| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Pathology, University of Southern California, Los Angeles, California, USA
2 C&C Research Laboratories, Kyunggi-do, Korea
3 Department of Biochemistry and Molecular Biology, University of Southern California, Los Angeles, California, USA
(Requests for offprints should be addressed to Jeong Hoon Kim, Department of Pathology, HMR 301, University of Southern California, 2011 Zonal Avenue, Los Angeles, California 90089-9092, USA; Email: jeongkim{at}usc.edu.)
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and methodsResultsDiscussionReferences |
|---|
(ER) depends on the conformation of helices 3 and 12 in the ligand-binding domain. To better understand the function of helix 3 in ER, we examined the role of charged residues, which are conserved in most steroid receptors in helix 3, in estrogen-dependent transactivation. The replacement of Asp-351 with lysine (D351K) or leucine (D351 L) completely abolished estrogen-dependent transactivation without affecting estrogen-binding, DNA-binding and homodimerization activities in ER. The mutations dramatically reduced the ligand-dependent activation function 2 activity and impaired the ability of ER to bind p160 coactivators. In addition, the D351K mutant effectively inhibited the transcriptional activation activity of wild-type ER. Furthermore Asp-351 was required not only for the estrogen-dependent conformational change of wild-type ER but also for the constitutive transcriptional activity and ligand-independent active conformation of ER mutant Y537N. Similarly, in the orphan nuclear receptor called estrogen-related receptor 3 (ERR3), the replacement of Asp-273 (the corresponding amino acid to Asp-351 in ER) with lysine abolished constitutive transcriptional activity of ERR3 without affecting DNA-binding activity and impaired the ability of the receptor to interact with p160 coactivators. These data suggest a role of Asp-351 in inducing and stabilizing the active conformation of ER, and our results experimentally confirm the concept that Asp-351 in helix 3 interacts with the amide hydrogen of L540 in helix 12 to form a transcriptionally competent surface for binding p160 coactivators.
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
and ERß, which share a high degree of homology in their DBD and LBD but contain divergent N-terminal domains (Kuiper et al. 1996). Transcriptional activation is mediated by two transcription activation domains, AF-1 and AF-2. The AF-2 activity is dependent on ligand binding, whereas the AF-1 activity is hormone-independent, but can be regulated by phosphorylation of specific serine residues (Kato et al. 1995). The ligand-dependent activation by AF-2 requires ligand-dependent association with coactivator complexes including p160 coactivators. The p160 coactivators, which include SRC-1, GRIP1 and AIB1, can interact with most NRs in a ligand-dependent manner, using NR boxes having the sequence LXXLL, and potentiate transcription of their specific target genes (Onate et al. 1995, Anzick et al. 1997, Heery et al. 1997, Hong et al. 1997, Darimont et al. 1998). The crystal structures of the LBDs of the ERs and other NRs, have revealed a conserved helical fold with 12-helices (numbered H1 to H12), and have shown that the most striking difference between unliganded (apo-) and liganded (holo-) receptors is the position of the H12 (Bourguet et al. 1995, Wurtz et al. 1996, Brzozowski et al. 1997, Moras & Gronemeyer 1998, Shiau et al. 1998, Pike et al. 1999, Kallenberger et al. 2003). In apo-retinoid X receptor and apo-peroxisome proliferator-activated receptor 
crystal structures, the helix H12 projects away from the LBD or is highly mobile in the absence of ligand but is immobilized and tightly packed against H3 of the LBD upon ligand binding (Bourguet et al. 1995, Kallenberger et al. 2003). The realignment of H12 over the entrance to the ligand-binding pocket after ligand binding generates transcriptionally active receptors by creating a surface that allows binding of p160 coactivators (Shiau et al. 1998). Together with H12, Lys-362 at the C-terminal end of H3 forms a hydrogen bond with the LXXLL -helix (Mak et al. 1999, Shiau et al. 1998).
ER is of particular interest because its ligands have important biological effects on the development and function of tissues of the reproductive, bone, liver and cardiovascular systems (Hall et al. 2001). The beneficial effects of estrogens in bone maintenance, on the blood lipid profile, and in the cardiovascular system, account for their importance in hormone replacement in postmenopausal women and underscore the physiological role of ER function in these systems. Estrogen, however, also stimulates the development and growth of breast cancers, especially those hormone-responsive tumors containing significant levels of the ER protein (Speirs et al. 1999). Thus, it is important to understand in detail how the ER is activated by ligand binding.
The actions of estrogens are antagonized by anti-estrogens, which compete with estrogen for binding to the ERs. Anti-estrogens vary in their biological actions. Selective ER modulators (SERMs) such as tamoxifen (TAM) and raloxifene (RAL) act as mixed or partial agonists/antagonists with the degree of antagonist activity dependent upon the cell type and promoter context (Jordan & Morrow 1999, Osborne et al. 2000). Other anti-estrogens such as ICI-182780 appear to be pure/complete antagonists (Howell 2000). The crystal structures of ER show a different final position of H12 when anti-estrogens are bound (Brzozowski et al. 1997, Shiau et al. 1998). The backbones of TAM and RAL concisely bind to the Glu-353 and His-524 that lock estradiol in the LBD. Interestingly, the side chains of TAM and RAL project out of the hydrophobic pocket of the LBD and the nitrogen in the side chain forms a hydrogen bond with Asp-351. This interaction between Asp-351 and the side chain of TAM and RAL plays an important role in displacing H12 and inducing the transcriptionally inactive conformation of this domain. Although clear proof of the essential role of Asp-351 in the anti-estrogenic action of TAM and RAL was obtained by crystallographic and biochemical studies (Wolf & Jordan 1994, Brzozowski et al. 1997, Levenson & Jordan 1998, Shiau et al. 1998, Liu et al. 2002), the precise role of Asp-351 in estrogen-dependent transcriptional activation of ER remains obscure.
