An estrogen receptor chimera senses ligands by nuclear translocation

Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

An estrogen receptor chimera senses ligands by nuclear translocation Elisabeth D. Martinez a , Geetha V. Rayasam a,1 , Angie B. Dull b , Dawn A. Walker a , Gordon L. Hager a,∗ a

b

Laboratory of Receptor Biology and Gene Expression, National Cancer Institute, National Institutes of Health (NIH), Bethesda, MD 20892-5055, USA Basic Research Program, SAIC-Frederick, Inc., Molecular Targets Development Program, National Cancer Institute, Frederick, MD, USA Received 5 April 2005; accepted 23 June 2005

Abstract We have developed a new mammalian cell-based assay to screen for ligands of the estrogen receptor. A fluorescently tagged chimera between the glucocorticoid and the estrogen receptors, unlike the constitutively nuclear estrogen receptor, is cytoplasmic in the absence of hormone and translocates to the nucleus in response to estradiol. The chimera maintains specificity for estrogen receptor ␣ ligands and does not show cross-reactivity with other steroids, providing a clean system for drug discovery. Natural and synthetic estrogen receptor ␣ agonists as well as phytoestrogens effectively translocate the receptor to the nucleus in a dose-dependent manner. Antagonists of the estrogen receptor can also transmit the structural signals that result in receptor nuclear translocation. The potency and efficacy of high-affinity ligands can be evaluated in our system by measuring the nuclear translocation of the fluorescently labeled receptor in response to increasing ligand concentrations. The chimera is transcriptionally competent on transient and replicating templates, and is inhibited by estrogen receptor antagonists. Interestingly, the nucleoplasmic mobility of the chimera, determined by FRAP analysis, is faster than that of the wild type estrogen receptor, and the chimera is resistant to ICI immobilization. The translocation properties of this chimera can be utilized in high content screens for novel estrogen receptor modulators. © 2005 Elsevier Ltd. All rights reserved. Keywords: Chimera; Nuclear translocation; Estrogen; Glucocorticoid; Drug discovery; Cell-based assay; High content screen; Mobility

1. Introduction The estrogen receptor (ER), a member of the steroid receptor family, mediates the primary biological effects of Abbreviations: ER, estrogen receptor; GR, glucocorticoid receptor; GFP, green fluorescent protein; Tet, tetracycline; LBD, ligand binding domain; Dex, dexamethasone; E2, 17-␤-estradiol; PPT, 4,4 ,4 -(4-propyl[1H]-pyrazole-1,3,5-triyl)trisphenol; DES, diethylstilbestrol; DPN, 2,3bis(4-hydroxyphenyl)-propionitrile; HSP90, heat shock protein 90; MMTV, mouse mammary tumor virus long terminal repeat; SERM, selective estrogen receptor modulator; YES, yeast estrogen screen ∗ Corresponding author at: Laboratory of Receptor Biology and Gene Expression, Building 41, Room B602, National Cancer Institute, National Institutes of Health (NIH), Bethesda, MD 20892-5055, USA. Tel.: +1 301 496 9867; fax: +1 301 496 4951. E-mail address: [email protected] (G.L. Hager). 1 Current address: Metabolic and Urology Group, NDDR, R&D II, Plot No. 20, Sector-18, Udyog Vihar Industrial Area, Guragon 122001, Haryana, India. 0960-0760/$ – see front matter © 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.jsbmb.2005.06.033

estrogens. Treatment of certain conditions such as breast and ovarian cancers, osteoporosis, and menopause rely on estrogen receptor ligands to either inhibit or reactivate the actions of the estrogen receptor. Because of their medical significance, the search for novel estrogen receptor ligands and selective receptor modulators has been a priority in both endocrinology and medical pharmacology. A number of groups have designed effective systems to monitor the presence of estrogenic activities in environmental and food samples [1–4]. There is a need for accurate methods to screen for new compounds that could modulate receptor activity in mammalian cells and thus serve therapeutic purposes. The most commonly used screening assays currently measure the proliferative effects of estrogens, or the transcriptional activity of the ER or evaluate in vitro or in yeast the estrogenic activity of unknown compounds [5–9]. We and others have previously shown that the glucocorticoid receptor (GR) fused to the green fluorescent protein

308

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

(GFP) is found predominantly in the cytoplasm, and translocates rapidly to the nucleus in response to ligand [10,11]. The movement of the GR from one cellular compartment to another can be monitored by fluorescence microscopy and provides a tool for drug discovery. Like other nuclear receptors, the estrogen receptor shuttles between the cytoplasm and the nucleus but its constitutive location is predominantly nuclear [12–16], so it does not undergo translocation to the nucleus in response to ligand. Our earlier studies showed that the translocation properties of GR can be conferred to the ligand binding domain of another nuclear receptor, the retinoic acid receptor [17]. We therefore initiated efforts to generalize the concept that the intrinsic translocation properties of GR can be transferred to non-translocating receptors. In this report, we describe the construction and characterization of a fluorescent glucocorticoid–estrogen receptor chimera that localizes to the cytoplasm in the absence of estrogens but translocates to the nucleus in response to all ER␣ ligands tested, while remaining unresponsive to glucocorticoids. The efficacy and potency of high-affinity ligands can be measured in our system through dose response analysis of translocation dynamics. Because the translocation of the chimera is monitored by the accumulation of fluorescence in the nucleus, the assay can be automated and adapted for high throughput screening applications. Furthermore, since the chimera is transcriptionally active on both transient and integrated templates, secondary functional assays can be performed on candidate ligands to determine their agonistic/antagonistic activity. Finally, the chimera described here can serve as a new tool in the study of nuclear receptor structure and function, particularly in elucidating the process of nuclear translocation and receptor mobility.

2. Materials and methods 2.1. Automated fluorescence measurements 3617 cells were plated at 2500 cells/well on an optical bottom 96-well plate in the absence of tetracycline to induce the expression of GFP–GR. The following day, cells were washed four times and fresh DMEM supplemented with 5% charcoal stripped serum was added. Twenty-four hours later, cells were washed and the media was replaced with PBS supplemented with 5% charcoal stripped serum and 2 ␮g/ml Hoechst 33342 stain. After 2 h, cells were treated with 0, 1, 10, or 100 nM dexamethasone and incubated for 15 min before scanning. Live cell scanning was carried out on a Q3DM platform composed of an inverted Nikon TE300 fluorescence microscope (equipped with a robotic stage, a CCD camera and LED phase contrast light source), a mercury arc lamp, fiber optic housing and a control unit. When using live cells, penetration of the Hoechst dye is incomplete and therefore the nuclear mask used to define the nucleus is not visible in all cells. Fluorescence in the GFP channel overlapping the areas of Hoechst fluorescence in

