Rapid effects of estrogen on intracellular Ca²⁺ regulation in human airway smooth muscle

Abstract

The severity of asthma, a disease characterized by airway hyperresponsiveness and inflammation, is enhanced in some women during the menstrual cycle and during pregnancy but relieved in others. These clinical findings suggest that sex steroids modulate airway tone. Based on well-known relaxant effects of estrogens on vascular smooth muscle, we hypothesized that estrogens relax airway smooth muscle (ASM), thus facilitating bronchodilation. In ASM tissues from female patients, Western and immunocytochemical analyses confirmed the presence of both estrogen receptor (ER) isoforms, ERα and ERβ. In fura 2-loaded, dissociated ASM cells maintained in culture, acute exposure to physiological concentrations of 17β-estradiol (E₂; 100 pM to 10 nM) decreased the intracellular Ca²⁺ ([Ca²⁺]_i) response to 1 μM histamine, an effect reversed by the ER antagonist ICI-182,780. The ERα-selective agonist (R,R)-THC had a greater reducing effect on [Ca²⁺]_i responses to histamine and 1 μM ACh compared with the ERβ-selective agonist (DPN). The effects of E₂ on [Ca²⁺]_i were mediated, at least in part, via decreased Ca²⁺ influx through l-type channels and store-operated Ca²⁺ entry but not via Ca²⁺-activated K⁺ channels, receptor-operated entry, or sarcoplasmic reticulum reuptake. Overall, these data support our hypothesis that estrogens relax ASM and suggest a potentially novel therapeutic target in airway hyperresponsiveness.

asthma, a disease characterized by airway hyperresponsiveness and inflammation, is more prevalent among adult women compared with men (11, 32, 45). Clinical data reveal an increased severity and frequency of exacerbations in women (2, 41). Cyclical variations in airway reactivity termed premenstrual asthma have been estimated to affect approximately 40–50 percent of women with preexisting disease (8, 37). Although these clinical data would suggest that sex steroids such as estrogens may have a detrimental effect in asthma, closer examination reveals that asthma exacerbation is actually greater during the late luteal phase when estrogen levels are lowest (8, 16, 18). Furthermore, some postmenopausal women experience decreased airway function (3). Finally, although asthma is the most prevalent airway disease endured by pregnant women (approximately 4–8%; Refs. 29, 51) during the third trimester of pregnancy (when circulating estrogen levels are high and the gravid uterus would be expected to compress the lungs), only one-third of women with preexisting disease will experience exacerbation of symptoms, and another one-third actually experience improvement in asthma symptoms (2, 24). Taken together, these contrasting clinical data suggest that sex steroids modulate airway tone. What is not clear is which sex steroid (estrogen vs. progesterone) and whether these hormones are detrimental or beneficial.

In both normals and asthmatics, airway tone represents a balance between bronchoconstriction and bronchodilation, mediated by airway smooth muscle (ASM) contractility vs. relaxation. ASM contractility is determined by both regulation of intracellular Ca²⁺ ([Ca²⁺]_i) and force (28, 40). Previous studies have already demonstrated that increased airway tone in diseases such as asthma involve, at least in part, increased [Ca²⁺]_i (25, 43, 50). Therefore, it is possible that sex steroids such as estrogens and progesterone modulate airway tone by altering [Ca²⁺]_i in ASM.

There are currently limited data on the role of sex steroids in regulation of airway tone or their effects on ASM, especially in humans. Measurements of airway resistance in mouse models of asthma have found both downregulation (5, 13, 31) as well as upregulation (6) of airway hyperresponsiveness by estrogens. Other studies in guinea pig have found that progesterone can induce ASM relaxation (44). However, in humans and mouse models, airway function is actually reduced when progesterone levels are high (20, 52). Based on the clinical data that asthma exacerbation is greater when estrogen levels are lower, we focused on estrogens and hypothesized that estrogens facilitate bronchodilation.

