Cancer Research Aziza Shad Sign up for Cancer Research eTOC's
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online
Cancer Research 67, 9549, October 1, 2007. doi: 10.1158/0008-5472.CAN-07-1590
© 2007 American Association for Cancer Research
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wittmann, B. M.
Right arrow Articles by McDonnell, D. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wittmann, B. M.
Right arrow Articles by McDonnell, D. P.

Endocrinology

Definition of Functionally Important Mechanistic Differences among Selective Estrogen Receptor Down-regulators

Bryan M. Wittmann, Andrea Sherk and Donald P. McDonnell

Department of Pharmacology and Cancer Biology, Duke University Medical Center, Durham, North Carolina

Requests for reprints: Donald P. McDonnell, Department of Pharmacology and Cancer Biology, Duke University Medical Center, Box 3813, Durham, NC 27710. Phone: 919-684-6035; Fax: 919-681-7139; E-mail: donald.mcdonnell{at}duke.edu.


   Abstract
Top
 Abstract
Introduction
 Materials and Methods
 Results
 Discussion
 References
 
One subclass of antiestrogens, the selective estrogen receptor down-regulators (SERDs), have received considerable attention of late as they competitively inhibit estrogen binding and induce a rapid, proteasome-dependent degradation of the receptor. Contained within this class of molecules is the steroidal antiestrogen ICI182,780 (faslodex), recently approved for the treatment of metastatic cancer, and GW5638/DPC974, a SERD that is currently being evaluated in the clinic. Given that mechanistic differences between different selective estrogen receptor modulators have been translated into important clinical profiles, it was of interest to determine if the SERD subclass of ligands were likewise functionally or mechanistically distinguishable. In this study, we show that although the steroidal and nonsteroidal SERDs target ER{alpha} for degradation, the underlying mechanism(s) are different. Of note was the identification of a specific protein-protein interaction surface presented on ER{alpha} in the presence of the ICI182,780-activated receptor which is required for degradation. Interestingly, this surface is also presented on ER{alpha} in the presence of RU58,668, a SERD that is chemically distinct from ICI182,780. This surface is not required for GW5638-mediated degradation, and thus, this SERD seems to affect ER{alpha} down-regulation by a different mechanism. These data suggest that sequencing of therapies using drugs of this class is likely to be possible. Finally, because of the unmet need for orally active SERDS that function similarly to ICI182,780, we have used the insights from these mechanistic studies to develop and validate a high-throughput screen for compounds of this class with improved pharmaceutical properties. [Cancer Res 2007;67(19):9549–60]


   Introduction
Top
 Abstract
 Introduction
Materials and Methods
 Results
 Discussion
 References
 
The biological actions of estrogens are manifested through two high-affinity estrogen receptors, ER{alpha} and ERß, which are expressed at different levels in target cells (1–3). In the absence of their cognate hormones, these receptors reside in the cytoplasm or nuclei of target cells associated with an inhibitory heat-shock protein complex (4). After binding estrogen, they undergo activating conformational changes that lead to the displacement of heat-shock proteins, an event that facilitates the formation of receptor homodimers or heterodimers and their subsequent interaction with specific DNA response elements within the regulatory regions of target genes. The DNA-bound receptor can then either positively or negatively regulate target gene transcription. Alternate nongenomic pathways by which estrogen-activated ER{alpha} or ERß can activate signaling processes at the membrane or in the cytoplasm have been identified, although the physiologic significance of these responses remains controversial (5).

Although generally considered a reproductive hormone, the expression of these ERs in nonreproductive tissues indicates that the regulatory actions of estrogens extend beyond reproduction (6, 7). Not surprisingly, therefore, pharmaceuticals that both positively and negatively regulate ER action are widely used in the clinic. Outside the realm of normal physiology, however, it has been shown that the majority of breast tumors express ER{alpha} and are responsive to the mitogenic actions of estrogen(s) (ref. 8). Furthermore, it has been shown that signaling pathways, initiated at the cell membrane, can impinge on ER{alpha} positively, regulating its activity in the presence or absence of exogenous estrogens (9). A protective role of ERß in breast cancer cells has been inferred from studies done in vitro, although its role(s) in tumor pathology has not been determined (10–12). Regardless, the clinical benefit of disrupting estrogen signaling in the breast using antiestrogens, compounds that competitively inhibit estrogen binding to ER{alpha}, or aromatase inhibitors, compounds that block estrogen production, are well established (13). Although it has recently become clear that aromatase inhibitors are superior to antiestrogens as first-line hormonal interventions in metastatic breast cancer, the chronic nature of the disease makes it likely that both types of drugs will play a role in the management of this disease (13–15). Not surprisingly, therefore, a considerable amount of attention is now being paid to the development of different sequencing regimens that may provide increased therapeutic benefit. Therefore, interest in identifying new antiestrogens with unique mechanisms of action has increased considerably over the past few years.

The currently available antiestrogens fall into two general classes (a) selective estrogen receptor modulators (SERMs), such as tamoxifen, and (b) selective estrogen receptor down-regulators (SERDs), such as the pure antiestrogen faslodex (ICI182,780; ref. 13). Tamoxifen functions by competitively displacing estradiol from the receptor, disrupting the primary coactivator-binding pocket on the receptor and blocking the access of ER, activated by residual estrogens, to its target gene promoters. The SERDs share these attributes with tamoxifen but, in addition, they induce a rapid and quantitative down-regulation of ER{alpha} (14, 16). It is likely that it is the latter activity which is responsible for the dramatic results observed in patients treated with faslodex who have failed at least one endocrine therapy and which has made it a first-in-class SERD approved for the treatment of metastatic breast cancer (16–19). Although clinical studies have shown some success with faslodex, its poor pharmacodynamic properties and lack of oral bioavailability have limited its clinical utility (16–20). Thus, although faslodex has provided very important proof-of-concept information, it is clear that there is an unmet medical need for (a) faslodex-like compounds with improved pharmaceutical properties and (b) other SERDs/antiestrogens with distinct mechanisms of action. The rationale for the continued development of novel classes of ER antagonists has been detailed recently in several comprehensive reviews (21, 22). For this reason, we have undertaken a comparative study of the mechanism of action of faslodex and that of GW5638/DPC974, a chemically distinct SERD with a view to developing mechanism-based screens that will allow the identification of new antiestrogens (23–25).


   Materials and Methods
Top
 Abstract
 Introduction
 Materials and Methods
Results
 Discussion
 References
 
Plasmids, chemicals, and antibodies. Plasmids were previously described (26–28) or constructed through site-directed mutagenesis (QuickChange kit; Stratagene), PCR amplification, restriction enzyme subcloning, and/or "Gateway" LR reactions (Invitrogen). More detailed instructions on plasmid construction are available on request. 17ß-Estradiol (E2), 4-hydroxy-tamoxifen (TOT), dexamethasone, and 9-cis-retinoic acid were purchased from Sigma. Raloxifene, EM-652, GW7604, RU58-668, and RU486 were provided as gifts from Dr. E. Larson (Pfizer, Inc., Groton, CT), Dr. Fernand Labrie (CHUL Research Center, Quebec City, Quebec, Canada), Dr. T. Willson (GlaxoSmithKline, Research Triangle Park, NC), Michel Renoir (UMR Centre National de la Recherche Scientifique, France), and Ligand Pharmaceuticals, Inc. (San Diego, CA), respectively. ICI182,780 was obtained from Tocris. Casodex was provided by GlaxoSmithKline. R1881 and R5020 were purchased from NEN Life Science Products. Antibodies used included ER{alpha}, lamin A, {alpha}-tubulin, cytokeratin 18, and GAL4-DBD antibody (Santa Cruz Biotechnology), V5 (Invitrogen), and anti–enhanced green fluorescent protein (EGFP) polyclonal antibody (Clontech).

