The Androgen Metabolite, 5-Androstane-3?, 17?-Diol, Is a Potent Modulator of Estrogen Receptor-?1-Mediated Gene Transcription in Neuronal Ce
http://www.100md.com
内分泌学杂志 2005年第1期
Department of Biomedical Science, College of Veterinary Medicine and Biomedical Sciences, Colorado State University, Fort Collins, Colorado 80523
Address all correspondence and requests for reprints to: Toni R. Pak, Department of Biomedical Sciences, College of Veterinary Medicine, Colorado State University, Fort Collins, Colorado 80523. E-mail: toni.pak@colostate.edu.
Abstract
5-Androstane-3?, 17?-diol (3?Adiol) is a metabolite of the potent androgen, 5-dihydrotestosterone. Recent studies showed that 3?Adiol binds to estrogen receptor (ER)-? and regulates growth of the prostate gland through an estrogen, and not androgen, receptor-mediated pathway. These data raise the possibility that 3?Adiol could regulate important physiological processes in other tissues that produce 3?Adiol, such as the brain. Although it is widely accepted that the brain is a target for 5-dihydrotestosterone action, there is no evidence that 3?Adiol has a direct action in neurons. To explore the molecular mechanisms by which 3?Adiol might act to modulate gene transcription in neuronal cells, we examined whether 3?Adiol activates ER-mediated promoter activity and whether ER transactivation is facilitated by a classical estrogen response element (ERE) or an AP-1 complex. The HT-22 neuronal cell line was cotransfected with an expression vector containing ER, ER-?1, or the ER? splice variant, ER-?2 and one of two luciferase-reporter constructs containing either a consensus ERE or an AP-1 enhancer site. Cells were treated with 100 nM 17?-estradiol, 100 nM 3?Adiol, or vehicle for 15 h. We show that 3?Adiol activated ER-?1-induced transcription mediated by an ERE equivalent to that of 17?-estradiol. By contrast, 3?Adiol had no effect on ER- or ER-?2-mediated promoter activity. Moreover, ER-?1 stimulated transcription mediated by an ERE and inhibited transcription by an AP-1 site in the absence of ligand binding. These data provide evidence for activation of ER signaling pathways by an androgen metabolite in neuronal cells.
Introduction
GONADAL STEROID HORMONES, such as androgens and estrogens, are critical regulators of a variety of physiological processes ranging from developmental sexual differentiation to adult reproduction. Historically, the biological actions of the androgen metabolite, 5-androstane-3?, 17?-diol (3?Adiol), have been largely ignored due to its very short circulating half-life (1). However, in rats, the administration of 3?Adiol stimulated prostate gland growth after castration-induced regression and, more recently, has been shown to play a critical role in the regulation of prostate gland epithelial cell proliferation, an important finding for understanding the regulatory mechanisms of prostate cancer (2, 3, 4). Recent reports have shown that 3?Adiol can be converted from its androgen precursor, 5-dihydrotestosterone (DHT), by a variety of steroid metabolizing enzymes, including: 17?-hydroxysteroid dehydrogenase (17?HSD), 3-hydroxysteroid oxidoreductase (3HSD), and 3?-hydroxysteroid oxidoreductase (3?HSD) (5, 6). These enzymes are ubiquitously expressed in many tissues, including the prostate gland and brain (7, 8, 9). To date, there is no direct evidence of a biological function for 3?Adiol in the brain, yet the abundance of steroid metabolizing enzymes and steroid receptors in the brain raises the possibility that 3?Adiol could be an important central regulator. Thus, in this study, we investigated the potential for 3?Adiol to mediate transcription in neuronal cells.
The biological actions of gonadal steroid hormones are mediated by receptors belonging to the superfamily of nuclear receptors (10). Although 3?Adiol is an androgen metabolite, in some tissues like the prostate gland it binds and acts through estrogen receptors (ERs) rather than androgen receptors (3, 4). There are two subtypes of ERs, termed ER and ER?, which are encoded by distinct genes but have highly homologous DNA-binding domains (97%) (11). Previous studies have demonstrated that these two subtypes differ in both structure and function. Further, the binding affinities of ER and ER? for their putative cognate ligand, 17?-estradiol (E2), are relatively similar despite only 57% homology in their ligand-binding domains (LBDs) (11, 12). Four splice variants of ER? (also called ER-?1) have been identified in rodents since its first detection in 1996, including: ER-?13, ER-?14, ER-?2, and ER-?23 (13, 14, 15, 16). Although a functional significance for these variant forms of ER? has yet to be elucidated, they are found in numerous tissues, including the brain (13, 15).
E2 is the presumed natural ligand for ER-?1, although other compounds exhibit either a high binding affinity or selectivity for ER-?1, including plant-derived and synthetic estrogens (17). Moreover, the androgen metabolite 3?Adiol binds to ER-?1 with a relatively high affinity and, in doing so, regulates prostate growth (3, 4, 17). These studies highlight the diversity of ligands that could bind and subsequently activate ER?. Consequently, whereas ER-?1 shares structural similarity to ER, it does not necessarily require estrogen as a ligand.
Ligand activation of steroid hormone receptors induces a conformational change that facilitates receptor dimerization and interaction with DNA at specific promoter enhancer sites. The classical DNA binding site for ERs is an estrogen response element (ERE) that contains two palindromic hexanucleotide repeats that bind a ligand-activated ER homodimer (18, 19). Both major ER subtypes (ER and ER-?1) activate gene transcription when mediated by an ERE (11, 20). However, direct binding to DNA is not an absolute requirement for ER-mediated transcription. For instance, the upstream promoter enhancer element, activator protein-1 (AP-1), also plays a role in ER-mediated gene transcription. Ligand-bound ERs interact with AP-1-associated Fos/Jun transcription factors to mediate transcription without binding directly to DNA (see Ref. 21 for review). However, it is particularly interesting that the two major ER subtypes (ER and ER-?1) act differentially at an AP-1 site, with ER- stimulating and ER? inhibiting transcription at that same site (20).
In the present study, we investigated the possibility that 3?Adiol has a biological function in neurons using a mouse hippocampal-derived neuronal cell line (HT-22). Specifically, we determined that 3?Adiol activates transcription mediated by ER-?1 and that this effect is mediated by an ERE and not an AP-1 complex. We provide evidence that 3?Adiol is a potent agonist for ER-?1-induced gene transcription in neurons. Notably, our data also revealed a ligand-independent regulation by ER-?1 at both an ERE and an AP-1 complex in a neuronal cell line.
Materials and Methods
Expression vectors
Plasmid expression vectors (pcDNA 3.0; Invitrogen, Carlsbad, CA) containing inserts for ER-?1 and ER-?2 were provided by Dr. Tom Brown (Pfizer Corp., Cambridge, MA). Full-length human rat ER was inserted into a PSG5 vector and provided by Dr. Richard Price (University of California San Francisco, San Francisco, CA).
Reporter constructs
The ERE-tk-luciferase reporter (generously provided by Dr. Paul Budworth, Case Western Reserve University, Cleveland, OH) contains two copies of vitellogenin ERE sequence coupled to the minimal tk-luciferase promoter and subcloned into pGL2-Basic plasmid (Promega Corp., Madison, WI). The AP1-tk-luciferase promoter (generously provided by Dr. Jennifer Nyborg, Colorado State University, Fort Collins, CO) contains three copies of the AP-1 sequence (TGACTCATGCTTT) coupled to the minimal tk-luciferase promoter and subcloned into pGL2-Basic plasmid. A ?-actin LacZ reporter construct, expressing ?-galactosidase under the control the human actin promoter, was used as an internal control for plasmid transfer efficiency.
Synthesis of hormone receptor proteins
Full-length rat ER, ER-?1, and ER-?2 expression vectors were used to synthesize receptor proteins in vitro using the TnT-coupled rabbit reticulocyte lysate system (Promega Corp.) with T7-RNA polymerase according to manufacturer’s directions.
Cell culture
The mouse hippocampus-derived cell line HT-22 (generously provided by Dr. Dave Schubert, Salk Institute, San Diego, CA) was used for all transient transfections. Cells were maintained in DMEM containing 4.5% glucose and L-glutamine (Invitrogen); supplemented with 1x nonessential amino acids, and 10% fetal bovine serum (FBS) (Gemini Bioproducts, Woodland, CA). Cells were grown to 70% confluency and used before passage 10 for all transient transfection experiments. HT-22 cells were selected for this study because they are a unique neuronal cell line that reportedly does not express ER (22) or ER-?1 (data not shown).
Transient transfections
Cells were plated at a density of 0.2 x 105 cells/well in 24-well plates 24 h before transfection. All constructs were transfected in triplicate wells within each assay, and each transfection assay was repeated a minimum of six times. Each experiment was performed using a minimum of three different preparations for each plasmid. Transfections were carried out using a lipid-mediated transfection reagent (Superfect; Qiagen Inc., Valencia, CA) according to manufacturer’s instructions) and a total DNA concentration of 2.2 μg/well. Cells were incubated with transfection media complex for 2.5 h, media was replaced with treatment media for 15 h, and then cells were lysed. All treatments were performed in DMEM containing 10% dextran-charcoal-stripped FBS to ensure estrogen-free culture conditions (Hyclone Laboratories, Logan, UT). In addition, each transfection experiment was performed a minimum of three times with phenol red-free media, and no differences were observed from the transfections performed with media containing phenol red (data not shown). Luciferase activity was measured using 20 μl lysate added to 100 μl luciferin substrate (Promega Corp.). ?-Galactosidase activity was measured using 40 μl lysate added to 200 μl galacton substrate (Tropix-GalactoLight kit; Applied Biosystems, Foster City, CA) according to manufacturer’s instructions). Relative light units (RLU) were measured using a 20/20 TD luminometer (Turner Designs).
Hormone treatments
The following compounds were diluted in dimethylsulfoxide (DMSO) and used at a final concentration of 100 nM in 0.01% DMSO: E2, 4OH-tamoxifen (TAM; Sigma-Aldrich Co., St. Louis, MO), 3?Adiol, 3Adiol (Steraloids, Newport, RI), and diarylpropionitrile (DPN; Tocris-Cookson, Inc., Ellisville, MO). Each compound was added to DMEM supplemented with 10% dextran-charcoal-stripped FBS (Hyclone Laboratories).
Saturation isotherms
To establish the binding affinity of 3?Adiol for various ERs, increasing concentrations (0.01–500 nM) of 3?Adiol were incubated with 100-μl aliquots of reticulocyte lysate supernatant for 90 min at room temperature (ER-?1), 4.5 h at 4 C (ER-?2), and 18 h at 4 C (ER), with 10-nM concentrations of [3H]-E2 in TEGMD (10 mM Tris-Cl, 1.5 mm EDTA, 10% glycerol, 25 mM molybdate, and 1.0 mM dithiothreitol, pH 7.4). These incubation times were determined empirically and represent optimal binding of receptor with E2. Nonspecific binding was assessed using a 200-fold excess of the ER agonist diethylstilbestrol in parallel tubes. After incubation, bound and unbound [3H]-E2 were separated by passing the reaction through a 1.0-ml lipophilic Sephadex LH-20 (Sigma-Aldrich Co.) column. Columns were constructed by packing a disposable pipette tip (1.0 ml; Labcraft, Curtin Matheson Scientific, Inc., Houston, TX) with TEGMD-saturated Sephadex according to previously published protocols (23). For chromatography, the columns were reequilibrated with TEGMD (100 μl), and the incubation reactions were added individually to each column and allowed to incubate on the column for an additional 30 min. After this incubation, 600 μl TEGMD was added to each column, flow-through was collected, 4 ml scintillation fluid was added, and samples were counted (5 min each) in a 2900 TR Packard scintillation counter (Packard Bioscience, Meriden, CT). Curve fitting, scientific graphing, and binding analyses were completed using GraphPad Software (GraphPad Prism 3.0, San Diego, CA).