the blue channel (pseudo-colored red in Fig. 1) defines the nuclear fluorescence. Algorithms built into the Q3DM system are used to define the full area occupied by each cell based on region expansion outward from the nucleus and cell boundary determinations. The percent nuclear over total fluorescence for each cell is then calculated. Cell line 3134 [18] containing the integrated GFP-tagged GRER construct were plated on Nunc 96-well glass bottom plates and grown overnight. Cells were treated with increasing concentrations of estradiol for 4 h, then washed in PBS, fixed with 4% formaldehyde and stained with 2 ␮g/ml Hoechst 33342 for 1 h in media containing DMSO or estradiol. Nuclear versus total cell fluorescence was measured on a Discovery1 system (Molecular Devices, Downingtown, PA). Images were captured using a 20X Plan Fluor objective at four sites per well. GFP expression was visualized using a FITC filter (470/535) with a 400 ms exposure. Nuclei were visualized using a DAPI filter (405/465) with a 30 ms exposure. Images were analyzed using a protein translocation journal from Molecular Devices. The nuclei are defined by the Hoechst stain, and the cytoplasm is measured by region expansion from the nucleus. Correlation coefficients were used for analysis of the nuclear translocation. Correlation coefficient values compare the overlap of the GFP and Hoechst signal, which have a theoretical range between −1 and +1. 2.2. Construction of chimeric receptors pCI-nGFP-C656G containing a polylinker replacing sites downstream of the coding region of helix 1 of the LBD of rGR [17] was digested sequentially with StuI and EcoRI at the polylinker. The ER␣ expression vector HEGO was digested with BlpI and the overhang filled in prior to digestion with EcoRI. The BlpI–EcoRI fragment of HEGO corresponding to partial ER␣ LBD sequences was inserted into the StuI–EcoRI digested vector downstream of partial GR coding sequences. The final construct is shown in Fig. 2 and results in a partial duplication of the loop between helices 1 and 3 of the receptor. GRER
loop was similarly made using a HindIII–EcoRI fragment of hER, which when inserted into the StuI–EcoRI digested vector results in a partial deletion of the loop between helices 1 and 3 of the LBD. 2.3. Cell culture, transient transfections and development of stable cell lines Cells were maintained in DMEM media supplemented with 10% calf serum, 2 mM glutamine, 100 U/ml penicillin G sodium, 100 ␮g/ml streptomycin sulfate, and 1 mM sodium pyruvate. For 3617 cells, the media was supplemented with 5 ␮g/ml tetracycline to inhibit expression of GFP–GR, unless otherwise stated. Transient transfections of all cells were performed following the manufacturer’s protocol using either Fugene 6 or Lipofectamine 2000 reagents. In the case of C127 derived cell lines, cells were transiently transfected

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

309

Fig. 1. Automated measurements of GR translocation in response to ligand. (A) 3617 cells were plated on glass bottom 96-well plates and grown in the absence of tetracycline to induce GFP–GR expression. Cells were treated with increasing amounts of dexamethasone for 15 min and nuclear fluorescence in live cells was measured in an automated fluorescent microscope system. Sample pictures of measured cell clusters are shown. Green represents the GFP–GR expressed in the cells. The nuclear stain has been pseudo-colored in red (the stain is incorporated into live cells only partly). (B) Nuclear fluorescence at each concentration was quantitated using built-in algorithms developed by Q3DM (now Beckman). Averages of n > 140 cells are shown for each concentration. Error bars correspond to standard deviations. * Increases in nuclear fluorescence at each concentration were significant compared to all other concentrations; p < 0.001 in all cases. This experiment was performed twice with similar results obtained in all cases. Equivalent results were also obtained on a separate experiment using a Discovery-1 system for fluorescence measurements. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

by electroporation. For the development of stable cell lines, 1361.5 cells, derived from NIH/3T3 cells [19] were transfected with an expression vector for the GRER chimera, an MMTV-luciferase reporter gene and a neomycin resistance marker at a 10:10:1 molar ratio. Forty-eight hours after transfection, cells were changed into media containing 1 mg/ml G418 and selected for a period of 2–3 weeks. A resistant colony was expanded and cell line 5624 was established. Cell line 3134 was similarly stably transfected, resulting in the establishment of a cell line containing the stably integrated GFP–GRER construct. 2.4. Transactivation assays and EC50 determinations After the corresponding hormone treatment, cells were harvested in 0.25 M Tris–HCl, pH 7.8 and subjected to three cycles of freeze/thawing. Cell lysates were cleared of debris by centrifugation and protein concentration was determined using a standard calorimetric assay. Equal amounts of protein across samples were used for either luciferase or CAT assays. For luciferase assays, cell lysates were combined with a reaction mix containing 0.1 M KPO4 supplemented with MgCl2 and ATP in a 96-well format. Luciferin substrate was

automatically injected into each well and luminescence was immediately measured in a luminometer. For CAT assays, cell lysates were combined in scintillation vials with 1 ␮l of tritium labeled acetyl Coenzyme A, 19.0 ␮l of 2 mg/ml chloramphenicol, and 80 ␮l of 0.25 M Tris–HCl, pH 7.8. Econoflour-2 scintillation mixture was overlaid on the reactions [20]. Labeled substrate in the organic layer was detected in a scintillation counter and the rate of the increase of counts per minute over time was calculated. For EC50 determinations, the transcriptional activity of the GRER chimera in cells treated with increasing concentrations of ligand was measured as above. The data was then plotted and fitted using DeltaGraph 5.0 software according to the following equation: y = ((max x)/(EC50 + x)) + b, where b is the background value and x is the ligand concentration [21]. 2.5. Immunoprecipitations Cells were transiently transfected with expression vectors for the GFP-tagged GRER chimeras, for GFP–GR (pCI-nGFP–GRC656G), or for the w.t. GFP protein. Immunoprecipitations were performed as described [17], using a monoclonal antibody against GFP (3EQ, Biogene) to

310

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

Fig. 2. Diagram of the composition of the GRER LBD chimeras used in this study. (A) GRER contains the rGR N-terminus, DBD and hinge regions upstream of a hybrid LBD composed of rGR helix 1 and partial loop 1–3 sequences linked to hER␣ LBD sequences starting with the C-terminus of loop 1–3. (B) GRER
loop contains a partial deletion of hER loop 1–3 sequences, shortening this loop, but is otherwise identical to GRER. (C) Structure of the GR construct expressing only helix 1 sequences of the LBD and eight amino acids corresponding to linker sequences. In all cases, the last rGR amino acid and the first hER amino acid expressed are shown. Red represents GR, blue ER. The underlined amino acids in the linker sequence are expressed in all the constructs. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

bring down complexes. Immunoprecipitates were subjected to SDS electrophoresis and blotted on nitrocellulose. Membrane blots were probed with a monoclonal anti-HSP 90 antibody (BD Transduction Laboratories 610418) and detected by chemiluminescence. 2.6. Fluorescence microscopy and translocation assays Transiently or stably transfected cells were plated on #1.5 glass-cover chambers (Lab-Tek II, Nalge Nunc) overnight and then treated with hormone as indicated in figure legends. Cells were viewed on an Olympus 1X70 system equipped with a Photometrics camera or a CARV Metamorph system equipped with an Orca II CCD camera. To calculate the EC50 values for translocation, cells were fixed after hormone treatment with 3.5% paraformaldehyde for 10 min and rinsed in PBS. Cells were then stained with Dapi to visualize the nucleus, rinsed with PBS and then viewed in a CARV-Metamorph system that monitors fluorescence in dual channels. Dapi signals were used to define regions corresponding to the nucleus and GFP intensity was measured for individual cells both within the nucleus and within the whole cell. After background subtraction, the intensity of GFP in the nucleus versus the total intensity per cell was calculated for each treatment. Averages across hormone concentrations were plotted and fitted as described above for EC50 determinations.

