The pleiotropic effects of estrogen are mediated via stimulation of two estrogen receptor (ER) subtypes, ERα and ERβ. Although a number of studies have identified expression of one or both subtypes in estrogen target tissues, fewer studies have correlated ER expression with a functional role of these proteins in regulating cellular excitability. In the present study, we have combined cellular fluorescence, immunocytochemistry, and molecular expression techniques with single-channel patch-clamp studies to determine which ER mediates estrogen-stimulated potassium channel activity in human coronary artery smooth muscle cells (HCASMC). We had demonstrated previously that estrogen stimulates activity of the large-conductance, calcium- and voltage-activated potassium (BKCa) channel in HCASMC via a nongenomic mechanism. We now demonstrate expression of both ERα and ERβ subtypes in HCASMC. Functionally, however, expression of ERα antisense plasmid abolished the acute effect of estrogen on these channels, whereas estrogen retained its ability to stimulate BKCa channels in cells transfected with only green fluorescence protein. In contrast, overexpression of ERα enhanced the stimulatory action of estrogen in HCASMC. Transfection with ERα antisense/sense plasmid did not alter ERβ expression. These findings indicate that the ERα isoform mediates estrogen-induced stimulation of BKCa channels in HCASMC and thereby provide evidence for a receptor-dependent signaling mechanism that can mediate estrogen-induced inhibition of cellular excitability.
In addition to its classic effects on reproduction and development, estrogen also exerts powerful influences on a variety of nonreproductive tissues. Interestingly, the molecular signaling mechanisms underlying these nonreproductive effects and their functional importance continue to be a source of controversy. For example, estrogens are reported to have both salutary and deleterious effects on the cardiovascular system (Stampfer et al., 1991; Rosano et al., 1993; Rossouw et al., 2002), such that postmenopausal hormone replacement therapy, once a highly prescribed regimen, has been largely discontinued in light of recent clinical trials (Simon et al., 2001; Rossouw et al., 2002). As underscored in a recent review (Turgeon et al., 2004), amid all of the debate concerning hormone replacement therapy one thing has become clear; further fundamental research is needed to better understand the complicated effects of estrogen and other gonadal hormones, particularly on the heart and blood vessels. Because coronary heart disease continues to be the leading cause of death for both women and men, it is especially important to understand estrogen signaling in coronary arteries. Elucidating the signal transduction mechanisms of estrogen action in target tissues is a primary goal of current research, and the first step in this process is identifying and characterizing the nature of estrogen receptor (ER) that serves as the initiator of estrogen responses.
Estrogen is a vasoactive hormone that is usually presumed to dilate arteries by stimulating nitric-oxide synthase activity in endothelial cells (Type III) (Mendelsohn, 2002); however, smooth muscle cells within the vascular wall also express both subtypes of ER (ERα and ERβ), indicating that these myocytes are targets of estrogen action. Nonetheless, comparatively few studies have focused on estrogen signaling and its functional significance in smooth muscle cells, particularly in coronary arteries (Mugge et al., 1993; White et al., 1995; Darkow et al., 1997). Interestingly, human vascular smooth muscle (VSM) expresses aromatase (Harada et al., 1999), thus providing a mechanism for de novo estrogen synthesis (from precursors such as testosterone) within the vascular wall. It seems highly probable then that estrogen functions as an acute, autocrine/paracrine regulator of VSM function in both men and women. Unfortunately, our knowledge of the direct effects of estrogen on VSM cells remains rather limited.
Although biochemical and molecular methods have characterized ER expression in coronary arteries, we have found essentially no studies providing direct evidence for a functional role of ER subtypes in mediating cellular excitability. Therefore, the purpose of the present study was to combine molecular biology techniques with molecular physiology (i.e., single-channel patch-clamp technology) to investigate the functional role of ER receptor expression in human coronary artery smooth muscle cells (HCASMC). After selective transfection of HCASMC, we then employed activity of the large-conductance, calcium-activated potassium (BKCa) channel as a sensitive molecular assay for ER signaling. We and others have demonstrated previously that BKCa channels are an important target of estrogen action in coronary arteries (White et al., 1995, 2002; Darkow et al., 1997; Node et al., 1997; Rosenfeld et al., 2000) and provide a powerful repolarizing negative feedback mechanism to limit calcium influx into HCASMC. To our knowledge, this study is the first to provide both biochemical and functional results, evidence that it is primarily the ERα that mediates acute (most likely nongenomic) estrogen signaling in HCASMC.
