| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Provides Evidence in Support of a Role for Corepressors in Regulating Cellular Responses to Agonists and Antagonists
Department of Pharmacology and Cancer Biology, Duke University Medical Center, Durham, North Carolina 27710
Address all correspondence and requests for reprints to: Dr. Donald P. McDonnell, Department of Pharmacology and Cancer Biology, Duke University Medical Center, Box 3813, Durham, North Carolina 27710. E-mail: donald.mcdonnell{at}duke.edu.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
mutants, we found that helix 12 was not required for the binding of CoRNR box peptides, whereas disruption of helixes 3 and 5 had a marked effect on peptide binding. One mutant, ER-L372R, lost the ability to interact with CoRNR box-containing peptides without affecting its binding to LXXLL motif-containing peptides. The estradiol- and tamoxifen-mediated transcriptional activity of ER-L372R was dramatically increased by 11- and 3-fold, respectively, compared with that of wild-type ER. The ICI 182,780-mediated repressional activity of this mutant was also reduced by 4-fold compared with that of wild-type ER. These results suggest that leucine 372 may be an important part of the interaction surface on ER that is responsible for corepressor binding. In addition, our data suggest that corepressors, other than NCoR/SMRT, may be involved in ER signaling. |
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
and ERß bind estrogens with high affinity and can activate the transcription of promoters containing a classical estrogen response element (ERE) (5). However, ERß has been shown to function as an inhibitor of ER transcriptional activity, decreasing cellular sensitivity to estradiol in target cells where both receptor subtypes are expressed (6). Consequently, the response of a cell to specific ER modulators is determined to a large extent by the relative and absolute expression levels of coactivators, corepressors, and each receptor subtype.
Crystallographic analysis of the ER ligand binding domain (LBD) has established that ligand binding has a dramatic effect on receptor structure. Agonists such as 17ß-estradiol and diethylstilbestrol induce a receptor conformation in which the carboxyl-terminal helix 12 (H12) is aligned over the hormone binding cavity, resulting in the formation of a specific binding site for the consensus LXXLL motif found within coactivators (7, 8). Selective ER modulators (SERMs) such as raloxifene and tamoxifen, on the other hand, induce conformational changes in ER that interfere with the repositioning of H12 and alter the coactivator recruitment surface (7, 8). More recently, the structure of the ERß LBD complexed with a pure antagonist, ICI 164,384, has been determined, indicating that binding of this antagonist completely destabilizes H12 and prevents it from adopting either agonist or SERM orientation (9). Proteolysis and peptide binding studies have also shown that pure antagonists facilitate a conformational change within ER that is unique and different from that induced by agonists and SERMs (10, 11, 12). Thus, SERMs and pure antagonists are mechanistically distinct classes of compounds that, not surprisingly, exhibit different pharmacological profiles in vivo. Of particular note in this regard are studies demonstrating that tamoxifen-resistant breast tumors are highly responsive to treatment with pure antiestrogens. An attractive model with which to explain this differential response is that the corepressor/coactivator ratio in breast tumors changes during the course of treatment, enabling tamoxifen to engage a suitable coactivator and to function as an agonist. However, as the pure antagonist-ER complex adopts a different conformation, its cofactor preferences are different and are not compatible with transcriptional activation (13, 14, 15). Clearly, there is a need to identify and characterize the coactivators and corepressors that interact with ER and determine how they interface with the receptor.
Two corepressors, nuclear receptor corepressor (NCoR) and silencing mediator of retinoid and thyroid receptors (SMRT), have been implicated in mediating ER signaling. Specifically, it has been noted that 1) ER-NCoR or ER-SMRT complexes can be formed in MCF-7 cells upon treatment with tamoxifen (13, 16); 2) tamoxifen manifests considerable agonist activity in mouse embryo fibroblasts derived from NCoR knockout mice (17); and 3) down-regulation of NCoR was observed in tamoxifen-resistant MCF-7 xenografts in an animal model (13). Despite this compelling functional data, attempts to demonstrate the direct interaction between ER and NCoR/SMRT have revealed that ER can bind to these corepressors regardless of agonist or antagonist treatment in vitro (14, 18). This difficulty in linking NCoR/SMRT-mediated regulation of ER pharmacology to a tamoxifen-dependent, direct protein-protein interaction may suggest that 1) the actions of these corepressors are indirect and require another protein to enable complex formation; or 2) there exists as yet unidentified corepressors that interact directly with ER. To address these issues we have constructed a focused phage display library that contains the CoRNR box motif, derived by aligning two receptor-interacting domains (ID) from NCoR and SMRT. The CoRNR box motif has been shown to be important for binding to thyroid hormone receptor (TR) and retinoic acid receptor (RAR) (19, 20), and the sequences adjacent to the CoRNR box motif have been suggested to play an important role in mediating the specificity of receptor interactions (21). In this study we have used phage display technology to identify CoRNR box-containing peptides that bind ER specifically. Furthermore, disruption of the surfaces on ER that facilitate CoRNR box peptide-receptor interactions have dramatic effects on ER pharmacology, leading to the conclusion that this receptor is subject to negative regulation either by an inhibitory intramolecular interaction or by an as yet unidentified corepressor.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
). The overall amino acid distribution for the random codons is shown in Table 2A. To eliminate the bias inherent in codon usage, the frequency of each amino acid was divided by the number of different codons encoding that particular amino acid (Table 2, last column). Acidic amino acids such as aspartate and glutamate were the most abundant amino acids in this library (
5%). Cysteine was very rare (0.3%). For the amino acid residue at position 4 in the peptide, an MWC codon (M = A or C; W = A or T) was used. This codon will encode not only histidine and isoleucine, which are found in the consensus CoRNR box sequence, but also asparagine and leucine. Table 2B shows that 29% of the peptides expressed the desired histidine/isoleucine. For the amino acid residue at position 9 in the peptide, an MTC codon was used, which only allows isoleucine and leucine to be expressed. The frequency of occurrence at this position is 56% and 44% for leucine and isoleucine, respectively, very close to the expected frequencies (Table 2B).