In this study, we introduced several mutations at charged residues in helix 3 and determined the functional consequences of these mutations on the ligand-dependent transactivation properties of ER. Our data confirmed that Lys-362 is required for ligand-dependent activation, and the K362A mutant exerts a dominant negative effect on the activity of wild-type ER. Asp-351 is also important for ligand-dependent transcription and the interaction of ER with p160 coactivators. Similar results were observed with constitutively active ER mutant Y537N and the constitutively active orphan NR called estrogen-related receptor 3 (ERR3). Specifically, Asp-351 is required for the active conformation of wild-type ER and constitutively active ER mutant Y537N. These data indicate that Asp-351 plays an important role in ligand-dependent intramolecular folding of ER LBD. Our observation may also help explain the anti-estrogenic mechanism of SERMs.
|
Materials and methods |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
The following plasmids have been previously described: p2 ERE-TK-LUC (Kim et al. 2003), pHEG0 (Kim et al. 2003), pSG5.HA-ERR3 (Hong et al. 1999) and pGEX-GRIP1(563–1121) (Ma et al. 1999). The ERß cDNA was amplified from a human brain total RNA (Ambion) by RT-PCR and cloned into BamHI and XhoI sites of pcDNA3.1-V5/His (Invitrogen). Mutations were introduced in the ER cDNA (HEG0), ERß, or ERR3 cDNA by site-directed mutagenesis using the GeneEditor in vitro oligonucleotide-directed mutagenesis system (Promega). The sequences of oligonucleotides used for mutagenesis are available upon request. The pCR3.1 full-length hSRC-1a expression plasmid was kindly provided by Drs M-J Tsai and B W O’Malley (Baylor College of Medicine, Houston, TX, USA). The ER LBDs (residues 301–595) or ERR3 LBDs (residues 214–458) were amplified with primers incorporating BamHI and KpnI or BamHI and EcoRV sites respectively, and inserted into the GAL4 DBD(1–147) expression vector pBIND (Promega) or the VP16 activation domain (411–456) expression vector pACT (Promega). Similarly, GRIP1(563–767) and SRC-1(395–879) were amplified by PCR for insertion into the pBIND plasmid with BamHI and KpnI restriction sites. SRC1(395–879) was also cloned into BamHI and KpnI sites of the pGEX-5X-1 (Amersham Pharmacia Biotech).
Cell culture and transient transfections
COS-7, COS-1, HeLa and CV-1 cells were maintained in Dulbecco’s modified Eagle’s medium (DMEM) plus phenol-red with 5% fetal bovine serum (FBS). These cells were then grown in phenol-red free DMEM containing 3% dextran-coated charcoal-stripped FBS (DCFBS) for 3 days prior to transfection. For transiently transfected reporter gene assays, COS-7, CV-1 or HeLa cells were plated into 24-well plates in phenol-red free DMEM containing 3% DCFBS. Cells were transfected by using TransFast (Promega) with 200 ng p2 ERE-TK-LUC and 4 ng pRL-SV (Promega) as an internal control, and ER, ERß or ERR3 expression plasmids as indicated in the Figure legends. For dominant negative mutant assays, cells were transfected with wild-type receptor plasmids, mutant receptor plasmids, or both as indicated in the Figure legends. For mammalian one-hybrid and two-hybrid assays, cells were transfected with 200 ng pG5 LUC (Promega), 4 ng pRL-SV and 100 ng GAL4 fusion plasmid or 100 ng of each GAL4 fusion plasmid and VP16 fusion plasmid. In some experiments pSG5-GRIP1 was added as indicated in the Figure legends. The overall amount of total DNA in all transfections was equalized (500 or 600 ng) by adding appropriate amounts of the empty vector, pSG5. After 2 h transfection, the cells were treated with phenol-red free DMEM containing 3% DCFBS and either the ethanol vehicle or various concentration of 17ß-estradiol (E2) as indicated in the figure legends for 24 or 48 h.
For biochemical analysis (hormone-binding assays and western blots), the wild-type and mutant receptors were expressed in COS-1 cells. Cells were plated at 1.0 x 106 cells per 100-mm dish and transfected by the calcium phosphate coprecipitation method. Transfections were performed using 10 µg ER or ERR3 expression vector, 2.5 µg pAdVantage (Promega), and 2.5 µg pGEM-3Z (Promega) carrier plasmid. After 48 h incubation, whole-cell extracts were prepared in TEG (20 mM Tris–HCl (pH 7.4), 1 mM EDTA, 0.5 mM EGTA, 0.15 M NaCl, 1 mM dithiothreitol (DTT), 10% glycerol, 1 mM Na3VO4, 0.5 mM leupeptin and 0.2 mM phenylmethylsulfonyl fluoride) buffer. The protein contents of cell extracts were determined by a colorimetric method (Bio-Rad). Western blotting was performed as described previously (Kim et al. 2003).
Hormone-binding assays
COS-1 cells transfected with either wild-type or mutant receptor expression vectors were harvested in Hanks’ balanced salt solution containing 1.2 mM EDTA, washed once in ice-cold TEG buffer and homogenized. The homogenate was centrifuged at 180.000 g (30 min at 4 °C) to yield the whole cell extract. Aliquots of the whole cell extract (50 µl, 1 mg/ml protein) were incubated at 25 °C for 2 h with [3H]estradiol (Amersham Pharmacia Biotech) at increasing concentrations from 3.25 x 10–11 to 4 x 10–9 M. Non-specific binding was determined in the presence of a 100-fold excess of unlabeled ligand. Unbound E2 was removed from the samples by treatment with dextran-coated charcoal for 10 min at 4 °C. An aliquot of each supernatant containing bound ligand was withdrawn for liquid scintillation counting. Affinity of wild-type or mutant receptors for E2 was determined by the method of Scatchard (1949) from at least three independent experiments.