EMSAs
Oligonucleotides.
Double-stranded oligonucleotides containing the vitellogenin consensus ERE sequence, or SP1 sequence, were 32P-end-labeled with T4 polynucleotide kinase. The percent incorporation was determined, and labeled probes with greater than 50% 32P incorporation were used for EMSAs.
Gel electrophoresis.
Receptor protein lysates were incubated with 100 nM E2, 100 nM 3?Adiol, or 0.01% DMSO (vehicle control) for 18 h before gel electrophoresis. After ligand binding, receptor lysates were incubated with 1 x gel shift binding buffer [20% glycerol, 5 mM MgCl2, 2.5 mM EDTA, 2.5 mM dithiothreitol, 250 mM NaCl, 50 mM Tris-HCl, 0.25 mg/ml poly(dI-dC)·poly(dI-dC)] for 10 min. Specific binding reactions were also incubated with 500- to 1000-fold excess of unlabeled ERE oligonucleotide. Nonspecific binding was tested using the 32P-SP1 oligonucleotide. After an initial 10-min incubation, 32P-ERE was added and incubated for an additional 20 min. DNA-protein complexes were resolved on a 6% Novex DNA retardation gel (Invitrogen) for 20 min at 250 V. Gels were dried on a vacuum gel dryer at 80 C for 1 h before autoradiography.
Autoradiography.
Dried gels were exposed to x-ray film (Biomax MS; Eastman Kodak Co., New Haven, CT) for 12 h at –70 C. Film images were captured using a CCD camera (Sony XC-77; Sony Corp., Tokyo, Japan) that was connected to a QuickCapture framegrabber board (Data Translation Inc., Marlboro, MA) on a PC computer. The captured images were analyzed with Scion Image software [W. Rasband, National Institutes of Health (NIH), Bethesda, MD]. Data are represented as mean density of pixels ± SD.
Statistics
Differences among hormone treatment groups for individual receptors were analyzed by one-way ANOVA followed by Tukey’s HSD test. Comparisons between control groups (empty vector + vehicle; receptor + vehicle) were analyzed using a t test. Differences were considered significant at P < 0.05. All transfection data are represented as percent change compared with vehicle-treated, empty-vector controls.
Results
Relative binding affinity of ERs for 3?Adiol
Competition binding studies were used to determine the relative binding affinities of the androgen metabolite 3?Adiol to ER, ER-?1, and ER-?2. In these experiments, displacement of 3H-E2 from the receptor provides a quantitative measurement of the affinity of the receptor for a specific ligand, such as 3?Adiol. Our results showed that, at optimal binding parameters, ER-?1 displayed the highest binding affinity for 3?Adiol compared with all other ERs tested (Table 1). The selectivity rank of 3?Adiol for all ERs was as follows: ER-?1 > ER > ER-?2. Although 3?Adiol bound all ERs tested, it was relatively weak compared with that of the cognate ER ligand E2 (Table 1).
TABLE 1. Competition assay depicting the relative binding affinities of various ER subtypes for the androgen metabolite 3?Adiol: a comparison with the cognate ligand E2
Constitutive (i.e. ligand-independent) regulation of ERE and AP-1-luciferase activity by ERs
ER-?1 and ER demonstrated constitutive alteration of ERE-luciferase promoter activity in the absence of any hormone treatment. Figure 1A illustrates a dose-dependent change in luciferase activity with varying concentrations of ER and ER-?1 expression vectors. ER-?1 significantly increased ERE-luciferase activity at concentrations as low as 0.01 μg/well, whereas ER did not significantly increase ERE-luciferase activity until concentrations reached 1.0 μg/well (Fig. 1A). Further, ER-?1 showed a very robust response, increasing ERE-luciferase activity by approximately 700% over empty vector controls. On the other hand, even at the highest dose used, the constitutive regulation by ER was relatively weak, increasing ERE-luciferase activity by only 160%.
FIG. 1. Ligand-independent effects of varying concentrations of ER and ER-?1 expression vectors on ERE- and AP1-luciferase (luc) promoter activity. Cotransfection of HT-22 cells with an ERE-tk-luc (A) or an AP-1-tk-luc (B) reporter construct and an expression vector containing varying concentrations of ER or ER-?1. After transfection, cells were treated with 0.01% DMSO (vehicle) in DMEM supplemented with 10% dextran-charcoal-stripped FBS for 15 h. Data are represented as percent change in RLU from vehicle-treated empty vector control ± SEM. The presence of a symbol denotes a significant difference from empty vector controls.
In cells cotransfected with an AP-1-luciferase reporter construct, ER-?1 significantly reduced promoter activity by about 72% at the highest concentration of 1.0 μg/well in a ligand-independent fashion (P < 0.05, Fig. 1B). A slightly lower, but statistically significant effect was also observed, with a lower concentration of 0.5 μg/well (Fig. 1B). By contrast, ER showed no ligand-independent regulation at an AP-1 site. Concentrations of ER expression vectors higher than 1.0 μg/well were not significantly more efficacious at altering ligand-independent activity of either the ERE or AP1 reporter constructs. Thus, in all subsequent experiments, a concentration of 1.0 μg was used for all ER expression vectors.
Comparison of 3?Adiol-activated transcription by ER and ER-?1 at an ERE and an AP-1 site
To determine whether 3?Adiol differentially activates transcription through ER and ER?, we cotransfected an ERE-luciferase reporter construct with an expression vector containing either ER or ER-?1 using the HT-22 neuronal cell line. Our results showed that treatment with 3?Adiol significantly stimulated ERE-driven luciferase activity through ER-?1 but had no effect on ER-mediated transcription (P < 0.05, Fig. 2A). Interestingly, 3?Adiol stimulated ER-?1-induced transcription at a level equivalent to that of E2, despite its much lower binding affinity (Table 1 and Fig. 2A). As expected, treatment with E2, the endogenous ligand for ER and ER?, significantly stimulated ERE-luciferase activity through both ERs (Fig. 2A). Similar results were observed, with hormone concentrations as low as 1.0 nM for E2 and 10 nM for 3?Adiol (data not shown). ER-?1 and ER also demonstrated constitutive (i.e. ligand-independent) up-regulation of ERE-luciferase promoter activity (Fig. 2A). In fact, ER-?1 significantly increased ERE-luciferase activity compared with that of ER in vehicle-treated control groups (Fig. 2A). Surprisingly, E2 did not significantly increase ER-?1-induced ERE-luciferase activity beyond the observed constitutive up-regulation (P = 0.06; Fig. 2A). Together, these results suggest that 3?Adiol is a selective agonist of ER? at an ERE.
FIG. 2. Effects of 3?Adiol on ERE- and AP-1-luc promoter activity mediated by ER and ER-?1. Cotransfection of HT-22 cells with an ERE-tk-luc (A) or an AP-1-tk-luc (B) reporter construct and an expression vector containing 1.0 μg/well ER or ER-?1. After transfection, cells were treated with 0.01% DMSO (vehicle), 100 nM E2, or 100 nM 3?Adiol for 15 h. Data are represented as percent change in RLU from vehicle-treated empty vector control ± SEM. The presence of a symbol (ER) or letter (ER-?1) denotes a significant difference from empty vector controls. Dissimilar symbols/letters denote a statistically significant difference among treatment groups (P < 0.05).
In cells cotransfected with the AP1-reporter, construct ER-?1, but not ER, dramatically reduced luciferase activity. Treatment with E2 and 3?Adiol did not alter this effect, unlike the observed agonistic effect of 3?Adiol at an ERE (Fig. 2B). Interestingly, E2 treatment did not significantly increase ER-mediated AP1-luciferase activity (Fig. 2B). It is possible that this response is specific to the HT-22 cell line, or it might represent a more general difference between peripheral and neuronal cell lines.
Comparison of 3?Adiol with other ER ligands on ER-?1-mediated ERE-luciferase activity
Selective ER modulators (SERMs) differentially regulate transcription at an ERE and an AP-1 site (see Ref. 24 for review). In fact, SERMs that antagonize ER can act as an agonist for ER-?1 (20). This experiment compared the ligand-activating profile of 3?Adiol with other ER ligands known to bind and activate ER-?1. We also investigated whether the co-metabolite and stereoisomer of 3?Adiol, 3Adiol could similarly stimulate ER-?1-mediated gene transcription. Cells cotransfected with an ERE-luciferase reporter and ER-?1 were exposed to three treatments: 1) a highly selective ER-?1 agonist, DPN; 2) the SERM, TAM; and 3) 3Adiol. DPN significantly stimulated ER-?1-mediated promoter activity above the observed constitutive increase but not greater than that of 3?Adiol (Fig. 3A). Further, treatment with TAM completely abolished the constitutive ER-?1-induced up-regulation of ERE-mediated promoter activity (Fig. 3A). Last, 3Adiol had no effect on promoter activity mediated by an ERE, despite its structure being similar to that of 3?Adiol (Fig. 3A). Hence, the ligand-activating profile of 3?Adiol mimics that of a selective ER? agonist in the HT-22 neuronal cell line.
FIG. 3. Effects of various ER ligands on ER-?1-induced promoter activity at an ERE or an AP-1-site. Cotransfection of HT-22 cells with an ERE-tk-luc (A) or an AP-1-tk-luc (B) reporter construct and an expression vector containing 1.0 μg/well ER-?1. After transfection, cells were treated with 0.01% DMSO (vehicle), 100 nM E2, 100 nM 3?Adiol, 100 nM DPN, 1.0 μM TAM, or 100 nM 3Adiol for 15 h. Data are represented as percent change in RLU from vehicle-treated empty vector control ± SEM. The presence of a letter denotes a significant difference from empty vector controls. Dissimilar letters denote a statistically significant difference among groups (P < 0.05).
Comparison of 3?Adiol to other ER ligands on ER-?1-mediated AP1-luciferase activity
As shown in Fig. 2B, 3?Adiol had no effect on ER-?1-induced transcription mediated by an AP-1 complex. Similarly, the ER-?1 agonist DPN failed to modulate ER-?1-mediated inhibition of AP-1-luciferase promoter activity (Fig. 3B). By contrast, treatment with TAM abolished the constitutive inhibitory effect of ER-?1 on AP-1-luciferase promoter activity, thereby antagonizing the ligand-independent effects of ER-?1 (Fig. 3B). Similar to what we observed at an ERE, 3Adiol had no effect at an AP-1 site (Fig. 3B). Taken together, these results demonstrate that: 1) ER agonists do not further attenuate the observed ER-?1-induced constitutive decrease in transcription mediated by an AP-1 complex; and 2) potential ER antagonists, such as TAM, abolish the ligand-independent effects of ER-?1.
Effect of 3?Adiol on ER-?2-induced transcription at an ERE and an AP-1 site
The ER-?1 splice variant ER-?2 contains an extra 18-amino-acid sequence in the LBD. Consequently, ER-?2 binds most ligands less effectively than ER-?1. Consistent with this observation, we showed that ER-?2 binds 3?Adiol with an approximately 50-fold lower affinity than that of ER-?1 [Ki (nM) = 1.7 ± 0.3 and 58.1 ± 5.5, respectively]. However, despite these lower binding affinities, in most systems, ER-?2 has been shown to mimic, albeit to a lesser degree, the actions of ER-?1 for most ligands tested (13). Thus, we examined whether 3?Adiol stimulates ER-?2-induced transcription mediated by an ERE or an AP-1 enhancer site. Most strikingly, ER-?2 did not constitutively increase ERE-luciferase promoter activity as was observed with ER-?1 (Fig. 4A). This demonstrates an inherent difference in the ER? splice variants that is apparently dependent on their differences in the LBD. Further, 3?Adiol had no effect on ER-?2-induced transcription despite ER-?2s ability to bind 3?Adiol, albeit weakly (Fig. 4A). The ER agonists E2 and DPN were effective at stimulating ER-?2-mediated transcription at an ERE, as expected (P < 0.05; Fig. 4A). Surprisingly, 3Adiol also significantly increased ER-?2-induced ERE-luciferase activity (P < 0.05; Fig. 4A).
