentanoic acid as previously described (29). Incubations were performed for 30 min at 34 C, after which 0.5 ml NaOH (0.5 m) was added. The mixture was extracted twice with 2 ml chloroform and mixed with neutral alumina to remove nonmetabolized substrate, and an aliquot was removed for measurement by liquid scintillation counting. Enzyme assays were conducted twice to ensure that the results were repeatable, and the values from intraassay replicate samples were then subjected to statistical analysis.

Steroidogenic enzyme activity

Persistent stimulation of Leydig cells by higher than normal LH levels is known to increase the volume of cellular smooth endoplasmic reticulum (SER); this organelle contains three of the enzymes involved in androgen biosynthesis (3␤-HSD, P45017␣, and 17␤-HSD) (30). We therefore hypothesized that increased LH stimulation might increase ␣ERKO Leydig cell size. To measure cell size, we performed immunocytochemistry to localize Leydig cells in the testicular interstitium, using an antibody specific to mouse P45017␣. After perfusion in Bouin’s fixative solution, testis from WT and ␣ERKO mice (n ⫽ 3) were postfixed in the same solution overnight and then embedded in paraffin. At least 5 sections were obtained from each testis, i.e. a minimum of 10 sections from each animal. Before immunostaining, antigen retrieval was carried out by microwave treatment for 10 min in 10 mm (pH 6.0) citrate buffer. Endogenous peroxidase activity was quenched by incubation with 0.3% H2O2 in absolute methanol for 20 min. A monoclonal antibody specific for mouse P45017␣ was used at a 1:100 dilution overnight at 4 C (31). The specificity of the antibody was determined by the observation of staining only in the interstitium and not in the seminiferous epithelium, and the lack of staining when the primary antibody was substituted by nonimmune mouse IgG. Cell size was determined by measuring the area of Leydig cells in 10 different interstitial areas/section using image analysis (32).

The activity of enzymes involved in androgen biosynthesis was measured in WT and ␣ERKO Leydig cells. The activities of steroidogenic enzymes 3␤-HSD, P45017␣, and 17␤-HSD were determined by incubation of Leydig cells with radiolabeled substrates and separation of products by thin layer chromatography as previously described (28). The amount of substrate used for each enzyme was maximized to ensure that substrate concentration was not rate limiting. Control samples of culture medium alone were run in parallel with each enzyme assay. Each reaction mixture contained 1 ␮m substrate (1 ␮Ci) in medium. Reactions were initiated by adding to the reaction mixture an aliquot of 0.1– 0.2 ⫻ 106 cells. The reaction mixtures, conducted in triplicate, were maintained at 34 C and pH 7.2 in a shaking water bath for 10 min. Adding ice-cold ethyl acetate to sample tubes terminated reactions, and steroids were rapidly extracted. The organic layer was dried under nitrogen. Steroids were separated on thin layer chromatography plates, and radioactivity was measured using a radiometric scanner (System 200/AC3000, Bioscan, Inc., Washington DC). The activity of 3␤-HSD was determined by measuring conversion of pregnenolone to progesterone. P45017␣ catalyzes two mixed function oxidase reactions: 17␣-hydroxylation and C17–20 cleavage. 17␣-Hydroxylation activity was determined by measuring the conversion of progesterone to 17␣-hydroxyprogesterone, an-

FIG. 1. Serum LH and T concentrations. Serum LH and T levels were 2-fold higher in ␣ERKO compared with WT mice (A; n ⫽ 18), and the two hormones were similarly higher in ICI 182,780treated mice than in corresponding controls (B; n ⫽ 15). *, P ⬍ 0.01.

Determination of Leydig cell size


Effect of estrogen on T production The effect of estrogen on androgen production in WT and ␣ERKO mice was investigated in vivo and in vitro. In the first set of experiments, mice received sc injections of 50 ␮g/kg䡠d estradiol benzoate for a total period of 4 d. Pilot experiments showed that this dosage regimen reduced serum T levels in C57BL/6J mice. Control animals received only the oil vehicle. At the end of treatment and within 24 h of the last estradiol benzoate administration, animals were anesthetized with methoxyflurane, and blood was obtained by cardiac puncture. Serum was stored at ⫺20 C until analyzed for LH and T concentrations. In another set of experiments we attempted to determine the effects of blockade of ER signaling on androgen biosynthesis without the confounding involvement of pituitary gonadotropin changes as would occur in vivo. Thus, WT and ␣ERKO Leydig cells were incubated with low (1 and 10 nm) and high (1 ␮m) doses of E2 and DES for 18 h. Incubations were also conducted with the estrogenic metabolite HPTE at similar doses. HPTE is known to bind ERs to a greater degree than its parent compound, methoxychlor, and was previously demonstrated to decrease T biosynthesis by rat Leydig cells (27, 33). Furthermore, purified WT and ␣ERKO Leydig cells were incubated with ICI 182,780 for 18 h (0.01–1000 nm). In general, incubations of Leydig cells were conducted in medium containing 10 ng/ml LH, and serum and spent medium were stored at ⫺20 C until analyzed for LH and/or T concentrations.