Mobility shift assays
Gel retardation assays were performed with the in vitro translated receptors and the oligonucleotide corresponding to a 35 bp fragment of the vitellogenin A2 gene promoter containing a consensus ERE (sense strand, 5'-gtccaaagtcaGGTCAcagTGACCtgatcaaagtt-3'). ER, ERR3 and their mutants were translated in vitro by using a TNT-Quick Coupled Transcription/Translation System (Promega). Translation efficiency was determined by western blotting using anti-ER antibody G-20 (Santa Cruz Biotechnology) or anti-HA antibody 3F10 (Roche). For the binding reaction, the proteins were incubated on ice for 15 min in a buffer (10 mM Tris–HCl (pH 7.4), 1 mM DTT, 5% glycerol, 100 µg/ml BSA, 80 mM KCl) containing 1 µg of poly(dI-dC) and protease inhibitor cocktail (Roche) in the presence of ethanol or 100 nM E2 and then for 30 min at room temperature with approximately 200 fmol double-stranded oligonucleotide end-labeled with digoxigenin (DIG) (Roche). DNA–protein complexes were resolved on 6% polyacrylamide gels in 0.5 x TBE buffer and visualized by chemiluminescence method (DIG Gel Shift Kit; Roche).
Glutathione-S-transferase (GST) pull-down assays
NR interaction domains of GRIP1 and SRC-1a were expressed as GST fusion proteins in Escherichia coli according to the manufacturer’s instructions. The expression of correctly sized proteins was monitored by SDS-PAGE. For GST pull-down assays, bacterially expressed GST or GST fusion proteins were bound to glutathione-Sepharose 4B beads (Amersham Pharmacia Biotech) for 2 h at 4 °C. Beads containing either GST alone or GST fusion proteins were incubated with in vitro translated, 35S-labeled receptors in the presence or absence of 100 nM E2 in NET-N buffer (50 mM Tris–HCl (pH 7.5), 5 mM EDTA, 0.3 M NaCl, 1 mM DTT, 0.01% NP-40) containing protease inhibitor cocktail (Roche). Free proteins were washed away from the beads with NET-N buffer. Bound proteins were extracted into SDS sample buffer and separated by SDS-PAGE. The gels were fixed with 50% (vol/vol) methanol–10% (vol/vol) acetic acid for 30 min, then soaked with Amplify (Amersham Pharmacia Biotech) for 30 min, and vacuum dried. Fluorography was performed overnight.
Limited proteolytic digestion assays
Aliquots of in vitro translated, 35S-labeled proteins (24 µl) were treated with control (10% ethanol) vehicle or E2 at a final concentration of 10 µM for 20 min at room temperature. Aliquots (4 µl) of the receptor were incubated without or with trypsin to a final concentration of 4, 8, 10, 12 or 16 µg/ml (Sigma). After 10 min incubation at room temperature, the digestion was stopped with SDS sample buffer, and the samples were boiled and then separated on a 4–20% or 10–20% Tris–glycine gel (Novex). The gels were fixed, dried and subjected to fluorography.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
As shown in Fig. 1
, the positions of most of the charged amino acids contained in helix H3 of the LBD are conserved among steroid hormone receptors. These charged residues in helix H3 of ER are also present in orphan NRs ERR1, ERR2 and ERR3 (Hong et al. 1999). To assess the contribution of these residues to transcriptional activation by ER, a series of point mutants in H3 of human ER (amino acid residues Asp-351 and Arg-352) were created by oligonucleotide-directed mutagenesis of the ER cDNA. Additionally, Lys-362, which is extremely well conserved among NRs, was changed to alanine (K362A). This lysine residue has been previously shown to contribute to transactivation and to be essential for interaction with coactivators (Henttu et al. 1997, Mak et al. 1999).
|
mutants containing 11 different amino acid substitutions for Asp-351 of the human ER. All mutant and wild-type receptors were expressed at similar levels in a Western blot analysis (data not shown). We then tested the ability of the ER mutants to activate transcription in response to E2. Transcriptional activity was measured in COS-7 cells transfected with an ER expression vector and an estrogen responsive promoter–reporter gene construct, p2 ERE-TK-LUC, containing two EREs, basal TK gene promoter, and luciferase reporter gene (Kim et al. 2003). Activity was measured over a range of ligand concentrations (1 x 10–13 to 1 x 10–5 M) to assess the potency and efficacy of E2 to induce transcriptional activity of the wild-type and mutant ERs. For wild-type ER, maximal activity was reached at the E2 concentration of 1 x 10–8 M. All of the mutations proved to be dose-shifted in their response to E2, requiring elevated levels of E2 relative to wild-type ER in order to reach half-maximal and maximal activities (Fig. 2A–D). Of these mutants, the D351K, D351 L and D351R were inactive at all E2 concentrations tested (Fig. 2A).
