Skip to main content
NCBI home page
Molecular Endocrinology logoLink to Molecular Endocrinology
. 2012 Dec 14;27(1):2–11. doi: 10.1210/me.2011-1378

Point Mutations in the ERα Gαi Binding Domain Segregate Nonnuclear from Nuclear Receptor Function

Qian Wu 1,*, Ken Chambliss 1,*, Wan-Ru Lee 1, Ivan S Yuhanna 1, Chieko Mineo 1, Philip W Shaul 1,✉
PMCID: PMC3545213  PMID: 23242705

Abstract

A subpopulation of plasma membrane-associated estrogen receptor (ER)α interact directly with G proteins and mediate nonnuclear receptor signaling. This mechanism underlies numerous processes, including important cardiovascular protective actions of estradiol (E2), such as the activation of endothelial NO synthase (eNOS) and endothelial cell growth and migration. In the present work we sought a genetic approach to differentiate nonnuclear from nuclear ERα actions. We generated single alanine substitutions within the Gαi-binding domain of ERα (amino acids 251–260) and tested signaling to eNOS or ERK1,2 and activation of luciferase (Luc) reporters signifying transactivation via direct or indirect ERα-DNA binding in HeLa cells. The point mutants ERα-R256A, ERα-K257A, ERα-D258A, and ERα-R260A were all incapable of activating eNOS in response to E2, and ERα-R256A and ERα-D258A also showed loss of ERK1,2 activation. In contrast, ERα-R256A, ERα-K257A, ERα-D258A, and ERα-R260A all displayed normal capacity to invoke E2-induced transactivation of estrogen response element (ERE)-Luc or Sp1-Luc. However, whereas activator protein 1-Luc activation by ERα-R256A and ERα-D258A was intact, ERα-K257A and ERα-R260A were incapable of activator protein 1-Luc activation. In in vitro pull-down assays with the two mutants that lack all nonnuclear functions tested and retain all nuclear functions tested, ERα-R256A and ERα-D258A, there was normal direct interaction between Gαi and ERα-R256A and an absence of interaction between Gαi and ERα-D258A. When expressed in endothelial cells, these two mutants prevented E2-induced migration and eNOS activation mediated by endogenous receptor, indicative of dominant-negative action. Thus, the point mutants ERα-R256A and ERα-D258A in the receptor GαI-binding domain provide genetic segregation of nonnuclear from nuclear ERα function.

As members of the nuclear receptor superfamily, estrogen receptors (ERs) function classically as transcription factors and mediate changes in gene expression in response to estrogen. More recently, we and others have shown that estrogen also activates a subpopulation of membrane-associated ER to cause rapid signaling in various cell types (1–3). The underlying mechanisms are exemplified in endothelial cells in which caveolae/lipid raft-associated ERα activate Src kinases, phosphatidylinositol-3 (PI3) kinase, Akt, and ERK1,2 to stimulate nitric oxide (NO) production by endothelial NO synthase (eNOS) (1). We further discovered that nonnuclear signaling by ERα is mediated by a novel direct interaction of Gαi with amino acids 251–260 of the receptor that is operative proximal to kinase activation (4).

Using an estrogen-dendrimer-conjugate (EDC), which is excluded from the nucleus, it has been possible to definitively demonstrate that nonnuclear ERα signaling to eNOS underlies the promotion of endothelial cell growth and migration by estradiol (E2). Studies with EDC treatment in mice have further indicated that EDC is a selective nonnuclear ER modulator in vivo, and that nonnuclear ERα signaling promotes cardiovascular protection in both a carotid artery reendothelialization model and a neointima formation model. In contrast, the discrete activation of nonnuclear ER does not promote uterine enlargement or MCF-7 cell xenograft growth in vivo, illustrating that nonnuclear ER signaling segregates favorable cardiovascular actions from detrimental effects on cancer cells (5).

EDC and other potential forms of tethered estrogen provide powerful tools for selective nonnuclear ER gain of function. However, because nuclear ERs are not activated by EDC, it does not provide insight into the well-recognized cross talk that occurs between nonnuclear and nuclear ER function (6–9). Furthermore, a gain-of-function strategy such as EDC does not address if or how nonnuclear ER signaling is operative in the setting of receptor activation by endogenous estrogens, the abundance of which is regulated in a complex and dynamic manner both spatially and temporally (10). For these multiple reasons genetic loss-of-function strategies are needed to effectively investigate the biological consequences of nonnuclear ERα actions.

The purpose of the present investigation was to leverage the prior identification of the Gαi binding domain (GBD) that mediates Src and ERK1,2 activation, as well as downstream signaling by ERα, to devise a genetic selective loss-of-function strategy to segregate nonnuclear from nuclear receptor actions. This was accomplished in studies of nonnuclear ERα activation of eNOS, which affords the evaluation of amplified kinase signaling, and nonnuclear ERα activation of ERK1,2. To evaluate the specificity of the impact of the mutatgenesis on nonnuclear ERα function, the activation of luciferase (Luc) reporters signifying transactivation via direct or indirect ERα-DNA binding was interrogated. Using these approaches, two single alanine substitutions in the GBD were ultimately identified that disrupt nonnuclear signaling while preserving the nuclear functions of ERα.