each set of primers, and the optimal cycle number was determined over a range of 15– 40 cycles with different amounts of cDNA. PCR of the synthesized product was initiated by the addition of Taq DNA polymerase (Promega Corp., Madison, WI). PCR products were size-fractionated by agarose gel electrophoresis (2%) and stained with ethidium bromide. These bands were excised, sequenced, and verified to be mouse StAR, P450scc, 3␤-HSD, P45017␣, 17␤-HSD type III, 17␤-HSD type VII, and RPS16. Semiquantification of PCR products was performed using a densitometer (Kodak Imaging Systems, New York, NY) using 0.65 ␮g of a 100-bp DNA ladder as a standard and normalizing specific gene products to RPS16 bands.

Statistics All data were analyzed by one-way ANOVA, with multiple comparisons performed by the Duncan multiple range test to identify differences between groups. Differences were considered significant at P ⱕ 0.05.

Semiquantitative RT-PCR During androgen biosynthesis LH-stimulated movement of the steroid substrate cholesterol from the cytosol into mitochondria for sidechain cleavage is facilitated primarily by a 37-kDa protein that is generated from steroidogenic acute regulatory protein (StAR), a 30-kDa mitochondrial protein. Increased LH stimulation is known to enhance StAR expression in Leydig cells (34). We hypothesized that steroidogenic enzyme expression levels were higher in the testis of ER␣-deficient mice than in WT siblings, because P45017␣ and 17␤-HSD activity were higher in ␣ERKO Leydig cells than in WT. We therefore measured steady state testicular mRNA levels for P450scc, 3␤-HSD, P45017␣, and 17␤-HSD in WT and ␣ERKO mice. Because LH stimulation induces StAR expression, and the largest increases in enzymatic activity were observed for P45017␣, we also measured steady state mRNA levels for StAR and P45017␣ in the testis of ICI 182,780-treated mice. Total RNA was extracted from the testis by a single step method, using phenol and guanidinium thiocyanate (TRI-Reagent method, Molecular Research Center, Inc., Cincinnati, OH) following the manufacturer’s instructions. Only RNA samples exhibiting an A260/280 of 1.8 or more were used for RT. Four hundred nanograms of total RNA were reverse transcribed with avian myeloblastosis virus reverse transcriptase, random primers, and deoxy-NTPs at 37 C for 75 min, and heating at 95 C for 5 min terminated the reaction. Target cDNA was coamplified in the presence of ribosomal protein S16 (RPS16) to serve as an internal control. Using published sequences, primers for the target cDNAs selected on a PRIMER 3 software (Whitehead Institute for Biomedical Research, Cambridge, MA) were designed to span at least intron-exon junctions and to generate products distinguishable in size from that of the internal control. The primer sequences used were based on previously published sequences (35–37). Although the type III is the major isoform involved in T biosynthesis two isoforms of 17␤-HSD (types III and VII) are expressed in the adult mouse testis (36, 37). The primer sequences were 5⬘-TGTCAAGGAGATCAAGGTCCTG-3⬘ (forward) and 5⬘-CGATAGGACCTGGTTGATGAT-3⬘ (reverse) for StAR (product size, 310 bp), 5⬘-AGGTGTAGCTCAGGACTTCA-3⬘ (forward) and 5⬘-AGGAGGCTATAAAGGACACC-3⬘ (reverse) for P450scc (product size, 370 bp), 5⬘ACTGCAGGAGGTCAGAGCT-3⬘ (forward) and 5⬘-GCCAGTAACACACAGAATACC-3⬘ (reverse) for 3␤-HSD type I (product size, 565 bp), 5⬘-CCAGGACCCAAGTGTGTTCT-3⬘ (forward) and 5⬘-CCTGATACGAAGCACTTCTCG-3⬘ (reverse) for P45017␣ (product size, 250 bp), 5⬘ATTTTACCAGAGAAGACATCT-3⬘ (forward) and 5⬘-GGGGTCAGCACCTGAATAATG-3⬘ (reverse) for 17␤-HSD type III (product size, 367 bp), 5-TGCAGAGGAAGTCAAGCAAAA-3⬘ (forward) and 5⬘-CTTCTTTGCATTGCGAGAGGA-3⬘ (reverse) for 17␤-HSD type VII (product size, 310 bp), and 5⬘-CGTGCTTGTGCTCGGAGCTA-3⬘ (forward) and 5⬘-GCTCCTTGCCCAGAAGCAAA-3⬘ (reverse) for RPS16 (product size, 210 bp). Optimization of the PCR reaction was performed separately for

FIG. 2. T production by Leydig cells. Basal (A), LH-stimulated (B), and dibutyryl cAMP-stimulated (C) T production were higher by ␣ERKO Leydig cells than WT controls. These experiments were conducted with Leydig cells pooled from at least 12 males/genotype on 4 different occasions. *, P ⬍ 0.01.