FIG. 4. Effects of various ER ligands on ER-?2-induced promoter activity at an ERE or an AP-1-site. Cotransfection of HT-22 cells with an ERE-tk-luc (A) or an AP-1-tk-luc (B) reporter construct and an expression vector containing 1.0 μg/well ER-?2. After transfection, cells were treated with 0.01% DMSO (vehicle), 100 nM E2, 100 nM 3?Adiol, 100 nM DPN, 1.0 μM TAM, or 100 nM 3Adiol for 15 h. Data are represented as percent change in RLU from vehicle-treated empty vector control ± SEM. The presence of a letter denotes a significant difference from empty vector controls. Dissimilar letters denote a statistically significant difference among groups (P < 0.05). An asterisk denotes an intermediate effect of 3Adiol.
Cotransfection of ER-?2 and an AP-1-luciferase reporter construct showed that ER-?2 decreased promoter activity independent of hormone treatment, similar to ER-?1 (P < 0.05; Fig. 4B). However, treatment with the ER agonists E2 and DPN significantly enhanced this response, in contrast to their lack of effect on the ER-?1-mediated constitutive decrease at an AP-1 site (Fig. 4B). The ER antagonist TAM abolished the ligand-independent reduction of ER-?2-mediated transcription at an AP-1 site (Fig. 4B). Overall, these data demonstrate that ER-?2 has important functional differences from the ER-?1 splice variant.
Effects of 3?Adiol on ER-?1 binding to an ERE
Classical models of ER-mediated transcription pathways dictate that ER binding to an ERE occurs after ligand activation of the receptor. However, our data show that ER and ER-?1’s ability to stimulate ERE-mediated transcription is independent of ligand binding. Moreover, a novel ER-?1 ligand, 3?Adiol, further increased the observed constitutive regulation by ER-?1 (Fig. 2). Thus, we used EMSAs to: 1) confirm that ER and ER-?1 binds an ERE in the absence of ligand; and 2) determine whether 3?Adiol enhances the binding potency of ER-?1 at an ERE.
Lysate containing translated ER, ER-?1, and ER-?2 protein was incubated with a 32P-labeled ERE oligonucleotide and resolved on a 6% DNA retardation gel. After autoradiography, the results demonstrated a strong shift for all ERs (ER, ER-?1, and ER?2), indicating that all receptors bind an ERE (Fig. 5). Interestingly, there was no significant difference in the mean density/pixel2 in the absence or presence of ligand (Fig. 5 and Table 2). Moreover, the addition of 3?Adiol did not significantly alter the binding of any ER tested to the ERE (Table 2). Unlabeled ERE oligonucleotide (1000-fold excess) effectively competed for each ER binding to an ERE, and there was no shift observed when receptor proteins were incubated with a 32P-labeled SP1 oligonucleotide (Fig. 5, lanes 1 and 2). These results demonstrate the specificity of ER:DNA binding at an ERE.
FIG. 5. Effects of 3?Adiol on ER, ER-?1, and ER-?2 DNA binding. Vitellogenin consensus ERE oligonucleotide (0.2 pM) was 32P-end-labeled and incubated with 2 μl ER (A and B; lanes 3–6), ER-?1 (A; lanes 7–10), or ER-?2 (B; lanes 7–10) receptor protein lysate. Receptor proteins were incubated with 0.01% DMSO (vehicle), 100 nM E2, 100 nM 3?Adiol, or 100 nM 3Adiol for 18 h at 4 C. After 32P-ERE incubation, reaction product was resolved on a 6% DNA retardation gel for 20 min at 250 V. Gels were dried and exposed to autoradiographic film for 12 h. Unlabeled ERE oligonucleotide was added to determine binding specificity 1000-fold excess of 32P-ERE (A and B; lane 1). 32P-labeled SP1 oligonucleotide (SP-1 olig) was used as a negative control (A and B; lane 2).
TABLE 2. EMSA showing the effects of various ligands on ER, ER-?1, and ER-?2 binding to an ERE
Discussion
With these studies, we determined that, in a neuronal cell line, the androgen metabolite 3?Adiol binds ER-?1 and subsequently alters promoter activity mediated by an ERE and an AP-1 complex. These findings demonstrate a possible mechanism for the direct regulation by androgens in neuronal cell types that lack a classical androgen receptor (AR). In addition, our data show that ER-?1 differentially regulated transcription mediated by an ERE and an AP-1 site independent of E2 ligand. Thus, these data highlight the potential for ER? to act as a basic transcription factor in ER-?1-expressing neuronal target cells.
The androgen metabolite 3?Adiol is an endogenous ligand for ER-?1
In this study, we used a hippocampus-derived cell line to verify that 3?Adiol is an endogenous ligand for ER-?1 in neurons. Anatomical evidence suggests that 3?Adiol is produced in the brain, raising the possibility that it is an important central regulator. For example, the major enzymes for converting DHT to 3?Adiol, (3?-HSD, 3-HSD, and 17?-HSD) are expressed in many regions of the rat brain throughout development, including the cortex, hypothalamus, thalamus, septum, hippocampus, and cerebellum (5, 6, 7, 8, 9). Notably, 3?-HSD is expressed at highest levels in the cerebellum, which is also rich in ER-?1 (25, 26, 27). Although cellular colocalization of ER-?1 and 3?-HSD has yet to be shown, it is reasonable to assume that local production of 3?Adiol in these brain regions could activate cells expressing ER-?1.
3?Adiol is likely the major androgen metabolite produced in the prostate gland and plays a significant role in regulating prostate epithelial cell proliferation (3, 4). Importantly, ER-?1, and not AR, mediated 3?Adiol’s effect on prostate growth (3, 4). Our results revealed a potential molecular pathway for this novel endogenous ligand by demonstrating that 3?Adiol significantly increased ER-?1-mediated transcription at an ERE. These studies verify those of Kuiper et al. (1998) (17) that 3?Adiol binds ERs with a selectivity for ER-?1 over ER. Further, our data also demonstrate that the transcriptional selectivity of 3?Adiol is much greater than its binding selectivity.
The current model describing the role of coregulatory proteins in the ER signaling pathway maintains that unbound ERs are tethered to corepressor proteins that inhibit transactivation (see Ref. 10 for review). Ligand-binding initiates the dissociation of corepressors, such as nuclear receptor corepressor and silencing mediator for retinoid and thyroid hormone receptors, followed by the recruitment of coactivators, such as glucocorticoid receptor interacting protein and steroid receptor coactivator-1 (SRC-1). The failure of 3?Adiol to activate transcription upon binding to ER and ER-?2 in the present study raises the possibility that 3?Adiol might prevent the dissociation of, or actively recruits, one or more corepressor proteins when bound to ER and ER-?2. This hypothesis is consistent with Shang et al. (28), who showed that nuclear receptor corepressor and silencing mediator for retinoid and thyroid hormone receptors are associated with ER when it is bound to the SERM, TAM. Thus, we propose that 3?Adiol functions as an antagonist for ER and ER-?2 but as an agonist for ER-?1.
Alternatively, 3?Adiol might affect components of the signaling pathway downstream from the dissociation of corepressor proteins. For instance, coactivator proteins bind to an LXXLL motif embedded in the LBD and mediate transcription by binding histone acetyl-transferases, such as p300/CBP, that induce conformational changes in the chromatin (29, 30, 31). It is possible that inherent variations in the LBD among ER subtypes may confer specificity for the recruitment of distinct coactivators. Thus, we hypothesize that 3?Adiol binds ER-?1 and subsequently recruits coactivator proteins distinct from that of ER and ER-?2, thereby potentiating ERE-mediated transcription.
Overall, our results indicate that 3?Adiol is a specific agonist for ER-?1 and not other ER subtypes. Although 3?Adiol binds all ERs tested, transcription is initiated only upon binding to ER-?1. In fact, our data demonstrate that 3?Adiol increased promoter activity equal to that of the cognate ligand E2, despite having a lower binding affinity for ER-?1. Moreover, 3Adiol had no effect on ER-?1 transactivation, which is consistent with its very low binding affinity (11). The specificity of 3?Adiol for ER-?1 is likely due to inherent differences in the LBDs of the various ERs. We speculate that 3?Adiol could also act in an antagonistic fashion for ER and ER-?2. Indeed, 3?Adiol significantly attenuated the ER-induced constitutive regulation of ERE-luciferase promoter activity, lending further evidence for a possible antagonistic action of 3?Adiol through ER (Fig. 2A). In contrast, 3Adiol seems to mediate transcription preferentially through ER-?2 (Fig. 4A).
ER-?1 constitutively regulates transcription mediated by ERE and AP-1 enhancer elements
Constitutive regulation of promoter activity has been demonstrated for several members of the superfamily of nuclear receptors. Of particular relevance to our study, both ER and ER? have been shown to constitutively regulate gene transcription. For instance, the tyrosine mutant ER, found in metastatic breast cancer cells, stimulated ERE-mediated transcription in HeLa cells and was unaffected by E2 treatment (32). Moreover, ER-?1 increased the activity of arginine vasopressin promoter in vitro in the absence of E2 (33). In that study, ER-?1 constitutively increased arginine vasopressin promoter activity in a concentration-dependent fashion. Our results confirm that ER-?1 does not require ligand activation to modulate transcription through classical ER pathways. Interestingly, E2 treatment did not significantly increase ER-?1-induced ERE-luciferase activity above that of the observed ligand-independent increase. Also, E2 treatment did not significantly increase ER-mediated AP1-luciferase activity, as has been reported by others using cell lines derived from peripheral tissues, such as HeLa cells (20). This might be cell-line specific for HT-22 cells or it might represent a more general difference between peripheral and neuronal cell types. Taken together our results demonstrate that ER-?1 can serve as a basic transcription factor in a ligand-independent fashion, as well as a classical ligand-activated steroid hormone receptor.
A variety of neurotransmitters and growth factors, such as dopamine, IGF-I, and EGF, as well as the membrane-associated protein caveolin-1, have been shown to activate ER-induced transcription in the absence of E2 (34, 35, 36, 37). Such ligand-independent activation of ERs is thought to occur at the N-terminal activator-function (AF)-1 domain of the ER, whereas ligand-dependent activation occurs at the C-terminal AF-2 domain (38, 39). Furthermore, phosphorylation of the AF-1 domain of ER-?1 elicits the recruitment of coregulatory proteins, such as SRC-1, potentiating the ligand-independent transcriptional regulation (40, 41, 42, 43). It is likely that some, or all, of these factors are involved in the ER-?1-induced constitutive regulation demonstrated in the present study.
The classical ER signaling pathway suggests that ligand-bound ERs form homodimers and activate transcription by binding directly to DNA at an ERE (18, 19). However, this paradigm is uncertain because studies have shown both the lack, and the occurrence, of ER:ERE interaction in the absence of ligand (13, 14, 43). Our results confirm those of Petersen et al. (13) that ERs can bind to an ERE in the absence of ligand, thus providing a mechanism for the ER- and ER-?1-induced constitutive regulation of ERE-mediated transcription shown in our study.
An alternative ER signaling pathway has also been described, whereby the transactivational domain of the ER can interact with other transcription factor proteins, such as members of the Fos and Jun families, and subsequently modulate transcription through these protein:protein interactions (44). Interestingly, the ligand-independent transactivation by ER-?1 had opposite effects when mediated by an ERE, compared with an AP-1 complex. Differential activation of ER-?1 at an ERE and an AP-1 site has been shown previously, although only in the presence of ligand (20). To our knowledge, this is the first report demonstrating constitutive promoter regulation by ER-?1 mediated at an AP-1 site (see Ref. 10 for review). Furthermore, these data highlight the potential confounds for discerning the molecular mechanisms of ER-regulated promoter activity, especially complex promoters comprised of both ERE and AP-1 regulatory elements.