Materials and Methods

Cell culture, transfection, and evaluations of transcriptional transactivation

COS-7, HeLa, and human embryonic kidney 293 cells were maintained in DMEM (Sigma Chemical Co., St. Louis, MO) supplemented with 10% fetal bovine serum, and bovine aortic endothelial cells (BAEC) were maintained in EGM2 (Clonetics Corp., San Diego, CA) supplemented with growth factors and 5% fetal bovine serum. Cells were transfected with wild-type human ERα, ERα-Δ251–260, or point mutants of ERα in pCDNA3.1 that were generated by site-directed mutagenesis (QuikChange Site-Dorected Mutagenesis Kit, Stratagene, La Jolla, CA) using LipofectAMINE Plus (Invitrogen, Carlsbad, CA). In studies of eNOS activation in HeLa or COS-7 cells, the cells were also cotransfected with cDNA encoding the enzyme, and studies of the activation of eNOS or ERK1,2 were performed 24–48 h after transfection, To evaluate transcriptional transactivation, cells were transfected with estrogen response element (3ERE)-Luc, 3Sp1-Luc, or 6AP-1 (activator protein 1)-Luc, wild-type ERα or mutant expression plasmids, and cDNA encoding β-galactosidase for internal control. Cells were treated 8 h later with vehicle or 10−8 m E2 for 24 h, and luciferase reporter activity was measured as previously reported (5). 3ERE-Luc was kindly provided by David Shapiro (Department of Biochemistry, University of Illinois at Urbana-Champaign, IL), and 3Sp1-Luc and 6AP-1-Luc were kindly provided by Stephen Safe (Department of Biochemistry, Texas A&M University, College Station, TX).

Immunoblot analyses

For evaluation of receptor protein expression, equal amounts of whole-cell lysate protein were loaded on SDS-PAGE, and immuoblotting for ERα was performed using monoclonal anti-ERα antibody Ab-15 (Labvision Corp., Fremont CA). To assess ERK1,2 activation, HeLa cells were seeded in DMEM/F12 without phenol red (Sigma) supplemented with 2% charcoal stripped serum and transfected with appropriate constructs 24 h later. Cells were incubated with serum-free DMEM/F12 for 18 h and then treated with 10−8 m E2 for 0, 5, 10, or 15 min. Cells were harvested in SDS-PAGE sample buffer, equal amounts were loaded on SDS-PAGE, and proteins were transferred and blotted with antibodies. pERK was blotted with polyclonal phospho-p44/42 MAPK ERK1/2 (Tyr202/204) antibody (Cell Signaling Technology, Danvers, MA), and total ERK was blotted with monoclonal MAPK (ERK1/2) antibody (L34F12, Cell Signaling Technology) (4).

NOS activation

BAEC or HeLa or COS-7 cells were transfected with appropriate constructs, and NOS activation was assessed by measuring [14C]-l-arginine conversion to [14C]-l-citrulline over 15 min incubation with E2 (4, 5).

Subcellular localization

To study the subcellular distribution of wild-type and mutant forms of ERα, cell subfractionation was performed as previously described (11). The protein contents of all samples were determined by Bradford assay, and equal amount of proteins were loaded for immunoblot analysis.

Protein interaction analyses using in vitro pull-down assays (4)

Purified myristoylated His-tagged Gαi1 (300 nm) was incubated in 300 μl of 20 mm HEPES buffer (pH 8.0) containing 150 mm NaCl, 5 mm MgCl2, 4% glycerol, 0.05% C12E10, and protease inhibitor cocktail (Calbiochem, La Jolla, CA), with 30 μm guanosine diphosphate added for 1 h at room temperature. Purified flag-tagged ERα proteins were then added, and reactions were incubated at 4 C for 1 h with gentle agitation. Further incubation was performed for 1 h with Ni-nitrilotriacetic acid resin (QIAGEN, Valencia, CA) plus 15 mm immidazole to allow binding of the His-tagged Gαi1 and associated proteins. Samples were washed with the 20 mm HEPES buffer plus immidazole, and the resin was pelleted and suspended in SDS-PAGE sample buffer. After resolution by 10% SDS-PAGE, immunoblot analyses were performed with the Gαi1/2-specific antiserum B087 (12), and mouse monoclonal antibody Ab-15 directed against ERα.

Endothelial cell migration

BAEC were transfected with various constructs, grown to near confluence, and incubated in serum-free DMEM overnight. After a defined region of cells was removed with a razor blade, cells were treated for 24 h with vehicle or E2 and fixed, and the number of cells that had migrated past the wound edge was quantified (5).

Statistical analysis

Comparisons were made between multiple groups by ANOVA with Neuman-Keuls post hoc testing. When indicated, nonparametric ANOVA (Kruskal-Wallace) and post hoc Dunn testing was performed. Significance was defined as P < 0.05.

Results

Impact of deletion of the GBD

The first attempt to segregate nonnuclear from nuclear ERα actions entailed evaluation of the function of a form of the receptor lacking the GBD (ERα-Δ251–260). Whereas wild-type receptor was capable of stimulating eNOS activity upon 15 min of E2 treatment of COS-7 cells coexpressing the receptor and the enzyme, ERα-Δ251–260 was incapable of stimulating eNOS (Fig. 1A). Thus, deletion of the GBD predictably prevented nonnuclear signaling by ERα. Because the GBD resides within a minor nuclear localization signal (NLS), NLS3 (13), the capacity to activate ERE-dependent transcriptional transactivation was then evaluated. Whereas wild-type receptor activated the ERE-Luc reporter, ERα-Δ251–260 was incapable of invoking ERE-mediated transactivation (Fig. 1B). As such, deletion of the GBD does not afford segregation of nonnuclear from nuclear ERα action because both types of receptor function are attenuated.