2.7. Three-dimensional modeling Amino acid sequences corresponding to GRER chimeras were submitted to the Swissprot modeling site and modeled against known structures for GR (pdb 1M2Z and 1NHZ). The models were viewed and simplified using the Swissprot viewer to show ribbon traces of only the region of interest of the LBD. 2.8. Fluorescence recovery after photobleaching Prior to live cell imaging, NIH/3T3 cells were plated on Lab-Tek II chambers (Nalge Nunc International, Naperville, IL) at 50,000 cells per well. The cells were grown for 1 or 2 days prior to transient transfection. Transfections were performed as described above, introducing expression vectors for the GRER chimera or for w.t. ER (pCl-GFP-HEGO). The day after transfection, cells were treated for 1–4 h with 100 nM estradiol or 1 ␮M ICI, and FRAP analysis was carried out on a Zeiss 510 laser scanning confocal microsocope using a 100×/1.3N.A. oil immersion objective and 40 mW argon laser. The stage temperature was maintained at 37 ◦ C with an ASI 400 Air Stream incubator (Nevtek). Five single imaging scans were acquired prior to bleaching by a bleach pulse of 160 ms using 458/488/514 nm laser lines at 100% laser power (laser output 75%) without attenuation. Bleached areas were set to circles of 60 × 60 pixels in diameter in all cases.

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

Single z-section images were then collected at 0.5 s intervals using a 488 nm laser line with laser power attenuated to 0.1%. Fluorescence intensities in the regions of interest were analyzed and FRAP recovery curves were generated using LSM software and Microsoft Excel as previously described [22]. Values in FRAP recovery curves represent averages from at least 20 cells imaged in 2–3 independent experiments.

3. Results To assess the feasibility of using the cytoplasmic to nuclear translocation of GFP-tagged receptors as an assay in nuclear receptor ligand screens, we monitored changes in nuclear fluorescence in response to glucocorticoids in a cell line expressing a GFP–GR fusion protein under Tet control [23]. The GR expressed in this cell line contains a single point mutation (C656G) in the ligand binding domain (LBD) that increases the sensitivity of the receptor to hormone [24] enabling the induction of the tagged GR at low concentrations of dexamethasone (Dex) while the endogenous GR is uninduced [25]. Nuclear fluorescence intensity (corresponding to the nuclear translocation of the tagged receptor) was monitored over a range of hormone concentrations using a platform developed by Q3DM equipped with an automated fluorescence microscope and data analysis software (see Section 2). Changes in nuclear fluorescence in measured cells (n > 140 for each concentration) showed a clear and significant dose dependence, with translocation becoming apparent even at relatively low concentrations of glucocorticoids (Fig. 1A and B). Encouraged by the possibility of using this approach for drug discovery and nuclear receptor functional studies, we decided to develop a system that would respond to estrogens. To create an estrogen receptor that would reside in the cytoplasm and only translocate to the nucleus in the presence of ligand, we designed a chimeric protein containing sequences of the LBD of ER␣ downstream of GFPtagged GR sequences corresponding to the N-terminus, DNA binding domain, hinge and partial LBD regions of the GR (Fig. 2A). The cellular localization of our novel GRER chimera was first analyzed in transiently transfected COS 1 cells. Unlike the parental w.t. ER, the chimera resided in the cytoplasm in the absence of hormone (Fig. 3A, left panel) and its localization was unaltered by GR ligands such as dexamethasone (not shown and Fig. 5A). Estradiol, however, caused a clear and rapid redistribution of the chimeric receptor to the nucleus (Fig. 3A, right panel) that was complete within 1–2 h. Both the cytoplasmic distribution of the unliganded chimera and its ability to respond to ligand by translocating to the nucleus were maintained in all cell lines tested (Fig. 3B and C; CV1, MCF7 and 1471.1 cells not shown), confirming that the phenotype is not cell-type-dependent. To evaluate the dose-dependence of the nuclear translocation process, 3134 cells containing the integrated GFP–GRER chimera construct, were treated for 4 h with increasing concentrations of estradiol and the nuclear versus total cell fluorescence was

311

measured as outlined in Section 2.1. As can be seen in Fig. 3D, the nuclear translocation process was dose-dependent, reaching a maximal efficiency of about 75%, similar to what was observed in the case of the GR (Fig. 1B) in the same cells. It is thought that untransformed receptors are retained in the cytoplasm, at least in part, through a mechanism involving chaperone proteins, including HSP90 [26,27]. Several studies have found that the chaperones interact with steroid receptors through the LBD, yet multiple regions of this domain seem to serve as interaction sites when expressed outside the context of the full receptor [28]. More recently, it has been proposed that the GR in particular, interacts with HSP90 through a region of helix 1 of the LBD that is conserved in our chimera [29]. We, therefore, expected GRER to interact with this chaperone. To confirm this, we performed immunoprecipitation studies in transfected cells. GFP-tagged proteins were transiently expressed in COS cells and precipitated with an anti-GFP antibody loaded on beads. The presence of HSP90 in the immunoprecipitates was determined by western analysis. As demonstrated in Fig. 4, the GRER chimera clearly interacts with HSP90, as does the positive control, GR. In contrast, GFP alone did not show a detectable interaction with the chaperone (Fig. 4A). It is thus clear that the untransformed GRER chimera is retained in the cytoplasm and that its hybrid LBD maintains the ability to interact with the HSP90 chaperone. We have demonstrated here that the GRER chimera mimics the GR in its cytoplasmic accumulation when uninduced and in its nuclear translocation when activated by estradiol (Fig. 3). This behavior allowed us to investigate whether other known ER ligands could reproduce, in the context of the GRER chimera, the structural signals that cause receptor translocation. This was an important step in testing whether a nuclear translocation assay could detect ER ligands other than estradiol in our cell-based system. To this end, we stably transfected 1361.5 cells (derived from NIH/3T3) with the GRER expression vector, an MMTV-luciferase reporter gene (for later measurements of transcriptional activity) and a neomycin resistance marker. A G418-resistant colony was expanded and cell line 5624 was established. The cellular localization of the constitutively expressed chimera was monitored by fluorescence microscopy in the presence or absence of various ER ligands. As can be seen in Fig. 5A and B, the constitutively expressed chimera is predominantly cytoplasmic even in the presence of dexamethasone, but it becomes mainly nuclear when induced by estradiol, just like the transiently expressed receptor. In the absence of estradiol nuclear fluorescence was approx. 20% of total cell fluorescence whereas 100 nM estradiol increased this nuclear partitioning of fluorescence to about 70–80%, similar to our results in 3134 cells. Importantly, we found that both natural and synthetic ER␣ agonists fully translocated the receptor to the nucleus (Fig. 5B). An ER␤-specific ligand, DPN [30,31], failed to elicit receptor translocation even at 100 ␮M, suggesting the chimeric LBD maintains ER␣ specificity (Fig. 5C). Most intriguingly, both full (ICI 182,780) and

312

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

Fig. 3. Cytoplasmic localization of GRER in various cell types. (A) COS 1 cells were transiently transfected with a GFP–GRER expression vector and were treated the next day for 1–4 h with 100 nM 17-␤-estradiol or were left untreated. Live cells were visualized under a fluorescent microscope. (B) 3134 cells. (C) NIH/3T3 cells. (A–C) Representative cells are shown. These experiments were carried out two to four times and random areas of the wells were analyzed by fluorescence microscopy. In all cases, the cytoplasmic to nuclear translocation of the chimera was apparent in response to estradiol. (D) 3134 cells expressing the GFP–GRER chimera from an integrated construct were exposed to increasing concentrations of estradiol for 4 h. Nuclear vs. total cell fluorescence was measured by automated fluorescence microscopy on a Discovery-1 system using a DNA stain to define the nucleus as described in Section 2. The estradiol dose dependence of the translocation process is shown for a representative experiment where at least 25 cells were measured per estradiol concentration. Bar graphs represent the standard deviation of the translocation response across cells treated with a particular amount of estradiol.