SERMs, such as TAM, can be both agonistic and antagonistic, depending on the cell type, ER subtype, and interacting promoter response element (45, 46). For instance, it has been reported that TAM stimulated ER-?1-induced transcription when mediated by an AP-1 complex (20). In contrast, we demonstrated that ER-?1 constitutively decreased promoter activity at an AP-1 site, and treatment with TAM abolished this response but did not stimulate promoter activity over baseline expression levels. However, without the inclusion of an empty vector control to demonstrate this, the effect of TAM could be interpreted as stimulating ER?1-induced promoter activity. In addition, for these studies, we used a neuronal cell line as opposed to the cervical epithelial-derived HeLa cell line used by Paech et al. (20). Such differences in experimental paradigms might further explain these discrepant results. Taken together, these data might also reflect some of the observed differences in SERM action in peripheral vs. neuronal tissue types.
In summary, this study demonstrated that 3?Adiol, an androgen metabolite, facilitated ER-mediated transcription through classical ER signaling pathways in a neuronal cell type. This raises important concerns for using the nonaromatizable androgen DHT as the penultimate ligand to study AR-dependent pathways. Further, the constitutive regulation by ERs suggests that cell-specific expression has important biological consequences independent of hormonal milieu. The importance of coregulatory proteins involved in these processes is currently under investigation.
Acknowledgments
The authors thank Matthew Ewert and Brian Cherrington for their technical assistance.
References
Grover PK, Odell WD 1975 Correlation of in vivo and in vitro activities of some naturally occurring androgens using a radioreceptor assay for 5-dihydrotestosterone with rat prostate cytosol receptor protein. J Steroid Biochem 6:1373–1379
Ahmad N, Warren DW, Haltmeyer GC 1978 The effects of 5-reduced androgens on maintenance and regeneration of prostate glands and seminal vesicles in castrated and hypophysectomized rats. Anat Rec 192:543–554
Weihua Z, M?kel? S, Andersson LC, Salmi S, Saji S, Webster JI, Jensen EV, Nilsson S, Warner M, Gustafsson JA 2001 A role for estrogen receptor ? in the regulation of growth of the ventral prostate. Proc Natl Acad Sci USA 98:6330–6335
Weihua Z, Lathe R, Warner M, Gustafsson JA 2002 An endocrine pathway in the prostate, ER?, AR, 5-androstane-3?, 17?-diol, and CYP7B1 regulates prostate growth. Proc Natl Acad Sci USA 99:13589–13594
Torn S, Nokelainen P, Kurkela R, Pulkka A, Menjivar M, Ghosh S, Cocaprados M, Peltoketo H, Isomaa V, Vihko P 2003 Production, purification, and functional analysis of recombinant human and mouse 17?-hydroxysteroid dehydrogenase type 7. Biochem Biophys Res Commun 305:37–45
Steckelbroeck S, Jin Y, Gopishetty S, Oyesanmi B, Penning TM 2004 Human cytosolic 3-hydroxysteroid dehydrogenase of the aldo-keto reductase superfamily display significant 3?-hydroxysteroid dehydrogenase activity: implications for steroid hormone metabolism and action. J Biol Chem 279:10784–10795
Guennoun R, Fiddles RJ, Gouezou M, Lombes M, Baulieu EE 1995 A key enzyme in the biosynthesis of neurosteroids, 3?-hydroxysteroid dehydrogenase/5- 4-isomerase (3?-HSD), is expressed in rat brain. Brain Res Mol Brain Res 30:287–300
Pelletier G, Luu-The V, Labrie F 1995 Immunocytochemical localization of type I 17-? hydroxysteroid dehydrogenase in the rat brain. Brain Res 704:233–239
Ibanez C, Guennoun R, Liere P, Eychenne B, Pianos A, El-Etr M, Baulieu EE, Schumacher M 2003 Developmental expression of genes involved in neurosteroidogenesis: 3?-hydroxysteroid dehydrogenase/5-4 isomerase in the rat brain. Endocrinology 144:2902–2911
McKenna NJ, O’Malley BW 2002 Combinatorial control of gene expression by nuclear receptors and coregulators. Cell 108:465–474
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
Mosselman S, Polman J, Dijkema R 1996 ER?: identification and characterization of a novel human estrogen receptor. FEBS Lett 392:49–53
Petersen DN, Tkalcevic GT, Koza-Taylor PH, Turi TG, Brown TA 1998 Identification of estrogen receptor ?2, a functional variant of estrogen receptor ? expressed in normal rat tissues. Endocrinology 139:1082–1092
Hanstein B, Liu H, Yancisin MC, Brown M 1999 Functional analysis of a novel estrogen receptor-? isoform. Mol Endocrinol 13:129–137
Price Jr RH, Lorenzon N, Handa RJ 2000 Differential expression of estrogen receptor ? splice variants in rat brain: identification and characterization of a novel variant missing exon 4. Mol Brain Res 80:260–268
Price Jr RH, Butler CA, Webb PA, Uht R, Kushner P, Handa RJ 2001 A splice variant of estrogen receptor ? missing exon 3 displays altered subnuclear localization and capacity for transcriptional activation. Endocrinology 142:2039–2049
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
Kumar V, Chambon P 1988 The estrogen receptor binds tightly to its responsive element as a ligand-induced homodimer. Cell 55:145–156
Klein-Hitpass L, Tsai SY, Greene GL, Clark JH, Tsai MJ, O’Malley BW 1989 Specific binding of estrogen receptor to the estrogen response element. Mol Cell Biol 9:43–49
Paech K, Webb P, Kuiper GG, Nilsson S, Gustafsson JA, Kushner PJ, Scanlan TS 1997 Differential ligand activation of estrogen receptors ER and ER? at AP1 sites. Science 277:1508–1510
Kushner PJ, Agard DA, Greene GL, Scanlan TS, Shiau AK, Uht RM, Webb P 2000 Estrogen receptor pathways to AP-1. J Steroid Biochem Mol Biol 74:311–317
Kim H, Bang OY, Jung MW, Ha SD, Hong HS, Huh K, Kim SU, Mook-Jung I 2001 Neuroprotective effects of estrogen against ?-amyloid toxicity are mediated by estrogen receptors in cultured neuronal cells. Neurosci Lett 302:58–62
Handa RJ, Stadelman HL, Resko JA 1987 Effect of estrogen on androgen receptor dynamics in female rat pituitary. Endocrinology 121:84–89
McDonnell DP, Wijayaratne A, Chang CY, Norris JD 2002 Elucidation of the molecular mechanism of action of selective estrogen receptor modulators. Am J Cardiol 90:35F–43F
Price Jr RH, Handa RJ 2000 Expression of estrogen receptor-? protein and mRNA in the cerebellum of the rat. Neurosci Lett 288:115–118
Shughrue PJ, Merchenthaler I 2001 Distribution of estrogen receptor ? immunoreactivity in the rat central nervous system. J Comp Neurol 436:64–81
Jakab RL, Wong JK, Belcher SM 2001 Estrogen receptor ? immunoreactivity in differentiating cells of the developing rat cerebellum. J Comp Neurol 430:396–409
Shang Y, Hu X, DiRenzo J, Lazar MA, Brown M 2000 Cofactor dynamics and sufficiency in estrogen receptor-regulated transcription. Cell 103:843–852
Heery DM, Kalkhoven E, Hoare S, Parker MG 1997 A signature motif in transcriptional co-activators mediates binding to nuclear receptors. Nature 387:733–736
Feng W, Ribeiro RC, Wagner RL, Nguyen H, Apriletti JW, Fletterick RJ, Baxter JD, Kushner PJ, West BL 1998 Hormone-dependent coactivator binding to a hydrophobic cleft on nuclear receptors. Science 280:1747–1749
Perissi V, Dasen JS, Kurokawa R, Wang Z, Korzus E, Rose DW, Glass CK, Rosenfeld MG 1999 Factor-specific modulation of CREB-binding protein acetyltransferase activity. Proc Natl Acad Sci USA 96:3652–3657
Zhang QX, Borg A, Wolf DM, Oesterreich S, Fuqua SA 1997 An estrogen receptor mutant with strong hormone-independent activity from a metastatic breast cancer. Cancer Res 57:1244–1249
Shapiro RA, Xu C, Dorsa DM 2000 Differential transcriptional regulation of rat vasopressin gene expression by estrogen receptor and ?. Endocrinology 141:4056–4064
Power RF, Mani SK, Codina J, Conneely OM, O’Malley BW 1991 Dopaminergic and ligand-independent activation of steroid hormone receptors. Science 254:1636–1639
Ignar-Trowbridge DM, Nelson KG, Bidwell MC, Curtis SW, Washburn TF, McLachlan JA, Korach KS 1992 Coupling of dual signaling pathways: epidermal growth factor action involves the estrogen receptor. Proc Natl Acad Sci USA 89:4658–4662
Klotz DM, Hewitt SC, Ciana P, Raviscioni M, Lindzey JK, Foley J, Maggi A, DiAugustine RP, Korach KS 2002 Requirement of estrogen receptor- in insulin-like growth factor-1 (IGF-1)-induced uterine responses and in vivo evidence for IGF-1/estrogen receptor cross-talk. J Biol Chem 277:8531–8537
Schlegel A, Wang C, Pestell RG, Lisanti MP 2001 Ligand-independent activation of oestrogen receptor by caveolin-1. Biochem J 359:203–210
Goff PL, Montano MM, Schodin DJ, Katzenellenbogen BS 1994 Phosphorylation of the human estrogen receptor. Identification of hormone-regulated sites and examination of their influence on transcriptional activity. J Biol Chem 269:4458–4466
Arnold SF, Obourn JD, Jaffe H, Notides AC 1994 Serine 167 is the major estradiol-induced phosphorylation site on human estrogen receptor. Mol Endocrinol 8:1208–1214
Tremblay A, Tremblay GB, Labrie F, Giguere V 1999 Ligand-independent recruitment of SRC-1 to estrogen receptor ? through phosphorylation of activation function AF-1. Mol Cell 3:513–519
Tremblay A, Giguere V 2001 Contribution of steroid receptor coactivator-1 and CREB binding protein in ligand-independent activity of estrogen receptor ?. J Steroid Biochem Mol Biol 77:19–27
Webb P, Nguyen P, Shinsako J, Anderson C, Feng W, Nguyen MP, Chen D, Huang SM, Subramanian S, McKinerney E, Katzenellenbogen BS, Stallcup MR, Kushner PJ 1998 Estrogen receptor activation function 1 works by binding p160 coactivator proteins. Mol Endocrinol 12:1605–1618
Murdoch FE, Meier DA, Furlow JD, Grunwald KA, Gorski J 1990 Estrogen receptor binding to a DNA response element in vitro is not dependent upon estradiol. Biochemistry 29:8377–8385
Webb P, Nguyen P, Valentine C, Lopez GN, Kwok GR, McInerney E, Katzenellenbogen BS, Enmark E, Gustafsson JA, Nilsson S, Kushner PJ 1999 The estrogen receptor enhances AP-1 activity by two distinct mechanisms with different requirements for receptor transactivation functions. Mol Endocrinol 13:1672–1685
Webb P, Lopez GN, Uht RM, Kushner PJ 1995 Tamoxifen activation of the estrogen receptor/AP-1 pathway: potential origin for the cell-specific estrogen-like effects of antiestrogens. Mol Endocrinol 9:443–456
Tremblay GB, Tremblay A, Labrie F, Giguere V 1998 Ligand-independent activation of the estrogen receptors and ? by mutations of a conserved tyrosine can be abolished by antiestrogens. Cancer Res 58:877–881(Toni R. Pak, Wilson C. J.)