88


Results Serum LH and T concentrations are higher in ␣ERKO and ICI 182,780-treated mice

Serum LH and T concentrations were 2-fold higher in ␣ERKO than in WT mice (Fig. 1A; P ⬍ 0.01). Similarly, C57BL/6J mice treated with the antiestrogen ICI 182,780 had 2-fold higher serum levels of LH and T compared with controls (Fig. 1B; P ⬍ 0.01). Androgen biosynthesis by Leydig cells is increased by ER␣ gene deficiency

Basal and LH-stimulated T production by ␣ERKO Leydig cells were approximately 80% and 50% greater compared with WT controls, respectively (Fig. 2, A and B; P ⬍ 0.01). Similar amounts of T were produced when ␣ERKO Leydig cells were either maximally stimulated with 100 ng/ml LH or incubated with 100 ␮m dibutyryl cAMP, and these levels were higher relative to WT controls (Fig. 2, B and C; P ⬍ 0.01). After incubation with T precursors, ␣ERKO Leydig cells produced greater amounts of T than WT controls in the presence of progesterone (6.8 ⫾ 0.6 vs. 4.3 ⫾ 0.4 ␮g/106 cells䡠3 h; P ⬍ 0.01) and androstenedione (3.4 ⫾ 0.1 vs. 2.5 ⫾ 0.2 ␮g/106 cells䡠3 h; P ⬍ 0.01) as substrates. However, ␣ERKO and WT Leydig cells produced similar amounts of T when incubations were conducted with 22(R)-hydroxycholesterol (5.1 ⫾ 0.4 vs. 4.4 ⫾ 0.5 ␮g/106 cells䡠3 h) or pregnenolone (4.0 ⫾ 0.4 vs. 3.2 ⫾ 0.3 ␮g/106 cells䡠3 h).

FIG. 3. Steroidogenic enzyme activities in Leydig cells. The activity of two enzymes, namely P45017␣ (C) and 17␤-HSD (D), were higher in ␣ERKO Leydig cells than in WT cells. Enzyme assays were conducted twice. *, P ⬍ 0.01.


Increased androgen biosynthesis in the ER␣-deficient mouse is associated with higher Leydig cell steroidogenic enzyme activity and greater cell size

The higher levels of androgen biosynthesis in ␣ERKO Leydig cells were associated with increased enzymatic activity, because P45017␣ and 17␤-HSD activities, measured by conversion of radiolabeled substrates to T or its precursors, were higher in ␣ERKO Leydig cells compared with WT controls (Fig. 3; P ⬍ 0.01). On the other hand, cytochrome P450scc and 3␤-HSD activities were similar in WT and ␣ERKO Leydig cells (Fig. 3). Increased androgen biosynthesis was also associated with higher testicular steady state mRNA levels for StAR, P45017␣, and 17␤-HSD type III (Fig. 4); these levels were 1.5- to 2-fold higher in ␣ERKO mice compared with WT controls. Mice treated with ICI 182,780 also had higher steady state mRNA levels for StAR and the P45017␣ enzyme (Fig. 5; P ⬍ 0.05). The average Leydig cell area is greater in ␣ERKO mice than in WT siblings (160.8 ⫾ 6 vs. 98.8 ⫾ 3 ␮m; P ⬍ 0.01), as represented in Fig. 6. Because volume is directly related to cell size, enhanced LH stimulation thus increases the volume of the steroidogenic machinery in the ER␣-deficient mouse Leydig cell. Administration of estrogen decreased serum T levels in WT, but not ␣ERKO, mice, and treatment with antiestrogen in vitro reduced T biosynthesis only by WT Leydig cells

Exogenous administration of estradiol benzoate decreased serum LH and T concentrations in WT mice to 50% of control



FIG. 4. Steady state mRNA levels for StAR and androgen biosynthetic enzymes. Compared with WT, the testis of ␣ERKO mice showed higher expression levels for StAR and two enzymes: P45017␣ and 17␤-HSD type III. Total RNA for RT-PCR was isolated from at least four testes from different animals in each group. *, P ⬍ 0.05.

values, but did not affect these parameters in the ␣ERKO (Fig. 7; P ⬍ 0.01). However, there were no E2-, DES-, or HPTE-related effects on T production by ␣ERKO or WT Leydig cells at 0.001, 0.01, and 1 ␮m in vitro. For example, T production after exposure of WT Leydig cells to 0.001 ␮m E2 for 18 h was 264 ⫾ 22 ng/106 cells vs. 254 ⫾ 12 for untreated WT Leydig cells, whereas ␣ERKO Leydig cells treated with E2 produced 775 ⫾ 88 ng/106 cells compared with 745 ⫾ 75 ng/106 cells for corresponding controls. Similar observations were made after incubation of Leydig cells with DES and HPTE (Table 1). On the other hand, incubation of Leydig cells with the antiestrogen ICI 182,780 decreased androgen biosynthesis by WT Leydig cells, whereas ␣ERKO Leydig cells were unaffected, as shown in Fig. 8. Discussion