Materials and Methods
Cell Culture. HCASMC were purchased from Clonetics/Cambrex (Walkersville, MD) and were grown as described previously (White et al., 2002). Cells (passages 3–7) were plated onto 12-mm coverslips for patch-clamp experiments. HEK293 cells were purchased from American Type Culture Collection (Manassas, VA).
Gene Subcloning and Transient Gene Transfection. Plasmid ERα, pCMV-ERα, was kindly provided by Drs. B. Katzenellenbogen and K. Weis (University of Illinois, Urbana, IL), and from it a forward-back orientation pCMV-ERα was constructed by inserting the ERα into the vector in reverted orientation. After digesting with Pstl, the 5′-flank (approximately 360 base pairs) of the ERα gene was reserved together with the vector and transformed to the cells (Escherichia coli Top 10 strain) to produce ERα antisense (ERα-AS) plasmid. Transfection was obtained according to the manufacturer's instruction (Mirus, Madison, WI). The pEGFP was designed for use as a cotransfection marker. Patch-clamp experiments were performed after a 24- to 48-h transfection of the HCASMC. Stable transfection cell line of ERα was established in HEK293 cells.
Immunofluorescence Studies. HCASMC were grown in monolayers on 22 × 22-mm coverslips in 35-mm-diameter dishes (Fisher-brand; Fisher Scientific, Atlanta, GA). After washing with TBS buffer (50 mM Tris-HCl, pH 7.4, and 150 mM NaCl) twice, the cells were fixed with 4% paraformaldehyde and permeabilized with 0.2% Triton X-100 in phosphate-buffered saline for 10 min. Reactive groups were quenched with 0.1% sodium borohydride in TBS for 5 min and then blocked for 1 h in blocking buffer [10% horse serum, 1% bovine serum albumin (BSA), and 0.02% NaN3 in phosphate-buffered saline]. Cells were incubated with primary antibody (ABR–Affinity BioReagents, Golden, CO): ERα (1:2000; catalog number PA1-309, anti-human, rabbit polyclonal to N-terminal residues 21–32 of human ERα); or ERβ (1:2000; catalog number PA1-311, anti-human, rabbit polyclonal to the amino acid residues 55–77 of ERβ) diluted in 1% BSA-TBS at 4°C overnight. After washing, a 1:1000 dilution in 1% BSA-TBS of goat anti-rabbit fluorescein isothiocyanate-conjugated secondary antibody was placed on the cells for 30 min. After washing three times, the coverslips were mounted on slides. The cellular distribution of the receptors was photographed with a cooled CCD camera attached to a Zeiss microscope (Carl Zeiss GmbH, Jena, Germany). In other experiments, cells were transfected with either ERα antisense or sense plasmid and the secondary antibody was goat anti-rabbit-conjugated Alexa Fluor 594 (Molecular Probes, Eugene, OR). Receptor expression was detected by using the deconvolution system on a Zeiss microscope. Control immunofluorescent studies were obtained either by omitting the primary antibody or by including a neutralizing peptide (PEP-011 for ERα; PEP-037 for ERβ).
Protein Extraction and Western Blotting. In brief, for isolation of whole-cell extracts, HCASMC were cultured in 100-mm dishes. Human breast adenocarcinoma cell line (MCF7) whole cell lysate was used as positive control for both ERα and ERβ detection. Proteins were extracted, and Western blotting was performed as described previously (Dimitropoulou et al., 2005). The protein (50 μg) of HCASMC lysate or MCF7 cell lysate (20 μg) was loaded into each well.