ER{alpha} protein and RNA levels. MCF7 cells were maintained in DMEM/F12 (Invitrogen) supplemented with 10% fetal bovine serum (HyClone), 0.1 mmol/L of nonessential amino acids, and 1 mmol/L of sodium pyruvate (Invitrogen). MCF7 cells were plated in six-well dishes at 500,000 cells per well containing phenol red–free DMEM/F12 supplemented with 5% charcoal-stripped fetal bovine serum for 36 h. Cells were subsequently treated with either vehicle or 100 nmol/L of ER{alpha} ligand for the times indicated in Fig. 1 . Whole cell lysates were harvested by boiling the cells in Laemmli buffer and were subjected to SDS-PAGE and Western analysis. Following ligand treatments, ER{alpha} mRNA levels were detected by quantitative PCR using primers specific to ER{alpha} and normalized to 36B4 mRNA levels.


Figure 1
View larger version (20K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 1. Estradiol, GW7604 and ICI182,780 induce a rapid and sustained decrease in ER{alpha} protein. A, compound structures of ER{alpha} ligands (21). MCF7 cells were grown as described in Materials and Methods and treated with either vehicle, or 100 nmol/L of 17ß-estradiol (E2), 4-hydroxy-tamoxifen (TOT), GW7604 (GW), raloxifene (Ral), EM-652 (EM), or ICI182,780 (ICI) for either 4 h (B) or 48 h (C). Whole cell extracts were obtained by boiling cells in Laemmli buffer and were subjected to Western analysis to detect ER{alpha} levels. Equal loading was determined by detection of {alpha}-tubulin and cytokeratin 18. The percentage of remaining ER{alpha} was determined by densitometry readings using Image J and comparing ER{alpha} density to cytokeratin 18 density from two independent experiments for each treatment. ER{alpha} RNA levels following 4 h (B) or 48 h (C) of treatment were detected by quantitative PCR and normalized to 36B4 mRNA levels. Columns, means from one of two independent experiments for each treatment.

 
Cell fractionation and analysis. MCF7 cells were plated as previously described and subsequently treated with either vehicle, 100 nmol/L of E2, ICI182,780, or GW7604 for the times specified in Fig. 2 . Cells were harvested and cell pellets were resuspended on ice for 15 min in 100 µL of cytoplasmic extraction buffer [10 mmol/L HEPES (pH 7.9), 10 mmol/L KCl, 0.1 mmol/L EDTA, 0.1 mmol/L EGTA] containing protease inhibitor cocktail set III (Calbiochem). Cells were permeabilized by adding IGEPAL (Sigma) to 0.63% total volume, vortexed for 10 s, and centrifuged for 1 min to obtain the cytoplasmic extract. To obtain soluble nuclear extracts, nuclear pellets were resuspended on ice for 20 min in 100 µL of nuclear extract buffer (20 mmol/L HEPES, 25% total volume glycerol, 0.4 mol/L NaCl, 1 mmol/L EDTA, and 1 mmol/L EGTA) with protease inhibitors, and centrifuged for 5 min. The remaining pellets were sequentially extracted in nuclear matrix digestion buffer containing protease inhibitors and DNase I (Roche), 0.25 mol/L of ammonium sulfate, and finally, 2.0 mol/L of NaCl as previously described (29), and combined to a total volume of 100 µL to compose the detergent-extractable nuclear matrix fraction. The remaining insoluble fraction was solubilized in 100 µL of 95% Laemmli buffer (Bio-Rad) with 5% ß-mercaptoethanol and boiled at 95°C for 20 min. The protein concentration of the cytoplasmic extracts was determined and the volume constituting 100 µg of cytoplasmic extract was analyzed by SDS-PAGE. In addition, equivalent volumes of the nuclear extract, detergent-extractable nuclear matrix, and insoluble fractions were loaded. Proteins were transferred to nitrocellulose membrane and subjected to Western analysis.


Figure 2
View larger version (28K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 2. Subcellular compartmentalization of ER{alpha} upon binding ICI182,780 is distinct from that observed in the presence of GW7604 and E2. MCF7 cells were grown as described in Materials and Methods and treated with either vehicle, 100 nmol/L of E2, 100 nmol/L of ICI182,780, or 100 nmol/L of GW7604 for the time periods indicated. Following treatment, the cells were harvested and lysed to obtain separate soluble cytoplasmic fractions (C), soluble nuclear fractions (N), soluble DNase/0.25 mol/L ammonium sulfate/2 mol/L NaCl fractions (D), and remaining insoluble fractions (I) as described in Materials and Methods. Fractions were subjected to Western analysis to detect ER{alpha}, {alpha}-tubulin (C fraction control), lamin A (N, D, I fraction controls), and cytokeratin 18 (loading and C and I fraction controls).

 
T7 phage display. mRNA was obtained from HepG2, MCF7, and HEC1B cells using the Straight A's mRNA kit (Novagen). T7 phage libraries were constructed using OrientExpress cDNA kit and T7 Select kit according to the instructions of the manufacturer (Novagen). Costar 96-well plates were coated with 20 µg of neutravidin in 100 µL of 100 mmol/L NaHCO3 overnight at 4°C. Plates were blocked with 150 µL of 2% milk in 100 mmol/L of NaHCO3 for 1 h at room temperature, then washed five times with PBST (PBS + 0.1% Tween 20). Two picomoles of biotinylated double-stranded DNA composed of the vitellogenin estrogen receptor response element in PBST were applied to each well for 1 h at room temperature and excess neutravidin was subsequently blocked for another hour with 10 µL of 1 mmol/L biotin. Wells were washed five times with PBST and ~0.25 µg of baculovirus-expressed recombinant ER{alpha} (Affinity Bioreagents) in 100 µL of PBST containing vehicle or 1 µmol/L of ER{alpha} ligand was added to each well and incubated overnight at 4°C. After washing, wells were blocked with 150 µL of 2% milk in PBS for 1 h at room temperature, washed, and subsequently incubated with 1 x 108 plaque-forming units of T7 library in 100 µL of PBST containing vehicle or 1 µmol/L of ER{alpha} ligand for 1 h at room temperature. Following five washes with PBST, bound phages were eluted in a DNase elution mix (99 µL PBST, 0.5 µL of 1 mol/L MgCl2, and 0.5 µL of 10 mg/mL DNase I) for 1 h at room temperature. Eluted phages were amplified in a 10 mL culture of Escherichia coli BLT5615 cells induced with 1 mmol/L of isopropyl-ß-D-thiogalactopyranoside until complete bacterial lysis occurred. Amplified phages were harvested following centrifugation and applied to a new plate containing ER{alpha}. After five rounds of panning, phage ELISAs were done to verify affinity enrichment of ER{alpha}-interacting phages. Specifically, amplified phages (100 µL/well) were added to wells containing ER{alpha} for 1 h at room temperature. Unbound phages were washed away and wells were incubated with 150 µL of a 1:5,000 dilution of anti–T7-HRP (Novagen) in PBST for 1 h at room temperature. After thoroughly washing the wells, bound phages were detected by incubating the wells with 100 µL of 2'-2'-azino-bis-ethylbenzthiazoline-6-sulfonic acid in the presence of 0.05% H2O2 and the color change was measured at 405 nmol/L with a plate reader (Multiskan MS; Labsystems). T7 phages from the rounds with the highest enrichment of ER{alpha}-interacting phages were plaque-purified and clones were randomly selected for PCR amplification of the cDNA inserts using T7 select UP and DOWN primer (Novagen). The inserts were sequenced, translated, and corresponding proteins were identified using the BLASTp program (National Center for Biotechnology Information).

Mammalian two-hybrid assays. HepG2 cells were maintained, plated, and transfected and mammalian two-hybrid assays were done as previously described (27).