|
|
or ERß in the absence of hormone or in the presence of tamoxifen or ICI 182,780. After three rounds of affinity selection, significant enrichment of phage was observed using ERß, whereas minimal enrichment of ER interactors was observed. We believe that this is due to differences in the stability of the purified recombinant receptors, as most of the peptides obtained using ERß as the target interact in an efficient manner with ER in a mammalian two-hybrid assay (see below). Because of the difficulty of using ER in the screen, we used ERß as the target for all subsequent primary screens. ELISA was used to identify the phage clones that bind ER with high affinity and ligand dependency. Of 48 random clones analyzed in each pool, 32 clones bound apo-ERß, 17 clones bound tamoxifen-ERß, and 31 clones bound ICI 182,780-ERß. The amino acid sequence of the interacting peptides was deduced after DNA sequencing of the phage inserts. Table 3 shows the peptide sequences and frequency of occurrence for the phage clones obtained. The bN, bT, and bI phage clones represent phage isolated using apo-ERß, tamoxifen-ERß, and ICI 182,780-ERß as targets, respectively. All of the bT and bI phage were isolated using receptor immobilized directly to plastic plates. However, enrichment of bN phage can only be obtained using receptor bound to immobilized EREs, suggesting that apo-ER may adopt a more favorable conformation for CoRNR box peptide binding when an ERE is present. Among the four unique bN clones obtained, clones bN1 and bN2 were found most frequently. It was noticed that bN1 and bN2 both possess arginine at position -2, tryptophan at position -1, and glycine at position 3, the significance of which remains to be determined. Seventeen clones showed specific binding to tamoxifen-bound ERß. However, 16 of the clones in this group share the same sequence (bT1). Importantly, bT17 has the same sequence as bN2, suggesting that ER may adopt a similar conformation at the binding site of this peptide in the absence of hormone or in the presence of tamoxifen. Twenty-four unique clones were obtained that showed specific binding to ICI 182,780-bound ER. However, only two of the bI clones (bI2 and bI29) showed specific binding to ER using the mammalian two-hybrid assay (see below), suggesting that ICI 182,780 treatment in vitro may have forced ER to adopt a conformation that is not normally seen in vivo.
|
or VP16-ERß occupied by different estrogens/antiestrogens in mammalian cells was determined by assaying the ability of the complex to activate a luciferase reporter construct under the control of five copies of a Gal4 upstream enhancer element (5xGal4-TATA-Luc). As a point of reference we also produced Gal4-DBD-peptide fusions corresponding to the previously identified IDs from NCoR and SMRT. A direct comparison of the sequences of all the peptides used in this study is shown in Fig. 1E. The fusion proteins were shown to be expressed at similar levels by Western blot analysis (data not shown). Figure 1, A and B, shows the results from a mammalian two-hybrid assay confirming the ER-peptide interactions in HepG2 cells. Interestingly, although the CoRNR box peptides chosen for this analysis were isolated using ERß as a target, most peptides interact with ER also (Fig. 1A). Most importantly, all of these CoRNR box peptides interact with ER or ERß in the absence of hormone or in the presence of tamoxifen or ICI 182,780, and none interacts with estradiol-bound ER. These results strengthen our hypothesis that the CoRNR box peptides bind to surfaces on ER that are important for corepressor binding. The mammalian two-hybrid assay was further validated by demonstrating that the four CoRNR box-containing peptides derived from NCoR/SMRT IDs were capable of interacting with apo-TRß or apo-RAR, and these interactions were disrupted upon addition of thyroid hormone or all-trans-retinoic acid (Fig. 1, C and D). However, neither TRß nor RAR was able to interact with the ER-interacting CoRNR box peptides identified in our screens (Fig. 1, C and D). Furthermore, we were unable to demonstrate an interaction of ER or ERß with any of the previously defined CoRNR box peptides found in NCoR/SMRT (Fig. 1, A and B). Identical results were obtained using HeLa cells (data not shown). In general, the ER binding profiles of the CoRNR box peptides evaluated in vivo reflect the characteristics observed using in vitro ELISA and confirm that there are surfaces on ER that can accommodate these CoRNR box-containing peptides. These results are consistent with the presence of a specific corepressor binding surface(s) on ER and ERß. However, the ER-binding corepressors may be different from the corepressors (NCoR/SMRT) bound to TR or RAR.