Patch-Clamp Recordings. Single potassium channels were measured in cell-attached patches on HCASMC as described previously. (White et al., 1995, 2002) In brief, patch pipettes (3–5 MΩ) were filled with a standard Ringer solution (140 mM NaCl, 5 mM KCl, 1 mM MgCl2, 2 mM CaCl2, and 10 mM HEPES, pH 7.4). Voltage across the patch was controlled by clamping the cell (outside the pipette) at 0 mV with a high concentration of potassium extracellular solution containing 140 mM KCl, 10 mM MgCl2, 0.1 mM CaCl2, 10 mM HEPES, and 30 mM glucose, pH 7.2. Currents were filtered at 2 kHz and digitized at 10 kHz by using an Axopatch 200B amplifier and pCLAMP 9.0 software (Axon Instruments Inc., Union City, CA). BKCa channels were identified by amplitude and sensitivity to calcium and to 1 mM tetraethylammonium. The effects of estrogen were tested over a range of membrane voltages, but channel open-time probabilities (NPo) were calculated from patches held at +40mV to enhance the reliability and accuracy of the recordings, as described previously (Han et al., 1999). Drugs were applied to the microscope chamber via perfusion, and potential responses to vehicle (e.g., 0.1% ethanol) were also obtained.
Statistical Analysis. All data were expressed as the mean ± S.E. Statistical significance between two groups was evaluated by Student's t test for paired data. Comparison between multiple groups was made by the one-way analysis of variance test. A probability of less than 0.05 was considered to indicate a significant difference.
The importance of estrogen receptors in mediating estrogen signaling in HCASMC was examined in intact cells. Using 4,5-diaminofluorescein diacetate (DAF-2 DA), a fluorescent indicator of NO production, we observed estrogen-stimulated NO generation in HCASMC (100 nM estrogen; n = 5; Fig. 1A). 17β-Estradiol significantly increased fluorescence intensity above basal levels by a factor of 1.52 ± 0.19 (P < 0.05; Fig. 1, A and B) at 15 min and nearly doubled the signal (1.90 ± 0.21) after 30 min (P < 0.05; n = 4; Fig. 1B). In contrast, background levels of fluorescence were unchanged for over 2 h in the absence of estrogen. Furthermore, estrogen-stimulated NO fluorescence was attenuated by 100 nM ICI 182,780 (fulvestrant), an estrogen receptor antagonist, thus indicating that the response of HCASMC to estrogen involved the ERα and/or ERβ. ICI 182,780 prevented the stimulatory effect of estrogen on NO production (n = 4; Fig. 1, A and B) and was also able to reverse the action of estrogen, as ICI 182,780 decreased estrogen-stimulated fluorescence intensity back to basal levels (1.01 ± 0.01, 15 min, P < 0.05; 1.04 ± 0.04, 30 min, P < 0.05, n = 4). In contrast, ICI 182,780 alone had no effect on NO production. Likewise, 10 μM NG-monomethyl-l-arginine (l-NMMA), an inhibitor of nitric-oxide synthase, prevented 100 nM 17β-estradiol from enhancing NO fluorescence (1.05 ± 0.05, 15 min; 1.12 ± 0.05, 30 min; n = 4; Fig. 1, A and B), thus controlling for potential nonspecific effects on fluorescence intensity.
Detection and Localization of ERα and ERβ in HCASMC. Immunofluorescent staining was performed to determine ER expression and intracellular location of ER. ERα and ERβ were detected in both cytosolic and nuclear compartments of human coronary artery smooth muscle cells (Fig. 2). ERα antibody PA1-309 was generated against the N-terminal residues 21 to 32 of human ERα. This antibody is specific for ERα versus ERβ, because the N-terminal region of ERβ is not conserved and is considerably shorter in length than that in ERα (Kuiper et al., 1996). ER-PA1-309 antibody does not cross-react with ERβ. ERβ antibody PA1-311 was developed against the N-terminal residues 55 to 70 of human, mouse, and rat ERβ and displays no cross-reactivity with human ERα (Rosenfeld et al., 2000). Both PA1-309 and PA1-311 indicated mixed cytoplasmic and nuclear staining of HCASMC (Fig. 2, B, C, E, and F). In neutralization experiments, peptide preabsorption of the antibodies with 10 μg/ml synthetic peptides (PEP-037 and PEP-011, respectively; ABR–Affinity BioReagents) resulted in insignificant labeling at similar background levels (data not shown). Control studies in which nonimmune serum was substituted for primary antibody revealed only faint and diffuse background staining (Fig. 2, A and B). These results demonstrate that HCASMC express both ERα and ERβ.