Address all correspondence and requests for reprints to: Toni R. Pak, Department of Biomedical Sciences, College of Veterinary Medicine, Colorado State University, Fort Collins, Colorado 80523. E-mail: toni.pak@colostate.edu.
Abstract
5-Androstane-3?, 17?-diol (3?Adiol) is a metabolite of the potent androgen, 5-dihydrotestosterone. Recent studies showed that 3?Adiol binds to estrogen receptor (ER)-? and regulates growth of the prostate gland through an estrogen, and not androgen, receptor-mediated pathway. These data raise the possibility that 3?Adiol could regulate important physiological processes in other tissues that produce 3?Adiol, such as the brain. Although it is widely accepted that the brain is a target for 5-dihydrotestosterone action, there is no evidence that 3?Adiol has a direct action in neurons. To explore the molecular mechanisms by which 3?Adiol might act to modulate gene transcription in neuronal cells, we examined whether 3?Adiol activates ER-mediated promoter activity and whether ER transactivation is facilitated by a classical estrogen response element (ERE) or an AP-1 complex. The HT-22 neuronal cell line was cotransfected with an expression vector containing ER, ER-?1, or the ER? splice variant, ER-?2 and one of two luciferase-reporter constructs containing either a consensus ERE or an AP-1 enhancer site. Cells were treated with 100 nM 17?-estradiol, 100 nM 3?Adiol, or vehicle for 15 h. We show that 3?Adiol activated ER-?1-induced transcription mediated by an ERE equivalent to that of 17?-estradiol. By contrast, 3?Adiol had no effect on ER- or ER-?2-mediated promoter activity. Moreover, ER-?1 stimulated transcription mediated by an ERE and inhibited transcription by an AP-1 site in the absence of ligand binding. These data provide evidence for activation of ER signaling pathways by an androgen metabolite in neuronal cells.
Introduction
GONADAL STEROID HORMONES, such as androgens and estrogens, are critical regulators of a variety of physiological processes ranging from developmental sexual differentiation to adult reproduction. Historically, the biological actions of the androgen metabolite, 5-androstane-3?, 17?-diol (3?Adiol), have been largely ignored due to its very short circulating half-life (1). However, in rats, the administration of 3?Adiol stimulated prostate gland growth after castration-induced regression and, more recently, has been shown to play a critical role in the regulation of prostate gland epithelial cell proliferation, an important finding for understanding the regulatory mechanisms of prostate cancer (2, 3, 4). Recent reports have shown that 3?Adiol can be converted from its androgen precursor, 5-dihydrotestosterone (DHT), by a variety of steroid metabolizing enzymes, including: 17?-hydroxysteroid dehydrogenase (17?HSD), 3-hydroxysteroid oxidoreductase (3HSD), and 3?-hydroxysteroid oxidoreductase (3?HSD) (5, 6). These enzymes are ubiquitously expressed in many tissues, including the prostate gland and brain (7, 8, 9). To date, there is no direct evidence of a biological function for 3?Adiol in the brain, yet the abundance of steroid metabolizing enzymes and steroid receptors in the brain raises the possibility that 3?Adiol could be an important central regulator. Thus, in this study, we investigated the potential for 3?Adiol to mediate transcription in neuronal cells.
The biological actions of gonadal steroid hormones are mediated by receptors belonging to the superfamily of nuclear receptors (10). Although 3?Adiol is an androgen metabolite, in some tissues like the prostate gland it binds and acts through estrogen receptors (ERs) rather than androgen receptors (3, 4). There are two subtypes of ERs, termed ER and ER?, which are encoded by distinct genes but have highly homologous DNA-binding domains (97%) (11). Previous studies have demonstrated that these two subtypes differ in both structure and function. Further, the binding affinities of ER and ER? for their putative cognate ligand, 17?-estradiol (E2), are relatively similar despite only 57% homology in their ligand-binding domains (LBDs) (11, 12). Four splice variants of ER? (also called ER-?1) have been identified in rodents since its first detection in 1996, including: ER-?13, ER-?14, ER-?2, and ER-?23 (13, 14, 15, 16). Although a functional significance for these variant forms of ER? has yet to be elucidated, they are found in numerous tissues, including the brain (13, 15).
E2 is the presumed natural ligand for ER-?1, although other compounds exhibit either a high binding affinity or selectivity for ER-?1, including plant-derived and synthetic estrogens (17). Moreover, the androgen metabolite 3?Adiol binds to ER-?1 with a relatively high affinity and, in doing so, regulates prostate growth (3, 4, 17). These studies highlight the diversity of ligands that could bind and subsequently activate ER?. Consequently, whereas ER-?1 shares structural similarity to ER, it does not necessarily require estrogen as a ligand.
Ligand activation of steroid hormone receptors induces a conformational change that facilitates receptor dimerization and interaction with DNA at specific promoter enhancer sites. The classical DNA binding site for ERs is an estrogen response element (ERE) that contains two palindromic hexanucleotide repeats that bind a ligand-activated ER homodimer (18, 19). Both major ER subtypes (ER and ER-?1) activate gene transcription when mediated by an ERE (11, 20). However, direct binding to DNA is not an absolute requirement for ER-mediated transcription. For instance, the upstream promoter enhancer element, activator protein-1 (AP-1), also plays a role in ER-mediated gene transcription. Ligand-bound ERs interact with AP-1-associated Fos/Jun transcription factors to mediate transcription without binding directly to DNA (see Ref. 21 for review). However, it is particularly interesting that the two major ER subtypes (ER and ER-?1) act differentially at an AP-1 site, with ER- stimulating and ER? inhibiting transcription at that same site (20).
In the present study, we investigated the possibility that 3?Adiol has a biological function in neurons using a mouse hippocampal-derived neuronal cell line (HT-22). Specifically, we determined that 3?Adiol activates transcription mediated by ER-?1 and that this effect is mediated by an ERE and not an AP-1 complex. We provide evidence that 3?Adiol is a potent agonist for ER-?1-induced gene transcription in neurons. Notably, our data also revealed a ligand-independent regulation by ER-?1 at both an ERE and an AP-1 complex in a neuronal cell line.
Materials and Methods
Expression vectors
Plasmid expression vectors (pcDNA 3.0; Invitrogen, Carlsbad, CA) containing inserts for ER-?1 and ER-?2 were provided by Dr. Tom Brown (Pfizer Corp., Cambridge, MA). Full-length human rat ER was inserted into a PSG5 vector and provided by Dr. Richard Price (University of California San Francisco, San Francisco, CA).
Reporter constructs
The ERE-tk-luciferase reporter (generously provided by Dr. Paul Budworth, Case Western Reserve University, Cleveland, OH) contains two copies of vitellogenin ERE sequence coupled to the minimal tk-luciferase promoter and subcloned into pGL2-Basic plasmid (Promega Corp., Madison, WI). The AP1-tk-luciferase promoter (generously provided by Dr. Jennifer Nyborg, Colorado State University, Fort Collins, CO) contains three copies of the AP-1 sequence (TGACTCATGCTTT) coupled to the minimal tk-luciferase promoter and subcloned into pGL2-Basic plasmid. A ?-actin LacZ reporter construct, expressing ?-galactosidase under the control the human actin promoter, was used as an internal control for plasmid transfer efficiency.
Synthesis of hormone receptor proteins
Full-length rat ER, ER-?1, and ER-?2 expression vectors were used to synthesize receptor proteins in vitro using the TnT-coupled rabbit reticulocyte lysate system (Promega Corp.) with T7-RNA polymerase according to manufacturer’s directions.
Cell culture
The mouse hippocampus-derived cell line HT-22 (generously provided by Dr. Dave Schubert, Salk Institute, San Diego, CA) was used for all transient transfections. Cells were maintained in DMEM containing 4.5% glucose and L-glutamine (Invitrogen); supplemented with 1x nonessential amino acids, and 10% fetal bovine serum (FBS) (Gemini Bioproducts, Woodland, CA). Cells were grown to 70% confluency and used before passage 10 for all transient transfection experiments. HT-22 cells were selected for this study because they are a unique neuronal cell line that reportedly does not express ER (22) or ER-?1 (data not shown).
Transient transfections
Cells were plated at a density of 0.2 x 105 cells/well in 24-well plates 24 h before transfection. All constructs were transfected in triplicate wells within each assay, and each transfection assay was repeated a minimum of six times. Each experiment was performed using a minimum of three different preparations for each plasmid. Transfections were carried out using a lipid-mediated transfection reagent (Superfect; Qiagen Inc., Valencia, CA) according to manufacturer’s instructions) and a total DNA concentration of 2.2 μg/well. Cells were incubated with transfection media complex for 2.5 h, media was replaced with treatment media for 15 h, and then cells were lysed. All treatments were performed in DMEM containing 10% dextran-charcoal-stripped FBS to ensure estrogen-free culture conditions (Hyclone Laboratories, Logan, UT). In addition, each transfection experiment was performed a minimum of three times with phenol red-free media, and no differences were observed from the transfections performed with media containing phenol red (data not shown). Luciferase activity was measured using 20 μl lysate added to 100 μl luciferin substrate (Promega Corp.). ?-Galactosidase activity was measured using 40 μl lysate added to 200 μl galacton substrate (Tropix-GalactoLight kit; Applied Biosystems, Foster City, CA) according to manufacturer’s instructions). Relative light units (RLU) were measured using a 20/20 TD luminometer (Turner Designs).
Hormone treatments
The following compounds were diluted in dimethylsulfoxide (DMSO) and used at a final concentration of 100 nM in 0.01% DMSO: E2, 4OH-tamoxifen (TAM; Sigma-Aldrich Co., St. Louis, MO), 3?Adiol, 3Adiol (Steraloids, Newport, RI), and diarylpropionitrile (DPN; Tocris-Cookson, Inc., Ellisville, MO). Each compound was added to DMEM supplemented with 10% dextran-charcoal-stripped FBS (Hyclone Laboratories).
Saturation isotherms
To establish the binding affinity of 3?Adiol for various ERs, increasing concentrations (0.01–500 nM) of 3?Adiol were incubated with 100-μl aliquots of reticulocyte lysate supernatant for 90 min at room temperature (ER-?1), 4.5 h at 4 C (ER-?2), and 18 h at 4 C (ER), with 10-nM concentrations of [3H]-E2 in TEGMD (10 mM Tris-Cl, 1.5 mm EDTA, 10% glycerol, 25 mM molybdate, and 1.0 mM dithiothreitol, pH 7.4). These incubation times were determined empirically and represent optimal binding of receptor with E2. Nonspecific binding was assessed using a 200-fold excess of the ER agonist diethylstilbestrol in parallel tubes. After incubation, bound and unbound [3H]-E2 were separated by passing the reaction through a 1.0-ml lipophilic Sephadex LH-20 (Sigma-Aldrich Co.) column. Columns were constructed by packing a disposable pipette tip (1.0 ml; Labcraft, Curtin Matheson Scientific, Inc., Houston, TX) with TEGMD-saturated Sephadex according to previously published protocols (23). For chromatography, the columns were reequilibrated with TEGMD (100 μl), and the incubation reactions were added individually to each column and allowed to incubate on the column for an additional 30 min. After this incubation, 600 μl TEGMD was added to each column, flow-through was collected, 4 ml scintillation fluid was added, and samples were counted (5 min each) in a 2900 TR Packard scintillation counter (Packard Bioscience, Meriden, CT). Curve fitting, scientific graphing, and binding analyses were completed using GraphPad Software (GraphPad Prism 3.0, San Diego, CA).
EMSAs
Oligonucleotides.