There is already a precedent for estrogen-induced relaxation of smooth muscle in the well-recognized vasodilatory effects of estrogens (34, 36). In this regard, as with vascular smooth muscle, estrogen may act via both acute (likely nongenomic) and genomic mechanisms in modulating ASM tone. In vascular smooth muscle, both of these effects are mediated through estrogen receptors (ERs), primarily ERα. Previous studies have shown that the ERα and ERβ isoforms are expressed in human lung (35); however, their functional role is unclear. Although ER expression has been noted in the lung, specific ER expression in ASM or its function have not been established. In the present study, we determined the expression of ERs in normal human ASM and their role in acute (nongenomic) regulation of [Ca²⁺]_i during agonist stimulation, thus setting the foundation for examining the role of estrogens in asthmatic ASM. In this regard, signaling pathways activated by the two ER isoforms are likely different [e.g., in other tissues, ERβ modulates ERα signaling, which, in turn, involves multiple signaling pathways such as mitogen-associated kinases and Src-phosphatidylinositol pathway (19, 30)]; however, we did not specifically examine these differential signaling pathways here.

MATERIALS AND METHODS

Isolation of Human ASM Cells

Studies were conducted using ASM cells enzymatically dissociated from 3rd to 6th generation bronchi of lung samples incidental to patient surgery at Mayo Clinic, Rochester, MN (approved and considered exempt by Mayo Institutional Review Board). This study was limited to tissues obtained from adult female patients, although no attempt was made to determine whether these patients were in menopause or not. Lung samples excised for empyema or other infectious or severe inflammatory causes were excluded. Visibly normal airway, as confirmed by pathological review, was excised and rapidly transferred to the laboratory in ice-cold HBSS. ASM cells were enzymatically dissociated as previously described by our group (46). Briefly, epithelium and connective tissues were removed by blunt dissection, and smooth muscle tissue was carefully minced followed by papain and collagenase dissociation and ovomucoid/albumin separation as per manufacturer's instructions (Worthington Biochemical, Lakewood, NJ). Cell pellets were resuspended and seeded into 75-cm² tissue culture flasks or eight-well Lab-Tek chambers (Nalge Nunc International, Rochester, NY). Cells were maintained at 37°C (5% CO₂-95% air) using phenol red-free DMEM/F-12 medium (phenol red is known to activate ERs, albeit weakly; Invitrogen, Carlsbad, CA) supplemented with 10% FBS until ∼80% confluent. Before experiments, the cells were washed in PBS, and medium was changed to phenol red-free DMEM/F-12 lacking serum for 48 h. All experiments were performed in cells from passages 1 to 3 of subculture. Periodic assessment of smooth muscle myosin heavy chain (or other contractile protein) expression, and lack of fibroblast markers, was performed to rule out dedifferentiation during the cell-processing period. ER expression in passaged cells was compared with expression in ASM tissue from which they were derived and found to be comparable.

Isolation of ASM Cell Fractions

Human ASM cells were obtained as described above and plated on 100-mm plates under the same conditions. Cells were grown to 80% confluence and harvested. These whole cell lysates were then separated into heavy (nuclear/Golgi), cytosolic, and membrane fractions using the Fraction-PREP Cell Fractionation System (BioVision, Mountain View, CA) according to the manufacturer's protocol.

Immunoprecipitation of Proteins from Cell Fractions

The above fractions were subjected to immunoprecipitation for either ERα or ERβ. Primary antibody (2 μg; mouse anti-ERα: cat. no. 2512, Cell Signaling Technology; mouse anti-ERβ: cat. no. SC-53494, Santa Cruz Biotechnology) was used per 200 μl of sample fraction and incubated overnight at 4°C with gentle rotation. Protein G agarose beads (50 μl) were added to the sample and incubated for 4 h at 4°C. Proteins were recovered through centrifugation and eluted from the beads at 100°C for 5 min. These samples were then processed as described below for Western analysis (goat-anti-ERβ SC-6822 was used for overnight incubation for ERβ detection). Purity of cell fractions was confirmed via Western analysis of proteins exclusive to each fraction (heavy: rabbit anti-c-Jun SC-1694; membrane: rabbit anti-G_αi-3 SC-262; cytosol: rabbit anti-ERK42 SC-94).