Western Blot Analysis of ERα and ERβ in HCASMC. To further confirm the presence of ER in human coronary artery smooth muscle cells, Western blot analysis was performed. The same antibodies as above were used for the detection of ERs in homogenized cell lysates. Using the ERα antibody PA1-309, a single dark immunoreactive ER band was observed at 64 kDa in the human coronary artery smooth muscle culture cell extract from cell passage 5 through passage 9 at the same molecular mass position as MCF7 cell lysate, a positive control (Fig. 3A). Likewise, the ERβ antibody PA1-311 detected a specific immunoreactive band at 52 kDa in lysates from either HCASMC or MCF7 cell lysates (Fig. 3B). These findings are consistent with our immunofluorescent staining of which indicated expression of both ER subtypes in HCASMC. Transfection and functional studies were now performed to provide insights into which receptor subtype(s) likely mediated the response to HCASMC to estrogen.
ERα Transfection Studies. To investigate the role of ERα in mediating 17β-estradiol-stimulated BKCa channel activity, ion channel-recording experiments were done on HCASMC transiently cotransfected with pEGFP and either pCMV-ERα antisense or sense plasmid. However, before these studies were done, we first tested the effectiveness of sense/antisense expression by generating a stable ERα transfection of HEK293 cells. Western blots revealed expression of ERα in these transfected HEK293 cells compared with wild-type cells (Fig. 4). Further transfection of these stable ERα-HEK293 cells with ERα antisense plasmid greatly attenuated expression of ERα; however, ERα expression was not diminished by subsequent transfection with the vector plasmid alone. These studies verified the effectiveness of our ERα sense and antisense plasmids, and these DNA plasmids were then employed to study the functional role of ERα in HCASMC.
To confirm the effectiveness of sense/antisense plasmid transfection in HCASMC, we cotransfected these cells with both pEGFP and pCMV-ERα antisense or sense plasmid (gene ratio was 1:1) and observed 5 to 10% transfection efficiency. We then performed immunocytochemistry on these transfected HCASMC. The same primary antibodies as above were used for the detection both ERα and ERβ, but the secondary antibody was goat anti-rabbit Alexa Fluor 594. The ERα expression pattern was similar to that identified in Fig. 2. Results from four typical experiments (n = 3 separate experiments for each of the four conditions) are illustrated in Fig. 5. The top panel identifies expression of ERα (red), whereas the bottom panel is a picture of the same field indicating only those cells successfully cotransfected (∼5–10% efficiency, as illustrated) with pEGFP (green) and either ERα antisense/sense or vector plasmid. Transfection with ERα-AS abolished ERα expression, as illustrated in A and E. In contrast, transfection with the ERα gene (ERα-SS; B and F) enhanced expression of ERα protein in HCASMC. Cotransfection of HCASMC with the pEGFP and the pCMV vector plasmid (C, G, D, and H) demonstrated that the transfection process itself did not influence the expression of ERα. The same level of ERα expression was demonstrated in the vector-transfected cells (C and G). Furthermore, neutralization of ERα primary antibody completely abolished detection of ERα (D and H), indicating the specificity of the immunostaining.
In contrast to what we observed with ERα receptor expression, transfecting HCASMC with ERα sense/antisense plasmid did not alter the expression of ERβ (Fig. 6). As above, results are from four typical experiments (n = 3 separate experiments for each of the four conditions). The top panel identifies expression of ERβ (red), whereas the bottom panel is a picture of the same field indicating only those cells successfully cotransfected (∼5–10% efficiency, as illustrated) with pEGFP (green) and either ERα antisense/sense or vector plasmid. Transfection with ERα antisense/sense plasmid did not modify fluorescence, thus indicating no significant effect on expression of ERβ protein (A and E; B and F; and C and G). Neutralizing peptide for ERβ primary antibody abolished all ERβ immunostaining (D and H).