ER{alpha} mutant degradation analysis. HeLa cells were maintained as previously described (30). For transient transfection and Western analysis of ER{alpha}, ER{alpha} mutants, and ERß, HeLa cells were plated in six-well dishes containing phenol red–free media supplemented with 8% charcoal-stripped fetal bovine serum at 500,000 cells per well for 24 h. Cells were subsequently transfected with 1.5 µg of pEGFP, 0.3 µg of CMV-ß-gal, and 1.5 µg of pCDNA 3.1-V5-ER{alpha}, -ER{alpha} mutant, or ERß using Fugene (Roche) according to the instructions of the manufacturer for 24 h, and subsequently treated with vehicle or 100 nmol/L of ICI182,780 for 3 h. HeLa cells were harvested and cell pellets were resuspended and mechanically disrupted by pipetting in low salt/low detergent radioimmunoprecipitation assay buffer [100 mmol/L Tris (pH 8.3), 2 mmol/L EDTA, 0.02% SDS, 0.5% IGEPAL, and 150 mmol/L NaCl] containing protease inhibitors, and incubating on ice for 10 min. Soluble fractions were harvested and insoluble pellets were solubilized in an equivalent amount of Laemlli buffer as previously described. Soluble protein lysates were subjected to ß-gal assays, and protein volumes were loaded based on ß-gal expression and subjected to Western analysis as previously described.

Antiestrogen-interacting peptide and control adenovirus construction and expression. GAL4-DBD–antiestrogen interacting peptide (AEIP) and GAL4-DBD-control (AEIP mutated within the ER{alpha}-binding region) adenoviruses were generated using the "Gateway" ViraPower Adenoviral Expression System according to the manufacturer's protocol (Invitrogen) and purified using the Vira Bind Adenovirus Purification Kit according to the manufacturer's instructions (Cell BioLabs, Inc.).

To measure the effects of adenovirally expressed AEIP on ER{alpha} translocation, MCF7 cells were plated at 500,000 cells per well in six-well dishes as previously described for 24 to 48 h. Cells were incubated with either pAD-DEST-GAL4-DBD-AEIP or -control virus at a multiplicity of infection of 150 in 500 µL of media for 2 h at room temperature. Following the 2 h incubation, 2 mL of media were added to the cells and the cells were returned to 37°C. Forty-eight hours later, the cells were treated with vehicle or 100 nmol/L of E2, ICI182,780, or GW7604 for the specified time periods. Cells were harvested and subjected to low salt/low detergent radioimmunoprecipitation assay buffer–mediated cell lysis. Soluble and insoluble extracts were subjected to Western analysis as previously described.


   Results
Top
 Abstract
 Introduction
 Materials and Methods
 Results
Discussion
 References
 
There is a high level of interest in developing SERDs, a subclass of antiestrogens that both disrupt ER{alpha}-cofactor interactions and induce receptor turnover in cells. This interest is driven primarily by the observation that a compound of this pharmacologic class, ICI182,780 (faslodex), has shown remarkable efficacy as a second-line treatment for tamoxifen-resistant breast cancers (17). Little is known, however, about the mechanism(s) by which ER{alpha} degradation is accomplished upon interacting with SERDs, an impediment to the development of molecules with improved therapeutic properties. Thus, the primary objective of this study was to define the characteristics of compounds that enable them to induce ER{alpha} turnover and determine if chemically distinct SERDs function in a similar manner. As a first step in addressing this issue, we examined the stability of endogenous ER{alpha} in MCF-7 cells following treatment with either estradiol or selected SERMs and SERDs (Fig. 1A). Specifically, total ER{alpha} protein and mRNA expression levels were examined in cells following acute (4 h) or chronic (48 h) treatments with each compound (Fig. 1B and C). Western immunoblots and quantitative densitometry were used to measure ER{alpha} protein levels. Not surprisingly, no ER{alpha} protein was recovered from cells following 4 h of treatment with estradiol (Fig. 1B). Unexpectedly, with the exception of TOT, all of the other ligands tested reduced ER{alpha} levels in cells albeit with different efficacies and these changes could not be attributed to changes in ER{alpha} mRNA levels. Most notably, ICI182,780 and the higher affinity 4-hydroxylated metabolite of GW5638 (GW7604) were as effective as estradiol in this acute experiment. No additional down-regulation of ER{alpha} protein was observed following extended (48 h chronic) treatment with either raloxifene or EM652 (Fig. 1C). However, a significant increase in ER{alpha} levels was observed when TOT-treated cells were examined at this time point. The reappearance of ER{alpha} in the cells treated for 48 h suggests that, despite the presence of saturating amounts of estradiol, constitutive suppression of ER{alpha} levels by this ligand was not achievable. However, both ICI182,780 and GW7604 distinguished themselves from the other ER ligands examined in this assay as they permitted a robust and sustained suppression of ER{alpha} protein levels and this rapid loss of receptor could not be solely attributed to decreased ER{alpha} mRNA levels (Fig. 1C). Given these findings, we proceeded to determine whether or not these two compounds achieved the end point of receptor turnover in the same manner.

The SERDs ICI182,780 and GW7604 have differential effects on subcellular localization of ER{alpha}. It has been shown previously that although treatment with either estradiol or ICI182,780 results in a quantitative down-regulation of ER{alpha} in cells, the mechanism(s) by which this occurs is not the same (30–32). Most notable is the observation that these compounds have discrete effects on receptor compartmentalization, which precedes the degradation of the receptor (33–36). This suggests that different ligand-induced changes in the structure of ER{alpha} result in the receptor entering the proteolytic pathways through alternate mechanisms. We were interested in determining if the distinct cellular compartmentalization of ER{alpha}, observed following treatment with ICI182,780, was a hallmark of SERDs or if it was specific to this steroidal antagonist. To this end, we compared the kinetics of receptor distribution in MCF-7 cells following treatment with vehicle alone, estradiol, or with the SERDs ICI182,780 or GW7604. The treated cells were harvested and cellular extracts were fractioned into soluble cytoplasmic, soluble nuclear, soluble DNase/ammonium sulfate/high NaCl treated, and the remaining insoluble fraction (Fig. 2). In this manner, it was observed that ER{alpha} was found in either the cytoplasmic or insoluble fractions following treatment with vehicle (Fig. 2A–C). Treatment of cells with estradiol resulted in a higher recovery of ER{alpha} in the soluble nuclear and DNase/ammonium sulfate/high NaCl fractions at earlier time points with progressively more receptor associating with the insoluble fraction as the duration of exposure increased (Fig. 2A). In contrast, the ICI182,780-occupied ER{alpha} was found to associate predominantly with the insoluble fraction (Fig. 2B). Using the same approach, we observed that the localization pattern for receptors treated with GW7604 was quite similar to that observed for estradiol and surprisingly distinct from the ICI182,780-treated receptor (Fig. 2C). Lamin A (nuclear) and {alpha}-tubulin (cytoplasmic) were measured in all extracts to assess the effectiveness of the fractionation protocol and to verify that these ligands did not have a nonspecific effect on protein compartmentalization. Thus, although ICI182,780 and GW7604 are classified as SERDs, they do not have the same effect on the compartmentalization of ER{alpha} within the cell. Although it is not possible to infer from this data that ICI182,780 and GW7604-dependent ER{alpha} turnover is related in any way to cellular localization, it does indicate that the cell does not handle the receptor in the same manner when occupied by these different SERDs. We conclude from these studies that the rapid repartitioning of ICI182,780-treated ER{alpha} from a soluble to an insoluble fraction precedes degradation and is a unique characteristic of this subclass of SERDs.