Fig. 1.

Fig. 1.

Open in a new tab

Deletion of the GBD blunts both nonnuclear and nuclear ERα function. A, COS-7 cells were cotransfected with eNOS cDNA and cDNA encoding wild-type ERα or ERαΔ251–260, and 48 h later NOS activation was evaluated. [14C]-l-arginine conversion to [14C]-l-citrulline was determined in intact cells during 15-min incubations under basal conditions or in the presence of 10−8 m E2. Relative expression of the two forms of ERα and eNOS was assessed by immunoblot analysis (inset). B, ERE-mediated gene transcription was evaluated in human embryonic kidney 293 cotransfected with receptor constructs (0–250 ng DNA/well), 3ERE-Luc reporter plasmid, and a plasmid containing simian virus 40-driven β-galactosidase to normalize for transfection efficiency. Reporter activity calculated as luciferase activity/β-galactosidase activity was determined after exposure to basal media containing the E2 vehicle dimethylsulfoxide (DMSO) or 10−8 m E2 for 24 h and expressed relative to activity with DMSO control. Values are mean ± sem, n = 4, *, P < 0.05 vs. basal. Similar findings were obtained in three independent experiments.

GBD point mutants and non-nuclear ERα signaling

Having observed that the deletion of the 10 residues within the GBD yields a receptor with blunted nuclear action, site-directed mutagenesis was performed to generate single alanine substitutions within the GBD. The mutants were cotransfected with eNOS in HeLa cells, and the activation of eNOS in response to 15 min incubation with E2 was evaluated. Despite expression to levels comparable to wild-type receptor, which displayed eNOS activation in response to E2, ERα-R256A, ERα-K257A, ERα-D258A, and ERα-R260A were incapable of activating eNOS in response to E2 (Fig. 2A). In contrast, ERα-G253A, ERα-G254A, ERα-I255A, and ERα-R259A displayed stimulation of eNOS by E2 to a degree that was comparable to wild-type receptor (data not shown). These observations were confirmed in COS-7 cells (data not shown). Summary findings regarding the capacity of the GBD point mutants to activate or not activate eNOS are given in Fig. 2B (alanine substitution shown in gray).

Fig. 2.

Fig. 2.

Open in a new tab

Select point mutations in the ERα GBD prevent E2 activation of eNOS. A, HeLa cells were cotransfected with eNOS cDNA and cDNA encoding wild-type ERα (WT) or mutant forms of the receptor (ERα-R256A, ERα-K257A, ERα-D258A or ERα-R260A), and 48 h later ERα and eNOS expression (upper panel) and NOS activation (lower panel) were evaluated. [14C]-l-arginine conversion to [14C]-l-citrulline was determined in intact cells during 15-min incubations under basal conditions or in the presence of 10−8 m E2. E2-stimulated NOS activity above basal activity was determined. In the multiple experiments performed, E2 activation of wild-type ERα typically caused a 50–150% increase in NOS activity above basal values. Values are mean ± sem; n = 4; *, P < 0.05 vs. WT. Similar findings were obtained in three independent experiments. B, Summary of findings. The amino acid sequence of the wild-type GBD (residues 251–260) is shown on top, and the single alanine substitutions in the sequence are designated beneath in gray. Forms of ERα yielding normal eNOS activation in response to E2 are indicated by +, and those lacking the capacity to activate eNOS are indicated by −.

To provide an additional assessment of nonnuclear signaling by the four point mutants displaying an inability to stimulate eNOS, the activation of ERK1,2 was evaluated in HeLa cells (Fig. 3). Wild-type ERα caused a 5-fold increase in ERK phosphorylation in response to E2, and ERα-R256A caused a lesser, 2.5- to 3-fold increase. In contrast, ERα-K257A displayed an exaggerated capacity to activate ERK, causing a 9-fold increase in ERK phosphorylation with E2 binding. Both ERα-D258A and ERα-R260A had negligible ability to induce ERK phosphorylation. As such, two of the four point mutants with attenuated capacity to stimulate eNOS also lacked the ability to mediate ERK1,2 activation.

Fig. 3.

Fig. 3.

Open in a new tab

Select point mutations in the ERα GBD prevent E2 activation of ERK1,2. HeLa cells were transfected with cDNA encoding wild-type ERα (WT) or a mutant form of the receptor that displayed inability to mediate E2 stimulation of eNOS (see Fig. 2). The activation of ERK1,2 by 10−8 m E2 (0–15 min) was evaluated by immunoblotting for phosphorylated ERK1,2 (pERK) or total ERK1,2 (tERK) 24 h after transfection. Representative immunoblots are shown in panel A, and panel B depicts summary data for four independent experiments. Values are mean ± sem; *, P < 0.05 vs. time 0.

GBD point mutants and nuclear ERα function

Seeking selective loss of nonnuclear receptor function, the impact of mutagenesis on ERE-mediated transcriptional transactivation was evaluated. Predictably, wild-type ERα activated the 3ERE-Luc reporter in response to E2, and similar responses were observed in cells expressing ERα-R256A, ERα-K257A, ERα-D258A, or ERα-R260A (Fig. 4A). The four mutant forms of the receptor also mediated normal activation of 3Sp1-Luc (Fig. 4B). However, in studies of 3AP-1-Luc activation, whereas ERα-R256A and ERα-D258A displayed normal function, AP-1-mediated transcriptional transactivation was not activated by E2 binding to ERα-K257A or ERα-R260A (Fig. 4C). Therefore, only ERα-R256A and ERα-D258A have normal capacity to activate gene transcription via either direct, ERE-mediated binding to DNA or indirect, tethered, DNA binding via the transcription factors Sp1 or AP-1.