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

313

Table 1 Comparison of EC50 values for the transactivation vs. the translocation process in response to ER␣ agonists

Translocation Transactivation

Estradiol (nM)

PPTa (␮m)

DES (nM)

EC50 = 46 ± 11 EC50 = 80 ± 25

EC50 = 1.4 ± 0.2 EC50 = 4.7 ± 0.8

EC50 = 30 EC50 = 149 ± 19

Values given are averages of two to four independent determinations with the exception of translocation EC50 for DES. Standard deviations follow average values. a Dose–response curves for PPT did not achieve full saturation at 10 ␮M.

Fig. 4. GRER and GRER
loop interact with the HSP90 chaperone. (A) COS1 cells were transiently transfected with expression vectors for GFP, GFP–GRER (GE) or GFP–GR. Complexes were immunoprecipitated from cytosolic extracts with an anti-GFP antibody loaded on beads, subjected to SDS electrophoresis, blotted on a nitrocellulose membrane and probed with an anti-HSP90 antibody, as detailed in Section 2. HSP90 present in 1% of input cytosolic extract is shown on the left panel. (B) COS1 cells transiently transfected with expression vectors for GFP–GR (GR), GFP–GRER (GE), or GFP–GRER
loop (
loop) were similarly processed.

partial (tamoxifen) ER antagonists were able to cause translocation, although a fraction of the receptor protein remained in the cytoplasm in many cells (Fig. 5D). Raloxifene, a selective estrogen receptor modulator (SERM), induced complete nuclear translocation at 10 ␮M (not shown). The ability of the antagonists and SERMS to translocate the receptor suggests that upon binding to the LBD, these ligands are able to cause the structural rearrangements in the receptor that result in receptor movement, similar to the action of agonists. We next measured the sensitivity of our system to low affinity ER ligands. We tested the ability of genestein, a natural isoflavone phytoestrogen [32,33] to translocate the receptor into the nucleus. Genestein was very effective in translocating the chimera with clear nuclear translocation observed at 10 ␮M (Fig. 5E). This shows that our assay is sensitive to phytoestrogens and potentially to other low affinity ligands. Thus this nuclear translocation assay can detect a variety of ER␣ ligands and could be used as a screening tool. In screens for novel ER modulators, it would be advantageous to determine the efficacy and potency of candidate compounds. To evaluate our system for efficacy and potency measurements, we performed comparative dose response translocation studies. Nuclear translocation was quantitated by fluorescence microscopy after overnight treatment with increasing concentrations of agonists (Fig. 5F). Nuclear translocation EC50 values were then calculated for each ago-

nist ligand (Table 1) as outlined in Section 2. Estradiol had a translocation EC50 of 46 nM. Synthetic agonist DES had a higher potency than PPT in moving the receptor into the nucleus. The three ligands showed similar efficacy since they translocated about 80% of the receptor at high concentrations. Antagonists, as mentioned above, had a weaker translocation efficacy. This demonstrates that our system can be used to screen for general ER ligands as well as to evaluate highaffinity ligands for potency and efficacy. Because full agonists, full antagonists and mixed agonist–antagonist ligands and SERMS can all translocate the receptor, although to varying degrees, our assay can act as a pre-screen for all these types of receptor ligands. Secondary assays can then be performed to characterize the potential compounds according to their ability to induce or inhibit the transcriptional activity of the receptor. To evaluate the ability of the chimera to activate transcription, a series of reporter assays were performed. NIH/3T3 cells were transiently transfected with an expression vector for the GRER chimera and an MMTV-luciferase reporter construct. Luciferase activity was enhanced over ten fold in the presence of 100 nM estradiol and the response was dose-dependent (Fig. 6A). The chimera was also transcriptionally active in COS1, CV1, MCF-7 and 1471.1 cells (not shown). This demonstrates that the hybrid structure of the receptor and the novel activity that the ER LBD acquires within this chimeric context do not interfere with the ability to activate transcription. Some steroid receptors have been reported to differ in their transcriptional behavior depending upon the structure of the promoter template [34,35]. We therefore evaluated whether the transiently expressed chimera would be able to activate transcription from integrated templates. GRER was transiently expressed in 1470.2 cells, which contain replicating copies of the MMTV-LTR promoter regulated by a family of stably positioned nucleosomes [36]. The transiently expressed chimera activated transcription from this integrated template in a dose-dependent manner (Fig. 6C). The constitutively expressed chimera also showed clear transcriptional activity on replicating MMTV templates in 5624 cells, with an EC50 of 80 nM (Fig. 6B). As expected for hybrids of modular proteins, the chimera is not as sensitive to hormone as the w.t. ER. Taken together, these results demonstrate that both the transiently expressed and the constitutively expressed GRER chimera are able to trigger downstream events at integrated templates in response to estradiol, resulting in transcriptional

314

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

Fig. 5. ER␣ ligands translocate the GRER chimera to the nucleus. (A) 5624 cells were plated on glass bottom slides and treated with 1 ␮M dexamethasone overnight before visualizing under a fluorescence microscope. (B) 5624 cells were plated on glass bottom slides and treated with various ER␣ agonists: E2, 100 nM 17-␤-estradiol; PPT, 10 ␮M 4,4 ,4 -(4-propyl-[1H]-pyrazole-1,3,5-triyl)trisphenol; DES, 1 ␮M diethylstilbestrol. (C) 5624 cells were treated with an ER␤ agonist; DPN, 0.1 ␮M (left panel) or 10 ␮M (right panel) 2,3-bis(4-hydroxyphenyl)-propionitrile. (D) 5624 cells were treated with ER␣ antagonists: TMX, 10 ␮M tamoxifen citrate; ICI, 1 ␮M ICI 182,780. (E) Cells were treated with a phytoestrogen; Gen, genestein (10 ␮M, left panel; 100 ␮M, right panel). (F) Translocation dynamics for ER␣ agonists were measured over a range of concentrations. All hormone treatments were overnight; translocation was apparent after 30 min to 4 h of treatment depending on the ligand and its concentration.

activation of the MMTV promoter. While robust transcriptional activity was also seen in cells treated with other agonists (Fig. 7A and B and Table 1), no transcriptional activity was detected in antagonist-induced cells (Fig. 7C and D),

except for high doses of tamoxifen as expected from its weak partial agonist activity. Indeed, the antagonists inhibited the agonist-induced transcriptional activity (Fig. 7E) as they do for w.t. ER.

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

315

Fig. 6. Transcriptional activity of GRER. (A) NIH/3T3 cells were transiently transfected with an expression vector for the GRER chimera and an MMTVluciferase reporter construct. After 24 h, cells were stimulated with increasing concentrations of 17-␤-estradiol overnight and assayed for luciferase activity. (B) 5624 cells, stably expressing the GRER chimera and containing replicating copies of an MMTV-luciferase reporter were treated with increasing concentrations of 17-␤-estradiol overnight. (C) 1470.2 cells, which contain replicating copies of MMTV-CAT, were transiently transfected with an expression vector for GRER and induced with increasing concentrations of estradiol as in (B). Cells were then assayed for Cat activity. (A–C) Cells were assayed for luciferase activity or CAT activity as described in Section 2. Hormone treatments were done overnight. Representative experiments are shown. Error bars correspond to standard deviations across equivalent samples within the same experiment.