|
Required for CoRNR Box Peptide Binding for a more extensive analysis in this study because its pharmacology is more clearly defined. A schematic diagram of ER and the mutants used in the current study is shown in Fig. 2. Mutations in helix 12 have been shown previously to decrease ER transcriptional activity and abolish the interaction of the receptor with the coactivator glucocorticoid receptor-interacting protein 1 (GRIP1) (22) and with LXXLL motif-containing peptides (23). Mutations of lysine 362, leucine 372, and valine 376 were chosen because amino acid substitutions at equivalent positions in TR or RXR are known to decrease their ability to interact with NCoR/SMRT and to abrogate ligand-independent repression activity (20). All of these mutations were created in the background of a full-length VP16-ER protein to enable evaluation of receptor-peptide interactions using a mammalian two-hybrid assay in HepG2 cells. The interactions of several different classes of LXXLL peptides with mutant ERs were also analyzed for comparison. Three LXXLL peptides D11, D47, and F6 with distinct receptor binding characteristics (23) were used. The Gal4-DBD-GRIP1 NR-box fusion contains the middle three copies of the LXXLL motif found in the coactivator GRIP1. As shown in Fig. 3, A–D, mutations in helix 12 did not affect the binding of the CoRNR box peptides to ER, but actually enhanced the binding of bT1 and bI2 peptides to the receptor. The interaction of the CoRNR box peptides and the ER helix 12 mutants was interesting in that the exquisite hormonal specificity observed on the wild-type ER (wtER) was not preserved (Fig. 3, A–C). This suggests that the positioning of helix 12, influenced by the nature of the ligand bound, is an important regulator of CoRNR box peptide binding. Mutations at the amino acid residues Lys362 and Val376 within ER decreased its ability to interact with both the CoRNR box and LXXLL peptides, suggesting that the binding surfaces for these two classes of peptides are close or overlapping. The results obtained with ER-L372R are the most interesting, as this specific mutation totally abolished the interaction between ER and two of the CoRNR box peptides (bT1 and bI2) while having no effect on binding of the LXXLL peptides to the receptor (Fig. 3). Additional LXXLL peptides (D2, D22, and SRC-1 NR-box) were also tested, and they all exhibit similar binding characteristics (data not shown). Thus, although the binding sites on ER for the CoRNR box and LXXLL peptides are closely linked, they can be functionally separated.
|
|
and all the CoRNR boxcontaining peptides were severely impaired (Fig. 3
, A–D). The interaction of ER-HBD with LXXLL motif-containing peptides was affected, but not as dramatically (Fig. 4, E–H). These results suggest that domains other than LBD may be involved in the formation of the CoRNR box peptide binding pocket in ER.
|
conformation. Next we determined whether the surfaces on the receptor with which these peptides interact could have a specific regulatory function. To this end, we compared the transcriptional activity of wtER with that of the ER-L372R mutant in a reconstituted transcriptional system in transfected cells. Specifically, the luciferase reporter plasmid 3xERE-TATA-Luc and increasing concentrations of wild-type or mutant ER expression vector were transiently transfected into ER-negative HepG2 cells. All transfections were performed in the absence or presence of 10-7 M 17ß-estradiol, 4-hydroxytamoxifen, or ICI 182,780. As shown in Fig. 4A, the constitutive transcriptional activity of ER-L372R was considerably lower than that displayed by the wild-type receptor. This effect is probably a general consequence of disturbances of the ER-LBD surface, as other mutations in ER helix 12 analyzed in the same manner also exhibit a lower level of constitutive activity (data not shown). Tamoxifen manifests significant partial agonist activity in HepG2 cells expressing exogenous wtER, the magnitude of which is increased by a maximal 3-fold when assayed in cells expressing the L372R mutant (Fig. 4C). In this cell system the pure antiestrogen ICI 182,780 functions as an inverse agonist, and in the absence of added agonist it caused a 90% repression of the basal transcriptional activity of the 3xERE-TATA-Luc reporter in cells expressing wtER. This repression activity is decreased by a maximal 4-fold when a similar assay is performed in cells expressing the ER-L372R mutant (Fig. 4D). The most dramatic and unexpected result observed, however, was that estradiol-stimulated transcriptional activity of the ER-L372R mutant was greatly enhanced compared with that of wtER (Fig. 4B). This observed increase in the efficacy of estradiol was most dramatic when 10 ng of the ER-L372R expression vector was used (11-fold more active than wtER). Transfection of higher concentrations of ER-L372 led to a progressive decrease in estradiol-dependent transcriptional activity, possibly due to sequestration of limiting transcription factors by overexpression of this highly active mutant receptor. Similar results were observed using a C3-Luc reporter that contains a bona fide estrogen-responsive human C3 complement promoter (data not shown). Western analysis indicated that the wild-type and mutant ERs used in this study were similarly expressed, ruling out differences in expression level as the basis for the observed differences in transcriptional activity (Fig. 4E). These results suggest that amino acid residue L372 is contained within a region that functions as a negative regulatory surface for ER.
The ER-L372R Mutant Is Not Hypersensitive to Estradiol
The dramatically enhanced estradiol-dependent transcriptional activity of ER-L372R prompted us to examine whether this mutant receptor displayed an increased sensitivity to estradiol. To address this issue, HepG2 cells were transfected with 10 ng of either the wtER or ER-L372R expression vector along with 3xERE-TATA-Luc reporter and subsequently treated with increasing concentrations of estradiol (Fig. 5). Contrary to what was expected, we observed that the mutant receptor was about one order of magnitude less sensitive to estradiol than the wild-type receptor. These results suggest that mutation at leucine 372 of ER did not render the receptor more sensitive to estradiol treatment.
|
and ER-L372R to interact with different classes of LXXLL-containing peptides were compared in the presence of increasing concentrations of estradiol using a mammalian two-hybrid assay (Fig. 6). The Gal4-DBD-D11, Gal4-DBD-D47, Gal4-DBD-F6, and Gal4-DBD-GRIP1 NR-box constructs were described above. The Gal4-DBD-SRC-1 NR-box fusion contains the middle three copies of the LXXLL motif found in the coactivator SRC-1. As shown in Fig. 6, compared with the wtER, the ER-L372R mutant did not show a higher affinity for any of the LXXLL peptides tested. However, ER-L372R did exhibit a higher capacity for binding to the GRIP1 NR-box peptide at higher concentrations of estradiol (Fig. 6D). Therefore, the enhanced transcriptional activity of ER-L372R can be explained in part by increased coactivaor binding.