Fig. 4.

Fig. 4.

Open in a new tab

Select point mutations in the ERα GBD do not disrupt transactivation via direct or indirect ERα-DNA binding. A, ERE-mediated gene transcription was evaluated in HeLa cells cotransfected with cDNA encoding wild-type ERα (WT) or a mutant form of the receptor that displayed inability to mediate E2 stimulation of eNOS (see Fig. 2), 3ERE-Luc reporter plasmid, and a plasmid containing simian virus 40-driven β-galactosidase to normalize for transfection efficiency. Reporter activity calculated as luciferase activity/β-galactosidase activity was determined after exposure to basal media containing the E2 vehicle dimethylsulfoxide (DMSO) or 10−8 m E2 for 24 h and expressed relative to activity with DMSO control. In a similar manner, using a 3Sp1-Luc reporter construct or a 6AP1-Luc reporter construct, Sp1-mediated and AP-1-mediated transcriptional transactivation were evaluated (panels B and C, respectively). In A–C, values are mean ± sem; n = 4; *, P < 0.05 vs. basal, and similar findings were obtained in three independent experiments.

There are numerous posttranslational modifications of ERα that influence its actions. This includes ERα ubiquitination and proteasomal-dependent degradation, which happen as a result of E2 binding to the receptor and are necessary for its nuclear function (14). We assessed E2-induced degradation of ERα-R256A, ERα-K257A, ERα-D258A, and ERα-R260A by evaluating receptor abundance before and after 48 h of ligand activation (Fig. 5A). Wild-type ERα and the four point mutants displayed similar declines in their abundance in response to E2 treatment, suggesting intact mechanisms of ubiquitin- and proteasome-dependent degradation despite altered receptor-specific modulation of ERK1,2.

Fig. 5.

Fig. 5.

Open in a new tab

Characterization of point mutations in the ERα GBD. A, Ligand-induced proteasomal degradation of wild-type ERα, ERα-R256A, ERα-K257A, ERα-D258A, and ERα-R260A was evaluated in transfected HEK293 cells by determining receptor abundance in whole-cell lysates by immunoblot analysis after 24 h of vehicle (Veh) or 10−8 m E2 treatment. B, Dose-response studies of ERE-mediated gene transcription activated by E2 (M) were performed in HeLa cells cotransfected with cDNA encoding wild-type ERα (WT), or ERα-R256A or ERα-D258A, which are the two mutant forms of the receptor that displayed impaired ability to mediate E2 stimulation of eNOS or ERK1,2 (Figs. 2 and 3) yet intact capacity to invoke gene transcription via direct or indirect ERα-DNA binding (Fig. 4). Values are mean ± sem; n = 4; *, P < 0.05 vs. DMSO control; and similar findings were obtained in three independent experiments.

To further characterize the two mutants that lack the capacity for nonnuclear signaling but retain all of the nuclear functions tested, ERα-R256A and ERα-D258A, dose-response studies of 3ERE-Luc reporter activation by E2 were performed (Fig. 5B). The dose responses were comparable for wild-type receptor and the two point mutants, with EC50 values of approximately 10−12 m. Because these were tested under identical conditions in the same cellular context such that cofactor availability is constant, these findings are consistent with equivalent affinity of E2 to the three forms of the receptor (15).

Basis for GBD point mutant loss of nonnuclear function

To determine why the two single alanine substitution mutants that retain all of the nuclear functions tested lack the capacity for nonnuclear signaling, their subcellular localization was evaluated by fractionation and immunoblot analysis. First confirming equal abundance in comparison with wild-type receptor in whole-cell lysates after transfection in COS-7 cells, the abundance of ERα-R256A and ERα-D258A in the nucleus was found to be similar to that observed for wild-type ERα (Fig. 6A). There was also equal abundance of the wild-type and mutant forms of the receptor in association with the plasma membrane. Thus, the inability of ERα-R256A and ERα-D258A to mediate nonnuclear signaling is not due to altered targeting to the plasma membrane.

Fig. 6.

Fig. 6.

Open in a new tab

Elucidation of the basis for nonnuclear loss of function by ERα-R256A and ERα-D258A. A, Subcellular localization of the two single alanine substitution mutants that lack the capacity for nonnuclear signaling but retain all of the nuclear functions tested was evaluated by fractionation and immunoblot analysis. Abundance was assessed in whole cell lysates (WC), a nuclear fraction, a cytosolic fraction, and in purified plasma membranes (PM). Sp1, tubulin, and caveolin-1 (Cav-1) abundance was assessed to evaluate the efficacy of subfractionation. B, The direct interaction between purified recombinant ERα-R256A or ERα-D258A and recombinant myristoylated His-tagged Gαi was evaluated in in vitro pull-down assays. For panels A and B, similar findings were obtained in three independent experiments. WT, Wild type.

The basis for a lack of nonnuclear signaling was further interrogated in studies of the direct interaction between purified recombinant ERα-R256A or ERα-D258A and recombinant myristoylated Gαi in in vitro pull-down assays. Consistent with prior observations, wild-type ERα displayed direct protein-protein interaction with Gαi (Fig. 6B). ERα-R256A displayed direct interaction with Gαi that was comparable in degree to that observed for wild-type receptor. In contrast, ERα-D258A did not interact with Gαi. Thus, the lack of direct coupling with Gαi likely explains why ERα-D258A is incapable of nonnuclear signaling. The basis for the inability of ERα-R256A to function outside the nucleus is yet to be determined.