|
activity (Fig. 2E
). Alignment of human ER and ERß indicated that Asp-351 in ERß corresponds to Asp-303 of ERß. We next replaced Asp-303 of ERß with lysine, leucine, alanine, glycine, tyrosine and asparagine. Indeed, the transcriptional activities of ERß mutant receptors were almost identical to those of ER mutants (Fig. 2F and data not shown). The D303K and D303 L mutants displayed negligible transcriptional activity. Thus we conclude that Asp-351, but not Arg-352, is required for transcriptional activation by ER. Mutation of Asp-351 does not affect E2 or DNA binding
To establish whether Asp-351 was directly involved in transactivational activation, we investigated whether the mutation affected its ability to bind E2 or DNA. The DNA-binding properties of the mutant receptors were determined in gel retardation assays with the vitellogenin A2 consensus ERE sequence and in vitro translated ERs. The mutant receptors retained their ability to bind to DNA (Fig. 3). Point mutations, D351 L and K362A, resulted in weaker interactions with the ERE. Importantly, D351K mutant bound DNA as strongly as wild-type receptor. We next analyzed their E2-binding activities. The affinities of the mutant receptors for E2 were determined by saturation binding and Scatchard analysis. The cell extracts were incubated with 3.25 x 10–11 to 4 x 10–9 M [3H]estradiol in the presence or absence of excess radioinert E2. The dissociation constant (Kd) values are given in Table 1. The Kd determined for wild-type receptor was 0.036 nM. The Kd values of most ER mutants were very similar to that of wild-type receptor (2- to 3-fold reduction), whereas D351E and D351 L had 4.8- and 10.5-fold reduced E2-binding affinity respectively. Thus, although modest reductions in binding affinity were observed for mutant receptors, all of the receptors would be expected to be saturated with the hormone at the concentration used in the transiently transfected reporter gene assays, namely, 10–8 M.
|
|
We also tested whether the mutation interferes with homodimerization of ER in mammalian two-hybrid assays. The Asp-351 mutants showed strong homodimerization and heterodimerization with wild-type receptor in an E2-dependent manner, similar to the wild-type ER (data not shown). Given that the mutations do not significantly affect these properties of the receptor, we concluded that the Asp-351 mutations do not interfere with dimerization, E2-binding and DNA-binding activities of ER.
Dominant negative effect of mutant receptors
Having established the functional properties of the receptor mutants, we next assessed the ability of ER mutants to interfere with transcriptional activation by wild-type ER. We conducted experiments in which plasmids encoding the wild-type and mutant ER were introduced separately or together into mammalian cells and assayed for the ability to transactivate an ERE-containing reporter gene. The dominant negative activity was assessed at 10 nM E2, which is a saturating dose for wild-type and mutant ER (Fig. 4).
|
alone in both cell lines. When an 8-fold molar excess of mutant over wild-type ER was employed, E2-stimulated transcriptional activity was further reduced to 35% of E2-stimulated wild-type ER, again in both cell lines. Another ER H3 mutant, K362A, also exhibited dominant negative activity. These data suggest that ER H3 mutants D351K and K362A are potent repressors of E2-stimulated ER activity. Asp-351 is required for AF-2 activity
To test whether the disruptive effect of the Asp-351 mutation on transcriptional activation by ER was restricted to AF-2 or dependent on the presence of AF-1, we fused the wild-type and mutated LBDs to the DBD of GAL4 to generate GAL4-LBD-WT, GAL4-LBD-D351K, GAL4-LBD-D351 L and GAL4-LBD-K362A. The ability of GAL4-LBD-WT to stimulate transcription of a GAL4 reporter in COS-7 cells was dependent on the presence of E2 (Fig. 5). In contrast, GAL4-LBD-D351 mutants were unable to activate transcription, in accord with the results obtained with full-length receptors (Fig. 2). We also tested for transcriptional activation by GAL4-LBD-K362A mutant. As expected, GAL4-LBD-K362A showed low activity. These data suggest that Asp-351, as well as Lys-362, is required for AF-2 activity in the ER.
|
Previous works have shown that the LBDs of NRs are able to bind a number of coactivators and that the interaction with mutant receptors correlates with their transcriptional activity. Therefore, we tested whether the difference in transcriptional activity was correlated with the ability of these mutants to bind p160 coactivators in GST pull-down experiments. Full length wild-type ER bound strongly to GST-GRIP1(563–1121) and GST-SRC-1(395–879) in an E2-dependent manner (Fig. 6A). In contrast, little if any interaction was detected with the liganded helix 3 mutants. No interaction was observed with GST. Thus, ligand-dependent interactions between H3 mutants and p160 coactivators were dramatically impaired compared with wild-type ER.
|
and mutant receptors to interact in vivo with p160 coactivators (Fig. 6B). We fused GRIP1(563–767) or SRC-1(395–879) to the GAL4 DBD and the LBD(301–595) of the wild-type or mutant ER to the VP16 acidic activation domain. Upon transient transfection into COS-7 cells, a ligand-dependent interaction between ER and p160 coactivator proteins (GRIP1 and SRC-1) caused activation of a GAL4 reporter gene. The H3 mutants had little or no interaction with coactivators. Control mammalian two-hybrid assays showed that the wild-type and H3 mutant LBDs had equivalent abilities to homodimerize (data not shown). These findings indicate that H3 mutants, D351K, D351 L and K362A, despite retaining the ability to bind hormone and DNA and to homodimerize, are selectively impaired in interaction with coactivator proteins. Thus, Asp-351 seems to be essential for AF-2 activity, due to its participation in coactivator binding.
In view of the differences in the ability of wild-type and helix 3 mutants to interact with p160 coactivators, we compared the ability of p160 coactivators to stimulate transcriptional activation by wild-type and mutant receptors in COS-7 cells. The E2-stimulated wild-type receptor activity was enhanced by increasing amounts of GRIP1 (Fig. 6C). In contrast, the transcriptional activities of helix 3 mutants were not significantly affected by expression of GRIP1, although their activities were slightly restored at the highest concentration of GRIP1. These findings also demonstrate that Asp-351 is a critical residue in ER LBD for mediating activated transcription through interaction with p160 coactivators.
Partial proteolytic analysis of wild-type and mutant receptors
Ligand-induced conformational changes in steroid receptors are critical for their ability to bind coactivators. These conformational changes also result in an increased resistance to limited proteolytic digestion (Lazennec et al. 1997). We therefore tested for differences in proteolytic digestion patterns as an indication of conformational differences between mutant and wild-type ER species.