Double-stranded oligonucleotides containing the vitellogenin consensus ERE sequence, or SP1 sequence, were 32P-end-labeled with T4 polynucleotide kinase. The percent incorporation was determined, and labeled probes with greater than 50% 32P incorporation were used for EMSAs.
Gel electrophoresis.
Receptor protein lysates were incubated with 100 nM E2, 100 nM 3?Adiol, or 0.01% DMSO (vehicle control) for 18 h before gel electrophoresis. After ligand binding, receptor lysates were incubated with 1 x gel shift binding buffer [20% glycerol, 5 mM MgCl2, 2.5 mM EDTA, 2.5 mM dithiothreitol, 250 mM NaCl, 50 mM Tris-HCl, 0.25 mg/ml poly(dI-dC)·poly(dI-dC)] for 10 min. Specific binding reactions were also incubated with 500- to 1000-fold excess of unlabeled ERE oligonucleotide. Nonspecific binding was tested using the 32P-SP1 oligonucleotide. After an initial 10-min incubation, 32P-ERE was added and incubated for an additional 20 min. DNA-protein complexes were resolved on a 6% Novex DNA retardation gel (Invitrogen) for 20 min at 250 V. Gels were dried on a vacuum gel dryer at 80 C for 1 h before autoradiography.
Autoradiography.
Dried gels were exposed to x-ray film (Biomax MS; Eastman Kodak Co., New Haven, CT) for 12 h at –70 C. Film images were captured using a CCD camera (Sony XC-77; Sony Corp., Tokyo, Japan) that was connected to a QuickCapture framegrabber board (Data Translation Inc., Marlboro, MA) on a PC computer. The captured images were analyzed with Scion Image software [W. Rasband, National Institutes of Health (NIH), Bethesda, MD]. Data are represented as mean density of pixels ± SD.
Statistics
Differences among hormone treatment groups for individual receptors were analyzed by one-way ANOVA followed by Tukey’s HSD test. Comparisons between control groups (empty vector + vehicle; receptor + vehicle) were analyzed using a t test. Differences were considered significant at P < 0.05. All transfection data are represented as percent change compared with vehicle-treated, empty-vector controls.
Results
Relative binding affinity of ERs for 3?Adiol
Competition binding studies were used to determine the relative binding affinities of the androgen metabolite 3?Adiol to ER, ER-?1, and ER-?2. In these experiments, displacement of 3H-E2 from the receptor provides a quantitative measurement of the affinity of the receptor for a specific ligand, such as 3?Adiol. Our results showed that, at optimal binding parameters, ER-?1 displayed the highest binding affinity for 3?Adiol compared with all other ERs tested (Table 1). The selectivity rank of 3?Adiol for all ERs was as follows: ER-?1 > ER > ER-?2. Although 3?Adiol bound all ERs tested, it was relatively weak compared with that of the cognate ER ligand E2 (Table 1).
TABLE 1. Competition assay depicting the relative binding affinities of various ER subtypes for the androgen metabolite 3?Adiol: a comparison with the cognate ligand E2
Constitutive (i.e. ligand-independent) regulation of ERE and AP-1-luciferase activity by ERs
ER-?1 and ER demonstrated constitutive alteration of ERE-luciferase promoter activity in the absence of any hormone treatment. Figure 1A illustrates a dose-dependent change in luciferase activity with varying concentrations of ER and ER-?1 expression vectors. ER-?1 significantly increased ERE-luciferase activity at concentrations as low as 0.01 μg/well, whereas ER did not significantly increase ERE-luciferase activity until concentrations reached 1.0 μg/well (Fig. 1A). Further, ER-?1 showed a very robust response, increasing ERE-luciferase activity by approximately 700% over empty vector controls. On the other hand, even at the highest dose used, the constitutive regulation by ER was relatively weak, increasing ERE-luciferase activity by only 160%.
FIG. 1. Ligand-independent effects of varying concentrations of ER and ER-?1 expression vectors on ERE- and AP1-luciferase (luc) promoter activity. Cotransfection of HT-22 cells with an ERE-tk-luc (A) or an AP-1-tk-luc (B) reporter construct and an expression vector containing varying concentrations of ER or ER-?1. After transfection, cells were treated with 0.01% DMSO (vehicle) in DMEM supplemented with 10% dextran-charcoal-stripped FBS for 15 h. Data are represented as percent change in RLU from vehicle-treated empty vector control ± SEM. The presence of a symbol denotes a significant difference from empty vector controls.
In cells cotransfected with an AP-1-luciferase reporter construct, ER-?1 significantly reduced promoter activity by about 72% at the highest concentration of 1.0 μg/well in a ligand-independent fashion (P < 0.05, Fig. 1B). A slightly lower, but statistically significant effect was also observed, with a lower concentration of 0.5 μg/well (Fig. 1B). By contrast, ER showed no ligand-independent regulation at an AP-1 site. Concentrations of ER expression vectors higher than 1.0 μg/well were not significantly more efficacious at altering ligand-independent activity of either the ERE or AP1 reporter constructs. Thus, in all subsequent experiments, a concentration of 1.0 μg was used for all ER expression vectors.
Comparison of 3?Adiol-activated transcription by ER and ER-?1 at an ERE and an AP-1 site
To determine whether 3?Adiol differentially activates transcription through ER and ER?, we cotransfected an ERE-luciferase reporter construct with an expression vector containing either ER or ER-?1 using the HT-22 neuronal cell line. Our results showed that treatment with 3?Adiol significantly stimulated ERE-driven luciferase activity through ER-?1 but had no effect on ER-mediated transcription (P < 0.05, Fig. 2A). Interestingly, 3?Adiol stimulated ER-?1-induced transcription at a level equivalent to that of E2, despite its much lower binding affinity (Table 1 and Fig. 2A). As expected, treatment with E2, the endogenous ligand for ER and ER?, significantly stimulated ERE-luciferase activity through both ERs (Fig. 2A). Similar results were observed, with hormone concentrations as low as 1.0 nM for E2 and 10 nM for 3?Adiol (data not shown). ER-?1 and ER also demonstrated constitutive (i.e. ligand-independent) up-regulation of ERE-luciferase promoter activity (Fig. 2A). In fact, ER-?1 significantly increased ERE-luciferase activity compared with that of ER in vehicle-treated control groups (Fig. 2A). Surprisingly, E2 did not significantly increase ER-?1-induced ERE-luciferase activity beyond the observed constitutive up-regulation (P = 0.06; Fig. 2A). Together, these results suggest that 3?Adiol is a selective agonist of ER? at an ERE.
FIG. 2. Effects of 3?Adiol on ERE- and AP-1-luc promoter activity mediated by ER and ER-?1. Cotransfection of HT-22 cells with an ERE-tk-luc (A) or an AP-1-tk-luc (B) reporter construct and an expression vector containing 1.0 μg/well ER or ER-?1. After transfection, cells were treated with 0.01% DMSO (vehicle), 100 nM E2, or 100 nM 3?Adiol for 15 h. Data are represented as percent change in RLU from vehicle-treated empty vector control ± SEM. The presence of a symbol (ER) or letter (ER-?1) denotes a significant difference from empty vector controls. Dissimilar symbols/letters denote a statistically significant difference among treatment groups (P < 0.05).
In cells cotransfected with the AP1-reporter, construct ER-?1, but not ER, dramatically reduced luciferase activity. Treatment with E2 and 3?Adiol did not alter this effect, unlike the observed agonistic effect of 3?Adiol at an ERE (Fig. 2B). Interestingly, E2 treatment did not significantly increase ER-mediated AP1-luciferase activity (Fig. 2B). It is possible that this response is specific to the HT-22 cell line, or it might represent a more general difference between peripheral and neuronal cell lines.
Comparison of 3?Adiol with other ER ligands on ER-?1-mediated ERE-luciferase activity
Selective ER modulators (SERMs) differentially regulate transcription at an ERE and an AP-1 site (see Ref. 24 for review). In fact, SERMs that antagonize ER can act as an agonist for ER-?1 (20). This experiment compared the ligand-activating profile of 3?Adiol with other ER ligands known to bind and activate ER-?1. We also investigated whether the co-metabolite and stereoisomer of 3?Adiol, 3Adiol could similarly stimulate ER-?1-mediated gene transcription. Cells cotransfected with an ERE-luciferase reporter and ER-?1 were exposed to three treatments: 1) a highly selective ER-?1 agonist, DPN; 2) the SERM, TAM; and 3) 3Adiol. DPN significantly stimulated ER-?1-mediated promoter activity above the observed constitutive increase but not greater than that of 3?Adiol (Fig. 3A). Further, treatment with TAM completely abolished the constitutive ER-?1-induced up-regulation of ERE-mediated promoter activity (Fig. 3A). Last, 3Adiol had no effect on promoter activity mediated by an ERE, despite its structure being similar to that of 3?Adiol (Fig. 3A). Hence, the ligand-activating profile of 3?Adiol mimics that of a selective ER? agonist in the HT-22 neuronal cell line.
FIG. 3. Effects of various ER ligands on ER-?1-induced promoter activity at an ERE or an AP-1-site. Cotransfection of HT-22 cells with an ERE-tk-luc (A) or an AP-1-tk-luc (B) reporter construct and an expression vector containing 1.0 μg/well ER-?1. After transfection, cells were treated with 0.01% DMSO (vehicle), 100 nM E2, 100 nM 3?Adiol, 100 nM DPN, 1.0 μM TAM, or 100 nM 3Adiol for 15 h. Data are represented as percent change in RLU from vehicle-treated empty vector control ± SEM. The presence of a letter denotes a significant difference from empty vector controls. Dissimilar letters denote a statistically significant difference among groups (P < 0.05).
Comparison of 3?Adiol to other ER ligands on ER-?1-mediated AP1-luciferase activity
As shown in Fig. 2B, 3?Adiol had no effect on ER-?1-induced transcription mediated by an AP-1 complex. Similarly, the ER-?1 agonist DPN failed to modulate ER-?1-mediated inhibition of AP-1-luciferase promoter activity (Fig. 3B). By contrast, treatment with TAM abolished the constitutive inhibitory effect of ER-?1 on AP-1-luciferase promoter activity, thereby antagonizing the ligand-independent effects of ER-?1 (Fig. 3B). Similar to what we observed at an ERE, 3Adiol had no effect at an AP-1 site (Fig. 3B). Taken together, these results demonstrate that: 1) ER agonists do not further attenuate the observed ER-?1-induced constitutive decrease in transcription mediated by an AP-1 complex; and 2) potential ER antagonists, such as TAM, abolish the ligand-independent effects of ER-?1.
Effect of 3?Adiol on ER-?2-induced transcription at an ERE and an AP-1 site
The ER-?1 splice variant ER-?2 contains an extra 18-amino-acid sequence in the LBD. Consequently, ER-?2 binds most ligands less effectively than ER-?1. Consistent with this observation, we showed that ER-?2 binds 3?Adiol with an approximately 50-fold lower affinity than that of ER-?1 [Ki (nM) = 1.7 ± 0.3 and 58.1 ± 5.5, respectively]. However, despite these lower binding affinities, in most systems, ER-?2 has been shown to mimic, albeit to a lesser degree, the actions of ER-?1 for most ligands tested (13). Thus, we examined whether 3?Adiol stimulates ER-?2-induced transcription mediated by an ERE or an AP-1 enhancer site. Most strikingly, ER-?2 did not constitutively increase ERE-luciferase promoter activity as was observed with ER-?1 (Fig. 4A). This demonstrates an inherent difference in the ER? splice variants that is apparently dependent on their differences in the LBD. Further, 3?Adiol had no effect on ER-?2-induced transcription despite ER-?2s ability to bind 3?Adiol, albeit weakly (Fig. 4A). The ER agonists E2 and DPN were effective at stimulating ER-?2-mediated transcription at an ERE, as expected (P < 0.05; Fig. 4A). Surprisingly, 3Adiol also significantly increased ER-?2-induced ERE-luciferase activity (P < 0.05; Fig. 4A).