ER{alpha} adopts a unique structural conformation in the presence of ICI182,780. It is now generally accepted that many of the functional differences exhibited by different ER ligands are related to (a) their ability to induce distinct alterations in receptor structure and (b) the ability of differently conformed receptors to engage different cofactors in target cells. We inferred, therefore, that if the SERDs ICI182,780 and GW7604 are indeed mechanistically distinct, as the data presented in Fig. 2 indicates, then the structure of the receptor in the presence of these molecules should be sufficiently different so as to enable the differential engagement of cofactors. This idea is supported by our previous observation that the degree of hydrophobicity exhibited by the ER{alpha} ligand-binding domain is not equivalent in the ICI182,780 and GW7604 complexes (23). Thus, we next examined whether or not these two mechanistically distinct SERDs induce different structural alterations in ER{alpha}. To address this issue, we used combinatorial T7 phage display to screen random peptide and cDNA expression libraries for peptides/proteins that interact in a specific manner with ICI182,780-activated ER{alpha}. This was accomplished using purified ER{alpha} immobilized in plastic to select for phages expressing either an interacting peptide or cofactor. Interestingly, although this approach led to the identification of a large number of peptides and proteins that interacted with estradiol-activated ER{alpha} (data not shown), we identified only a single peptide (AEIP) that interacted specifically with the ICI182,780/ER{alpha} complex. Further characterization of the receptor-interacting properties of this peptide was done using a mammalian two-hybrid assay (Fig. 3 ). For this assay, we expressed AEIP as a fusion protein with the GAL4-DBD and assessed its ability to interact with a full-length VP16-ER{alpha} fusion protein in the presence of different ligands. The results of this analysis indicate that this peptide recognizes a surface presented on the ICI182,780/ER{alpha} complex that is not present on the receptor when it is occupied by GW7604 (a SERD) or other SERMs (tamoxifen, raloxifene, and EM-652; Fig. 3A). Furthermore, we showed that the interaction of AEIP with ER{alpha} could be reversed by the addition of increasing concentrations of estradiol, confirming that ICI182,780 and estradiol-activated ER{alpha} are structurally distinct (Fig. 3B). Finally, GAL4-DBD-AEIP did not interact with any of the other nuclear receptors that were tested, including both the long (amino acids 1–530) and short (amino acids 1–477) VP16-ERß isoforms (Fig. 3C). This was an extremely important result, as we and others have shown that ERß expression is not influenced by ICI182,780 (37). These data provide compelling evidence that potentially, a specific surface on ER{alpha}, presented upon binding ICI182,780, results in distinct receptor re-compartmentalization and subsequent degradation. Thus, within the confines of the resolution of this assay, we conclude that the structure of ICI182,780-activated ER{alpha} is distinct from the nonsteroidal antiestrogens evaluated in this assay.


Figure 3
View larger version (23K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 3. Identification of a peptide that interacts specifically with the ICI182,780-bound ER{alpha}. A, mammalian two-hybrid assays were done in HepG2 cells as described in Materials and Methods by transfecting a 5XGal4-TATA-luc reporter, a VP16-ER{alpha} expression plasmid, and either GAL4-DBD alone or GAL4-DBD-AEIP expression plasmid. Cells were treated with vehicle or 100 nmol/L of the following ligands: E2, TOT, GW7604, raloxifene, EM-652, or ICI182,780, for 24 to 36 h. Transcriptional activation of 5XGal4-TATA-luc, and therefore ER{alpha}/AEIP interaction, was measured with a luciferase assay and activity was normalized for transfection efficiency with transfected CMV-ß-gal plasmid. Columns, average of at least three triplicate experiments for all ligands except raloxifene (two triplicates); bars, SE. B, mammalian two-hybrid assays were done as previously described and cells were treated with vehicle, ICI182,780 alone, or with ICI182,780 in combination with E2 at various concentrations from 0.1 to 100 nmol/L. ER{alpha}/AEIP interaction was measured as in (A). Columns, averages of two triplicate experiments; bars, SEs. C, mammalian two-hybrid assays were done as described using GAL4-DBD-AEIP and VP16-ER{alpha}, -ERß long form (L), -ERß short form (S), -androgen receptor (AR), -glucocorticoid receptor (GR), -progesterone receptor A or B form (PRA or PRB), or retinoic acid receptor (RAR). Cells were treated with either vehicle (V), 100 nmol/L of ICI182,780 (ICI), E2, casodex (Cas), R1881, RU486 (486), dexamethasone (Dex), RU5020 (5020), or 9-cis retinoic acid (9-cis) depending on the transfected receptor. GAL4-DBD-AEIP interaction with various receptors was determined by luciferase expression as in (A), and compared with GAL4-DBD interaction alone. Columns, average of ER{alpha}, ERß L, AR, and PRA interaction with AEIP from two triplicate experiments and the average of ERß S, GR, PRB, and RAR interaction with AEIP from one triplicate experiment; bars, SE. D, diagram of AEIP amino acid sequence and amino acid sequences of mutated peptides. Wt AEIP, amino acids of AEIP beyond the GAL4-DBD; NAAIRS mutants, the seven regions (1, 2, 3, 4, 5, 6, 7) of the AEIP peptide that were individually mutated to the NAAIRS amino acid sequence in the context of the wild-type peptide. Alanine mutants, the amino acids that were mutated to alanine either individually (1, 2, 3, 4, 5, 6) or together (1 & 2). The specific amino acids in AEIP necessary for the pure antiestrogen-induced AEIP/ER{alpha} interaction (underlined). Mammalian two-hybrid assays were done in HepG2 cells as previously described in Materials and Methods to determine the interaction of VP16-ER{alpha} with various AEIP mutants. Bottom left, AEIP-NAAIRS mutants in the presence of vehicle or 100 nmol/L of ICI182,780. Columns, the average interactions from three triplicate experiments; bars, SEs. Bottom right, AEIP alanine mutants in the presence of vehicle or 100 nmol/L of ICI182,780. Columns, the average interactions from three triplicate experiments; bars, SEs.

 
Sequence analysis of AEIP revealed that it was composed of a total of 42 amino acids (Fig. 3D). In order to define the ER{alpha}-interacting domain of this extended peptide, we sequentially mutated six amino acid regions across the entire sequence using a NAAIRS replacement strategy (38). A schematic illustrating the regions mutated is shown in Fig. 3D. When the AEIP NAAIRS mutants were used in M2H assays, it was observed that the amino acids within region 5 were responsible for the interaction with ER{alpha} (Fig. 3D). This region contained a high number of hydrophobic amino acids, which is interesting given that E2-bound ER{alpha} has a high affinity for the hydrophobic amino acid sequence LxxLL found in coactivator proteins (26, 39, 40). To further determine which of the six amino acids were required for the interaction, each amino acid was individually mutated to alanine (A) or in combination at positions 1 and 2 because we thought it was possible these hydrophobic amino acids might compensate for one another. M2H assays revealed that the last three amino acids (S28, P29, and M30) were key amino acids involved in the interaction (Fig. 3D). Thus, AEIP contains a discrete hydrophobic domain that is required for interaction with ICI182,780-activated ER{alpha}.

Conformation as a predictor of ER{alpha} stability. Over the past few years, it has become apparent that there is a strong relationship between the structure of an ER-ligand complex and resultant biological activity (26, 27, 39–41). Thus, we felt that it was likely that any compound that facilitates the interaction of AEIP with ER{alpha} would induce its turnover. To test this hypothesis, we used a mammalian two-hybrid assay as a primary screen to identify from a library of ER ligands those that could facilitate the interaction of ER{alpha} with AEIP. Only a single compound, the pure antiestrogen RU58,668, facilitated a significant interaction of ER{alpha} with AEIP (Supplemental Fig. S1A; ref. 42). As expected, RU58,668 induced a rapid and sustained decrease in ER{alpha} protein levels that could not be attributed to a decrease in ER{alpha} mRNA levels alone (Supplemental Fig. S1B and C). These findings provide a strong link between receptor conformation and stability and provide the validation for a mechanism-based screen with which to identify SERDs with improved pharmaceutical properties.

Definition of the functional significance of the AEIP binding surface on ER{alpha}. It is our assumption that the surfaces presented on ER{alpha} in the presence of ICI182,780 allow its interaction with specific proteins that target it for degradation. As of yet, we have been unable to identify proteins that are capable of interacting with this specific receptor-ligand complex, and thus, are unable to test this hypothesis directly. Furthermore, because the crystal structure of ER{alpha} in the presence of ICI182,780 has not yet been solved, we have had to rely on mutagenesis to determine what region and amino acids of ER{alpha} constitute the AEIP interaction surface(s). This is an important next step in understanding the mechanism of action of this class of SERDs as it will hopefully lead to the definition of the mechanistic link between ER{alpha} conformation and stability. The goal in the next series of studies, therefore, was to characterize the AEIP interaction domain on ER{alpha} and determine if this region is directly involved in the regulation of receptor stability or if its presentation is merely a surrogate for more functionally important structural alterations that occur at another surface on the protein.