35S-labeled wild-type and mutant ERs were digested to small peptides for 10 min with increasing amounts of trypsin (Fig. 7). In the absence of ligand, ER is highly sensitive to trypsin. In agreement with previous findings (Lazennec et al. 1997), limited treatment of wild-type receptor with trypsin in the absence of ligand produced a proteolytic digestion pattern in which the fragment sizes decreased rapidly with increasing concentrations of trypsin, until two bands (28 and 30 kDa) of approximately equal intensity appeared and remained relatively stable. When the receptor was occupied with E2, only the upper band (30 kDa) was strongly stabilized, suggesting that ligand-occupied receptor is protected from further digestion by trypsin. In addition, a transiently stabilized fragment of about 38 kDa appeared with E2 treatment when receptor was digested at lower concentrations of trypsin.
|
. In contrast, the D351K and D351 L mutants were more sensitive to trypsin. The transiently resistant fragment of 38 kDa was almost completely absent. These results suggest that Asp-351 is required for the correct E2-dependent intramolecular folding of ER LBD, and the inability of the D351K mutant to activate transcription is therefore a consequence of an altered receptor conformation which does not allow coactivator binding. In contrast, Lys-362 may not be involved in E2-dependent folding of the LBD but rather makes direct contact with the coactivator.
Suppression of constitutive activity of ER mutant Y537N by Asp-351 mutation
To confirm that mutations at position Asp-351 can induce a non-productive conformation and repress transcriptional activity of ER in the presence of E2, we introduced mutation D351K into the Y537N mutant receptor. Y537N is a naturally occurring and constitutively active ER mutant that can recruit p160 coactivators in the absence of hormone (Zhang et al. 1997, Tremblay et al. 1998). The constitutive activity of the Y537N ER mutant can be further stimulated by E2 to a level essentially equal to that of wild-type receptor (Fig. 8A). Interestingly, introduction of the D351K mutation to generate the double mutation, D351K/Y537N, abolished the high level of basal activity and hormone-independent GRIP1 binding observed with constitutively active Y537N receptor, but retained partial E2-dependent transcriptional activity and GRIP1 binding (Fig. 8A and B).
|
(Fig. 7). The transiently resistant band at 38 kDa, which in wild-type receptor is stabilized only in the presence of E2, is strongly stabilized in Y537N without or with E2. Surprisingly, trypsin digestion of the double mutant without and with E2 generated proteolytic patterns indistinguishable from that obtained with the native ER. These results suggest that, compared with the Y537N ER, the D351K/Y537N double mutation restores ligand-dependent conformational change of the receptor, and the overall structure of the double mutant resembles that of the wild-type receptor. These results also confirm that Asp-351 plays an important role in the active conformation of ER. Functional analysis of H3 and H12 ERR3 mutants
ERRs are constitutively active orphan NRs closely related to ERs. Despite their high homology with ERs, ERRs do not respond to E2. Recently, the crystal structure of ERR3 has been solved and revealed the mechanism by which the active conformation is stabilized in the absence of ligand (Greschik et al. 2002). The conserved aspartate in H3 is also present in ERRs (Fig. 1). In the crystal structure, Asp-273 is predicted to interact with the amide of the peptide bond between Leu-449 and Leu-450 in H12. To elucidate the functional role of this aspartate residue in the constitutive transcriptional actions of the ERRs, we introduced a D273K mutation (corresponding to D351K mutation in ER) into ERR3. Additionally, we also tested the functional roles of K284 and E452 residues (corresponding to K362 and E542 in ER), which are extremely conserved among NRs and have been shown to form hydrogen bonds with the LXXLL-helix of SRC-1 NR box II in the crystal structures of ERR3 (Greschik et al. 2002).
We examined the transcriptional activity of wild-type and mutant ERR3s in CV-1 cells (Fig. 9A). ERR3 activated a 2 ERE-TK-LUC reporter gene expression in a dose-dependent manner. In contrast, ERR3 mutants, D273K, K284A and E452Q, were inactive at all expression plasmid doses tested. When they were coexpressed with the wild-type ERR3, the ERR3 mutants functioned as dominant negative mutants by repressing transcriptional activity of ERR3 (Fig. 9B). Since the transcriptional activity of ERR3 is almost entirely dependent on its AF-2 domain (Hong et al. 1999), we next tested whether these amino acids are required for AF-2 activity of ERR3 (Fig. 9C). The wild-type and mutant ERR3 LBDs were fused to GAL4 DBD and their transcriptional activities were tested in COS-7 cells. The wild-type ERR3 LBD activated transcription and, as expected, mutant ERR3 LBDs failed to activate transcription. These results suggest that the well-conserved, charged amino acids in helices H3 and H12 are also essential for the ligand-independent AF-2 activity of the ERR3. The wild-type and mutant ERR3s were tested in gel shift assays for DNA-binding activity (Fig. 9D). The DNA-binding activities of ERR3 mutants were similar to or stronger than that of wild-type receptor, indicating that the low level of activity seen in reporter gene assays was not the result of disruption of DNA binding.
|
). The ligand-independent interaction between ERR3 and GRIP1 was eliminated by D273K mutation. The mutations, K284A and E452Q, also strongly reduced or eliminated the interaction between ERR3 and GRIP1. These results showed a strong correlation between the abilities of ERR3 to activate transcription and bind coactivators. Taken together, these charged amino acids (D351, K362 and E542 in ER and D273, K284 and E452 in ERR3) are required for and functionally conserved in ligand-dependent and ligand-independent transactivation in ER and ERR3. |
Discussion |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
(Henttu et al. 1997, Mak et al. 1999). H3 contains Lys-362, a conserved residue in NRs, which is an important part of the interaction surface for many coactivators. In the structure of the agonist-bound ER LBD, Lys-362 and Glu-542 in helix H12 were shown to form hydrogen bonds with the polypeptide backbone of NR box II of GRIP1 (Shiau et al. 1998). In this study we confirmed the requirement of Lys-362 for coactivator binding and E2-dependent transcriptional activation by ER, and also demonstrated a dominant negative effect of the K362A mutant on the transcriptional activity of the wild-type receptor. In addition, we demonstrated that the conserved lysine (Lys-284) in H3 and glutamate (Glu-452) in H12 are essential for transactivation by ERR3 and interaction with p160 coactivators.