Impact of GBD point mutants on endothelial cell responses to E2

To test how the GBD point mutations affect a change in cell behavior that is known to occur upon nonnuclear ERα activation (5), E2-induced migration was evaluated in transfected BAECs. Scratch assays showed that E2 stimulated the migration of sham-transfected cells (Fig. 7A). In contrast, paralleling the prior observation of blunted E2-induced eNOS activation (4), cells expressing ERα-Δ251–260 had attenuated migration in response to E2. Whereas the overexpression of wild-type receptor resulted in E2-stimulated migration similar to that observed with sham transfection, BAEC expressing ERα-R256A or ERα-D258A did not migrate with E2. Levels of receptor expression are shown in Fig. 7B, and cumulative findings are in Fig. 7C.

Fig. 7.

Fig. 7.

Open in a new tab

ERα-R256A and ERα-D258A have dominant-negative impact on E2-stimulated endothelial cell migration and eNOS activation. A, Using scratch assays, migration was assessed in endothelial cells transfected with sham plasmid, ERαΔ251–260, wild-type ERα (WT), ERα-R256A, or ERα-D258A and treated with dimethylsulfoxide (DMSO) vehicle or 10−8 m E2 for 24 h. B, Relative abundance of ERα was evaluated by immunoblot analysis. C, Summary data are shown for six experiments. Values are mean ± sem; *, P < 0.05 vs. DMSO control. D and E, Endothelial cells were transfected with varying amounts of sham or receptor cDNA (values shown are micrograms per well), and eNOS activation by 10−8 m E2 was measured 24 h later. Receptor abundance was quantified by immunoblot analysis (D), and eNOS activation by E2 expressed relative to basal activity is shown in panel E. Values are mean ± sem; n = 4; *, P < 0.05 vs. basal.

To evaluate a second endothelial response to E2 that entails nonnuclear ERα activation, eNOS stimulation was also studied in transfected BAEC. To determine how the abundance of the exogenous form of ERα influences this process, cells were transfected with varying amounts of receptor cDNA (0.1–1.0 μg/well). Immunoblot analysis revealed that comparable expression was achieved for the different forms of the receptor (Fig. 7D), and at the lowest level of transfection, exogenous receptor abundance was approximately 3-fold greater than endogenous receptor. Whereas the overexpression of wild-type receptor resulted in E2 activation of eNOS similar to that observed with sham transfection (Fig. 7E), E2 did not stimulate the enzyme in endothelial cells expressing ERα-R256A or ERα-D258A, even at the lowest amount of exogenous receptor attained. Thus, the introduction of ERα-R256A or ERα-D258A prevented E2-induced migration and E2 activation of eNOS mediated by endogenous receptor, indicative of dominant-negative action, and the two mutant forms of the receptor had similar potency in that regard.

Discussion

It is now readily appreciated that ERs function as transcription factors altering gene expression and as mediators of nonnuclear signaling to kinases, and that the ultimate cellular response to estrogen is governed by both these types of processes and the impact of cross talk between them (1–3). Recently, stable forms of conjugated estrogen incapable of nuclear entry have provided a pharmacological means to selectively activate nonnuclear ERs (5, 16). It has been far more challenging to obtain selective loss of function of nonnuclear ER signaling such that nuclear receptor actions remain intact. Leveraging the prior discovery of a Gαi-binding domain within ERα, we now report the identification of two point mutants of the receptor that lack nonnuclear function but retain nuclear action. These forms of the receptor can potentially be used to provide genetic segregation of the two primary categories of ERα function.

In earlier attempts to manipulate ERα structure to distinguish the features required for nonnuclear vs. nuclear processes, we assessed the role of domains necessary for nuclear action in nonnuclear signaling to kinases and eNOS in transfected COS-7 cells. ERα mutants lacking the NLSs NLS2,3 (ERαΔ250–274) or the DNA-binding domain (ERαΔ185–251), which targeted normally to plasma membrane and caveolae/lipid rafts, were incapable of activating eNOS. The loss of NLS2/NLS3 prevented Src and ERK1,2 activation, and it altered ligand-induced PI3 kinase-ERα interaction and prevented eNOS phosphorylation. Loss of the DNA-binding domain did not change E2 activation of Src or ERK1,2, but ligand-induced PI3 kinase-ERα binding and eNOS phosphorylation did not occur (11). These previous findings revealed that there is complex functional overlap between domains of ERα involved in nuclear and nonnuclear receptor actions. The complexity is now even more apparent regarding NLS2,3 (amino acids 250–274) and the overlapping Gαi-binding domain (amino acids 251–260) because whereas the deletion of amino acids 250–274 prevents both ERK1,2 and eNOS enzyme activation (11), we now find that alanine substitution of Lys257 retards eNOS stimulation, yet ERK1,2 activation is actually enhanced. Although the molecular basis for this finding is yet to be elucidated, the contrasting effects on eNOS and ERK1,2 that result from this single amino acid substitution indicate that the signaling pathway from ERα through G proteins and ERK1,2 to eNOS is not linear, as may have been previously implied (17).