In the first series of studies, we compared the effect of a series of receptor mutations, created in the background of VP16-ER{alpha}, on the receptor-binding activities of AEIP (Fig. 4A ). These specific mutations were selected as they have previously been shown to alter the interaction of ER{alpha} with coactivators and corepressors (26, 27, 39, 43). For comparative purposes, we also examined the interaction of these mutants with (a) the NR-box of SRC-1 and (b) a peptide that interacts with tamoxifen-activated ER{alpha} (Fig. 4B and C; ref. 27). As observed in Fig. 4A, the interaction of AEIP with ICI182,780-bound ER{alpha} was abrogated when the amino acid I358 within helix 3, L379 in helix 5, and the hydrophobic amino acids of helix 12 were mutated. Interestingly, with the exception of amino acid K362, which mediates a capping interaction with LxxLL-containing coactivators (i.e., SRC1), the mutations shown to abrogate the interaction of AEIP with ICI182,780-bound ER{alpha} were the same as those necessary for E2-induced interaction of the SRC1 NRID (peptide region of SRC1 containing all three LxxLL; Fig. 4A and B; ref. 39). These mutant receptors were capable of interacting with a peptide that binds to ER{alpha} in a ligand-independent manner, indicating that the selected amino acid changes did not induce a global alteration in receptor structure (data not shown). We have previously identified a peptide (bT1) whose binding surface is presented only on tamoxifen-activated ER{alpha} (27). Analysis of the effects of the selected mutants on the interaction of this peptide with ER{alpha} has led us to conclude that the surfaces presented on the ICI182,780-activated ER{alpha} are dramatically different from those presented upon tamoxifen binding (Fig. 4C). Finally, Kong et al. have used crystallography to identify a novel protein-protein interaction domain that is presented on ER{alpha} following its interaction with any ligand (43). The functionality of this domain is disrupted by the G442H mutation. However, this mutation has no effect on the interaction of the receptor with AEIP. Thus, we conclude from these mutagenesis experiments that the conformation of ER{alpha} adopted in the presence of estradiol and ICI182,780 are unexpectedly quite similar. Furthermore, we consider it likely that the AEIP and coactivator binding surfaces overlap.


Figure 4
View larger version (13K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 4. ICI182,780 and E2 induce similar conformational changes within ER{alpha} as assessed by AEIP binding. Previously described mammalian two-hybrid assays were done using VP16-ER{alpha} wild-type (WT), ligand-binding domain (LBD), or specific mutations within the ER{alpha} ligand-binding domain previously shown to affect either coactivator and/or corepressor binding to ER{alpha} within helix 3 (I358R, K362A), helix 5 (L372R, V376R, L379R, E380Q), G442H, and helix 12 [L540Q, 3x (D538N, E542Q, D545N), LL (L539A, L540A), or stop (ER535 helix 12 deletion)]. A, ligand-induced interactions were compared for each ER{alpha} mutant with Gal4-DBD-AEIP in the presence of vehicle or 100 nmol/L of ICI182,780. Columns, average of four triplicate experiments; bars, SEs. B, Gal4-DBD-SRC1-NRID (expression plasmid of Gal4-DBD fused to a portion of the p160 coactivator SRC1 containing the three LxxLL motifs previously shown to be necessary for E2-induced interaction between ER{alpha} and SRC1) in the presence of vehicle or 100 nmol/L of E2. Columns, the average of four independent experiments done in triplicate; bars, SEs. C, Gal4DBD-bT1 (previously identified peptide containing a transcriptional corepressor-like CoRNR box motif that interacts with ER{alpha} in the presence of TOT) in the presence of vehicle or 100 nmol/L of TOT. Columns, the average of three triplicate experiments; bars, SEs.

 
The integrity of the AEIP binding site is required for ICI182,780-mediated subcellular trafficking of ER{alpha}. The earliest response to ICI182,780 we have observed is the repartitioning of ER{alpha} from a soluble to an insoluble fraction; an activity that tracks with degradation (Fig. 2B). Thus, in order to test the functional role of the AEIP binding surface directly, we examined the cellular partitioning of each mutant in the presence of ICI182,780 (Fig. 5 ). To avoid the confounding effect of endogenous ER{alpha} that may dimerize with some of the mutants, we did these studies in transiently transfected ER{alpha} (–) HeLa cells. For these studies, a V5 tag was added to the NH2 terminus of each protein to allow quantitation by Western immunoblot. We first confirmed that we could observe the ICI182,780-induced subcellular compartmentalization of exogenously expressed ER{alpha} in HeLa cells, as was observed with the endogenous receptor in MCF-7 cells (Fig. 5A). In these studies, because ER{alpha} was never observed in the soluble nuclear or DNase fractions following ICI182,780 treatment (Fig. 2B), we only assessed its partitioning between the low salt–soluble and low salt–insoluble fractions. As observed previously, we did not see a decrease in total ER{alpha} levels following 90 min of treatment, allowing us to examine the early compartmentalization events (Fig. 5A; ref. 36). However, we did observe a rapid loss of ER{alpha} in the low salt–soluble fractions and a concomitant increase in insoluble receptor following 15 min of treatment (Fig. 5A). Using this system, we proceeded to evaluate the effect of mutating the AEIP binding site on ER{alpha} compartmentalization. To this end, we expressed ER{alpha}, ERß, or each ER{alpha} mutant individually in HeLa cells and assessed their compartmentalization in the presence of ICI182,780 as a function of time. As can been seen in Fig. 5B, those mutants that interacted with AEIP were all found to rapidly associate with the insoluble fraction. In contrast, those mutants that did not interact with AEIP failed to recompartmentalize under the conditions of our assay. The data presented in Fig. 5C and D represent the average level of soluble and insoluble ER{alpha}, ER{alpha} mutants, and ERß following ICI182,780 treatment from three independent experiments. We believe that the robust nature of this data highlights the importance of the AEIP-interacting domain in regulating ER{alpha} trafficking, and likely, its stability.


Figure 5
View larger version (35K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 5. Specific amino acids responsible for the ICI182,780-bound ER{alpha} interaction with AEIP are also necessary for ICI182,780-mediated subcellular compartmentalization. A, HeLa cells were transfected with CMV-ß-gal, pEGFP, and pCDNA-V5–tagged ER{alpha}, treated with vehicle or 100 nmol/L of ICI182,780 for the specified time period, and lysed to obtain low salt–soluble and -insoluble fractions as previously described in Materials and Methods. The two fractions were subjected to Western analysis to detect V5-tagged ER{alpha}, and EGFP. Loading was initially controlled based on ß-gal expression followed by EGFP expression. B, HeLa cells were transfected with pEGFP, CMV-ß-gal, and either pCDNA-V5-ER{alpha} [wild-type (WT) or mutants] or -V5-ERß expression plasmids, treated for 3 h with either vehicle (V) or 100 nmol/L of ICI182,780 (ICI), harvested and lysed as in (A). The soluble and insoluble fractions were subjected to Western analysis to detect V5-tagged ER{alpha} WT, mutants, or ERß, and EGFP. Loading was controlled as in (A). Western blots are representative of three independent experiments for each receptor. Relative soluble (C) and insoluble (D) ER levels following treatment versus vehicle as determined by densitometry readings using Image J and comparing ER density to EGFP density. Columns, average from three independent experiments for each receptor; bars, SEs.