FIG. 4. Effects of various ER ligands on ER-?2-induced promoter activity at an ERE or an AP-1-site. Cotransfection of HT-22 cells with an ERE-tk-luc (A) or an AP-1-tk-luc (B) reporter construct and an expression vector containing 1.0 μg/well ER-?2. After transfection, cells were treated with 0.01% DMSO (vehicle), 100 nM E2, 100 nM 3?Adiol, 100 nM DPN, 1.0 μM TAM, or 100 nM 3Adiol for 15 h. Data are represented as percent change in RLU from vehicle-treated empty vector control ± SEM. The presence of a letter denotes a significant difference from empty vector controls. Dissimilar letters denote a statistically significant difference among groups (P < 0.05). An asterisk denotes an intermediate effect of 3Adiol.
Cotransfection of ER-?2 and an AP-1-luciferase reporter construct showed that ER-?2 decreased promoter activity independent of hormone treatment, similar to ER-?1 (P < 0.05; Fig. 4B). However, treatment with the ER agonists E2 and DPN significantly enhanced this response, in contrast to their lack of effect on the ER-?1-mediated constitutive decrease at an AP-1 site (Fig. 4B). The ER antagonist TAM abolished the ligand-independent reduction of ER-?2-mediated transcription at an AP-1 site (Fig. 4B). Overall, these data demonstrate that ER-?2 has important functional differences from the ER-?1 splice variant.
Effects of 3?Adiol on ER-?1 binding to an ERE
Classical models of ER-mediated transcription pathways dictate that ER binding to an ERE occurs after ligand activation of the receptor. However, our data show that ER and ER-?1’s ability to stimulate ERE-mediated transcription is independent of ligand binding. Moreover, a novel ER-?1 ligand, 3?Adiol, further increased the observed constitutive regulation by ER-?1 (Fig. 2). Thus, we used EMSAs to: 1) confirm that ER and ER-?1 binds an ERE in the absence of ligand; and 2) determine whether 3?Adiol enhances the binding potency of ER-?1 at an ERE.
Lysate containing translated ER, ER-?1, and ER-?2 protein was incubated with a 32P-labeled ERE oligonucleotide and resolved on a 6% DNA retardation gel. After autoradiography, the results demonstrated a strong shift for all ERs (ER, ER-?1, and ER?2), indicating that all receptors bind an ERE (Fig. 5). Interestingly, there was no significant difference in the mean density/pixel2 in the absence or presence of ligand (Fig. 5 and Table 2). Moreover, the addition of 3?Adiol did not significantly alter the binding of any ER tested to the ERE (Table 2). Unlabeled ERE oligonucleotide (1000-fold excess) effectively competed for each ER binding to an ERE, and there was no shift observed when receptor proteins were incubated with a 32P-labeled SP1 oligonucleotide (Fig. 5, lanes 1 and 2). These results demonstrate the specificity of ER:DNA binding at an ERE.
FIG. 5. Effects of 3?Adiol on ER, ER-?1, and ER-?2 DNA binding. Vitellogenin consensus ERE oligonucleotide (0.2 pM) was 32P-end-labeled and incubated with 2 μl ER (A and B; lanes 3–6), ER-?1 (A; lanes 7–10), or ER-?2 (B; lanes 7–10) receptor protein lysate. Receptor proteins were incubated with 0.01% DMSO (vehicle), 100 nM E2, 100 nM 3?Adiol, or 100 nM 3Adiol for 18 h at 4 C. After 32P-ERE incubation, reaction product was resolved on a 6% DNA retardation gel for 20 min at 250 V. Gels were dried and exposed to autoradiographic film for 12 h. Unlabeled ERE oligonucleotide was added to determine binding specificity 1000-fold excess of 32P-ERE (A and B; lane 1). 32P-labeled SP1 oligonucleotide (SP-1 olig) was used as a negative control (A and B; lane 2).
TABLE 2. EMSA showing the effects of various ligands on ER, ER-?1, and ER-?2 binding to an ERE
Discussion
With these studies, we determined that, in a neuronal cell line, the androgen metabolite 3?Adiol binds ER-?1 and subsequently alters promoter activity mediated by an ERE and an AP-1 complex. These findings demonstrate a possible mechanism for the direct regulation by androgens in neuronal cell types that lack a classical androgen receptor (AR). In addition, our data show that ER-?1 differentially regulated transcription mediated by an ERE and an AP-1 site independent of E2 ligand. Thus, these data highlight the potential for ER? to act as a basic transcription factor in ER-?1-expressing neuronal target cells.
The androgen metabolite 3?Adiol is an endogenous ligand for ER-?1
In this study, we used a hippocampus-derived cell line to verify that 3?Adiol is an endogenous ligand for ER-?1 in neurons. Anatomical evidence suggests that 3?Adiol is produced in the brain, raising the possibility that it is an important central regulator. For example, the major enzymes for converting DHT to 3?Adiol, (3?-HSD, 3-HSD, and 17?-HSD) are expressed in many regions of the rat brain throughout development, including the cortex, hypothalamus, thalamus, septum, hippocampus, and cerebellum (5, 6, 7, 8, 9). Notably, 3?-HSD is expressed at highest levels in the cerebellum, which is also rich in ER-?1 (25, 26, 27). Although cellular colocalization of ER-?1 and 3?-HSD has yet to be shown, it is reasonable to assume that local production of 3?Adiol in these brain regions could activate cells expressing ER-?1.
3?Adiol is likely the major androgen metabolite produced in the prostate gland and plays a significant role in regulating prostate epithelial cell proliferation (3, 4). Importantly, ER-?1, and not AR, mediated 3?Adiol’s effect on prostate growth (3, 4). Our results revealed a potential molecular pathway for this novel endogenous ligand by demonstrating that 3?Adiol significantly increased ER-?1-mediated transcription at an ERE. These studies verify those of Kuiper et al. (1998) (17) that 3?Adiol binds ERs with a selectivity for ER-?1 over ER. Further, our data also demonstrate that the transcriptional selectivity of 3?Adiol is much greater than its binding selectivity.
The current model describing the role of coregulatory proteins in the ER signaling pathway maintains that unbound ERs are tethered to corepressor proteins that inhibit transactivation (see Ref. 10 for review). Ligand-binding initiates the dissociation of corepressors, such as nuclear receptor corepressor and silencing mediator for retinoid and thyroid hormone receptors, followed by the recruitment of coactivators, such as glucocorticoid receptor interacting protein and steroid receptor coactivator-1 (SRC-1). The failure of 3?Adiol to activate transcription upon binding to ER and ER-?2 in the present study raises the possibility that 3?Adiol might prevent the dissociation of, or actively recruits, one or more corepressor proteins when bound to ER and ER-?2. This hypothesis is consistent with Shang et al. (28), who showed that nuclear receptor corepressor and silencing mediator for retinoid and thyroid hormone receptors are associated with ER when it is bound to the SERM, TAM. Thus, we propose that 3?Adiol functions as an antagonist for ER and ER-?2 but as an agonist for ER-?1.
Alternatively, 3?Adiol might affect components of the signaling pathway downstream from the dissociation of corepressor proteins. For instance, coactivator proteins bind to an LXXLL motif embedded in the LBD and mediate transcription by binding histone acetyl-transferases, such as p300/CBP, that induce conformational changes in the chromatin (29, 30, 31). It is possible that inherent variations in the LBD among ER subtypes may confer specificity for the recruitment of distinct coactivators. Thus, we hypothesize that 3?Adiol binds ER-?1 and subsequently recruits coactivator proteins distinct from that of ER and ER-?2, thereby potentiating ERE-mediated transcription.
Overall, our results indicate that 3?Adiol is a specific agonist for ER-?1 and not other ER subtypes. Although 3?Adiol binds all ERs tested, transcription is initiated only upon binding to ER-?1. In fact, our data demonstrate that 3?Adiol increased promoter activity equal to that of the cognate ligand E2, despite having a lower binding affinity for ER-?1. Moreover, 3Adiol had no effect on ER-?1 transactivation, which is consistent with its very low binding affinity (11). The specificity of 3?Adiol for ER-?1 is likely due to inherent differences in the LBDs of the various ERs. We speculate that 3?Adiol could also act in an antagonistic fashion for ER and ER-?2. Indeed, 3?Adiol significantly attenuated the ER-induced constitutive regulation of ERE-luciferase promoter activity, lending further evidence for a possible antagonistic action of 3?Adiol through ER (Fig. 2A). In contrast, 3Adiol seems to mediate transcription preferentially through ER-?2 (Fig. 4A).
ER-?1 constitutively regulates transcription mediated by ERE and AP-1 enhancer elements
Constitutive regulation of promoter activity has been demonstrated for several members of the superfamily of nuclear receptors. Of particular relevance to our study, both ER and ER? have been shown to constitutively regulate gene transcription. For instance, the tyrosine mutant ER, found in metastatic breast cancer cells, stimulated ERE-mediated transcription in HeLa cells and was unaffected by E2 treatment (32). Moreover, ER-?1 increased the activity of arginine vasopressin promoter in vitro in the absence of E2 (33). In that study, ER-?1 constitutively increased arginine vasopressin promoter activity in a concentration-dependent fashion. Our results confirm that ER-?1 does not require ligand activation to modulate transcription through classical ER pathways. Interestingly, E2 treatment did not significantly increase ER-?1-induced ERE-luciferase activity above that of the observed ligand-independent increase. Also, E2 treatment did not significantly increase ER-mediated AP1-luciferase activity, as has been reported by others using cell lines derived from peripheral tissues, such as HeLa cells (20). This might be cell-line specific for HT-22 cells or it might represent a more general difference between peripheral and neuronal cell types. Taken together our results demonstrate that ER-?1 can serve as a basic transcription factor in a ligand-independent fashion, as well as a classical ligand-activated steroid hormone receptor.
A variety of neurotransmitters and growth factors, such as dopamine, IGF-I, and EGF, as well as the membrane-associated protein caveolin-1, have been shown to activate ER-induced transcription in the absence of E2 (34, 35, 36, 37). Such ligand-independent activation of ERs is thought to occur at the N-terminal activator-function (AF)-1 domain of the ER, whereas ligand-dependent activation occurs at the C-terminal AF-2 domain (38, 39). Furthermore, phosphorylation of the AF-1 domain of ER-?1 elicits the recruitment of coregulatory proteins, such as SRC-1, potentiating the ligand-independent transcriptional regulation (40, 41, 42, 43). It is likely that some, or all, of these factors are involved in the ER-?1-induced constitutive regulation demonstrated in the present study.
The classical ER signaling pathway suggests that ligand-bound ERs form homodimers and activate transcription by binding directly to DNA at an ERE (18, 19). However, this paradigm is uncertain because studies have shown both the lack, and the occurrence, of ER:ERE interaction in the absence of ligand (13, 14, 43). Our results confirm those of Petersen et al. (13) that ERs can bind to an ERE in the absence of ligand, thus providing a mechanism for the ER- and ER-?1-induced constitutive regulation of ERE-mediated transcription shown in our study.
An alternative ER signaling pathway has also been described, whereby the transactivational domain of the ER can interact with other transcription factor proteins, such as members of the Fos and Jun families, and subsequently modulate transcription through these protein:protein interactions (44). Interestingly, the ligand-independent transactivation by ER-?1 had opposite effects when mediated by an ERE, compared with an AP-1 complex. Differential activation of ER-?1 at an ERE and an AP-1 site has been shown previously, although only in the presence of ligand (20). To our knowledge, this is the first report demonstrating constitutive promoter regulation by ER-?1 mediated at an AP-1 site (see Ref. 10 for review). Furthermore, these data highlight the potential confounds for discerning the molecular mechanisms of ER-regulated promoter activity, especially complex promoters comprised of both ERE and AP-1 regulatory elements.