More targeted interrogations have demonstrated that numerous posttranslational modifications and protein-protein interactions are required for nonnuclear signaling by ERα. Although not found in all studies, there is evidence that ERα interacts with caveolin-1 to facilitate receptor trafficking to caveolae/lipid rafts (18, 19). A palmitoylation site has also been identified within the ligand-binding domain of ERα (Cys447) that is important for membrane localization, and additional residues flanking Cys447 are also required for effective receptor palmitoylation (20–22). Regarding protein-protein interactions, it has been demonstrated that ERα binds in a ligand-dependent fashion to the PI3 kinase p85a regulatory subunit (23), and the SH2 domain of c-Src kinase interacts with phosphorylated Tyr-537 of ERα (24–26). In addition, the arginine methyltransferase PRMT1 methylates ERα Arg-260 to promote the interaction of the receptor with the p85 subunit of PI3 kinase and with c-Src (27). A scaffolding protein designated as modulator of nongenomic activity of ER has been described that enhances E2-induced interaction between ERα and Src family kinase(s) (28). In endothelial cells striatin, which is a member of the WD (tryptophan-aspartic acid) repeat protein family, has direct interaction with ERα that involves receptor amino acids 183–253, and a peptide representing amino acids 176–253 prevents E2 activation of ERK1,2, Akt, and eNOS, suggesting that the ERα-striatin interaction participates in nonnuclear receptor signaling (29). Work in HeLa and MCF-7 cells has further revealed that ERα interacts with hematopoietic pre-B-cell leukemia transcription factor-interacting protein, which mediates the binding of the receptor with tubulins and also recruits the p85 subunit of PI3K and Src kinases to an ERα complex upon E2 binding, leading to the stimulation of Akt and ERK1,2. There is a direct interaction of ERα with hematopoietic pre-B-cell leukemia transcription factor-interacting protein that requires amino acids 264–302 and 552–595 of the receptor (30). Shc (Src homology and collagen homology), which has a direct interaction with ERα that requires residues between amino acids 1 and 367 of the receptor, is critically involved in linking E2 binding to nonnuclear ERα with morphological changes and the growth of breast cancer cells (31). Of these various processes participating in nonnuclear ERα function, point mutagenesis to modify ERα interaction with caveolin-1 or to blunt receptor palmitoylation or methylation yielded attenuation of nonnuclear signaling. In studies of other ERα mutants made to disrupt nonnuclear mechanism, ERE-Luc activation was tested and was intact; however, the impact of the modifications on other nuclear functions of the receptor was not explored (19, 22, 27). Thus, although we have recently learned a great deal about the characteristics of ERα that enable the receptor to enact signaling outside the nucleus, the present work is the first to modify a feature of ERα required for such function in a manner that does not disrupt transactivation via direct or indirect ERα-DNA binding as assessed using luciferase assays. How these discrete modifications impact the regulation of endogenous genes is worthy of interrogation in the future.

In addition to identifying ERα-R256A and ERα-D258A as point mutants lacking nonnuclear function but retaining diverse means of nuclear function, their direct interaction with Gαi was evaluated. Whereas ERα-D258A was incapable of interaction, ERα-R256A displayed normal binding to Gαi. Because two interventions that prevent the ERα-Gαi interaction, namely pertussis toxin treatment or expression of a peptide representing the GBD (amino acids 251–260) of ERα, both inhibit nonnuclear receptor signaling (4, 32), the loss of function of ERα-D258A is likely explained by its inability to associate with Gαi. The mechanism underlying the loss of function of the point mutant ERα-R256A is yet to be clarified. When expressed in endothelial cells along with endogenous ERα, two known cellular responses to nonnuclear ERα activation, namely cell migration and the activation of eNOS, were blunted by either ERα-R256A or ERα-D258A, and this is indicative of dominant-negative action. The basis for the dominant-negative effect is currently unknown, but it may entail interference with one or more of the other mechanisms noted above that mediate nonnuclear receptor function.

Although agonists have now been created that selectively activate nonnuclear ER, their use does not address the critical cross talk that exists between nonnuclear and nuclear receptor actions. Serines in ERα are targets for estrogen-induced phosphorylation that modifies the regulation of gene transcription by the receptor (6–8). There is also nonnuclear ER-dependent phosphorylation of coactivators that leads to their recruitment to the transcriptional apparatus (33). In addition, there is NO-induced S-nitrosylation of ERα that causes impaired DNA binding (9), such that nonnuclear ER activation in NOS-expressing cells including endothelium potentially attenuates nuclear ER function. Pharmacological interventions altering kinase activity have revealed that nonnuclear ER signaling plays an important role in the promotion of breast cancer cell proliferation, migration, and survival both in vitro and in vivo (34), but nonnuclear ER activation alone is insufficient to invoke breast cancer cell or tumor growth (5, 16). In breast cancer cells it has also been recently recognized that kinases with functions that are modified by nonnuclear ERα activation, such as ERK2, are recruited along with their substrates and with ERα to chromatin-binding sites (35). Therefore, it is quite apparent that there are important implications of nonnuclear-to-nuclear ER cross talk in the context of cancer, and means to interrogate that cross talk that entirely avoid kinase antagonism should provide new insights. We also know little about the contributions of nonnuclear ER actions to normal physiological processes. By providing selective disruption of nonnuclear processes with optimal specificity to the receptor, the ERα-R256A and ERα-D258A point mutants will be potentially helpful tools with which to better understand the complex nature of ERα biology.

Acknowledgments

This work was supported by National Institutes of Health Grant HD030276 (to P.W.S.), American Heart Association Postdoctoral Fellowship Award 09POST2250860 (to Q.W.), and the Crystal Charity Ball Center for Pediatric Critical Care Research (to P.W.S.).