Western Blot Analysis

Proteins were separated by SDS-PAGE (10% gradient gels; Criterion Gel System; Bio-Rad, Hercules, CA) at 200 V for 1 h and transferred to polyvinylidene difluoride (PVDF) membranes (Bio-Rad) for 60–75 min. Membranes were blocked for 1 h at room temperature with 5% milk in Tris-buffered saline (TBS) containing 0.1% Tween (TBST) and then incubated overnight at 4°C with 1 μg/ml rabbit anti-ERα (SC-542; Santa Cruz Biotechnology) or mouse anti-ERβ (SC-53494; Santa Cruz Biotechnology). GAPDH (cat. no. 2118; Cell Signaling Technology) was used for normalization. Following three washes with TBST, primary antibodies were detected using horseradish peroxidase-conjugated secondary antibodies and signals developed by SuperSignal West Pico Chemiluminescent Substrate (Pierce Chemical, Rockford, IL). Blots were imaged on a Kodak Image Station 4000MM (Carestream Health, New Haven, CT) and quantified using densitometry.

Human ASM samples were processed as described above. Samples of postpartum rat uterus were used as positive controls for ER expression. To determine applicability of the primary antibodies to other human tissues, human pulmonary artery samples were also analyzed.

Immunofluorescence Microscopy

ER expression in ASM cells was determined using immunofluorescence as described for other proteins (47). ASM grown on glass slides to 50% confluence were fixed with 2% paraformaldehyde for 15 min at room temperature and permeabilized with 0.1% Triton X-100 in 0.1 M TBS for 10 s. After washing gently with TBS, cells were blocked in 4% normal donkey serum for 1 h at room temperature and incubated overnight at 4°C with primary antibody (1 μg/ml rabbit anti-ERα SC-542, goat anti-ERβ SC-6822). After thoroughly washing with TBS, cells were incubated with appropriate Cy3 or Alexa Fluor 488 fluorescent secondary antibodies (1:200 dilution; donkey anti-rabbit or anti-goat IgG; Jackson ImmunoResearch) for 1 h at room temperature.

Labeled ASM cells were visualized using an Olympus FluoView laser scanning confocal microscope mounted on an Olympus BX50WI equipped with Ar and Kr lasers and appropriate filters. Cells were imaged at 1,024 × 1,024 pixels and a 0.4-μm optical section thickness using a ×40 oil-immersion lens and different hardware zooms.

Real-Time [Ca²⁺]_i Imaging

The techniques for [Ca²⁺]_i imaging of human ASM cells has been previously described (47). ASM cells were incubated in 5 μM fura 2-AM (Molecular Probes, Eugene, OR) for 45 min at room temperature and visualized using a fluorescence imaging system (MetaFluor; Universal Imaging, Downingtown, PA) on a Nikon Diaphot Inverted Microscope (Fryer Instruments, Edina, MN) equipped with a Photometric Cascade digital camera system (Roper Scientific, Tucson, AZ). [Ca²⁺]_i responses of ∼15 cells per visual field were obtained using individual, software-defined regions of interest (×40 oil-immersion lens, 512 × 512 pixel resolution, 0.33-Hz image acquisition of 510-nm emissions following alternative excitation at 340 vs. 380 nm). Cells were initially perfused with 2.5 mM Ca²⁺ HBSS, and baseline fluorescence was established. Actual [Ca²⁺]_i levels were estimated using previously described calibration techniques for fura 2 (39, 47).

Materials

(R,R)-5,11-diethyl-5,6,11,12-tetrahydro-2,8-chrysenediol [(R,R)-THC] and diarylpropionitrile (DPN) were obtained from Tocris Biosciences (Ellisville, MO). 17β-Estradiol (E₂), ICI-182,780 (ICI), and other chemicals were obtained from Sigma Chemical (St. Louis, MO) unless mentioned otherwise. Tissue culture reagents, including DMEM/F-12 and FBS, were obtained from Invitrogen.