Also from the crystal structure of ER bound to E2, Asp-351 in H3 is predicted to interact with the amide of the peptide bond between Leu-539 and Leu-540 in H12 (Fig. 10). We tested this prediction experimentally, and our data indicate the precise role for this interaction in E2-induced transactivation. D351 is essential for ligand-dependent coactivator binding and transactivation by ER. Replacement of this aspartic acid with lysine or leucine dramatically reduced AF-2 activity (Fig. 5) without affecting homodimerization (data not shown), ligand binding (Table 1) or DNA binding (Fig. 3). Consistent with these results, Asp-273, which is the corresponding residue in H3 of ERR3, is also required for the transcriptional activity of the receptor and interaction with p160 coactivators; furthermore the D273K mutant acted as a dominant negative receptor and repressed the activity of wild-type ERR3.
|
). The transcriptional activities and coactivator-binding activities of D351N and D351Q mutants were reduced relative to those of wild-type receptor (Fig. 2 and data not shown). This suggests that the negative charge of Asp-351 is required for optimal transcriptional activity. When Asp-351 was replaced by leucine, with a side chain that mimics the aliphatic stem of lysine but lacks the terminal positive charge, the transcriptional activity of the D351 L was abolished. Asp-351 is required for the interaction with SRC-1 and GRIP1. Our in vitro and in vivo interaction assays indicate that in the presence of E2 the p160 coactivators fail to associate in a ligand-dependent manner with ER containing D351K and D351 L substitutions (Fig. 6A and B). Functional studies also show that overexpression of GRIP1 enhanced ligand-dependent transcriptional activation by ER and that Asp-351 is essential for this function (Fig. 6C). This defect can explain the lack of E2-dependent transcriptional activation and the dominant negative activity of D351K. Taken together, these results suggest that the exact length and the terminal negative charge of the Asp-351 side chain are involved in mediating the AF-2 activity of the receptor. In contrast, Arg-352, another well-conserved charged residue in H3 of steroid receptors, is likely to be dispensable, since removal of the positive charge or shortening the side chain of this residue (by substituting Glu or Ala) had little effect on receptor function (Fig. 2).
Our studies reveal that not only the Lys-362 in H3 but also the conserved Asp-351 are required for interaction with p160 coactivators. It is possible that the inactivity of H3 mutants could be a consequence of an altered receptor conformation upon ligand binding. In our proteolytic digestion experiments, wild-type and K362A receptors produced similar digestion patterns, with a 38 kDa fragment becoming partially resistant to trypsin after E2 binding (Fig. 7). Given that K362A does not show any change in this property, it can be concluded that the overall structure of K362A resembles that of the wild-type receptor and that both undergo similar conformational changes after binding of E2. In contrast, the proteolytic digestion pattern of D351K and D351 L were clearly different from those of wild-type and K362A receptors. Asp-351 mutants exhibited a proteolytic digestion pattern in the presence of E2 which was very similar to the pattern of the wild-type receptor in the absence of E2, suggesting that these mutant receptors were in the conformationally inactive state in the presence of E2. These results suggest that loss of hydrogen bonding between residue 351 and the amide hydrogen of Leu-540 leads to opening of the triangle formed by H12, the bottom of H11, and the loop between them in the Asp-351 mutants. These data indicate that Asp-351 and Lys-362 play distinct roles in the transactivation process: the Asp-351 residue is involved in E2-dependent intramolecular folding of the AF-2 region; in contrast, the Lys-362 residue is not involved in intramolecular folding of the AF-2 domain, but instead functions as part of the exposed surface required for coactivator interaction.
In the conformation of the agonist-bound ER, Tyr-537 is positioned at the N-terminus of H12 and forms a hydrogen bond with Asn-348 in H3 (Fig. 10). In unliganded ER it was proposed that Tyr-537 may stabilize the transcriptionally inactive conformation by interaction with the loop between H2 and H3 of the receptor, thereby minimizing ER basal activity (Carlson et al. 1997, White et al. 1997). Conversely, replacement of tyrosine at this position with other amino acids less capable of this specific stabilization would allow ER to adopt a transcriptionally active conformation even in the absence of ligand, giving rise to constitutive activity. For example, mutation of Tyr-537 to serine produced ligand-independent activity (Weis et al. 1996, Lazennec et al. 1997), and serine in this position is postulated to form a hydrogen bond with Asp-351 based on molecular modeling (Nettles & Greene 2005). This tyrosine is conserved in all known ER and ERß sequences from diverse species but not in ERRs, which are constitutively active orphan receptors (White et al. 1997, Hong et al. 1999). This leads to the suggestion that constraining intramolecular interactions have been conserved during evolution to maintain the ERs preferentially in an inactive conformation in the absence of agonist. Conceivably, these inactivating constraints could be released as a part of the receptor activation mechanism after agonist binding, or due to specific mutations.