It is known that GR molecules lacking the entire LBD localize to the nucleus and exhibit constitutive transcriptional activity [37–39]. Thus, LBD sequences are necessary for cytoplasmic localization and hormone responsiveness and are thought to function at least partly by masking the NLS of the GR. To begin addressing the structural features in the LBD that may contribute to cytoplasmic targeting of unliganded steroid receptors, several constructs were analyzed. First, we tested whether GR helix 1 sequences, proposed to provide a site for HSP90 binding [29], are sufficient for cytoplasmic retention of the receptor in the absence of downstream LBD sequences. We analyzed the cellular localization of a GFPtagged, C-terminally truncated GR expressing only helix 1 of the LBD downstream of the N-terminus and DBD (Fig. 2C). This truncated receptor was found exclusively in the nuclear compartment (Fig. 8A), demonstrating that the signals contained in helix 1 of the LBD are not sufficient for cytoplasmic retention. In contrast, partial deletion of the loop between helices 1 and 3 (Fig. 2B), resulted in nuclear exclusion even in the presence of high estradiol concentrations (Fig. 8B). Longer exposure to hormone did induce the nuclear translocation of this altered receptor, but the process was severely delayed, requiring about 24 h for nuclear accumulation in a stable cell line (compared to 2–4 h for detection of nuclear GRER in a stable cell line). According to crystallographic data, the 1–3 loop is not in direct contact with ligand but in the periphery of the folded receptor [40–43]. Nonethe-

less, it is possible that this loop deletion decreases ligand affinity, indirectly affecting the translocation process. Efforts to directly determine binding Kd values were unsuccessful because of high background binding, possibly due to the presence of other nuclear receptors in these cells. Interestingly, this loop mutant retained the ability to interact with HSP90 in immunoprecipitation studies (Fig. 4B), suggesting that its slow nuclear translocation is not likely the result of an aberrant LBD conformation, as might occur if interactions with chaperones were abolished. Whether the loop between helices 1 and 3 plays a direct or indirect role in nuclear translocation remains an open question. Like other steroid receptors, ER exhibits rapid nuclear dynamics in its unliganded form [44–47]. The w.t. ER has remarkably different mobilities in the nucleus depending on whether it is bound to agonists or antagonists. The most striking difference in receptor mobility is that seen between the rapidly moving estradiol-bound ER and the almost immobile ICI-bound ER, as reported by Mancini and colleagues [47]. Since both ER␣ agonists and antagonists were able to induce the translocation of the chimeric receptor into the nucleus, we characterized the effects of these ligands on the nuclear mobility of our chimera, and compared the response to that of w.t. ER. Fluorescence recovery after photobleaching (FRAP) was used to measure the movement of the w.t. and chimeric receptors in the nucleoplasm of transiently transfected NIH/3T3 cells. As expected, w.t. ER fluorescence was

316

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

Fig. 7. Transcriptional activity dose response of the GRER chimera after stimulation with various ER␣ ligands. 5624 cells were treated with increasing concentrations of PPT (A), DES (B), tamoxifen (C), ICI (D), or with a combination of 100 nM 17-␤-estradiol and 10 ␮M antagonists (E). Luciferase activity is expressed as fold induction over uninduced cells. Error bars represent standard deviations across two independent representative determinations.

recovered in the bleached area within seconds in the case of the estradiol liganded receptor, with half the fluorescence in the region regained within the first 5 s after bleaching (Fig. 9A, top panel and Fig. 9B, light blue curve). In contrast, the ICI-bound receptor repopulated the bleached area very slowly, showing very poor recovery even after 50 s postbleaching (Fig. 9A, second panel and Fig. 9B, dark grey curve), consistent with earlier observations [47]. Remarkably, the GRER chimera moved rapidly in the nucleoplasm, repopulating the bleached area with half the initial fluorescence in less than 5 s regardless of the bound ligand (Fig. 9A bottom two panels and Fig. 9C). ICI did slow down the chimera’s mobility with respect to estradiol (Fig. 9C pink versus red

curves), but this small decrease in mobility was in strong contrast with the immobility of the ICI-bound w.t. ER. These studies suggest the possible involvement of regions outside the LBD in receptor mobility and/or the active engagement of sequences in helix 1 and the 1–3 loop of the LBD in modulating receptor movement in the nucleoplasm.

4. Discussion Currently available estrogen receptor agonists, antagonists and selective estrogen receptor modulators have proven effective in the treatment of conditions ranging from breast and

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

317

Fig. 8. Effects of mutants on receptor localization. (A) COS 1 cells were transiently transfected with an expression vector for a GFP-tagged GR truncated after helix 1 of the LBD. (B) NIH/3T3 cells stably expressing GRER
loop were treated with 100 nM 17-␤-estradiol overnight. (C) A three-dimensional ribbon model comparing the structure of the loop between helices 1 and 3 in GRER and GRER
loop. The red trace represents GRER; the green trace represents GRER
loop. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

ovarian cancer to osteoporosis and menopause. Their use, however, has not been without controversy because in some cases the advantages provided to some tissues are compromised by side effects in other tissues. New modulators of the estrogen receptor that may have tissue specific activity are needed to treat a broader set of disorders and to minimize undesired side effects. We have developed a new mammalian cell-based assay to screen for ligands of the estrogen receptor. We designed a chimera between the glucocorticoid and the estrogen receptors that, unlike the parental constitutively nuclear estrogen receptor, is cytoplasmic in the absence of hormone and translocates to the nucleus in response to estradiol in all tested cell lines. By tagging this receptor with a fluorescent protein, we can follow the receptor’s cellular localization in response to compounds with potential estrogen-like activity. In vivo translocation studies of the fluorescent chimera stably expressed in mammalian cells revealed that estrogen receptor ␣ agonists, antagonists and phytoestrogens induce its nuclear transport, while nonestrogenic steroids or estrogen receptor ␤-specific agonists do not. We have shown that the estrogenic potency and efficacy of high-affinity ligands can be measured in our system, and that translocation EC50 values can be calculated and correlate to transactivation values. Relevant secondary assays can be performed in the same cell system, since the chimera is transcriptionally active on an integrated reporter gene in response to estrogen receptor agonists but is inhibited by estrogen receptor antagonists. We have demonstrated that the assay is suitable for use in high throughput platforms and suggest its future use in screens for novel estrogen receptor modulators. Both the transiently transfected and the constitutively expressed chimera are found in the cytoplasm, demonstrating that this phenotype is independent of the mode of receptor expression and of possible differences in the processing of

the chimeric protein. Cytoplasmic retention of receptors in the cytoplasm is thought to be related to their mode of association with chaperone proteins. In designing the chimera, we chose to maintain helix 1 of the GR LBD in its normal context because this region has been recently shown to be necessary for interactions with HSP90 [29]. Although this region may contribute to the cytoplasmic phenotype of the chimera [48], we found that GR helix 1 sequences alone are insufficient to keep the uninduced receptor in the cytoplasm (Fig. 8A). Downstream sequences are necessary for cytoplasmic accumulation. A partial deletion of the loop between helices 1 and 3 hindered efficient nuclear translocation. Three-dimensional modeling of GRER and GRER
loop using published ER␣ and GR X-ray crystal structures as templates (Fig. 8C) suggests that the partial deletion of the loop between helices 1 and 3 seen in GRER
loop could affect the formation of a pseudo-beta strand implicated in w.t. GR in a hydrogen bond network at the dimer interface [40,41]. The orientation of the 1–3 loop has been shown to undergo rearrangements upon treatment of the w.t. ER with tamoxifen [49]. This structural change is thought to contribute to the formation of antagonistic forms of the receptor and is also seen in crystals of certain ER mutants [49]. In addition, there has been a suggestion that this region is important in the GR for functional interactions with immunophilins and that changes in its sequence affect receptor responses [57]. In the context of these earlier studies, our results suggest that the 1–3 loop may directly or indirectly be involved in the structural signals that accompany nuclear translocation in response to ligand, and that these signals may partly overlap with dimerization signals. Differences have been noted in the transcriptional behavior of transiently versus constitutively expressed receptors [35,34]. Our chimera is transcriptionally active on replicating templates whether it is transiently or constitutively expressed.