 
Expression of the AEIP peptide in cells partially inhibits ICI182,780-mediated recompartmentalization of ER{alpha}. The data presented thus far support the hypothesis that the AEIP binding site on ER{alpha} is a protein-protein interaction surface required for ICI182,780-mediated subcellular translocation. Definitive proof of this hypothesis will require the identification of the protein(s) that use this surface; an ongoing project in our laboratory. In the interim, however, we have tested this idea by expressing AEIP in cells and observing the effect of this manipulation on the fate of ICI182,780-occupied ER{alpha} in target cells. Specifically, we constructed adenoviruses expressing either a GAL4-DBD-AEIP fusion protein or a GAL4-DBD-control (AEIP interaction domain mutant). We first confirmed that expressed GAL4-DBD-AEIP, but not the mutant protein, interacted with ER{alpha} using a M2H assay. For this assay, increasing amounts of viruses expressing each protein were used to infect HeLa cells and their ability to interact with a transiently transfected VP16-ER{alpha} was assessed using a GAL4-RE-Luc reporter. Equivalent levels of both proteins were expressed in cells. However, only the AEIP protein was found to interact with ER{alpha} in an ICI182,780-dependent manner (Fig. 6A ).


Figure 6
View larger version (18K):
[in this window]
[in a new window]
[Download PPT slide]
  		Figure 6. Adenovirus-expressed GAL4-DBD-AEIP can partially inhibit ICI182,780-mediated subcellular compartmentalization while having no effect on GW7604-mediated subcellular compartmentalization. A, HepG2 cells were infected at various multiplicities of infection with either GAL4-DBD-AEIP or negative control viruses and subsequently transfected with a 5XGal4-TATA-luc reporter and a VP16-ER{alpha} expression plasmid. Cells were treated with vehicle or 100 nmol/L of ICI182,780 for 36 h. Transcriptional activation of 5XGal4-TATA-luc, and therefore ER{alpha}/AEIP interaction, was determined as previously described. An additional well of cells were infected and transfected as above, lysates harvested, and GAL4-DBD-AEIP or negative control expression was determined by Western analysis using a GAL4-DBD antibody. B, and C, MCF7 cells were infected with GAL4-DBD-AEIP or negative control viruses at a multiplicity of infection of 150. Cells were treated 48 h later with vehicle (V), 100 nmol/L of E2, 100 nmol/L of ICI182,780 (ICI), or 100 nmol/L of GW7604 (GW) for the times stated. Cells were harvested and subjected to Western analysis to detect low salt–soluble (B) or low salt–insoluble (C) ER{alpha}, GAL4-DBD-AEIP or -control, and cytokeratin 18. Using the Image J program, densitometry readings of ER{alpha} and cytokeratin 18 were determined and the remaining ER{alpha} was presented graphically as a percentage of vehicle-treated ER{alpha}. ER{alpha} densities were divided by cytokeratin 18 densities for each treatment. *, P = 0.0068, values from four experiments analyzing the effects of AEIP versus negative control adenovirus on the ICI182,780-mediated decrease in soluble ER{alpha} were statistically different using an unpaired two-tailed t test analysis.

 
We have validated the use of the recompartmentalization assay to study the fate of ICI182,780-activated ER{alpha}, and thus, again used this approach to assess the effect of AEIP expression on the partitioning of endogenous ER{alpha}. To this end, we infected MCF-7 cells with viruses expressing either AEIP or the mutant and measured ER{alpha} levels in the soluble and insoluble fractions of the cell following treatment with different ligands. The data obtained following 1 h of ICI182,780, shown in Fig. 6B and C, indicate that AEIP, but not the mutant, increases the amount of ER{alpha} in the soluble fraction and decreases its recovery from the insoluble fraction. The levels of ER{alpha} in either compartment was unaffected by AEIP expression following treatment with ligands other than ICI182,780. Each assay was done thrice with similar results, confirming the ability of AEIP to block the movement of ICI182,780-treated ER{alpha} to the insoluble fraction. We conclude from these experiments that AEIP recognizes a specific surface on ER{alpha} that is presented upon its interaction with ICI182,780, which enables a protein-protein interaction required for receptor recompartmentalization, and likely, its degradation. We also conclude that although GW7604 is a SERD and resembles ER{alpha} in its ability to down-regulate ER{alpha}, it does so by a different mechanism as it is unaffected by the expression of AEIP.


   Discussion
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
References
 
The utility of the antiestrogen tamoxifen in the treatment and prevention of breast cancer is well established (13). However, as with most first-in-class drugs, the insights provided from years of clinical exposure has revealed activities of this drug that may limit its clinical utility. With respect to tamoxifen, the most notable deficiencies are (a) de novo resistance in some ER{alpha}-positive tumors, (b) acquired resistance with continued administration of the drug, (c) partial agonist activity in the uterus, and (d) increased risk of thromboembolic events (13, 21, 22). Thus, there is a high level of interest in developing novel antiestrogens which will display improved therapeutic activity and which will be used either in place of, or subsequent to, tamoxifen (22). Unfortunately, despite promising results in preclinical models, there has been relatively little success in identifying antiestrogens that either show superior activity to tamoxifen or which can inhibit tamoxifen-resistant breast tumors. For instance, whereas the SERM raloxifene, a molecule that is both chemically and functionally distinct from tamoxifen, exhibited a favorable profile in preclinical studies, it was shown to be ineffective as a treatment for tamoxifen refractory metastatic breast cancer in the clinic (44). Indeed, these and other disappointing clinical trials led to the widely held view that ER{alpha} was not a useful therapeutic target in patients whose tumors progressed on tamoxifen. However, this opinion was changed by the dramatic results observed in a phase II clinical trial of ICI182,780, a SERD (14, 16–20). The efficacy of this particular drug has been attributed to its ability to both inhibit ER transcriptional activity and its ability to induce receptor turnover. Given that ER can be activated in a ligand-independent manner by signaling pathways that modulate the receptor, or its associated cofactors, it is likely that the ability of SERDs to induce receptor turnover is a key contributor to its clinical efficacy (14, 16, 24, 30). Faslodex (ICI182,780) is now approved for use in the treatment of metastatic disease, and although it is clearly a breakthrough medicine, its clinical utility is limited by its poor bioavailability and the length of time that it takes to achieve a useful therapeutic concentration in target tissues (14, 18, 20, 22). It was within the framework of this clinical problem that we embarked several years ago on the identification of new molecules that exhibit SERD activity. This work led to the identification of GW5638, an orally active, nonsteroidal tamoxifen derivative, which, like ICI182,780, facilitates a rapid turnover of ER{alpha} and effectively inhibits the growth of tamoxifen-resistant tumors propagated in athymic nude mice (22, 45). The current study extends this initial work and shows that although both ICI182,780 and GW5638, and its bioavailable metabolite GW7604 are members of the SERD subclass of ER ligands, they effect a down-regulation of the receptor by different mechanisms. This is an important finding as it suggests the possibility that these drugs, or molecules with similar mechanism(s) of action, could be used sequentially and/or they may be useful in patients whose tumors exhibit different molecular characteristics. Furthermore, given the limitations of ICI182,780, we considered that by understanding its mechanism of action, we may be able to develop predictive screens that would enable the identification of molecules with similar activities but show improved pharmaceutical properties.

It is now well established that the pharmacologic activities of different ER ligands are determined by the effects of each ligand on receptor structure and the secondary effect that this has on cofactor recruitment (26, 27, 39–41). We have previously shown, using a variety of biochemical and biophysical techniques, that the structures of ER{alpha} in the presence of ICI182,780 or GW5638 (GW7604) were not identical and suspected that this may translate into important mechanistic differences, secondary to differential cofactor recruitment (23, 24, 30). The identification of AEIP, a peptide that interacted specifically with ICI182,780-bound ER{alpha} supports this link between a specific molecular structure and receptor degradation. We have also observed that RU58,668, a steroidal pure antiestrogen, also facilitates ER{alpha}-AEIP interaction and induces receptor turnover. This important information provides compelling support for the link between a specific ER{alpha} conformation and stability. Importantly, AEIP did not bind to GW7604-activated ER{alpha}. However, using combinatorial peptide phage display, we have shown in the past that the surfaces presented on GW7604 or GW5638-activated ER{alpha} are generally not present when the receptor is occupied by ICI182,780 (24). We conclude therefore that ICI182,780 and GW5638 (GW7604) are members of two distinct subclasses of SERDs.