Disclosure Summary: The authors have nothing to disclose.

Footnotes

Abbreviations:
AP-1
Activator protein 1
BAEC
bovine aortic endothelial cells
E2
estradiol
EDC
estrogen-dendrimer-conjugate
eNOS
endothelial NOS
ER
estrogen receptor
ERE
estrogen response element
GBD
Gαi-binding domain
NLS
nuclear localization signal
NOS
nitric oxide synthase
PI3
phosphatidylinositol-3.

References

  • 1. Wu Q, Chambliss K, Umetani M, Mineo C, Shaul PW. 2011. Non-nuclear estrogen receptor signaling in the endothelium. J Biol Chem 286:14737–14743 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2. Hammes SR, Levin ER. 2007. Extranuclear steroid receptors: nature and actions. Endocr Rev 28:726–741 [DOI] [PubMed] [Google Scholar]
  • 3. Osborne CK, Schiff R. 2005. Estrogen-receptor biology: continuing progress and therapeutic implications. J Clin Oncol 23:1616–1622 [DOI] [PubMed] [Google Scholar]
  • 4. Kumar P, Wu Q, Chambliss KL, Yuhanna IS, Mumby SM, Mineo C, Tall GG, Shaul PW. 2007. Direct Interactions with G alpha i and G betagamma mediate nongenomic signaling by estrogen receptor α. Mol Endocrinol 21:1370–1380 [DOI] [PubMed] [Google Scholar]
  • 5. Chambliss KL, Wu Q, Oltmann S, Konaniah ES, Umetani M, Korach KS, Thomas GD, Mineo C, Yuhanna IS, Kim SH, Madak-Erdogan Z, Maggi A, Dineen SP, Roland CL, Hui DY, Brekken RA, Katzenellenbogen JA, Katzenellenbogen BS, Shaul PW. 2010. Non-nuclear estrogen receptor alpha signaling promotes cardiovascular protection but not uterine or breast cancer growth in mice. J Clin Invest 120:2319–2330 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Rogatsky I, Trowbridge JM, Garabedian MJ. 1999. Potentiation of human estrogen receptor alpha transcriptional activation through phosphorylation of serines 104 and 106 by the cyclin A-CDK2 complex. J Biol Chem 274:22296–22302 [DOI] [PubMed] [Google Scholar]
  • 7. Chen D, Washbrook E, Sarwar N, Bates GJ, Pace PE, Thirunuvakkarasu V, Taylor J, Epstein RJ, Fuller-Pace FV, Egly JM, Coombes RC, Ali S. 2002. Phosphorylation of human estrogen receptor alpha at serine 118 by two distinct signal transduction pathways revealed by phosphorylation-specific antisera. Oncogene 21:4921–4931 [DOI] [PubMed] [Google Scholar]
  • 8. Cheng J, Zhang C, Shapiro DJ. 2007. A functional serine 118 phosphorylation site in estrogen receptor-α is required for down-regulation of gene expression by 17β-estradiol and 4-hydroxytamoxifen. Endocrinology 148:4634–4641 [DOI] [PubMed] [Google Scholar]
  • 9. Garbán HJ, Márquez-Garbán DC, Pietras RJ, Ignarro LJ. 2005. Rapid nitric oxide-mediated S-nitrosylation of estrogen receptor: regulation of estrogen-dependent gene transcription. Proc Natl Acad Sci USA 102:2632–2636 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10. Santen RJ, Brodie H, Simpson ER, Siiteri PK, Brodie A. 2009. History of aromatase: saga of an important biological mediator and therapeutic target. Endocr Rev 30:343–375 [DOI] [PubMed] [Google Scholar]
  • 11. Chambliss KL, Simon L, Yuhanna IS, Mineo C, Shaul PW. 2005. Dissecting the basis of nongenomic activation of endothelial nitric oxide synthase by estradiol: role of ERα domains with known nuclear functions. Mol Endocrinol 19:277–289 [DOI] [PubMed] [Google Scholar]
  • 12. Linder ME, Middleton P, Hepler JR, Taussig R, Gilman AG, Mumby SM. 1993. Lipid modifications of G proteins: α subunits are palmitoylated. Proc Natl Acad Sci USA 90:3675–3679 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13. Ylikomi T, Bocquel MT, Berry M, Gronemeyer H, Chambon P. 1992. Cooperation of proto-signals for nuclear accumulation of estrogen and progesterone receptors. EMBO J 11:3681–3694 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14. La RP, Acconcia F. 2011. Signaling functions of ubiquitin in the 17β-estradiol (E2):estrogen receptor (ER) α network. J Steroid Biochem Mol Biol 127:223–230 [DOI] [PubMed] [Google Scholar]
  • 15. Simons SS, Jr, Chow CC. 2012. The road less traveled: new views of steroid receptor action from the path of dose-response curves. Mol Cell Endocrinol 348:373–382 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16. Harrington WR, Kim SH, Funk CC, Madak-Erdogan Z, Schiff R, Katzenellenbogen JA, Katzenellenbogen BS. 2005. Estrogen dendrimer conjugates that preferentially activate extranuclear, non-genomic versus genomic pathways of estrogen action. Mol Endocrinol 20:491–502 [DOI] [PubMed] [Google Scholar]
  • 17. Guo X, Razandi M, Pedram A, Kassab G, Levin ER. 2006. Estrogen induces vascular wall dilation: mediation through kinase signaling to nitric oxide and estrogen receptors α and β. J Biol Chem 280:19704–19710 [DOI] [PubMed] [Google Scholar]