Statistical Analysis

Twelve bronchial samples from different female patients were used to obtain ASM cells. [Ca²⁺]_i experiments were performed in at least 25 cells each for 4 different bronchial (patient) samples, although not all protocols were performed in each sample obtained. In experiments where responses were compared in the presence or absence of specific drugs, paired t-test was used, whereas population studies were compared using unpaired t-test or 1-way ANOVA with repeated measures as appropriate. Bonferroni correction was applied for multiple comparisons. Statistical significance was established at P < 0.05. All values are expressed as means ± SE.

RESULTS

Expression of ERs in Human ASM

Western blot analysis of ASM (i.e., epithelium denuded, ASM isolated) from bronchial samples of female patients revealed that both full-length ERα and ERβ are expressed to a considerable extent, even compared with tissues known to have these isoforms such as postpartum rat uterus (n = 6; Fig. 1). Furthermore, human pulmonary artery also showed expression of these ER isoforms, albeit to different extents compared with ASM (Fig. 1A). Overall, in human ASM, ERα expression (calculated relative to GAPDH expression) was greater than that of ERβ (Fig. 1B). In other tissues, lower molecular weight ER isoforms (that lack specific domains) have been reported (15, 26). Such isoforms were also noted in human ASM (data not shown).

Fig. 1.

Expression of estrogen receptors (ERs) in human (h) airway smooth muscle (ASM). A: full-length ER isoforms -α (ERα-66) and -β (ERβ-56) are expressed in both ASM and pulmonary artery (PA) from lungs of female patients. Rat uterus (Ut.) was used as a positive control in both cases. B: densitometric analysis of relative ER expression normalized to GAPDH levels shows ERα is more prominent than ERβ in ASM. Values are means ± SE. *Significant difference between ERα and ERβ (P < 0.05). AU, arbitrary units.

Immunostaining of enzymatically dissociated ASM cells demonstrated plasma membrane, cytosolic, as well as diffuse nuclear expression of ERα. In comparison, ERβ was expressed at the plasma membrane also but to a lesser extent; however, distinct expression of this isoform was detected at the nucleus. Overall, ERα expression was more ubiquitous than ERβ with ERβ preferentially localizing to the nucleus (Fig. 2A). These results were confirmed by Western analysis of cell fractions illustrating ERα expression in heavy (nuclear/Golgi), cytosolic, and membrane fractions, whereas ERβ expression was expressed largely in the heavy fraction (Fig. 2B).

Fig. 2.

Localization of ERs in ASM cells. A: immunocytochemical staining of enzymatically dissociated ASM cells from female patients followed by confocal imaging demonstrated expression of both ERα (Cy3-conjugated secondary antibody; red) and ERβ (Alexa Fluor 488-conjugated secondary antibody; green). Top shows secondary staining controls, whereas middle and bottom shows ERα and ERβ expression (left, middle) as well as colocalization (merged images with yellow) at 2 different magnifications. ERα and ERβ are present at the plasma membrane as well as in the cytosol and nucleus, albeit to different extents. B: ER expression was determined in whole ASM cell lysates separated into heavy (nuclear/Golgi), membrane, and cytosolic fractions. Complementary to immunocytochemical staining, ERα was expressed by all 3 fractions, whereas ERβ expression was limited to the heavy fraction. Purity of the cell fractions was tested by blotting for proteins exclusive to each fraction. B represents original scan contrasts and brightness, which differed between blots.