318

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

Fig. 9. The GRER chimera is resistant to ICI-mediated immobilization. (A) z-Sections of representative individual cells transiently expressing w.t. ER tagged with GFP (top two panels) or the GFP-tagged GRER chimera (bottom two panels) were subjected to FRAP analysis after treatment with 100 nM estradiol or 1 ␮M ICI. (B–C) FRAP curves were derived from qualitative analysis of at least 20 cells as detailed in Section 2. For each FRAP series, the bleached zone is indicated by the dashed yellow line. The great reduction in ER mobility caused by ICI (dark grey curve, panel B) is not observed with the chimeric receptor (pink curve, panel C). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of the article.)

In contrast, a chimera that combined the GR N-terminus and DNA binding domains upstream of the ER hinge and LBD did not have the ability to activate integrated templates [50]. This difference could potentially reside in our unique design of the hybrid LBD. In comparison with the molecule described by Bonavich et al., the GRER chimera we have designed contains the hinge, LBD helix 1, and partial helix 1–3 loop sequences of the GR. It is therefore possible that these regions confer onto the chimera the ability to interact with chromatin remodeling activities not present in equivalent ER subdomains. The fact that ER antagonists induced partial translocation of the chimeric receptor but did not enhance its transcriptional activity indicates that the translocation and transactivation events require at least partly distinct structural rearrangements in the receptor molecule. While the antagonist-bound receptors are competent for nuclear translocation, they are unable to activate gene transcription. Our chimera offers for

the first time an approach for evaluating ER antagonists in terms of their ability to cause receptor movement into the nucleus. In addition, because the translocation EC50 values are closely related to the transactivation EC50 values (Table 1), the potency and efficacy of strong agonists can be established by automatic measurement of the translocation dynamics over a range of agonist concentrations. The observation that a pure antagonist such as ICI can induce the translocation of the chimera into the nucleus suggests that the antagonist-bound LBD can adapt a conformation that is competent for nuclear retention. The ICI-bound chimera in the nucleus, however, can be distinguished from the w.t. receptor, since the pure antagonist only partly reduces the chimera’s mobility compared to the almost complete immobilization of w.t. ER, as shown by our FRAP studies. The generally faster mobility of the chimera compared to the w.t. ER may reflect the chimera’s shorter residence time on or decreased interaction with DNA targets. This is consistent

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

with the lower transcriptional activity of the chimeric receptor reflected on its higher EC50 , compared to w.t. ER. Mancini and colleagues first reported the observation that ICI immobilizes the ER in the nucleus [47]. These investigators proposed that this effect was mediated by helix 12 sequences of the ER LBD. Our results show that the ICI effect can be overcome even in the presence of ER LBD helix 3 through helix 12 sequences, by expressing upstream GR amino acids, suggesting that receptor regions N-terminal to helix 3 contribute to the mobility of the receptor. Over recent years, an effort has been made to develop systems for the accurate detection of environmental estrogens and phytoestrogens [4,2,1]. The yeast estrogen screen (YES) has been successfully used to test compounds for estrogenic activity as well as to analyze wastewater and water bodies for the presence of estrogenic substances [4,3,9]. In contrast with cell proliferation assays [3], the YES assay is relatively simple. Use of a yeast system is significantly limited, however, for screening of potentially therapeutic estrogen receptor modulators, since Saccharomyces cerevisiae lacks most of the coactivators and coregulatory proteins that function in the mammalian steroid receptor pathway. Indeed, certain ER ligands such as resveratrol, known to have estrogenic activity in mammalian cells, do not show activity in the YES assay [7]. As demonstrated in Fig. 1 and in Fig. 3D, the mammalian cell-based fluorescent translocation assay described here can be easily automated. Thus, high throughput screens can be implemented with this approach. In addition, the GRER chimera reflects the biological ligand specificity of the native receptor, since it responds to all ER␣ ligands tested but not to ER␤ ligands or glucocorticoids. Thus, it is reasonable to predict that our live mammalian cell assay will facilitate the rapid analysis of certain environmental samples as well as the identification of both novel agonists and antagonists of the hER␣ protein. The chimeric receptor translocation system also offers the possibility of identifying a new set of ligands and/or translocation effectors that may trigger the movement or retention of receptors in or out of the nucleus without altering the expression of hormone responsive genes. Screens can be performed in combination with estradiol treatment to find compounds that inhibit the translocation of the receptor into the nucleus, thus acting as a novel class of antagonists. These compounds could be particularly useful to inhibit the ligand-independent activity of the ER that occurs in certain cancer cells [51–54]. In summary, we have shown that a member of the steroid receptor family can be converted from an exclusively nuclear localization to a form that is ligand-dependent for translocation, extending our earlier observations on the retinoic acid receptor. Since the helix 1–helix 3 region from the LBD of the nuclear receptors is the most highly conserved domain in the superfamily [55,56], our findings suggest that other, perhaps most, members of the family can be converted to translocating versions by similar fusion with GR. This in turn opens the possibility for wide ranging efforts in new ligand discovery for the receptor superfamily, including orphan receptors,