Mice with ER␣ deficiency are known to produce fewer epididymal sperm that are less motile and have reduced fertilizing capacity; these animals also have diminished male sexual behavior compared with WT controls, although se-

rum LH and T concentrations are elevated (5). By measuring T production normalized to Leydig cell number, the present study demonstrates that higher serum T concentrations in ␣ERKO mice is due in part to enhanced capacity for androgen biosynthesis by individual Leydig cells, and is related to increased steroidogenic enzyme activity. Furthermore, WT mice treated with the pure antiestrogen ICI 182,780, effectively abrogating ER-mediated activity, showed similar serum LH and T profiles as ␣ERKO mice. Estrogen modulates GnRH neural function, thereby acting as a classic homeostatic feedback molecule between the gonad and the brain (38). Thus, higher serum LH concentrations in ER␣deficient mice presumably result from estrogen insensitivity (39), because ER enhancement of the negative feedback effects of androgen in the male hypothalamic-pituitary axis is thought to occur via aromatization of T to estradiol and subsequent activation of ER␣-mediated pathways (40). We also observed that exogenous estradiol did not affect serum LH and T concentrations in ␣ERKO mice, whereas both parameters were decreased in WT mice, in agreement with our

90



FIG. 5. Steady state mRNA levels for StAR and P45017␣ in mice treated with the antiestrogen ICI 182,780 (ICI). Compared with corresponding controls, StAR (A) and P45017␣ (B) mRNA levels were expressed at higher levels in ICI-treated mice. RT-PCR was conducted on at least four testes from different animals in each group. *, P ⬍ 0.05.

earlier report that estrogen suppressed LH␤ mRNA and serum LH levels in WT castrates whereas these effects were absent in castrated ␣ERKOs (39). Although androgen receptor-mediated activity plays a role in mediating steroid negative feedback (6), it is generally agreed that the majority of steroid negative feedback on the anterior pituitary in male and female mice occurs by estradiol binding to ER␣ (40, 41). Therefore, our observation of elevated serum LH levels after administration of ICI 182,780 highlights a prominent role for the pituitary in mediating estrogen negative feedback effects on the hypothalamo-pituitary axis, because the antiestrogen is unable to cross the blood-brain barrier to act at the level of the hypothalamus (42). These findings support the hypothesis that ER␣ is the primary, if not the exclusive, mediator of estrogen action in the mouse pituitary (43). However, it appears that the ability of ICI 182,780 to affect pituitary LH secretion, and hence androgen biosynthesis, is dose dependent because chronic exposure to this agent (weekly sc injections of 10 mg for 100 –150 d) did not affect prostate and seminal vesicle weights (44) as would be expected if serum T levels were elevated. It could also be that the changes in serum T levels, if any, were not dramatic enough to affect accessory sex organ weights. Previous studies indicate that activity and expression of cytochrome P450 enzymes are sensitive to modulation by LH and estrogen (17, 34, 45). In the present study, however, the activities of P45017␣ and 17␤-HSD, but not those of P450scc and 3␤-HSD, were elevated in ␣ERKO compared with WT

FIG. 6. Leydig cell size in ␣ERKO mice. Leydig cells were immunolocalized in the interstitium by staining with a monoclonal antibody specific for mouse P45017␣, and the areas of Leydig cells were measured by image analysis (A). Sections were obtained from both testis of three animals per group as described in Materials and Methods. Leydig cells were larger in the testes of ␣ERKO than in WT mice (B). Bar, 20 ␮m. *, P ⬍ 0.01.

Leydig cells. Enhanced androgen biosynthesis in ␣ERKO mice is therefore related to higher steroidogenic enzyme activity. We also measured higher steady state mRNA levels for StAR and the two enzymes exhibiting increased activity in the ␣ERKO testis, namely, P45017␣ and 17␤-HSD type III. These findings are not totally unexpected, because increased LH stimulation of Leydig cells is known to enhance StAR and steroidogenic enzyme gene expression (34). Moreover, chronic LH stimulation of Leydig cells increases the volume of steroidogenic organelles, i.e. mitochondria and SER, and the total SER surface area is positively correlated with the capacity for T production (46). In this regard, we observed



TABLE 1. Testosterone production by Leydig cells exposed to different estrogenic agents Estrogen

WT

␣ERKO

0 1 ␮M E2 0.01 ␮M E2 0.001 ␮M E2 1 ␮M DES 0.01 ␮M DES 0.001 ␮M DES 1 ␮M HPTE 0.01 ␮M HPTE 0.001 ␮M HPTE

254 ⫾ 12 262 ⫾ 11 224 ⫾ 24 264 ⫾ 22 261 ⫾ 23 255 ⫾ 26 266 ⫾ 25 265 ⫾ 24 248 ⫾ 26 259 ⫾ 25

745 ⫾ 75 792 ⫾ 108 756 ⫾ 75 775 ⫾ 88 734 ⫾ 33 837 ⫾ 103 845 ⫾ 128 714 ⫾ 25 787 ⫾ 46 795 ⫾ 81

Incubation of Leydig cells with estrogenic agents was for 18 h, and testosterone concentrations were measured in aliquots of spent medium (nanograms per 106 cells per 18 h). Values are the mean ⫾ SEM and represent the average for three experiments.