Effect of E₂ on [Ca²⁺]_i Response to Agonist

In the first set of studies, fura 2-loaded human ASM cells were exposed to 1 μM histamine, resulting in the typical transient [Ca²⁺]_i elevation to a peak level followed by a lower plateau that was still greater than baseline Ca²⁺ (Fig. 3A). Baseline [Ca²⁺]_i levels ranged from 100 to 150 nM. With exposure to histamine, the amplitude of the [Ca²⁺]_i response was ∼750 nM. Following a thorough washout of histamine with 2.5 mM Ca²⁺ HBSS for at least 15 min, cells were exposed to 1 nM E₂ for 15 min. This resulted in a small but insignificant rise in [Ca²⁺]_i basal levels. Subsequent exposure to histamine resulted in a significantly smaller [Ca²⁺]_i response compared with vehicle controls (n = 4; P < 0.05; Fig. 3, A and B). To determine whether these results were specific to histamine, the above experiment was repeated in separate sets of cells using ACh (1 μM) as an alternative agonist and KCl (110 mM) as a nonreceptor agonist that changes membrane potential. In general, the inhibitory effects of E₂ on [Ca²⁺]_i responses to ACh persisted (n = 4; P < 0.05; Fig. 6A); however, there was no difference in the peak [Ca²⁺]_i response between E₂-treated cells and vehicle controls with KCl depolarization (data not shown). Further analysis of [Ca²⁺]_i response to KCl showed that E₂ extended the time required to reach peak [Ca²⁺]_i levels (n = 4; P < 0.05; Fig. 6B).

Fig. 3.

Estrogen decreases intracellular calcium ([Ca²⁺]_i) response to histamine in human ASM cells. A: in cells loaded with fura 2 and perfused with 2.5 mM Ca²⁺ HBSS, 1 μM histamine resulted in a rise in [Ca²⁺]_i (black). In separate cells, pretreatment with 1 nM 17β-estradiol (E₂) decreased the response to histamine by ∼40% (dotted). Pretreatment with the isoform-nonspecific ER antagonist ICI-182,780 (ICI; 1 μM; Faslodex) reversed E₂ effects (gray), returning peak response to that of vehicle controls. See materials and methods for details of protocol. B: summary of E₂ and ICI effects (n = 4). Values are means ± SE. *Significant effect of E₂ (P < 0.05).

Fig. 6.

Role of ER isoforms in [Ca²⁺]_i responses to ACh and KCl. A: in human ASM cells, pretreatment with 1 nM E₂ or 10 nM (R,R)-THC (but not 10 nM DPN) significantly reduced peak Ca²⁺ response to 1 μM ACh (n = 4). These effects were qualitatively similar to E₂ and (R,R)-THC effects on responses to histamine. B: in human ASM cells depolarized with 110 mM KCl, pretreatment with 1 nM E₂ or 10 nM (R,R)-THC prolonged the time required to reach peak [Ca²⁺]_i compared with vehicle controls (n = 4). Values are means ± SE. *Significant drug effect compared with vehicle control (P < 0.05).

In a subsequent set of studies, seven concentrations of E₂ that span the physiological range were used to establish a dose-response relationship for effects on [Ca²⁺]_i response to histamine. Because of known tachyphylaxis with histamine, as well as the number of interventions, these studies had to be performed as comparisons of cell populations. ASM cells were exposed to 0 (vehicle control), 1 pM, 10 pM, 100 pM, 1 nM, 10 nM, 100 nM, and 1 μM E₂ for 15 min and then stimulated with 1 μM histamine. Compared with vehicle controls, there was considerable decrease in amplitude of [Ca²⁺]_i responses for all E₂ concentrations, forming a biphasic dose-response curve (Fig. 4). All E₂ concentrations in the physiological range (100 pM to 10 nM) significantly decreased the [Ca²⁺]_i amplitude, with the maximum reduction in [Ca²⁺]_i occurring at 1 nM E₂ (n = 4; P < 0.05). Accordingly, for subsequent studies, we used 1 nM E₂ to further explore estrogen signaling.

Fig. 4.

Dose-response of E₂ effects on [Ca²⁺]_i response to histamine. In separate cell populations, pretreatment with E₂ in concentrations ranging from 1 pM to 1 μM resulted in a consistent, significant decrease in [Ca²⁺]_i in response to histamine. However, maximum reduction in [Ca²⁺]_i was observed at 1 nM, whereas higher concentrations were less effective in reducing [Ca²⁺]_i levels. Overall, E₂ effects were most prominent within the physiological range 100 pM to 10 nM. Values are means ± SE. *Significant effect of E₂ (P < 0.05).