319

by an entirely novel approach in the context of live receptor biology in mammalian cells. Acknowledgments We wish to thank Catharine L. Smith, Sam John, and Tom Johnson for critical reading of the manuscript and Cem Elbi for insightful discussions. We are grateful to Tim Moran and Q3DM (now Beckman) for the use of their equipment and for technical advice. This publication has been funded in whole or in part with Federal funds from the National Cancer Institute, National Institutes of Health, under contract #NO1-CO-12400. The content of this publication does not necessarily reflect the views or policies of the Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government. This research was supported by the Intramural Research Program of the NIH, National Cancer Institute, Center for Cancer Research. References [1] R. Bolger, T.E. Wiese, K. Ervin, S. Nestich, W. Checovich, Rapid screening of environmental chemicals for estrogen receptor binding capacity, Environ. Health Perspect. 106 (9) (1998) 551–557. [2] S.D. Garrett, H.A. Lee, M.R. Morgan, A nonisotopic estrogen receptor-based assay to detect estrogenic compounds, Nat. Biotechnol. 17 (12) (1999) 1219–1222. [3] T.H. Rasmussen, F. Nielsen, H.R. Andersen, J.B. Nielsen, P. Weihe, P. Grandjean, Assessment of xenoestrogenic exposure by a biomarker approach: application of the E-Screen bioassay to determine estrogenic response of serum extracts, Environ. Health 2 (1) (2003) 12–18. [4] S.F. Arnold, M.K. Robinson, A.C. Notides, L.J. Guillette Jr., J.A. McLachlan, A yeast estrogen screen for examining the relative exposure of cells to natural and xenoestrogens, Environ. Health Perspect. 104 (5) (1996) 544–548. [5] G.F. Allan, A. Hutchins, J. Clancy, An ultrahigh-throughput screening assay for estrogen receptor ligands, Anal. Biochem. 275 (2) (1999) 243–247. [6] D. Shao, T.J. Berrodin, E. Manas, D. Hauze, R. Powers, A. Bapat, D. Gonder, R.C. Winneker, D.E. Frail, Identification of novel estrogen receptor alpha antagonists, J. Steroid Biochem. Mol. Biol. 88 (4/5) (2004) 351–360. [7] C.M. Klinge, K.E. Risinger, M.B. Watts, V. Beck, R. Eder, A. Jungbauer, D. Shao, T.J. Berrodin, E. Manas, D. Hauze, R. Powers, A. Bapat, D. Gonder, R.C. Winneker, D.E. Frail, Estrogenic activity in white and red wine extracts Identification of novel estrogen receptor alpha antagonists, J. Agric. Food Chem. 51 (7) (2003) 1850–1857. [8] G.J. Parker, T.L. Law, F.J. Lenoch, R.E. Bolger, Development of high throughput screening assays using fluorescence polarization: nuclear receptor–ligand-binding and kinase/phosphatase assays, J. Biomol. Screen. 5 (2) (2000) 77–88. [9] D.M. Klotz, B.S. Beckman, S.M. Hill, J.A. McLachlan, M.R. Walters, S.F. Arnold, Identification of environmental chemicals with estrogenic activity using a combination of in vitro assays, Environ. Health Perspect. 104 (10) (1996) 1084–1089. [10] H. Ogawa, S. Inouye, F.I. Tsuji, K. Yasuda, K. Umesono, Localization, trafficking, and temperature-dependence of the Aequorea green fluorescent protein in cultured vertebrate cells, Proc. Natl. Acad. Sci. U.S.A. 92 (25) (1995) 11899–11903.

320

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321

[11] H. Htun, J. Barsony, I. Renyi, D.J. Gould, G.L. Hager, Visualization of glucocorticoid receptor translocation and intranuclear organization in living cells with a green fluorescent protein chimera, Proc. Natl. Acad. Sci. U.S.A. 93 (1996) 4845–4850. [12] D. Picard, V. Kumar, P. Chambon, K.R. Yamamoto, Signal transduction by steroid hormones: nuclear localization is differentially regulated in estrogen and glucocorticoid receptors, Cell Regul. 1 (3) (1990) 291–299. [13] M.F. Press, S.H. Xu, J.D. Wang, G.L. Greene, Subcellular distribution of estrogen receptor and progesterone receptor with and without specific ligand, Am. J. Pathol. 135 (5) (1989) 857–864. [14] D.L. Stenoien, M.G. Mancini, K. Patel, E.A. Allegretto, C.L. Smith, M.A. Mancini, Subnuclear trafficking of estrogen receptor-alpha and steroid receptor coactivator-1, Mol. Endocrinol. 14 (4) (2000) 518–534. [15] E.V. Jensen, Steroid hormones, receptors, and antagonists, Ann. NY Acad. Sci. 784 (1996) 1–17. [16] H. Htun, L.T. Holth, D. Walker, J.R. Davie, G.L. Hager, Direct visualization of ligand occupied and unoccupied human estrogen receptor alpha reveals a role for ligand in the nuclear distribution of the receptor, Mol. Biol. Cell 10 (1999) 471–486. [17] S. Mackem, C.T. Baumann, G.L. Hager, A glucocorticoid receptor/RAR receptor chimera that supports cytoplasmic/nuclear translocation in response to retinoic acid: a real-time sensing assay for nuclear receptor ligands, J. Biol. Chem. 276 (49) (2001) 45501–45504. [18] P. Kramer, G. Fragoso, W.D. Pennie, H. Htun, G.L. Hager, R.R. Sinden, Transcriptional state of the mouse mammary tumor virus promoter can effect topological domain size in vivo, J. Biol. Chem. 274 (40) (1999) 28590–28597. [19] M.C. Ostrowski, H. Richard-Foy, R.G. Wolford, D.S. Berard, G.L. Hager, Glucocorticoid regulation of transcription at an amplified, episomal promoter, Mol. Cell. Biol. 3 (11) (1983) 2045–2057. [20] S. Zhang, M. Danielsen, Cotransfection assays and steroid receptor biology, Meth. Mol. Biol. 176 (2001) 297–316. [21] S. Zhang, M. Danielsen, Selective effects of 8-Br-cAMP on agonists and antagonists of the glucocorticoid receptor, Endocrine 3 (1995) 5–12. [22] C. Elbi, D.A. Walker, G. Romero, W.P. Sullivan, D.O. Toft, G.L. Hager, D.B. DeFranco, Molecular chaperones function as steroid receptor nuclear mobility factors, Proc. Natl. Acad. Sci. U.S.A. 101 (9) (2004) 2876–2881. [23] D. Walker, H. Htun, G.L. Hager, Using inducible vectors to study intracellular trafficking of GFP-tagged steroid/nuclear receptors in living cells, Methods (Compan. Meth. Enzymol.) 19 (3) (1999) 386–393. [24] P.K. Chakraborti, M.J. Garabedian, K.R. Yamamoto, S.S. Simons Jr., Role of cysteines 640, 656 and 661 in steroid binding to rat glucocorticoid receptors, J. Biol. Chem. 267 (1992) 11366– 11373. [25] C.L. Smith, H. Htun, R.G. Wolford, G.L. Hager, Differential activity of progesterone and glucocorticoid receptors on mouse mammary tumor virus templates differing in chromatin structure, J. Biol. Chem. 272 (22) (1997) 14227–14235. [26] W.B. Pratt, D.O. Toft, Steroid receptor interactions with heat shock protein and immunophilin chaperones, Endocr. Rev. 18 (3) (1997) 306–360. [27] W.B. Pratt, The role of heat shock proteins in regulating the function, folding, and trafficking of the glucocorticoid receptor, J. Biol. Chem. 268 (29) (1993) 21455–21458. [28] F.C. Dalman, L.C. Scherrer, L.P. Taylor, H. Akil, W.B. Pratt, Localization of the 90-kDa heat shock protein-binding site within the hormone-binding domain of the glucocorticoid receptor by peptide competition, J. Biol. Chem. 266 (6) (1991) 3482–3490. [29] S. Kaul, P.J. Murphy, J. Chen, L. Brown, W.B. Pratt, S.S. Simons Jr., Mutations at positions 547–553 of rat glucocorticoid receptors reveal that hsp90 binding requires the presence, but not defined com-

[30]

[31]