  • 18. Chambliss KL, Yuhanna IS, Mineo C, Liu P, German Z, Sherman TS, Mendelsohn ME, Anderson RG, Shaul PW. 2000. Estrogen receptor α and endothelial nitric oxide synthase are organized into a functional signaling module in caveolae. Circ Res 87:E44–E52 [DOI] [PubMed] [Google Scholar]
  • 19. Razandi M, Alton G, Pedram A, Ghonshani S, Webb P, Levin ER. 2003. Identification of a structural determinant necessary for the localization and function of estrogen receptor α at the plasma membrane. Mol Cell Biol 23:1633–1646 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20. Li L, Haynes MP, Bender JR. 2003. Plasma membrane localization and function of the estrogen receptor α variant (ER46) in human endothelial cells. Proc Natl Acad Sci USA 100:4807–4812 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Acconcia F, Ascenzi P, Fabozzi G, Visca P, Marino M. 2004. S-palmitoylation modulates human estrogen receptor-α functions. Biochem Biophys Res Commun 316:878–883 [DOI] [PubMed] [Google Scholar]
  • 22. Pedram A, Razandi M, Sainson RC, Kim JK, Hughes CC, Levin ER. 2007. A conserved mechanism for steroid receptor translocation to the plasma membrane. J Biol Chem 282:22278–22288 [DOI] [PubMed] [Google Scholar]
  • 23. Simoncini T, Hafezi-Moghadam A, Brazil DP, Ley K, Chin WW, Liao JK. 2000. Interaction of oestrogen receptor with the regulatory subunit of phosphatidylinositol-3-OH kinase. Nature 407:538–541 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24. Haynes MP, Li L, Sinha D, Russell KS, Hisamoto K, Baron R, Collinge M, Sessa WC, Bender JR. 2003. Src kinase mediates phosphatidylinositol 3-kinase/Akt-dependent rapid endothelial nitric-oxide synthase activation by estrogen. J Biol Chem 278:2118–2123 [DOI] [PubMed] [Google Scholar]
  • 25. Migliaccio A, Castoria G, Di Domenico M, de Falco A, Bilancio A, Lombardi M, Barone MV, Ametrano D, Zannini MS, Abbondanza C, Auricchio F. 2000. Steroid-induced androgen receptor-oestradiol receptor β-Src complex triggers prostate cancer cell proliferation. EMBO J 19:5406–5417 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. Li L, Hisamoto K, Kim KH, Haynes MP, Bauer PM, Sanjay A, Collinge M, Baron R, Sessa WC, Bender JR. 2007. Variant estrogen receptor-c-Src molecular interdependence and c-Src structural requirements for endothelial NO synthase activation. Proc Natl Acad Sci USA 104:16468–16473 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. Le Romancer M, Treilleux I, Leconte N, Robin-Lespinasse Y, Sentis S, Bouchekioua-Bouzaghou K, Goddard S, Gobert-Gosse S, Corbo L. 2008. Regulation of estrogen rapid signaling through arginine methylation by PRMT1. Mol Cell 31:212–221 [DOI] [PubMed] [Google Scholar]
  • 28. Barletta F, Wong CW, McNally C, Komm BS, Katzenellenbogen B, Cheskis BJ. 2004. Characterization of the interactions of estrogen receptor and MNAR in the activation of cSrc. Mol Endocrinol 18:1096–1108 [DOI] [PubMed] [Google Scholar]
  • 29. Lu Q, Pallas DC, Surks HK, Baur WE, Mendelsohn ME, Karas RH. 2004. Striatin assembles a membrane signaling complex necessary for rapid, nongenomic activation of endothelial NO synthase by estrogen receptor α. Proc Natl Acad Sci USA 101:17126–17131 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Manavathi B, Acconcia F, Rayala SK, Kumar R. 2006. An inherent role of microtubule network in the action of nuclear receptor. Proc Natl Acad Sci USA 103:15981–15986 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Song RX, McPherson RA, Adam L, Bao Y, Shupnik M, Kumar R, Santen RJ. 2002. Linkage of rapid estrogen action to MAPK activation by ERα-Shc association and Shc pathway activation. Mol Endocrinol 16:116–127 [DOI] [PubMed] [Google Scholar]
  • 32. Wyckoff MH, Chambliss KL, Mineo C, Yuhanna IS, Mendelsohn ME, Mumby SM, Shaul PW. 2001. Plasma membrane estrogen receptors are coupled to endothelial nitric-oxide synthase through Gα(i). J Biol Chem 276:27071–27076 [DOI] [PubMed] [Google Scholar]
  • 33. Levin ER. 2005. Integration of the extranuclear and nuclear actions of estrogen. Mol Endocrinol 19:1951–1959 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Levin ER, Pietras RJ. 2008. Estrogen receptors outside the nucleus in breast cancer. Breast Cancer Res Treat 108:351–361 [DOI] [PubMed] [Google Scholar]
  • 35. Madak-Erdogan Z, Lupien M, Stossi F, Brown M, Katzenellenbogen BS. 2011. Genomic collaboration of estrogen receptor α and extracellular signal-regulated kinase 2 in regulating gene and proliferation programs. Mol Cell Biol 31:226–236 [DOI] [PMC free article] [PubMed] [Google Scholar]

Articles from Molecular Endocrinology are provided here courtesy of The Endocrine Society

RESOURCES