We had demonstrated previously that estrogen stimulates activity of BKCa channels (microscopic conductance of 186.5 ± 3 pS) expressed in wild-type HCASMC (White et al., 2002). We then employed single-channel patch-clamp studies of this protein as a sensitive molecular assay for ERα function. Recordings from cell-attached patches were obtained after a 24-h recovery from transfection, and experiments were performed only on HCASMC expressing EGFP (Fig. 7). In cells transfected only with EGFP, 100 nM 17β-estradiol increased BKCa channel activity (NPo) significantly (from 0.008 ± 0.004 to 0.120 ± 0.034; n = 11, P < 0.005; left). In contrast, estrogen had no effect on BKCa channels in HCASMC expressing ERα antisense plasmid (NPo from 0.003 ± 0.001 to 0.014 ± 0.009, n = 16; middle). The response of these transfected cells to 17β-estradiol is illustrated by the typical traces obtained from cell-attached patches, and the average effects are summarized by the bar graphs in Fig. 7B. To further confirm a role for ERα in mediating estrogen-stimulated channel activity, HCASMC were transfected with ERα sense plasmid to overexpress the receptor. Cell-attached patch-clamp studies on these cells revealed a highly significant stimulatory response to 100 nM 17β-estradiol, as average channel activity was increased from 0.0027 ± 0.001 to 0.293 ± 0.078 (n = 15; P = 0.001; right).
Cloning of a novel estrogen receptor subtype (ERβ) in 1996 (Kuiper et al., 1996) and the development of an ERα knockout mouse (ERαKO) the following year (Iafrati et al., 1997) brought with them the hope that much of the complexity of estrogen signaling could now be understood in terms of differences in ER expression and distribution. Unfortunately, our knowledge has not progressed as rapidly as originally predicted. One complicating factor is that, whereas in some tissues expression of only a single subtype of ER predominates (e.g., ERα in endometrium; ERβ in prostate) (Gustafsson, 2003), other tissues (e.g., VSM) express both ERα and ERβ. Delineating physiological roles for these receptors, especially in vivo, has been hampered by the unavailability of pharmacological agonists/antagonists exhibiting high selectivity for one subtype against the other. In contrast, the combination of cellular transfection techniques with single-cell/molecular physiology has allowed us to demonstrate that it is the ERα receptor that is both sufficient and necessary to regulate excitability of HCASMC in vitro (via production of NO and stimulation of BKCa channels). Indeed, it remains an open question how directly these experimental findings reflect estrogen action on the coronary circulation in vivo; however, support for our findings is gained from studies of ERαKO mice exhibiting a depressed coronary flow rate compared with wild-type mice (Zhai et al., 2000), thus suggesting an important role for ERα in regulating coronary artery function.
ERα and ERβ are present in both the nucleus and cytoplasm of HCASMC. Although ERβ was reported to be the more prevalent wild-type ER mRNA in human VSM cells (Hodges et al., 2000), our findings reveal prominent expression of ERα protein in HCASMC. Consistent with these detection studies, patch-clamp experiments revealed an important role for ERα in modulating cellular excitability. Abolishing ERα mRNA with ERα antisense plasmid fully eliminated the stimulatory effect of estrogen on BKCa channels. In contrast, overexpression of ERα protein enhanced the stimulatory effect of estrogen on channel gating. The specificity of ERα antisense/sense plasmid for ERα was verified by immunocytochemistry experiments on HCASMC that indicated modification of ERα protein expression after transfection but no effect of these probes on ERβ expression.
Taken together, these findings indicate that estrogen stimulates BKCa channel activity via ERα. Because this stimulation occurred within minutes, not hours, it is most likely a nongenomic effect of estrogen, which would be consistent with previous clinical studies demonstrating acute effects of estrogen on coronary blood flow and/or relief of myocardial ischemia in patients of both sexes (Rosano et al., 1993; Reis et al., 1994; Alpaslan et al., 1997; Blumenthal et al., 1997). These findings constitute molecular evidence for a mechanism that could help explain ER-mediated, endothelium-independent effects of estrogen on human arteries. Nonetheless, these findings will need to be repeated in primary coronary artery smooth muscle cells in order to confirm this hypothesis.