|
and apo-TRß, none of the CoRNR box-containing peptides obtained in this study interacted with these two receptors (see Fig. 1, C and D). However, we noticed that the bN2 peptide can interact with apo-RXR, and treatment with 9-cis-retinoic acid abolished this interaction (Fig. 7A). Under the same conditions apo-RXR showed only minimal binding to NCoR/SMRT IDs, and the interaction between SMRT ID-C and apo-RXR was actually slightly increased, rather than inhibited, by retinoic acid treatment (Fig. 7A), consistent with previous observations that RXR only weakly interacts with SMRT, and retinoic acid treatment can strengthen this interaction (21, 24). The binding of progesterone receptor (PR) to ER-interacting CoRNR box peptides was also examined. Figure 7, B and C, showed that no interaction of PRA and PRB with the CoRNR box peptides was observed in the absence of hormone or in the presence of progesterone. However, treatment with RU486 (a mixed agonist/antagonist) or ZK98299 (a pure antagonist) facilitated a robust interaction between both PR isoforms and the bT1 peptide. As antagonist-bound PRs can interact with NCoR/SMRT IDs, as shown in Fig. 7, B and C, and in previous studies (25), it is not surprising that CoRNR box-containing peptides with this binding specificity were found in our screens. However, there are differences in the CoRNR box binding patterns of PR treated with RU486 or ZK98299. The bT1 peptide, which binds tamoxifen-bound ER and ERß, associates more strongly with PRs treated with RU486 than ZK98299, whereas NCoR/SMRT ID-Cs exhibit a stronger interaction with ZK98299-bound PRs than RU486-bound PRs. These results demonstrated that the flanking sequences of the CoRNR box can dictate the receptor binding specificity and that an unidentified corepressor(s) other than NCoR/SMRT might exist that shares the CoRNR box motif, but differentially interacts with different nuclear receptors.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
(26, 27). The generality of the requirement of helix 1 for corepressor binding, however, is unclear, because this domain is not highly conserved in ER. It now appears more likely that the CoR box contributes to the overall structure of a distal corepressor binding site on TR, but does not provide the primary point of contact between the receptor and corepressors (28). This conclusion is further supported by the recent crystallographic analysis of peroxisome proliferator-activated receptor (PPAR) complexed with antagonist GW6471 and SMRT ID-C, which reveals that the primary corepressor binding pocket overlaps that required for coactivator binding in helixes 3, 4, and 5 (29). No evidence for a contribution of sequences in helix 1 is apparent from this structure. Indeed, before the solution of this informative crystal structure the involvement of helixes 3 and 5 within TR and RXR in corepressor binding was implicated by mutagenesis (20). In the present study mutations of these corresponding amino acid residues in ER were made, and the resultant mutants were tested for their ability to interact with a series of CoRNR box- and LXXLL motif-containing peptides. Two of the ER mutants, ER-K362A and ER-V376R, exhibited impaired interactions with both LXXLL- and CoRNR box-containing peptides, confirming that the surfaces required for coactivator and corepressor binding overlap. The most informative mutation, however, is ER-L372R, which retained the LXXLL peptide binding characteristics of wtER, but was unable to interact with two of the CoRNR box peptides (bT1 and bI2) identified in our studies. Using this mutant we were able to dissociate corepressor and coactivator binding by demonstrating that the agonist efficacy of tamoxifen and estradiol could be dramatically enhanced when the binding site for these CoRNR box peptides was disrupted. The conclusion that ER-L372 was required for corepressor binding was supported by the additional observations that the inverse agonist activity of the pure antiestrogen ICI 182,780 was decreased when assayed on this mutant. Cumulatively, these findings provide strong evidence for the involvement of corepressors in ER action.
It has been shown that the corepressors NCoR and SMRT can associate with ER in the presence of tamoxifen (13, 16, 30, 31). However, in the current study we were unable to detect an interaction between ER and the CoRNR box peptides, which contain the NCoR/SMRT IDs, although these peptides were capable of binding apo-TRß and apo-RAR. It is possible that the interaction interface between ER and NCoR/SMRT lies outside of the CoRNR box motif. However, identification of CoRNR box peptides that bind ER in the presence of antagonists also suggests that there exist other CoRNR box-containing proteins that may be involved in ER signaling.
There is strong evidence that corepressors NCoR and SMRT can modulate the agonist activity of tamoxifen, but they play little, if any, role in regulating the physiological actions of estradiol (14, 17). Predictably, we observed that the agonist activity of tamoxifen was enhanced by disruption of the putative corepressor binding surface (defined by the L372R mutation) on ER. However, one of the most interesting findings of our study was that this mutation also enhanced the agonist efficacy of estradiol. One possible explanation is that this mutation disrupts the interaction of ER with a corepressor (or a competitive repressor) that, unlike NCoR and SMRT, can recognize the agonist-activated structure of ER. Potential candidates that could function in this manner include repressor of estrogen receptor activity (REA), short heterodimer partner (SHP), and dosage-sensitive sex reversal-adrenal hypoplasia congenita critical region on the X chromosome (DAX-1) (32, 33, 34). We believe that the data are also consistent with the hypothesis that there exist corepressors that can interact with apo-ER or ER bound to an antagonist and that the L372R mutation disrupts this interaction and in doing so reduces the energy of activation of the receptor by removing the "brake" afforded by the corepressor. This hypothesis will be tested in future studies.