Introduction of mutation D351K abolishes ligand-independent transcriptional activity of the constitutively active Y537N receptor (Fig. 8A) and prevents folding of the unliganded receptor in an active conformation (Fig. 7); nevertheless, E2 binding stimulates transcriptional activity of the Y537N/D351K double mutant and stabilizes the active conformation despite the D351K mutation. Accurate positioning of H12 is crucial for recruitment of coregulators (for a review, see Nettles & Greene 2005). We propose that Tyr-537 and Asp-351 are the ‘ying’ and ‘yang’ of the active conformation of ER. In this model, Asp-351 stabilizes the active conformation of ER, while Tyr-537 stabilizes the transcriptionally inactive conformation. Disruption of the interaction between Asp-351 and H12 by mutations or by antagonist binding would be sufficient to fully destabilize the active conformation in the liganded receptor; on the other hand, E2 binding or mutations of Tyr-537 would release the inactivating constraint of the Tyr-537 intramolecular interactions with the loop between H2 and H3. Thus the conformation of E2-bound D351K or D351 L receptor mimics the unliganded state; the structure of unliganded Y537N receptor mimics the E2-activated state of the receptor; and in the double mutant the D351K and Y537N mutations compensate for each other and restore the E2-dependent transcriptional activity. These data support our hypothesis that the active conformation of ER is negatively (‘ying’) regulated by Tyr-537 in the absence of E2 and is positively (‘yang’) regulated by Asp-351 or both (Asp-351 and Tyr-537) in the presence of E2. The active positioning of H12 caused by E2 binding is brought about by a series of intramolecular interactions between H3 and H12. The hydrogen bonds between Asp-351 and the amide hydrogen of Leu-540 and between residues Tyr-537 and Asn-348 might serve as a ‘zipper’ to close the ligand-binding pocket (Fig. 10).
A previous study with five Asp-351 (Y, A, V, E and G) mutants demonstrated that mutations at this position did not affect ligand- and DNA-binding properties of the receptor, but Asp-351 is necessary for constitutive activities of Y537A and L536P mutants (Anghel et al. 2000). Our results are consistent with these previous observations and also demonstrate that replacement of Asp-351 with other hydrophobic and positively charged amino acids did not significantly affect DNA- and ligand-binding properties of the receptor (Fig. 3 and Table 1). We also confirmed the requirement of Asp-351 for the ligand-independent activity of a constitutively active ER mutant, Y537N, by making D351K/Y537N double mutations. Anghel et al. also suggested that Asp-351 was necessary for basal activity of the unliganded ER, but not for ligand-dependent transactivation, and that Asp-351 was important for the active structure of the unliganded receptor, but less important for the liganded receptor. However, our findings contradict these conclusions of Anghel et al. The activities of our D351K, D351L and D351R mutants were not induced by all E2 concentrations tested, but we could not see any significant reductions in the basal activity of these Asp-351 mutants in our reporter gene assays (Fig. 2A, Fig. 5, Fig. 6C, Fig. 8A, and data not shown). Furthermore the facts that mutation of this conserved aspartate also abolishes ligand-independent transactivation by the Y537N mutant and ERR3 and that mutation of Asp-351 disrupts the active conformations of wild-type ER and the Y537N mutant strongly support our idea that this conserved aspartate in ERs and ERRs is essential for forming the active structures of both ligand-dependent and ligand-independent receptors.
The position of H12 is dramatically different in TAM-and RAL-bound ER compared with that of H12 in E2-bound ER (Brzozowski et al. 1997, Shiau et al. 1998). The position of H12 induced by TAM and RAL blocks coactivator binding to AF-2 and renders the receptor transcriptionally inactive. TAM and RAL both possess a bulky side chain extension that protrudes out of the ligand-binding pocket near the base of H12. This extension prevents H12 from sealing the binding pocket and occupying the normal E2-induced position. Interestingly, the tertiary amine present at the end of the side chains of TAM and RAL forms a hydrogen bond with Asp-351, thereby suggesting the importance of Asp-351 for anti-estrogen action (Brzozowski et al. 1997, Levenson & Jordan 1998, Shiau et al. 1998, Levenson et al. 2001). D351Y ER, isolated from a TAM-stimulated MCF-7 breast cancer cell line, has been demonstrated to change the pharmacology of TAM and RAL from antagonists to partial agonists (Wolf & Jordan 1994, Webb et al. 2000). This D351Y mutant exhibited increased TAM- and RAL-induced activation of ERE-controlled reporter genes, and both TAM and RAL also behaved as agonists for expression of the endogenous estrogen target gene, transforming growth factor-, in MDA-MB-231 cells stably transfected with this D351Y mutant (Levenson & Jordan 1998, Levenson et al. 2001). A recent study has suggested that one of the mechanisms for the enhanced partial agonistic effect of TAM and RAL on the D351Y receptor is the reduced interaction with corepressors, NCoR and SMRT (Yamamoto et al. 2001). The D351Y mutation might change the position of H12 induced by TAM or RAL and prevent the formation of the interaction surface for corepressors.
Therefore, we postulate that Asp-351 plays two pivotal roles in ER function: its interaction with H12 is critical for the intramolecular folding of ER and the resulting completion of the coactivator interaction surface upon agonist binding; and it stabilizes the corepressor interaction surface upon antagonist binding, through its interaction with the side chain of TAM or RAL. In this light, our observations may help to explain the key role of Asp-351 in the antagonistic character of TAM and RAL. The side chains of TAM and RAL block the intramolecular interaction between Asp-351 and the amide hydrogen of Leu-540, and the interaction of Asp-351 with the side chains of TAM or RAL might inhibit the zippering up of the ligand-binding pocket and result in a transcriptionally inactive conformation of ER. This model also suggests that the anti-estrogens which can block zippering function of Asp-351 and Tyr-537 simultaneously may be more potent antagonists in breast cancer cells than TAM and RAL. GW5638, a SERM, is a TAM analog which has more potent anti-estrogenic activities in breast cancer cells and beneficial estrogenic activities (Willson et al. 1997). Recently, the crystal structure of GW5638-bound ER LBD has been solved (Wu et al. 2005). Interestingly, in the crystal structure, the side chain of GW5638 directly interacts not only with Asp-351 but also with Leu-536 and Tyr-537, resulting in a subtle alteration in orientation for H12 in the GW5638-ER LBD and TAM- or RAL-ER LBD structures. The increase in the surface hydrophobicity by the altered conformation of H12 correlates with a significant destabilization of ER in MCF-7 cells. These findings are in good agreement with the idea that anti-estrogens blocking both Asp-351 and Tyr-537 functions might be more potent anti-estrogens than SERMs inhibiting just one of them.