FIG. 7. Serum LH and T levels after exogenous administration of estrogen (n ⫽ 8). Injection of WT mice with estradiol benzoate decreased serum LH and T concentrations, but these effects were absent in ␣ERKO mice. *, P ⬍ 0.01.

that Leydig cells were larger in size in the ␣ERKO than in WT mice. Altogether the increased capacity for androgen biosynthesis in ␣ERKO Leydig cells, related to the absence of ER␣-mediated activity, is associated with increased LH stimulation, enhanced StAR and steroidogenic enzyme activities, and greater cell volume. Interestingly, T production by WT and ␣ERKO Leydig cells in vitro was unaffected by E2 and DES at the dosages used in our experiments. It could be that these concentrations lie outside the range of doses that affect Leydig cell steroidogenesis or that E2 and DES effects require potentiation by paracrine or other factors that are not present in vitro. It may also be that the period of exposure to estrogen was not long enough. Androgen production by ␣ERKO and WT Leydig cells was also not affected by exposure to the estrogenic metabolite HPTE. This finding contrasts with our previous observation showing that this agent caused half-maximal inhibition of androgen biosynthesis at 1.11 ⫾ 0.02 ␮m in rat Leydig cells (27). The factors responsible for the lack of HPTE effects on mouse Leydig cell androgen biosynthesis, unlike those in the rat, are not clear. However, administration of

methoxychlor (the parent compound of HPTE) affects mouse reproductive tissues, increasing adult prostate size in mice after exposure in utero (47). It is possible that there are genetic differences in the responses of mouse and rat strains to chemical exposures (48), because there is evidence showing that the ability of a compound to bind a particular ER subtype is species dependent, and the agonist or antagonist activity of synthetic estrogens varies for each target tissue (49). On the other hand, incubation with the pure antiestrogen ICI 182,780 decreased T production by WT Leydig cells, whereas ␣ERKO Leydig cells were unaffected. This compound has a 40-fold greater potency on mouse ER␣ than ER␤ (49), implying that ICI 182,780-induced inhibition of T production is primarily ER␣ mediated. Estradiol is known to act via the ER to regulate P45017␣ gene expression and activity (45, 50), and passive immunization of rams with the antiserum to estradiol was also found to increase StAR mRNA levels and 17␤-HSD activity (51). As the antiestrogen also exhibits aromataseinhibiting properties along with ER antagonist activity (52), ICI 182,780-induced inhibition of T production by WT Leydig cells, and not ␣ERKO, indicates that estrogen signaling via ER␣ is a requirement for the maintenance of Leydig cell steroidogenic function. This observation of a direct LHindependent effect of estrogen signaling on Leydig cell steroidogenesis is in agreement with previous reports showing that low level estradiol immunoneutralization induces higher blood T levels not associated with changes in pituitary LH secretion (51, 53). Interestingly, the presence of ER␤ in ␣ERKO Leydig cells did not render them susceptible to estrogen action in vivo and in vitro, at least when T production was the measured end point. However, it has been proposed that disruption of ER␣ probably ablates any role that ER␤ may have in steroidogenesis, because both ER subtypes are known to form heterodimers in vivo. Although ER␣ has a tendency to form homodimers, ER␤ preferentially heterodimerizes with and probably modifies ER␣ action (54, 55). Indeed, there is evidence that in the uterus, ER␤ is a modulator of ER␣ action (56). The present study indicates that higher serum T levels in the ␣ERKO mouse are due in part to increased androgen biosynthesis by individual Leydig cells. The present data also show that maintenance of Leydig cell steroidogenesis requires ER␣-mediated activity. This conclusion is supported

92



FIG. 8. Leydig cell androgen biosynthesis after antiestrogen treatment in vitro. Incubation with ICI 182,780 caused a decrease in T production by WT, but not ␣ERKO, Leydig cells. Incubations were conducted in medium containing 10 ng/ml ovine LH for 18 h, and these experiments were performed twice. *, P ⬍ 0.01.