ERα and ERβ share a high degree of amino acid homology; however, the homology of N-terminal A/B domain is only 30%. In the present studies, 360-base pair ERα antisense gene from human ERα was designed to block only the 5′-flanking region of ERα mRNA, not ERβ mRNA. Transfection of ERα antisense plasmid fully blocked 17β-estradiol stimulation of the BKCa activity in HCASMC. In contrast, overexpression of ERα increased the responsiveness of BKCa channels to 17β-estradiol by over 10-fold. These findings strongly support an important role for ERα in mediating the influence of estrogen upon cellular excitability of HCASMC. Similar observations substantiating an important role for ERα in mediating estrogen responses have been made in other estrogen target tissue or cells. Overexpression of ERα led to marked enhancement of the acute activation of endothelial nitric-oxide synthase induced by 17β-estradiol in cultured endothelial cells (Rossouw et al., 2002). In addition, physiological levels of estradiol greatly reduced the extent of cerebral infarct in both wild-type mice and in ERβKO mice, but not in ERαKO mice (Paganini-Hill, 1995; Schmidt et al., 1996). The deletion of ERα totally abolished the protection afforded by estradiol in the brain, whereas deletion of ERβ did not diminish the ability of estradiol to protect the brain against injury (Dubal et al., 2001). Similar findings were obtained in osteoblasts, where estrogen inhibits cytokine production by repressing NF-κB activity via the ERα (Ray et al., 1994; Galien et al., 1996). Interestingly, a recent study has concluded that it is the ERα and not ERβ that mediates the ability of estrogen to protect against myocardial ischemia-reperfusion injury (Booth et al., 2005). Although the precise mechanism of estrogen action was not identified in this study of intact hearts, a protective role of ERα in myocardial function is perfectly consistent with the present findings demonstrating ERα-mediated effects on coronary artery cellular physiology. Therefore, the present findings and results from previous studies provide increasing evidence that ERα mediates many beneficial effects of estrogen in a variety of cell types and tissues.
In summary, our findings from cellular fluorescence, molecular transfection, and single-channel electrophysiology are entirely consistent with the conclusion that it is the ERα subtype that is the functionally dominant receptor mediating estrogen action in cultured HCASMC. These findings are supported by a number of studies suggesting that the classic ER, ERα, mediates estrogen action in a variety of target cells and tissues. Our findings in HCASMC are also consistent with studies revealing that the ERα is the functionally dominant receptor subtype when ERα/β heterodimers are formed via coexpression of ERα and ERβ (Li et al., 2004); however, we have not conclusively ruled out the possibility that other effects of estrogen on coronary arteries might be mediated via ERβ. Nonetheless, our findings provide evidence for the potential of ERα to serve as a target for therapeutic strategies to treat coronary heart disease or other abnormalities in coronary artery function. Future studies are needed to confirm the importance of ERα in regulating NO production and BKCa channel activity in intact arteries and primary cells.
We thank Dr. Benita Katzenellenbogen for the generous gift of ERα cDNA for these studies. We also acknowledge the most helpful technical support of Nicole L. Arevalo (ABR–Affinity BioReagents).
- Received July 29, 2005.
- Accepted November 16, 2005.
This work was supported by a Scientist Development Grant from the American Heart Association (to G.H.) and National Heart, Lung, and Blood Institute, National Institutes of Health Grant HL073890 (to R.E.W.).
ABBREVIATIONS: ER, estrogen receptor; VSM, vascular smooth muscle; HCASMC, human coronary artery smooth muscle cells; BKCa, large conductance, calcium- and voltage-activated potassium channel; HEK293, human embryonic kidney 293; TBS, Tris-buffered saline; BSA, bovine serum albumin; ERαKO, ERα knockout mouse; NPo, open-time probabilities; DAF-2 DA, 4,5-diaminofluorescein diacetate; l-NMMA, NG-monomethyl-l-arginine; ICI 182,780, 7-α-[9-(4,4,5,5,5-penta fluoropentylsulphinyl)nonyl]estra-,3,5-(10)-triene-3,17-β-diol.
- The American Society for Pharmacology and Experimental Therapeutics