Mutational analysis of AEIP identified a hydrophobic region within the peptide that interacts with ICI182,780-bound ER{alpha}. This was reminiscent of the hydrophobic regions of cofactors which are known to be involved in binding to both agonist- and SERM-bound ER{alpha} (26, 27, 39, 40). This led us to further examine the region of ER{alpha} with which AEIP was binding. It was observed that AEIP interacts with a similar, but slightly different, hydrophobic pocket than that used by estrogen-bound ER{alpha} for coactivator binding and was completely different from that used for interaction with CoRNR box–like peptides (27, 39). This suggested that the ICI182,780-induced conformation is more similar to that induced by estrogen than SERMs. Furthermore, based on these studies, we suspect that the extended sulfoxide-containing alkyl side chain of ICI182,780 is pushing helix 12 into the coactivator binding pocket, blocking coactivator binding but allowing for the AEIP peptide to bind. The dependence on helix 12 within ER{alpha} for AEIP interaction indicates why this peptide was unable to bind ICI182,780-bound ERß. Crystal structures of ICI182,780-bound ERß have been unable to resolve helix 12, suggesting that this region of ERß is highly mobile and unstructured when bound to ICI182,780 (46), and thus, it is likely that the conformation induced by steroidal SERDs within ER{alpha} is different from that induced in ERß. Studies are under way to crystallize ICI182,780-bound ER{alpha} in the presence of AEIP, which should help resolve this issue.

Our studies and those published by others have shown that following ICI182,780 treatment, ER{alpha} is rapidly translocated to a highly insoluble, nuclear matrix–associated fraction of the cell, whereupon it undergoes degradation (34–36). The recent demonstration that the solubility of ER{alpha} bound to ICI182,780 is influenced by the integrity of helix 12 supports the conclusions of our studies (36, 47). However, we have shown that additional mutations within the ligand-binding domain of ER{alpha} that disrupt AEIP-binding alter receptor compartmentalization and increase its stability. Specifically, we have shown that helices 3 and 5 are also involved in ER{alpha}/ICI182,780 complex and the binding of AEIP. In addition, these findings support the existence of an endogenous protein, containing a motif similar to that observed in AEIP, that regulates the trafficking and stability of the receptor. This hypothesis was further supported by the observation that expression of AEIP in cells was able to partially block ICI182,780-mediated turnover of ER{alpha}. Computational analysis using the National Center for Biotechnology Information BLAST program resulted in the identification of only one protein (ZMYM4) with an exact match to the AEIP motif (VPNSPM). In addition, a large number of many unknown proteins, and some known to be potentially involved in cellular trafficking such as trafficking kinesin-binding protein 1, were identified as containing the PNSPM sequence. It will be of interest to determine if any of these proteins are involved in determining the molecular pharmacology of ICI182,780. Regardless, the identification of AEIP and the demonstration that its binding determined the fate of ER{alpha} has, at a minimum, provided a facile assay for the identification of novel ICI,182,780-like SERDs.

In summary, we have determined that the currently available SERDs can be divided into two mechanistically distinct classes based on their differential effects on receptor structure and in the manner by which these differently conformed receptors are recognized within cells. This important finding predicts that, notwithstanding non–mechanism-related activities, the currently available steroidal antiestrogens will exhibit cross-resistance in breast tumors (48), but that the benzothiophene-derived SERD, GW5638, is sufficiently unique as to expect that it can be used in patients that are resistant to molecules like ICI182,780, as has been previously shown in tumors resistant to tamoxifen (24, 45).


   Acknowledgments
 
Grant support: NIH grant 5 R37 DK048807 (D.P. McDonnell).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

We thank Rachid Safi, Ching-yi Chang, John Norris, James Joseph, and Martin Tochacek for helpful advice during the preparation of this manuscript.


   Footnotes
 
Note: Supplementary data for this article are available at Cancer Research Online (http://cancerres.aacrjournals.org/).

Received 5/ 2/07. Revised 6/25/07. Accepted 7/19/07.


   References
Top
 Abstract
 Introduction
 Materials and Methods
 Results
 Discussion
 References
 