In conclusion, we have analyzed the functional role of Asp-351 in H3 of ER LBD. Our results provide strong evidence that Asp-351 is required for ligand-dependent coactivator interaction, transactivation and the active conformation of ER. Importantly, these data support previous conclusions from structural studies that ligand binding induces a dramatic conformational change in the LBD mediated through an intramolecular refolding of the LBD, which creates a transcriptional activation surface for p160 coactivator binding (Brzozowski et al. 1997, Shiau et al. 1998, Pike et al. 1999). Our results suggest that H3 functions in concert with H12, not only to form an active conformation through intramolecular interaction, but also to generate the hydrophilic surface for intermolecular interaction with the coactivators required for AF-2 activity. Moreover, the fact that the conserved aspartate is also required for ligand-independent transactivation by the Y537N mutant and ERR3 suggests the existence of common mechanisms for ligand-dependent and ligand-independent intra-molecular folding and transactivation in ERs and ERRs.
|
Acknowledgements |
|---|
|
References |
|---|
|
TopAbstractIntroductionMaterials and methodsResultsDiscussionReferences |
|---|
Anzick SL, Kononen J, Walker RL, Azorsa DO, Tanner MM, Guan XY, Sauter G, Kallioniemi OP, Trent JM & Meltzer PS 1997 AIB1, a steroid receptor coactivator amplified in breast and ovarian cancer. Science 277 965–968.
Benko S, Love JD, Beladi M, Schwabe JW & Nagy L 2003 Molecular determinants of the balance between co-repressor and co-activator recruitment to the retinoic acid receptor. Journal of Biological Chemistry 278 43797–43806.
Bourguet W, Ruff M, Chambon P, Gronemeyer H & Moras D 1995 Crystal structure of the ligand-binding domain of the human nuclear receptor RXR-alpha. Nature 375 377–382.[CrossRef][Medline]
Brzozowski AM, Pike AC, Dauter Z, Hubbard RE, Bonn T, Engstrom O, Ohman L, Greene GL, Gustafsson JA & Carlquist M 1997 Molecular basis of agonism and antagonism in the oestrogen receptor. Nature 389 753–758.[CrossRef][Medline]
Carlson KE, Choi I, Gee A, Katzenellenbogen BS & Katzenellenbogen JA 1997 Altered ligand binding properties and enhanced stability of a constitutively active estrogen receptor: evidence that an open pocket conformation is required for ligand interaction. Biochemistry 36 14897–14905.[CrossRef][Medline]
Darimont BD, Wagner RL, Apriletti JW, Stallcup MR, Kushner PJ, Baxter JD, Fletterick RJ & Yamamoto KR 1998 Structure and specificity of nuclear receptor-coactivator interactions. Genes and Development 12 3343–3356.
Greschik H, Wurtz JM, Sanglier S, Bourguet W, van Dorsselaer A, Moras D & Renaud JP 2002 Structural and functional evidence for ligand-independent transcriptional activation by the estrogen-related receptor 3. Molecular Cell 9 303–313.[CrossRef][ISI][Medline]
Hall JM, Couse JF & Korach KS 2001 The multifaceted mechanisms of estradiol and estrogen receptor signaling. Journal of Biological Chemistry 276 36869–36872.
Heery DM, Kalkhoven E, Hoare S & Parker MG 1997 A signature motif in transcriptional co-activators mediates binding to nuclear receptors. Nature 387 733–736.[CrossRef][Medline]
Henttu PM, Kalkhoven E & Parker MG 1997 AF-2 activity and recruitment of steroid receptor coactivator 1 to the estrogen receptor depend on a lysine residue conserved in nuclear receptors. Molecular and Cellular Biology 17 1832–1839.[Abstract]
Hong H, Kohli K, Garabedian MJ & Stallcup MR 1997 GRIP1, a transcriptional coactivator for the AF-2 transactivation domain of steroid, thyroid, retinoid, and vitamin D receptors. Molecular and Cellular Biology 17 2735–2744.[Abstract]
Hong H, Yang L & Stallcup MR 1999 Hormone-independent transcriptional activation and coactivator binding by novel orphan nuclear receptor ERR3. Journal of Biological Chemistry 274 22618–22626.
Howell A 2000 Faslodex (ICI 182780): an oestrogen receptor downregulator. European Journal of Cancer 36 (Suppl 4) S87–S88.
Jordan VC & Morrow M 1999 Tamoxifen, raloxifene, and the prevention of breast cancer. Endocrine Reviews 20 253–278.
Kallenberger BC, Love JD, Chatterjee VK & Schwabe JW 2003 A dynamic mechanism of nuclear receptor activation and its perturbation in a human disease. Nature Structural Biology 10 136–140.[CrossRef][ISI][Medline]
Kato S, Endoh H, Masuhiro Y, Kitamoto T, Uchiyama S, Sasaki H, Masushige S, Gotoh Y, Nishida E, Kawashima H et al. 1995 Activation of the estrogen receptor through phosphorylation by mitogen-activated protein kinase. Science 270 1491–1494.