by two facts: 1) the presence of higher serum LH and T levels and the lack of estradiol-induced effects on these measurements in the ␣ERKO, in contrast to observations in WT mice; and 2) the decrease in T production by WT Leydig cells, but not ␣ERKO, after treatment with ICI 182,780 in vitro. In addition, mice homozygous for the targeted deletion of the ER␤ gene exhibit similar serum LH and T profiles as their WT counterparts, implying that abrogation of ER␤-mediated activity does not affect LH and T levels in ␤ERKO mice as in ␣ERKO mice (5, 7). Thus, the function of ER␤ in the mouse Leydig cell remains to be clarified and may be through differential modulation of genes (57) that are not directly involved in T production. In this regard, some xenoestrogens are known to preferentially bind ER␤ in vitro (e.g. bisphenol A and phytoestrogens), and liganded ER␤ may have inhibitory effects on ER␣ signaling (58), but the extent to which these phenomena occur in vivo is not known. Finally, our data demonstrate that the effects of ER␣ deficiency in the mouse resulted in elevated pituitary LH output, which increased androgen biosynthesis, and suggests the absence of estrogen negative feedback effects on gonadotropin secretion. The present study also provides evidence showing that ER␣ has a regulatory role in androgen biosynthesis acting independently of LH action. Acknowledgments The technical assistance of Ms. Chantal Mannon Sottas and Dr. Guimin Wang is gratefully acknowledged. We are also thankful to Dr. V. K. Ganjam for his comments on earlier drafts of the manuscript, and to Evan Read for assistance with manuscript preparation. Received March 11, 2002. Accepted September 10, 2002. Address all correspondence and requests for reprints to: Matthew P. Hardy, Ph.D., Population Council, 1230 York Avenue, New York, New York 10021. E-mail: [email protected]. This work was supported in part by NIH Fogarty Award TWO-5350 (to B.T.A.) and NIH Grants HD-32588 and ES-10233 (to M.P.H.). Access to the Cell Culture Core Facility was provided in part by NICHHD/NIH support through a cooperative agreement (U54-HD-13541) as part of the Specialized Cooperative Centers Program in Reproduction Research.

References 1. Couse JF, Korach KS 1999 Estrogen receptor null mice: what have we learned and where will they lead us? Endocr Rev 20:358 – 417 2. Colborn T, vom Saal FS, Soto AM 1993 Developmental effects of endocrinedisrupting chemicals in wildlife and humans. Environ Health Perspect 101: 378 –384 3. Sharpe RM, Skakkebaek NE 1993 Are oestrogens involved in falling sperm counts and disorders of the male reproductive tract? Lancet 341:1392–1395 4. Gustafson JA 1999 Estrogen receptor ␤: a new dimension in estrogen mechanism of action. J Endocrinol 16:379 –383 5. Eddy EM, Washburn TF, Bunch DO, Goulding EH, Gladen BC, Lubahn DB, Korach KS 1996 Targeted disruption of the estrogen receptor gene in male mice causes alteration of spermatogenesis and infertility. Endocrinology 137:4796 – 4805 6. Wersinger SR, Haisenleder DJ, Lubahn DB, Rissman EF 1999 Steroid feedback on gonadotropin release and pituitary gonadotropin subunit mRNA in mice lacking a functional estrogen receptor ␣. Endocrine 11:137–143 7. Prins GS, Birch L, Couse JF, Choi I, Katzenellenbogen B, Korach KS 2001 Estrogen imprinting of the developing prostate gland is mediated through stromal estrogen receptor ␣: studies with ␣ERKO and ␤ERKO mice. Cancer Res 61:6089 – 6097 8. Krege JH, Hodgin JB, Couse JF, Enmark E, Warner M, Mahler JF, Sar M, Korach KS, Gustafsson JA, Smithies O 1998 Generation and reproductive phenotypes of mice lacking estrogen receptor ␤. Proc Natl Acad Sci USA 95:15677–15682 9. Drouin J, Labrie F 1976 Selective effect of androgens on LH and FSH release in anterior pituitary cells in culture. Endocrinology 98:1528 –1534 10. Shupnik MA 1996 Gonadotropin gene modulation by steroids and gonadotropin-releasing hormone. Biol Reprod 54:279 –286 11. Handa RJ, Resko JA 1988 Effects of gonadectomy and hormone replacement on steroid hormone receptors and 5␣-reductase activity in pituitaries of male rhesus macaques. J Clin Endocrinol Metab 66:1251–1258 12. Shughrue PJ, Lane MV, Merchenthaler I 1997 Comparative distribution of estrogen receptor-␣ and -␤ mRNA in the rat central nervous system. J Comp Neurol 388:507–525 13. Shughrue PJ, Komm B, Merchenthaler I 1996 The distribution of estrogen receptor-␤ mRNA in the rat hypothalamus. Steroids 61:678 – 681 14. Li X, Schwartz PE, Rissman EF 1997 Distribution of estrogen receptor-␤-like immunoreactivity in rat forebrain. Neuroendocrinology 66:63– 67 15. Gharib SD, Wierman ME, Shupnik MA, Chin WW 1990 Molecular biology of the pituitary gonadotropins. Endocr Rev 11:177–199 16. Abney TO, Myers RB 1991 17␤-Estradiol inhibition of Leydig cell regeneration in the ethane dimethylsulfonate-treated mature rat. J Androl 12:295–304 17. Saunders PT, Majdic G, Parte P, Millar MR, Fisher JS, Turner KJ, Sharpe RM 1997 Fetal and perinatal influence of xenoestrogens on testis gene expression. Adv Exp Med Biol 424:99 –110 18. Ewing LL and Zirkin BR 1983 Leydig cell structure and steroidogenic function. Recent Prog Horm Res 39:599 – 635 19. Lubahn DB, Moyer JS, Golding TS, Couse JF, Korach KS, Smithies O 1993 Alteration of reproductive function but not prenatal sexual development after


20.