SERMs, such as TAM, can be both agonistic and antagonistic, depending on the cell type, ER subtype, and interacting promoter response element (45, 46). For instance, it has been reported that TAM stimulated ER-?1-induced transcription when mediated by an AP-1 complex (20). In contrast, we demonstrated that ER-?1 constitutively decreased promoter activity at an AP-1 site, and treatment with TAM abolished this response but did not stimulate promoter activity over baseline expression levels. However, without the inclusion of an empty vector control to demonstrate this, the effect of TAM could be interpreted as stimulating ER?1-induced promoter activity. In addition, for these studies, we used a neuronal cell line as opposed to the cervical epithelial-derived HeLa cell line used by Paech et al. (20). Such differences in experimental paradigms might further explain these discrepant results. Taken together, these data might also reflect some of the observed differences in SERM action in peripheral vs. neuronal tissue types.
In summary, this study demonstrated that 3?Adiol, an androgen metabolite, facilitated ER-mediated transcription through classical ER signaling pathways in a neuronal cell type. This raises important concerns for using the nonaromatizable androgen DHT as the penultimate ligand to study AR-dependent pathways. Further, the constitutive regulation by ERs suggests that cell-specific expression has important biological consequences independent of hormonal milieu. The importance of coregulatory proteins involved in these processes is currently under investigation.
Acknowledgments
The authors thank Matthew Ewert and Brian Cherrington for their technical assistance.
References
Grover PK, Odell WD 1975 Correlation of in vivo and in vitro activities of some naturally occurring androgens using a radioreceptor assay for 5-dihydrotestosterone with rat prostate cytosol receptor protein. J Steroid Biochem 6:1373–1379
Ahmad N, Warren DW, Haltmeyer GC 1978 The effects of 5-reduced androgens on maintenance and regeneration of prostate glands and seminal vesicles in castrated and hypophysectomized rats. Anat Rec 192:543–554
Weihua Z, M?kel? S, Andersson LC, Salmi S, Saji S, Webster JI, Jensen EV, Nilsson S, Warner M, Gustafsson JA 2001 A role for estrogen receptor ? in the regulation of growth of the ventral prostate. Proc Natl Acad Sci USA 98:6330–6335
Weihua Z, Lathe R, Warner M, Gustafsson JA 2002 An endocrine pathway in the prostate, ER?, AR, 5-androstane-3?, 17?-diol, and CYP7B1 regulates prostate growth. Proc Natl Acad Sci USA 99:13589–13594
Torn S, Nokelainen P, Kurkela R, Pulkka A, Menjivar M, Ghosh S, Cocaprados M, Peltoketo H, Isomaa V, Vihko P 2003 Production, purification, and functional analysis of recombinant human and mouse 17?-hydroxysteroid dehydrogenase type 7. Biochem Biophys Res Commun 305:37–45
Steckelbroeck S, Jin Y, Gopishetty S, Oyesanmi B, Penning TM 2004 Human cytosolic 3-hydroxysteroid dehydrogenase of the aldo-keto reductase superfamily display significant 3?-hydroxysteroid dehydrogenase activity: implications for steroid hormone metabolism and action. J Biol Chem 279:10784–10795
Guennoun R, Fiddles RJ, Gouezou M, Lombes M, Baulieu EE 1995 A key enzyme in the biosynthesis of neurosteroids, 3?-hydroxysteroid dehydrogenase/5- 4-isomerase (3?-HSD), is expressed in rat brain. Brain Res Mol Brain Res 30:287–300
Pelletier G, Luu-The V, Labrie F 1995 Immunocytochemical localization of type I 17-? hydroxysteroid dehydrogenase in the rat brain. Brain Res 704:233–239
Ibanez C, Guennoun R, Liere P, Eychenne B, Pianos A, El-Etr M, Baulieu EE, Schumacher M 2003 Developmental expression of genes involved in neurosteroidogenesis: 3?-hydroxysteroid dehydrogenase/5-4 isomerase in the rat brain. Endocrinology 144:2902–2911
McKenna NJ, O’Malley BW 2002 Combinatorial control of gene expression by nuclear receptors and coregulators. Cell 108:465–474
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
Mosselman S, Polman J, Dijkema R 1996 ER?: identification and characterization of a novel human estrogen receptor. FEBS Lett 392:49–53
Petersen DN, Tkalcevic GT, Koza-Taylor PH, Turi TG, Brown TA 1998 Identification of estrogen receptor ?2, a functional variant of estrogen receptor ? expressed in normal rat tissues. Endocrinology 139:1082–1092
Hanstein B, Liu H, Yancisin MC, Brown M 1999 Functional analysis of a novel estrogen receptor-? isoform. Mol Endocrinol 13:129–137
Price Jr RH, Lorenzon N, Handa RJ 2000 Differential expression of estrogen receptor ? splice variants in rat brain: identification and characterization of a novel variant missing exon 4. Mol Brain Res 80:260–268
Price Jr RH, Butler CA, Webb PA, Uht R, Kushner P, Handa RJ 2001 A splice variant of estrogen receptor ? missing exon 3 displays altered subnuclear localization and capacity for transcriptional activation. Endocrinology 142:2039–2049
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
Kumar V, Chambon P 1988 The estrogen receptor binds tightly to its responsive element as a ligand-induced homodimer. Cell 55:145–156
Klein-Hitpass L, Tsai SY, Greene GL, Clark JH, Tsai MJ, O’Malley BW 1989 Specific binding of estrogen receptor to the estrogen response element. Mol Cell Biol 9:43–49
Paech K, Webb P, Kuiper GG, Nilsson S, Gustafsson JA, Kushner PJ, Scanlan TS 1997 Differential ligand activation of estrogen receptors ER and ER? at AP1 sites. Science 277:1508–1510
Kushner PJ, Agard DA, Greene GL, Scanlan TS, Shiau AK, Uht RM, Webb P 2000 Estrogen receptor pathways to AP-1. J Steroid Biochem Mol Biol 74:311–317
Kim H, Bang OY, Jung MW, Ha SD, Hong HS, Huh K, Kim SU, Mook-Jung I 2001 Neuroprotective effects of estrogen against ?-amyloid toxicity are mediated by estrogen receptors in cultured neuronal cells. Neurosci Lett 302:58–62
Handa RJ, Stadelman HL, Resko JA 1987 Effect of estrogen on androgen receptor dynamics in female rat pituitary. Endocrinology 121:84–89
McDonnell DP, Wijayaratne A, Chang CY, Norris JD 2002 Elucidation of the molecular mechanism of action of selective estrogen receptor modulators. Am J Cardiol 90:35F–43F
Price Jr RH, Handa RJ 2000 Expression of estrogen receptor-? protein and mRNA in the cerebellum of the rat. Neurosci Lett 288:115–118
Shughrue PJ, Merchenthaler I 2001 Distribution of estrogen receptor ? immunoreactivity in the rat central nervous system. J Comp Neurol 436:64–81
Jakab RL, Wong JK, Belcher SM 2001 Estrogen receptor ? immunoreactivity in differentiating cells of the developing rat cerebellum. J Comp Neurol 430:396–409
Shang Y, Hu X, DiRenzo J, Lazar MA, Brown M 2000 Cofactor dynamics and sufficiency in estrogen receptor-regulated transcription. Cell 103:843–852
Heery DM, Kalkhoven E, Hoare S, Parker MG 1997 A signature motif in transcriptional co-activators mediates binding to nuclear receptors. Nature 387:733–736
Feng W, Ribeiro RC, Wagner RL, Nguyen H, Apriletti JW, Fletterick RJ, Baxter JD, Kushner PJ, West BL 1998 Hormone-dependent coactivator binding to a hydrophobic cleft on nuclear receptors. Science 280:1747–1749
Perissi V, Dasen JS, Kurokawa R, Wang Z, Korzus E, Rose DW, Glass CK, Rosenfeld MG 1999 Factor-specific modulation of CREB-binding protein acetyltransferase activity. Proc Natl Acad Sci USA 96:3652–3657
Zhang QX, Borg A, Wolf DM, Oesterreich S, Fuqua SA 1997 An estrogen receptor mutant with strong hormone-independent activity from a metastatic breast cancer. Cancer Res 57:1244–1249
Shapiro RA, Xu C, Dorsa DM 2000 Differential transcriptional regulation of rat vasopressin gene expression by estrogen receptor and ?. Endocrinology 141:4056–4064
Power RF, Mani SK, Codina J, Conneely OM, O’Malley BW 1991 Dopaminergic and ligand-independent activation of steroid hormone receptors. Science 254:1636–1639
Ignar-Trowbridge DM, Nelson KG, Bidwell MC, Curtis SW, Washburn TF, McLachlan JA, Korach KS 1992 Coupling of dual signaling pathways: epidermal growth factor action involves the estrogen receptor. Proc Natl Acad Sci USA 89:4658–4662
Klotz DM, Hewitt SC, Ciana P, Raviscioni M, Lindzey JK, Foley J, Maggi A, DiAugustine RP, Korach KS 2002 Requirement of estrogen receptor- in insulin-like growth factor-1 (IGF-1)-induced uterine responses and in vivo evidence for IGF-1/estrogen receptor cross-talk. J Biol Chem 277:8531–8537
Schlegel A, Wang C, Pestell RG, Lisanti MP 2001 Ligand-independent activation of oestrogen receptor by caveolin-1. Biochem J 359:203–210
Goff PL, Montano MM, Schodin DJ, Katzenellenbogen BS 1994 Phosphorylation of the human estrogen receptor. Identification of hormone-regulated sites and examination of their influence on transcriptional activity. J Biol Chem 269:4458–4466
Arnold SF, Obourn JD, Jaffe H, Notides AC 1994 Serine 167 is the major estradiol-induced phosphorylation site on human estrogen receptor. Mol Endocrinol 8:1208–1214
Tremblay A, Tremblay GB, Labrie F, Giguere V 1999 Ligand-independent recruitment of SRC-1 to estrogen receptor ? through phosphorylation of activation function AF-1. Mol Cell 3:513–519
Tremblay A, Giguere V 2001 Contribution of steroid receptor coactivator-1 and CREB binding protein in ligand-independent activity of estrogen receptor ?. J Steroid Biochem Mol Biol 77:19–27
Webb P, Nguyen P, Shinsako J, Anderson C, Feng W, Nguyen MP, Chen D, Huang SM, Subramanian S, McKinerney E, Katzenellenbogen BS, Stallcup MR, Kushner PJ 1998 Estrogen receptor activation function 1 works by binding p160 coactivator proteins. Mol Endocrinol 12:1605–1618
Murdoch FE, Meier DA, Furlow JD, Grunwald KA, Gorski J 1990 Estrogen receptor binding to a DNA response element in vitro is not dependent upon estradiol. Biochemistry 29:8377–8385
Webb P, Nguyen P, Valentine C, Lopez GN, Kwok GR, McInerney E, Katzenellenbogen BS, Enmark E, Gustafsson JA, Nilsson S, Kushner PJ 1999 The estrogen receptor enhances AP-1 activity by two distinct mechanisms with different requirements for receptor transactivation functions. Mol Endocrinol 13:1672–1685
Webb P, Lopez GN, Uht RM, Kushner PJ 1995 Tamoxifen activation of the estrogen receptor/AP-1 pathway: potential origin for the cell-specific estrogen-like effects of antiestrogens. Mol Endocrinol 9:443–456
Tremblay GB, Tremblay A, Labrie F, Giguere V 1998 Ligand-independent activation of the estrogen receptors and ? by mutations of a conserved tyrosine can be abolished by antiestrogens. Cancer Res 58:877–881(Toni R. Pak, Wilson C. J.)