[32] [33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

[45]

position, of a seven-amino acid sequence at the amino terminus of the ligand binding domain, J. Biol. Chem. 277 (39) (2002) 36223– 36232. M.J. Meyers, J. Sun, K.E. Carlson, G.A. Marriner, B.S. Katzenellenbogen, J.A. Katzenellenbogen, Estrogen receptor-beta potencyselective ligands: structure–activity relationship studies of diarylpropionitriles and their acetylene and polar analogues, J. Med. Chem. 44 (24) (2001) 4230–4251. S.R. Stauffer, C.J. Coletta, R. Tedesco, G. Nishiguchi, K. Carlson, J. Sun, B.S. Katzenellenbogen, J.A. Katzenellenbogen, Pyrazole ligands: structure–affinity/activity relationships and estrogen receptoralpha-selective agonists, J. Med. Chem. 43 (26) (2000) 4934– 4947. S. Barnes, Soy isoflavones—phytoestrogens and what else? J. Nutr. 134 (5) (2004) 1225S–1228S. E. Farmakalidis, P.A. Murphy, Oestrogenic response of the CD-1 mouse to the soya-bean isoflavones genistein, genistin and daidzin, Food Chem. Toxicol. 22 (3) (1984) 237–239. T.K. Archer, C.J. Fryer, H.L. Lee, E. Zaniewski, T. Liang, J.S. Mymryk, Steroid hormone receptor status defines the MMTV promoter chromatin structure in vivo, J. Steroid Biochem. Mol. Biol. 53 (1–6) (1995) 421–429. C.L. Smith, R.G. Wolford, T.B. O’Neill, G.L. Hager, Characterization of transiently- and constitutively-expressed progesterone receptors: evidence for two functional states, Mol. Endocrinol. 14 (7) (2000) 956–971. H. Richard-Foy, G.L. Hager, Sequence specific positioning of nucleosomes over the steroid-inducible MMTV promoter, EMBO J. 6 (8) (1987) 2321–2328. M. Danielsen, J.P. Northrop, J. Jonklaas, G.M. Ringold, Domains of the glucocorticoid receptor involved in specific and nonspecific deoxyribonucleic acid binding, hormone activation, and transcriptional enhancement, Mol. Endocrinol. 1 (11) (1987) 816–822. P.J. Godowski, D. Picard, K.R. Yamamoto, Signal transduction and transcriptional regulation by glucocorticoid receptor–LexA fusion proteins, Science 241 (1988) 812–816. S.M. Hollenberg, V. Giguere, P. Segui, R.M. Evans, Colocalization of DNA-binding and transcriptional activation functions in the human glucocorticoid receptor, Cell 49 (1987) 39–46. R.K. Bledsoe, V.G. Montana, T.B. Stanley, C.J. Delves, C.J. Apolito, D.D. McKee, T.G. Consler, D.J. Parks, E.L. Stewart, T.M. Willson, M.H. Lambert, J.T. Moore, K.H. Pearce, H.E. Xu, Crystal structure of the glucocorticoid receptor ligand binding domain reveals a novel mode of receptor dimerization and coactivator recognition, Cell 110 (1) (2002) 93–105. B. Kauppi, C. Jakob, M. Farnegardh, J. Yang, H. Ahola, M. Alarcon, K. Calles, O. Engstrom, J. Harlan, S. Muchmore, A.K. Ramqvist, S. Thorell, L. Ohman, J. Greer, J.A. Gustafsson, J. Carlstedt-Duke, M. Carlquist, The three-dimensional structures of antagonistic and agonistic forms of the glucocorticoid receptor ligand-binding domain: RU-486 induces a transconformation that leads to active antagonism, J. Biol. Chem. 278 (25) (2003) 22748–22754. A.M. Brzozowski, A.C. Pike, Z. Dauter, R.E. Hubbard, T. Bonn, O. Engstrom, L. Ohman, G.L. Greene, J.A. Gustafsson, M. Carlquist, Molecular basis of agonism and antagonism in the oestrogen receptor, Nature 389 (6652) (1997) 753–758. D.M. Tanenbaum, Y. Wang, S.P. Williams, P.B. Sigler, Crystallographic comparison of the estrogen and progesterone receptor’s ligand binding domains, Proc. Natl. Acad. Sci. U.S.A. 95 (11) (1998) 5998–6003. G.V. Rayasam, C. Elbi, D.A. Walker, R.G. Wolford, T.M. Fletcher, D.P. Edwards, G.L. Hager, Ligand specific dynamics of the progesterone receptor in living cells and during chromatin remodeling in vitro, Mol. Cell. Biol. 25 (2005) 2406–2418. J.G. McNally, W.G. Mueller, D. Walker, R.G. Wolford, G.L. Hager, The glucocorticoid receptor: rapid exchange with regulatory sites in living cells, Science 287 (5456) (2000) 1262–1265.

E.D. Martinez et al. / Journal of Steroid Biochemistry & Molecular Biology 97 (2005) 307–321 [46] P. Maruvada, C.T. Baumann, G.L. Hager, P.M. Yen, Dynamic shuttling and intranuclear mobility of nuclear hormone receptors, J. Biol. Chem. 278 (14) (2003) 12425–12432. [47] D.L. Stenoien, K. Patel, M.G. Mancini, M. Dutertre, C.L. Smith, B.W. O’Malley, M.A. Mancini, FRAP reveals that mobility of oestrogen receptor-alpha is ligand- and proteasome-dependent, Nat. Cell Biol. 3 (1) (2001) 15–23. [48] Y. Wan, K.K. Coxe, V.G. Thackray, P.R. Housley, S.K. Nordeen, Separable features of the ligand-binding domain determine the differential subcellular localization and ligand-binding specificity of glucocorticoid receptor and progesterone receptor, Mol. Endocrinol. 15 (1) (2001) 17–31. [49] M. Gangloff, M. Ruff, S. Eiler, S. Duclaud, J.M. Wurtz, D. Moras, Crystal structure of a mutant hERalpha ligand-binding domain reveals key structural features for the mechanism of partial agonism, J. Biol. Chem. 276 (18) (2001) 15059–15065. [50] M.T. Bonovich, H.J. List, S. Zhang, M. Danielsen, A.T. Riegel, Identification of glucocorticoid receptor domains necessary for transcriptional activation of the mouse mammary tumor virus promoter integrated in the genome, Exp. Cell Res. 239 (2) (1998) 454–462. [51] S. Balasenthil, C.J. Barnes, S.K. Rayala, R. Kumar, Estrogen receptor activation at serine 305 is sufficient to upregulate cyclin D1 in breast cancer cells, FEBS Lett. 567 (2/3) (2004) 243–247.

321

[52] A.M. Fowler, N. Solodin, M.T. Preisler-Mashek, P. Zhang, A.V. Lee, E.T. Alarid, Increases in estrogen receptor-alpha concentration in breast cancer cells promote serine 118/104/106-independent AF-1 transactivation and growth in the absence of estrogen, FASEB J. 18 (1) (2004) 81–93. [53] N. Atanaskova, V.G. Keshamouni, J.S. Krueger, J.A. Schwartz, F. Miller, K.B. Reddy, MAP kinase/estrogen receptor cross-talk enhances estrogen-mediated signaling and tumor growth but does not confer tamoxifen resistance, Oncogene 21 (25) (2002) 4000–4008. [54] M.J. Font de, M. Brown, AIB1 is a conduit for kinase-mediated growth factor signaling to the estrogen receptor, Mol. Cell. Biol. 20 (14) (2000) 5041–5047. [55] S. Folkertsma, P. van Noort, J. Van Durme, H.J. Joosten, E. Bettler, W. Fleuren, L. Oliveira, F. Horn, J. de Vlieg, G. Vriend, A family-based approach reveals the function of residues in the nuclear receptor ligand-binding domain, J. Mol. Biol. 341 (2) (2004) 321–335. [56] J.J. Van Durme, E. Bettler, S. Folkertsma, F. Horn, G. Vriend, NRMD: nuclear receptor mutation database, Nucl. Acids Res. 31 (1) (2003) 331–333. [57] P.J. Fuller, B.J. Smith, F.M. Rogerson, Cortisol resistance in the New World revisited, Trends Endocrinol. Metab. 15 (7) (2004) 296– 299.