It is interesting to note that the cocrystal structure of ER with NR box 2 of GRIP1 predicts that K362, L372, and V376 make van der Waals contacts with the leucine side-chains within the LXXLL motif, and that K362 can also form hydrogen bonds with the backbone of this peptide (8). Although the transcriptional activity of the ER-V376R and ER-K362A mutants was compromised (data not shown), we were surprised to find that mutation of amino acid leucine 372 to arginine within ER did not decrease, but greatly enhanced, estradiol efficacy. Moreover, ER-L372R retained the ability to interact with all three classes of LXXLL motif-containing peptides and actually showed an increased binding capacity for the GRIP1 NR box peptides. It is likely that L372 of ER, which corresponds to L302 in PPAR, plays a significant role in interacting with CoRNR box-containing corepressors, as suggested by the recent crystal structure of the antagonist-PPAR bound to the SMRT ID-C (29). Alternatively, L372 could be important for mediating the autoinhibitory activity of ER through an intramolecular interaction. Whatever the mechanism, mutation at L372 disrupts the negative regulatory surface on ER, allowing the receptor to be more susceptible to agonist activation. Another possible explanation is that mutation of L372 to an arginine may inadvertently increase the affinity of the receptor for coactivators by making additional contacts with the amino acid residues adjacent to the LXXLL motif. Given the overlapping nature of the corepressor and coactivator binding site on ER, it is likely that the increased transcriptional activity of ER-L372R is due to both a loss of corepressor binding and an increase in the ability of coactivators to bind to the receptor.
Inspection of the crystal structure of human ER LBD in the presence of estradiol indicated that amino acid residues in helixes 3, 4, 5, and 12 can form a hydrophobic surface that constitutes the primary contact point for the LXXLL motif found in the p160 coactivators (8). However, helix 12 is not required for corepressor binding, as its deletion or mutation only serves to enhance the interaction of both NCoR and SMRT with nuclear receptors (27, 28, 35). One of the most interesting observations of the current study was that the binding of the CoRNR box peptides to ER was not negatively affected by mutations in helix 12. In fact, most of these CoRNR box peptides interacted better with ER mutants in which the helix 12 structure was compromised, demonstrating a negative regulatory function of helix 12 for corepresssor binding. In addition, it was found that the exquisite ligand specificity demonstrated by the different classes of CoRNR box peptides identified was lost when helix 12 was disrupted. Consistent with these results, a recent report demonstrated that a mutant ER deleted for the C-terminal 58 amino acids can be coimmunoprecipitated with SMRT in the absence of hormone or in the presence of estradiol and tamoxifen, whereas under the same conditions wtER can only interact with SMRT in a tamoxifen-dependent manner (36). Thus, although ER helix 12 is not required for corepressor binding, its relative positioning by different agonists/antagonists can regulate the interactions between corepressors and the receptor.
In summary, using combinatorial phage display, three different classes of CoRNR box-containing peptides have been identified that bind selectively to apo-ER, tamoxifen-bound ER, and ICI 182,780-bound ER. These peptides have helped map a negative regulatory surface within ER. Our findings are consistent with the existence of corepressors that interact with and modulate ER activity. Although NCoR/SMRT may be important for mediating the tamoxifen antagonist activity of ER, we cannot rule out the possibility that additional corepressors may be involved in regulating ER activity in the presence of antagonists/agonists.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Construction of the Phage Display Library
A focused CoRNR box phage display library was constructed using a degenerate oligonucleotide able to code for 23 amino acids in the format of X7-L-X-X-H/I-I-X-X-X-I/L-X7, where L is leucine, H is histidine, I is isoleucine, and X is any amino acid. A 100-bp single-stranded oligonucleotide containing terminal XhoI and XbaI sites and fully degenerate at the remaining (NNK) positions was synthesized, converted to double-stranded form, and purified by gel electrophoresis. The restricted oligonucleotides were ligated into the mBAX vector, and the ligated DNA was used to transform Escherichia coli DH5F' competent cells. The transformants were harvested, and the phage were purified from the transformants. The library has a complexity of 1.3 x 107 different peptide sequences.
Panning of CoRNR Box Peptide Library
Enrichment and amplification of the library using a target protein were carried out as described previously with minor modifications (23). To isolate phage that bind tamoxifen- or ICI 182,780-bound ER, 4 pmol ER or ERß were diluted in NaHCO3 plus 10-6 M antiestrogen and adsorbed onto an Immulon 4 plate (Thermo Labsystems, Helsinki, Finland) for 3 h at room temperature. To isolate apo-ER binding phage, Immulon 4 plates were coated with streptavidin in 0.1 M NaHCO3 and then incubated for 1 h with 2 pmol biotinylated vitellogenin ERE, followed by incubation for 1 h with 4 pmol ER or ERß. After blocking the ER-coated or ER/ERE-coated plates for 1 h with 0.1% BSA in NaHCO3, aliquots of the phage library (1010 plaque-forming units) were diluted in PBS containing 0.1% Tween 20, and phage were allowed to bind overnight at 4 C. Unbound phage were subsequently removed by washing with PBS containing 0.1% Tween 20, and bound phage were eluted with 50 mM glycine-HCl (pH 2.0), followed by 100 mM ethanolamine (pH 11.0). The pH of the first eluate was immediately neutralized with 200 mM Na2HPO4 (pH 8.5) and combined with the first eluate. Eluted phage were amplified in DH5F' cells for 5 h, and the supernatant containing amplified phage was collected for use in subsequent rounds of panning. After the third round of panning, randomly selected phage plaques were amplified, and their single-stranded DNA was sequenced. Phage amplified from single plaques and purified were tested for binding to ER or ERß with or without hormones using ELISA, as described previously (23).