21. 22.

23. 24.

25. 26.

27.

28. 29. 30.

31. 32. 33. 34. 35. 36.

37. 38. 39.

insertional disruption of the mouse estrogen receptor gene. Proc Natl Acad Sci USA 90:11162–11166 Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA 1997 Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors ␣ and ␤. Endocrinology 138: 863– 870 Dauvois S, White R, Parker MG 1993 The antiestrogen ICI 182780 disrupts estrogen receptor nucleocytoplasmic shuttling. J Cell Sci 106:1377–1388 Cochran RC, Ewing LL, Niswender GD 1981 Serum levels of follicle stimulating hormone, 5-androstane-3,17␣-diol, and 17␤-estradiol from male beagles with spontaneous or induced benign prostatic hyperplasia. Invest Urol 19: 142–147 Klinefelter GR, Kelce WR, Hardy MP1993 Isolation and culture of Leydig cells from adult rats. Methods Toxicol 3A:137–165. Rosenfeld CS, Ganjam VK, Taylor JA, Yuan X, Stiehr JR, Hardy MP, Lubahn DB 1998 Transcription and translation of estrogen receptor-␤ in the male reproductive tract of estrogen receptor-␣ knock-out and wild-type mice. Endocrinology 139:2982–2987 Payne AH, Downing JR, Wong KL 1980 Luteinizing hormone receptors and testosterone synthesis in two distinct populations of Leydig cells. Endocrinology 106:1424 –1429 Van Haren L, Flinterman JF, Orly J, Rommerts FF 1992 Luteinizing hormone induction of the cholesterol side-chain cleavage enzyme in cultured immature rat Leydig cells: no role of insulin-like growth factor-I? Mol Cell Endocrinol 87:57– 67 Akingbemi BT, Ge RS, Klinefelter GR, Gunsalus GL, Hardy MP 2000 A metabolite of methoxychlor, 2,2-bis (p-hydroxyphenyl)-1,1,1-trichloroethane, reduces testosterone biosynthesis in rat Leydig cells through suppression of steady-state messenger ribonucleic acid levels of the cholesterol side-chain cleavage enzyme. Biol Reprod 62:571–578 Ge RS, Hardy MP 1998 Variation in the end products of androgen biosynthesis and metabolism during postnatal differentiation of rat Leydig cells. Endocrinology 139:3787–3795 Georgiou M, Perkins LM, Payne AH 1987 Steroid synthesis-dependent, oxygen-mediated damage of mitochondrial and microsomal cytochrome P-450 enzymes in rat Leydig cell cultures. Endocrinology 12:1390 –1399 Wing TY, Ewing LL, Zirkin BR 1984 Effects of luteinizing hormone withdrawal on Leydig cell smooth endoplasmic reticulum and steroidogenic reactions which convert pregnenolone to testosterone. Endocrinology 115: 2290 –2296 Cascio C, Brown RC, Liu Y, Han Z, Hales DB, Papadopoulos V 2000 Pathways of dehydroepiandrosterone formation in rat brain glia. J Steroid Biochem Mol Biol 75:177–186 Wang GM, Ge RS, Latif SA, Morris DJ, Hardy MP 2002 Expression of 11␤-hydroxylase in rat Leydig cells. Endocrinology 143:621– 626 Shelby MD, Newbold RR, Tully DB, Chae K, Davis VL 1996 Assessing environmental chemicals for estrogenicity using a combination of in vitro and in vivo assays. Environ Health Perspect 104:1296 –1300 Reinhart AJ, Williams SC, Stocco DM 1999 Transcriptional regulation of the StAR gene. Mol Cell Endocrinol 151:161–169 Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ 1990 Basic local alignment search tool. J Mol Biol 215:403– 410 O’Shaughnessy PJ, Baker PJ, Heikkila M, Vainio S, McMahon AP 2000 Localization of 17␤-hydroxysteroid dehydrogenase/17-ketosteroid reductase isoform expression in the developing mouse testis: androstenedione is the major androgen secreted by fetal/neonatal Leydig cells. Endocrinology 141: 2631–2637 Baker PJ, Sha JH, O’Shaughnessy PJ 1997 Localisation and regulation of 17␤-hydroxysteroid dehydrogenase type 3 mRNA during development in the mouse testis. Mol Cell Endocrinol 133:127–133 Herbison AE 1998 Multimodal influence of estrogen upon gonadotropinreleasing hormone neurons. Endocr Rev 19:302–330 Couse JF, Lindzey J, Grandien K, Gustafsson JA, Korach KS 1997 Tissue


40.