  1. Green S, Walter P, Kumar V, et al. Human oestrogen receptor cDNA: sequence, expression and homology to v-erb-A. Nature 1986;320:134–9.[CrossRef][Medline]
  2. Greene GL, Gilna P, Waterfield M, Baker A, Hort Y, Shine J. Sequence and expression of human estrogen receptor complementary DNA. Science 1986;231:1150–4.[Abstract/Free Full Text]
  3. Kuiper GG, Enmark E, Pelto-Huikko M, Nilsson S, Gustafsson JA. Cloning of a novel receptor expressed in rat prostate and ovary. Proc Natl Acad Sci U S A 1996;93:5925–30.[Abstract/Free Full Text]
  4. McDonnell DP, Norris JD. Connections and regulation of the human estrogen receptor. Science 2002;296:1642–4.[Abstract/Free Full Text]
  5. Levin ER. Integration of extranuclear and nuclear actions of estrogen. Mol Endocrinol 2005;19:1951–9.[Abstract/Free Full Text]
  6. Nagel SC, Hagelbarger JL, McDonnell DP. Development of an ER action indicator mouse for the study of estrogens, selective ER modulators (SERMs), and xenobiotics. Endocrinology 2001;142:4721–8.[Abstract/Free Full Text]
  7. Hewitt SC, Korach KS. Oestrogen receptor knockout mice: roles for oestrogen receptors {alpha} and ß in reproductive tissues. Reproduction 2003;125:143–9.[Abstract]
  8. Epstein FH. Estrogens and the risk of breast cancer. N Engl J Med 2001;344:276–85.[Free Full Text]
  9. Levin ER. Bidirectional signaling between the estrogen receptor and the epidermal growth factor receptor. Mol Endocrinol 2003;17:309–17.[Abstract/Free Full Text]
  10. Leygue E, Dotzlaw H, Watson PH, Murphy LC. Altered estrogen receptor {alpha} and ß mRNA expression during human breast tumorigenesis. Cancer Res 1998;58:3197–201.[Abstract/Free Full Text]
  11. Shaw JA, Udokang K, Mosquera JM, Chauhan H, Jones JL, Walker RA. Oestrogen receptors {alpha} and ß differ in normal human breast and breast carcinomas. J Pathol 2002;198:450–7.[CrossRef][Medline]
  12. Roger P, Sahla ME, Makela S, Gustafsson JA, Baldet P, Rochefort H. Decreased expression of estrogen receptor ß protein in proliferative preinvasive mammary tumors. Cancer Res 2001;61:2537–41.[Abstract/Free Full Text]
  13. Gradishar WJ. Tamoxifen—what next? Oncologist 2004;9:378–84.[Abstract/Free Full Text]
  14. Buzdar AU, Robertson JF. Fulvestrant: pharmacologic profile versus existing endocrine agents for the treatment of breast cancer. Ann Pharmacother 2006;40:1572–83.[Abstract/Free Full Text]
  15. Chung CT, Carlson RW. The role of aromatase inhibitors in early breast cancer. Curr Treat Options Oncol 2003;4:133–40.[CrossRef][Medline]
  16. Bross PF, Baird A, Chen G, et al. Fulvestrant in postmenopausal women with advanced breast cancer. Clin Cancer Res 2003;9:4309–17.[Abstract/Free Full Text]
  17. Robertson JF, Come SE, Jones SE, et al. Endocrine treatment options for advanced breast cancer—the role of fulvestrant. Eur J Cancer 2005;41:346–56.[CrossRef][Medline]
  18. Howell A. The future of fulvestrant (‘Faslodex’). Cancer Treat Rev 2005;31:S26–33.[CrossRef][Medline]
  19. Robertson JF, Semiglazov V, Nemsadze G, et al. Effects of fulvestrant 250 mg in premenopausal women with oestrogen receptor-positive primary breast cancer. Eur J Cancer 2007;43:64–70.[CrossRef][Medline]
  20. Steger GG, Gips M, Simon SD, et al. Fulvestrant ("Faslodex"): clinical experience from the Compassionate Use Programme. Cancer Treat Rev 2005;31:S10–6.[CrossRef][Medline]
  21. Jordan VC. Antiestrogens and selective estrogen receptor modulators as multifunctional medicines. 1. Receptor interactions. J Med Chem 2003;46:883–908.[CrossRef][Medline]
  22. Jordan VC. Antiestrogens and selective estrogen receptor modulators as multifunctional medicines. 2. Clinical considerations and new agents. J Med Chem 2003;46:1081–111.[CrossRef][Medline]
  23. Wu Y-L, Yang X, Ren Z, et al. Structural basis for an unexpected mode of SERM-mediated ER antagonism. Mol Cell 2005;18:413–24.[CrossRef][Medline]
  24. Conner CE, Norris JD, Broadwater G, et al. Circumventing Tamoxifen resistance in breast cancers using antiestrogens that induce unique conformational changes in the estrogen receptor. Cancer Res 2001;61:2917–22.[Abstract/Free Full Text]
  25. Willson TM, Norris JD, Wagner BL, et al. Dissection of the molecular mechanism of action of GW5638, a novel estrogen receptor ligand, provides insights into the role of estrogen receptor in bone. Endocrinology 1997;138:3901–11.[Abstract/Free Full Text]
  26. Chang CY, Norris JD, Gron H, et al. Dissection of the LXXLL nuclear receptor-coactivator interaction motif using combinatorial peptide libraries: discovery of peptide antagonists of estrogen receptors {alpha} and ß. Mol Cell Biol 1999;19:8226–39.[Abstract/Free Full Text]
  27. Huang H-J, Norris JD, McDonnell DP. Identification of a negative regulatory surface within estrogen receptor {alpha} provides evidence in support of a role for corepressors in regulating responses to agonists and antagonists. Mol Endocrinol 2002;16:1778–92.[Abstract/Free Full Text]
  28. Hall JM, McDonnell DP. The estrogen receptor ß-isoform (ERß) of the human estrogen receptor modulates ER{alpha} transcriptional activity and is a key regulator of the cellular response to estrogens and antiestrogens. Endocrinology 1999;140:5566–78.[Abstract/Free Full Text]
  29. Coutts AS, Davie JR, Dotzlaw H, Murphy LC. Estrogen regulation of nuclear matrix-intermediate filament proteins in human breast cancer cells. J Cell Biochem 1996;63:174–84.[CrossRef][Medline]
  30. Wijayarante AL, McDonnell DP. The human estrogen receptor-{alpha} is a ubiquinated protein whose stability is affected differentially by agonists, antagonists, and selective estrogen receptor modulators. J Biol Chem 2001;276:35684–92.[Abstract/Free Full Text]
  31. Lonard DM, Nawaz Z, Smith CL, O'Malley BW. The 26S Proteasome is required for estrogen receptor-{alpha} and coactivator turnover and for efficient estrogen receptor-{alpha} transactivation. Mol Cell 2000;5:939–48.[CrossRef][Medline]
  32. Shao W, Keeton EK, McDonnell DP, Brown M. Coactivator AIB1 links estrogen receptor transcriptional activity and stability. Proc Natl Acad Sci U S A 2004;101:11599–604.[Abstract/Free Full Text]
  33. Htun H, Holth LT, Walker D, Davie JR, Hager GL. Direct visualization of the human estrogen receptor {alpha} reveals a role for ligand in the nuclear distribution of the receptor. Mol Biol Cell 1999;10:471–86.[Abstract/Free Full Text]
  34. Stenoien DL, Mancini MG, Patel K, Allegretto EA, Smith CL, Mancini MA. Subnuclear trafficking of estrogen receptor-{alpha} and steroid receptor coactivator-1. Mol Endocrinol 2000;14:518–34.[Abstract/Free Full Text]
  35. Stenoien DL, Patel K, Mancini MG, et al. FRAP reveals that mobility of oestrogen receptor-{alpha} is ligand- and proteasome-dependent. Nat Cell Biol 2001;3:15–23.[CrossRef][Medline]
  36. Long X, Nephew KP. Fulvestrant (ICI182,780)-dependent interacting proteins mediate immobilization and degradation of estrogen receptor-{alpha}. J Biol Chem 2006;281:9607–15.[Abstract/Free Full Text]
  37. Peekhaus NT, Chang T, Hayes EC, et al. Distinct effects of the antiestrogen Faslodex on the stability of estrogen receptors-{alpha} and -ß in the breast cancer cell line MCF-7. J Mol Endocrinol 2004;32:987–95.[Abstract]
  38. Banik SS, Guo C, Smith AC, et al. C-terminal regions of the human telomerase catalytic subunit essential for in vivo enzyme activity. Mol Cell Biol 2002;22:6234–46.[Abstract/Free Full Text]
  39. Shiau AK, Barstad D, Loria PM, et al. The structural basis of estrogen receptor/coactivator recognition and the antagonism of this interaction by Tamoxifen. Cell 1998;95:927–37.[CrossRef][Medline]
  40. Norris JD, Paige LA, Christensen DJ, et al. Peptide antagonists of the human estrogen receptor. Science 1999;285:744–6.[Abstract/Free Full Text]
  41. Brzozowski AM, Pike AC, Dauter Z, et al. Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 1997;389:753–8.[CrossRef][Medline]
  42. Laios I, Journe F, Nonclercq D, et al. Role of the proteasome in the regulation of estrogen receptor {alpha} turnover and function in MCF-7 breast carcinoma cells. J Steroid Biochem Mol Biol 2005;94:347–59.[CrossRef][Medline]
  43. Kong EH, Heldring N, Gustafsson J-A, Treuter E, Hubbard RE, Pike AC. Delineation of a unique protein-protein interaction site on the surface of the estrogen receptor. Proc Natl Acad Sci U S A 2005;102:3593–8.[Abstract/Free Full Text]
  44. Buzdar AU, Marcus C, Holmes F, Hug V, Hortobagyi G. Phase II evaluation of L Y156758 in metastatic breast cancer. Oncology 1988;45:344–5.[Medline]
  45. Dardes RC, O'Regan RM, Gajdos C, et al. Effects of a new clinically relevant antiestrogen (GW5638) related to tamoxifen on breast and endometrial cancer growth in vivo. Clin Cancer Res 2002;8:1995–2001.[Abstract/Free Full Text]
  46. Pike AC, Brzozowski AM, Walton J, et al. Structural insights into the mode of action of a pure antiestrogen. Structure 2001;9:145–53.[Medline]
  47. Lupien M, Jeyakumar M, Hebert E, et al. Raloxifene and ICI182,780 increase estrogen receptor-{alpha} association with a nuclear compartment via overlapping sets of hydrophobic amino acids in activation function 2 helix 12. Mol Endocrinol 2007;21:797–816.[Abstract/Free Full Text]
  48. Fog CK, Christensen IJ, Lykkesfeldt AE. Characterization of a human breast cancer cell line, MCF7/RU58R-1, resistant to the pure antiestrogen RU58,668. Breast Cancer Res Treat 2005;91:133–44.[CrossRef][Medline]



This article has been cited by other articles:


Home page
 J Mol EndocrinolHome page
A. M Davis, J. Mao, B. Naz, J. A Kohl, and C. S Rosenfeld
Comparative effects of estradiol, methyl-piperidino-pyrazole, raloxifene, and ICI 182 780 on gene expression in the murine uterus
J. Mol. Endocrinol., October 1, 2008; 41(4): 205 - 217.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Wittmann, B. M.
Right arrow Articles by McDonnell, D. P.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Wittmann, B. M.
Right arrow Articles by McDonnell, D. P.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Cancer Research Clinical Cancer Research
Cancer Epidemiology Biomarkers & Prevention Molecular Cancer Therapeutics
Molecular Cancer Research Cancer Prevention Research
Cancer Prevention Journals Portal Cancer Reviews Online
Annual Meeting Education Book Meeting Abstracts Online