Cell Culture and Transient Transfection
HepG2 (hepatocellular carcinoma) and HeLa (cervical carcinoma) cells were maintained in MEM (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum (HyClone Laboratories, Inc., Logan, UT), 0.1 mM nonessential amino acids, and 1 mM sodium pyruvate (Invitrogen) and maintained in a humidified 37 C incubator with 5% CO2. Cells were plated into 24-well plates and transfected using lipofectin reagent according to the manufacturer’s instructions (Invitrogen). A transfection mixture containing a total of 3000 ng plasmid in each of triplicate samples was incubated with cells for 3 h (for HeLa) or 5–16 h (for HepG2). Hormones were added to the cells 24 h before harvest. Luciferase and ßgalactosidase activities were measured as previously described (37). For mammalian two-hybrid assays, 1500 ng 5xGal4-TATA-Luc reporter plasmid, 500 ng VP16-receptor fusion, 500 ng Gal4-DBD-peptide fusion, and 500 ng normalization plasmid pCMVßgal were used. For receptor transcriptional activity assays, 1500 ng 3xERE-TATA-Luc, 100 ng pCMVßgal, and 0–640 ng wild-type or mutant pRST7-ER were used. A pBluescript II vector (Stratagene, La Jolla, CA) was added in these experiments to balance the amount of input DNA in transfections.
Plasmids
The Gal4-DBD-peptide fusions were constructed by excising DNA sequences coding for the peptides from mBAX vector with XhoI and XbaI and subcloning it into the SalI and XbaI sites of the pMsx vector (23). To generate the Gal4-DBD-NCoR ID-N, Gal4-DBD-SMRT ID-N, Gal4-DBD-NCoR ID-C, or Gal4-DBD-SMRT ID-C fusions, single-stranded oligonucleotides containing the DNA sequences coding for the individual ID and terminal XhoI and XbaI sites were synthesized and converted to double-stranded form, and the restricted oligonucleotides were ligated into the pMsx vector. Gal4-DBD-D11, Gal4-DBD-D47, Gal4-DBD-F6, Gal4-DBD-GRIP1 NR-box, and Gal4-DBD-SRC-1 NR-box were described previously (23). Mammalian expression plasmid pRST7-ER has been described previously (38). Mutant receptors pRST7-ER-K362A, pRST7-ER-L372R, and pRST7-ER-V376R were generated by using the QuikChange sitedirected mutagenesis kit (Stratagene), with wild-type pRST7-ER as the template. The reporter 3xERE-TATA-Luc has been described previously (22). Construction of VP16-ER, VP16-ERß, VP16-ER-LL, VP16-ER-3x, VP16-ER-535 stop, VP16-RAR, and VP16-RXR were described previously (23). VP16-TRß was provided by D. D. Moore (Baylor College of Medicine, Houston, TX). VP16-PRA and VP16-PRB were gifts from C. X. Wen (Ligand Pharmaceuticals, Inc.). VP16-ER-HBD was generated by PCR of the full-length human ER cDNA with primers containing EcoRI and SalI sites and subsequently subcloning the PCR product into the EcoRI and SalI sites in the pVP16 vector (CLONTECH Laboratories, Inc., Palo Alto, CA). Other VP16-ER point mutants (ER-K362A, ER-L372R, and ER-V376R) were constructed by excision of mutant ER cDNAs from their corresponding mammalian expression plasmids and subcloned into the pVP16 vector. The reporter 5xGal4-TATA-Luc was a gift from X. F. Wang (Duke University Medical Center, Durham, NC).
Western Blot Analysis
HepG2 cells were plated into 100-mm plates and transfected with 3 µg of the different forms of receptor treated with different hormones together with 1 µg of a green fluorescent protein expression vector for normalization. Nuclear extracts were prepared as described previously (39). Proteins (20-µg samples) were separated on sodium dodecyl sulfate-polyacrylamide gel and transferred to nitrocellulose. The receptors were detected with the monoclonal antibody H222 (provided by Geoffrey Greene, Ben May Institute, Chicago, IL). Green fluorescent protein was probed with an anti-green fluorescent protein polyclonal antibody (CLONTECH Laboratories, Inc.). The immunocomplexes were visualized by enhanced chemiluminescence (Amersham Pharmacia Biotech, Piscataway, NJ) as described by the manufacturer.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Abbreviations: DAX-1, Dosage-sensitive sex reversaladrenal hypoplasia congenita critical region on the X chromosome; ER, estrogen receptor; ERE, estrogen response element; Gal4-DBD, Gal4 DNA binding domain; GRIP1, glucocorticoid receptor-interacting protein 1; H, helix; HBD, hormone binding domain; ID, receptor-interacting domain; LBD, ligand binding domain; NCoR, nuclear receptor corepressor; PPAR, peroxisome proliferator-activated receptor; PR, progesterone receptor; RAR, retinoic acid receptor; REA, repressor of estrogen receptor activity; RXR, retinoid X receptor; SERM, selective estrogen receptor modulator; SHP, short heterodimer partner; SMRT, silencing mediator of retinoid and thyroid receptors; SRC-1, steroid receptor coactivator 1; TR, thyroid hormone receptor; wt, wild-type.