41.

42. 43. 44. 45. 46. 47.

48. 49. 50. 51.

52. 53.

54. 55. 56. 57. 58.

distribution and quantitative analysis of estrogen receptor-␣ (ER␣) and estrogen receptor-␤ (ER␤) messenger ribonucleic acid in the wild type and ER␣knockout mouse. Endocrinology 138:4613– 4621 Lindzey J, Wetsel WC, Couse JF, Stoker T, Cooper R, Korach KS 1998 Effects of castration and chronic steroid treatments on hypothalamic gonadotropinreleasing hormone content and pituitary gonadotropins in male wild type and estrogen receptor-␣ knockout mice. Endocrinology 139:4092– 4101 Stefaneanu L, Kovacs K, Horvath E, Lloyd RV, Buchfelder M, Fahlbusch R, Smyth H 1994 In situ hybridization study of estrogen receptor messenger ribonucleic acid in human adenohypophysial cells and pituitary adenomas. J Clin Endocrinol Metab 78:83– 88 Wade GN, Blaustein JD, Gray JM, Meredith JM 1993 ICI 182,780: a pure antiestrogen that affects behaviors and energy balance in rats without acting in the brain. Am J Physiol 265:R1392–R1398 Scully KM, Gleiberman AS, Lindzey J, Lubahn DB, Korach KS, Rosenfeld MG 1997 Role of estrogen receptor-␣ in the anterior pituitary gland. Mol Endocrinol 11:674 – 681 Oliveira CA, Carnes K, Franca LR, Hess RA 2001 Infertility and testicular atrophy in the antiestrogen-treated adult male rat. Biol Reprod 65:913–920 Majdic G, Sharpe RM, O’Shaughnessy PJ, Saunders PT 1996 Expression of cytochrome P450 17␣-hydroxylase/C17–20 lyase in the fetal rat testis is reduced by maternal exposure to exogenous estrogens. Endocrinology 137:1063–1070 Ewing LL, Wing TY, Cochran RC, Kromann N, Zirkin BR 1983 Effect of luteinizing hormone on Leydig cell structure and testosterone secretion. Endocrinology 112:1763–1769 Welshons WV, Nagel SC, Thayer KA, Judy BM, Vom Saal FS 1999 Low-dose bioactivity of xenoestrogens in animals: fetal exposure to low doses of methoxychlor and other xenoestrogens increases adult prostate size in mice. Toxicol Ind Health 15:12–25 Spearow JL, Doemeny P, Sera R, Leffler R, Barkley M 1999 Genetic variation in susceptibility to endocrine disruption by estrogen in mice. Science 285: 1259 –1261 Harris HA, Bapat AR, Gonder DS, Frail DE 2002 The ligand binding profiles of estrogen receptors ␣ and ␤ are species dependent. Steroids 67:379 –384 Nozu K, Dufau ML, Catt KJ 1981 Estradiol receptor-mediated regulation of steroidogenesis in gonadotropin-desensitized Leydig cells. J Biol Chem 256: 1915–22 Price CA, Cooke GM, Sanford LM 2000 Influence of season and low-level oestradiol immunoneutralization on episodic LH and testosterone secretion and testicular steroidogenic enzymes and steroidogenic acute regulatory protein in the adult ram. J Reprod Fertil 118:251–262 Long BJ, Tilghman SL, Yue W, Thiantanawat A, Grigoryev DN, Brodie AM 1998 The steroidal antiestrogen ICI 182,780 is an inhibitor of cellular aromatase activity. J Steroid Biochem Mol Biol 67:293–304 Sanford LM 1989 Increases in testosterone secretion in adult rams with immunoneutralization of endogenous estradiol occur in the absence of increases in pulsatile LH release or testicular LH receptors. Acta Endocrinol (Copenh) 120:180 –186 Cowley SM, Parker MG 1999 A comparison of transcriptional activation by ER␣ and ER␤. J Steroid Biochem Mol Biol 69:165–175 Couse JF, Curtis Hewitt S, Korach KS 2000 Receptor null mice reveal contrasting roles for estrogen receptor ␣ and ␤ in reproductive tissues. J Steroid Biochem Mol Biol 74:287–296 Weihua Z, Saji S, Makinen S, Cheng G, Jensen EV, Warner M, Gustafsson JA 2000 Estrogen receptor (ER) ␤, a modulator of ER␣ in the uterus. Proc Natl Acad Sci USA 97:5936 –5941 Paech K, Webb P, Kuiper GG, Nilsson S, Gustafsson J, Kushner PJ, Scanlan TS 1997 Differential ligand activation of estrogen receptors ER␣ and ER␤ at AP1 sites. Science 277:1508 –1510 Kuiper GG, Lemmen JG, Carlsson B, Corton JC, Safe SH, van der Saag PT, van der Burg B, Gustafsson JA 1998 Interaction of estrogenic chemicals and phytoestrogens with estrogen receptor ␤. Endocrinology 139:4252– 4263