Received for publication March 5, 2002. Accepted for publication April 18, 2002.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
. Mol Endocrinol 11:1486–1496 transcriptional activity and is a key regulator of the cellular response to estrogens and antiestrogens. Endocrinology 140:5566–5578 and ERß. Proc Natl Acad Sci USA 96:3999–4004 and ß. Mol Cell Biol 19:8226–8239 mutant D351Y shows reduced tamoxifen-dependent interaction with corepressor complexes. J Biol Chem 276:42684–42691 and ERß. J Biol Chem 274:345–353
(PPAR) mutant is a constitutive repressor and inhibits PPAR-mediated adipogenesis. J Biol Chem 275:5754–5759 that lacks the activation function 2 core domain. J Biol Chem 276:37280–37283NURSA Molecule Pages Link:
This article has been cited by other articles:
 |
|
 |
|
  R. Safi, G. G. Muramoto, A. B. Salter, S. Meadows, H. Himburg, L. Russell, P. Daher, P. Doan, M. D. Leibowitz, N. J. Chao, et al. Pharmacological Manipulation of the RAR/RXR Signaling Pathway Maintains the Repopulating Capacity of Hematopoietic Stem Cells in Culture Mol. Endocrinol., February 1, 2009; 23(2): 188 - 201. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. J Kojetin, T. P Burris, E. V Jensen, and S. A Khan Implications of the binding of tamoxifen to the coactivator recognition site of the estrogen receptor Endocr. Relat. Cancer, December 1, 2008; 15(4): 851 - 870. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. D. DuSell, M. Umetani, P. W. Shaul, D. J. Mangelsdorf, and D. P. McDonnell 27-Hydroxycholesterol Is an Endogenous Selective Estrogen Receptor Modulator Mol. Endocrinol., January 1, 2008; 22(1): 65 - 77. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. M. Wittmann, A. Sherk, and D. P. McDonnell Definition of Functionally Important Mechanistic Differences among Selective Estrogen Receptor Down-regulators Cancer Res., October 1, 2007; 67(19): 9549 - 9560. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. B. Mettu, T. B. Stanley, M. A. Dwyer, M. S. Jansen, J. E. Allen, J. M. Hall, and D. P. McDonnell The Nuclear Receptor-Coactivator Interaction Surface as a Target for Peptide Antagonists of the Peroxisome Proliferator-Activated Receptors Mol. Endocrinol., October 1, 2007; 21(10): 2361 - 2377. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Kressler, M. B. Hock, and A. Kralli Coactivators PGC-1beta and SRC-1 Interact Functionally to Promote the Agonist Activity of the Selective Estrogen Receptor Modulator Tamoxifen J. Biol. Chem., September 14, 2007; 282(37): 26897 - 26907. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. J. Peterson, S. Karmakar, M. C. Pace, T. Gao, and C. L. Smith The Silencing Mediator of Retinoic Acid and Thyroid Hormone Receptor (SMRT) Corepressor Is Required for Full Estrogen Receptor {alpha} Transcriptional Activity Mol. Cell. Biol., September 1, 2007; 27(17): 5933 - 5948. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. Heldring, A. Pike, S. Andersson, J. Matthews, G. Cheng, J. Hartman, M. Tujague, A. Strom, E. Treuter, M. Warner, et al. Estrogen Receptors: How Do They Signal and What Are Their Targets Physiol Rev, July 1, 2007; 87(3): 905 - 931. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Heldring, T. Pawson, D. McDonnell, E. Treuter, J.-A. Gustafsson, and A. C. W. Pike Structural Insights into Corepressor Recognition by Antagonist-bound Estrogen Receptors J. Biol. Chem., April 6, 2007; 282(14): 10449 - 10455. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. K S Ku and D. L Crowe Coactivator-mediated estrogen response in human squamous cell carcinoma lines J. Endocrinol., April 1, 2007; 193(1): 147 - 155. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Folkertsma, P. I. van Noort, A. de Heer, P. Carati, R. Brandt, A. Visser, G. Vriend, and J. de Vlieg The Use of in Vitro Peptide Binding Profiles and in Silico Ligand-Receptor Interaction Profiles to Describe Ligand-Induced Conformations of the Retinoid X Receptor {alpha} Ligand-Binding Domain Mol. Endocrinol., January 1, 2007; 21(1): 30 - 48. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Gaillard, M. A. Dwyer, and D. P. McDonnell Definition of the Molecular Basis for Estrogen Receptor-Related Receptor-{alpha}-Cofactor Interactions Mol. Endocrinol., January 1, 2007; 21(1): 62 - 76. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Wang, W. J. Zuercher, T. G. Consler, M. H. Lambert, A. B. Miller, L. A. Orband-Miller, D. D. McKee, T. M. Willson, and R. T. Nolte X-ray Crystal Structures of the Estrogen-related Receptor-{gamma} Ligand Binding Domain in Three Functional States Reveal the Molecular Basis of Small Molecule Regulation J. Biol. Chem., December 8, 2006; 281(49): 37773 - 37781. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Cui, A. Niu, R. Pestell, R. Kumar, E. M. Curran, Y. Liu, and S. A. W. Fuqua Metastasis-Associated Protein 2 Is a Repressor of Estrogen Receptor {alpha} Whose Overexpression Leads to Estrogen-Independent Growth of Human Breast Cancer Cells Mol. Endocrinol., September 1, 2006; 20(9): 2020 - 2035. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Stossi, V. S. Likhite, J. A. Katzenellenbogen, and B. S. Katzenellenbogen Estrogen-occupied Estrogen Receptor Represses Cyclin G2 Gene Expression and Recruits a Repressor Complex at the Cyclin G2 Promoter J. Biol. Chem., June 16, 2006; 281(24): 16272 - 16278. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Fowler, N. M. Solodin, C. C. Valley, and E. T. Alarid Altered Target Gene Regulation Controlled by Estrogen Receptor-{alpha} Concentration Mol. Endocrinol., February 1, 2006; 20(2): 291 - 301. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Hall and D. P. McDonnell Coregulators in Nuclear Estrogen Receptor Action: From Concept to Therapeutic Targeting Mol. Interv., December 1, 2005; 5(6): 343 - 357. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. K. Keeton and M. Brown Cell Cycle Progression Stimulated by Tamoxifen-Bound Estrogen Receptor-{alpha} and Promoter-Specific Effects in Breast Cancer Cells Deficient in N-CoR and SMRT Mol. Endocrinol., June 1, 2005; 19(6): 1543 - 1554. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R.-C. Wu, C. L. Smith, and B. W. O'Malley Transcriptional Regulation by Steroid Receptor Coactivator Phosphorylation Endocr. Rev., May 1, 2005; 26(3): 393 - 399. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. H Steel, R. White, and M. G Parker Role of the RIP140 corepressor in ovulation and adipose biology J. Endocrinol., April 1, 2005; 185(1): 1 - 9. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. H. Kong, N. Heldring, J.-A. Gustafsson, E. Treuter, R. E. Hubbard, and A. C. W. Pike Delineation of a unique protein-protein interaction site on the surface of the estrogen receptor PNAS, March 8, 2005; 102(10): 3593 - 3598. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. H. Herynk and S. A. W. Fuqua Estrogen Receptor Mutations in Human Disease Endocr. Rev., December 1, 2004; 25(6): 869 - 898. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
G. A. Finlay, B. York, R. H. Karas, B. L. Fanburg, H. Zhang, D. J. Kwiatkowski, and D. J. Noonan Estrogen-induced Smooth Muscle Cell Growth Is Regulated by Tuberin and Associated with Altered Activation of Platelet-derived Growth Factor Receptor-{beta} and ERK-1/2 J. Biol. Chem., May 28, 2004; 279(22): 23114 - 23122. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. A. Iannone, C. A. Simmons, S. H. Kadwell, D. L. Svoboda, D. E. Vanderwall, S.-J. Deng, T. G. Consler, J. Shearin, J. G. Gray, and K. H. Pearce Correlation between in Vitro Peptide Binding Profiles and Cellular Activities for Estrogen Receptor-Modulating Compounds Mol. Endocrinol., May 1, 2004; 18(5): 1064 - 1081. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. Heldring, M. Nilsson, B. Buehrer, E. Treuter, and J.-A. Gustafsson Identification of Tamoxifen-Induced Coregulator Interaction Surfaces within the Ligand-Binding Domain of Estrogen Receptors Mol. Cell. Biol., April 15, 2004; 24(8): 3445 - 3459. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
X.-f. Liu and M. K. Bagchi Recruitment of Distinct Chromatin-modifying Complexes by Tamoxifen-complexed Estrogen Receptor at Natural Target Gene Promoters in Vivo J. Biol. Chem., April 9, 2004; 279(15): 15050 - 15058. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. L. Smith and B. W. O'Malley Coregulator Function: A Key to Understanding Tissue Specificity of Selective Receptor Modulators Endocr. Rev., February 1, 2004; 25(1): 45 - 71. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Caporali, M. Imai, L. Altucci, M. Cancemi, S. Caristi, L. Cicatiello, F. Matarese, R. Penta, D. K. Sarkar, F. Bresciani, et al. Distinct Signaling Pathways Mediate Stimulation of Cell Cycle Progression and Prevention of Apoptotic Cell Death by Estrogen in Rat Pituitary Tumor PR1 Cells Mol. Biol. Cell, December 1, 2003; 14(12): 5051 - 5059. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. J. Morrison, R. E. Herrera, E. C. Heinsohn, R. Schiff, and C. K. Osborne Dominant-Negative Nuclear Receptor Corepressor Relieves Transcriptional Inhibition of Retinoic Acid Receptor but Does Not Alter the Agonist/Antagonist Activities of the Tamoxifen-Bound Estrogen Receptor Mol. Endocrinol., August 1, 2003; 17(8): 1543 - 1554. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
E. R. Levin Bidirectional Signaling between the Estrogen Receptor and the Epidermal Growth Factor Receptor Mol. Endocrinol., March 1, 2003; 17(3): 309 - 317. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
Y. Wu, W. W. Chin, Y. Wang, and T. P. Burris Ligand and Coactivator Identity Determines the Requirement of the Charge Clamp for Coactivation of the Peroxisome Proliferator-activated Receptor gamma J. Biol. Chem., February 28, 2003; 278(10): 8637 - 8644. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. Webb, P. Nguyen, and P. J. Kushner Differential SERM Effects on Corepressor Binding Dictate ERalpha Activity in Vivo J. Biol. Chem., February 21, 2003; 278(9): 6912 - 6920. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Shiozawa, H.-C. Shih, T. Miyamoto, Y.-Z. Feng, J. Uchikawa, K. Itoh, and I. Konishi Cyclic Changes in the Expression of Steroid Receptor Coactivators and Corepressors in the Normal Human Endometrium J. Clin. Endocrinol. Metab., February 1, 2003; 88(2): 871 - 878. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