Effect of ER Inhibition on E₂ Effects

To ascertain whether the blunting of [Ca²⁺]_i responses by E₂ was mediated by ERs, in an additional set of experiments, 1 μM isoform-nonspecific ER antagonist ICI (Faslodex) was administered 15 min before 1 nM E₂ exposure and was continued during the subsequent 15 min in combination with E₂ (n = 4). In the presence of ICI, the previously observed blunting of [Ca²⁺]_i response by E₂ was abolished, and [Ca²⁺]_i response to histamine approximated that of vehicle controls (n = 4; Fig. 3).

Effect of ER Isoform-Selective Agonists

E₂ is a nonspecific ER agonist and does not differentiate between ERα and ERβ activation. To determine which receptor subtype is involved in modulation of [Ca²⁺]_i, ERα- and ERβ-selective agonists were used. (R,R)-THC is a nonsteroidal, ERα-selective agonist (EC₅₀ = 10 nM; Ref. 53) but also an ERβ-selective antagonist. DPN is an ERβ-selective agonist (EC₅₀ = 0.85 nM; Ref. 33). The protocol described above to test E₂ effects was modified by replacing E₂ with either (R,R)-THC or DPN.

Acute exposure of ASM cells to 10 nM (R,R)-THC decreased [Ca²⁺]_i response to ∼50% of vehicle control (n = 4; P < 0.05; Fig. 5). The reduction of [Ca²⁺]_i by (R,R)-THC approximated the observed effects of E₂. In contrast, 10 nM DPN decreased the [Ca²⁺]_i response to a much lesser extent than (R,R)-THC and was statistically comparable with vehicle control (Fig. 5; n = 4). The combination of (R,R)-THC and DPN did not appear to have additive effects on [Ca²⁺]_i responses.

Fig. 5.

Role of ER isoforms in estrogen-induced reduction of [Ca²⁺]_i. A: in human ASM cells, pretreatment with 10 nM ERα-specific agonist (R,R)-THC (THC; see results for details) caused ∼45% reduction in peak Ca²⁺ response to histamine, comparable with the effect of 1 nM E₂ alone. Pretreatment with ERβ-specific agonist DPN (10 nM) resulted in a smaller decrement of [Ca²⁺]_i responses to histamine. B: summary of (R,R)-THC and DPN effects on [Ca²⁺]_i responses (n = 4). Values are means ± SE. *Significant drug effect compared with vehicle control (P < 0.05).

These experiments were repeated using ACh and KCl as described previously. In the case of ACh, the results were qualitatively similar to those with histamine. [Ca²⁺]_i was decreased in both cases by (R,R)-THC to ∼50% of control, whereas DPN did not decrease [Ca²⁺]_i significantly (n = 4; P < 0.05; Fig. 6A). There was no difference in the peak [Ca²⁺]_i response between (R,R)-THC- or DPN-treated cells and vehicle controls with KCl depolarization (data not shown). Analysis of [Ca²⁺]_i response to KCl showed that (R,R)-THC further extended the time required to reach peak [Ca²⁺]_i levels compared with vehicle controls and E₂ treatment, whereas DPN response approximated that of controls (n = 4; P < 0.05; Fig. 6B). The use of these additional agents illustrates that the E₂ effect is not specific to the agonist histamine but rather is likely related to Ca²⁺ influx triggered by either receptor-activated channels or membrane depolarization.

Mechanisms of E₂ Effects on [Ca²⁺]_i Regulation

Ca²⁺ influx.

Previous studies (including our own) in vascular smooth muscle have demonstrated that estrogens decrease Ca²⁺ influx (27, 48). Accordingly, we tested the role of influx mechanisms in estrogen effects in human ASM. The protocol described above for ICI was modified by replacing this compound with the l-type Ca²⁺ channel inhibitor nifedipine (1 μM). In the presence of nifedipine, we found that 1 nM E₂ effects on [Ca²⁺]_i plateau levels to 1 μM histamine were considerably smaller, compared with effects with E₂ alone (P < 0.05; n = 4; Fig. 7A). However, when similar studies were conducted in the presence of 100 nM iberiotoxin (to inhibit BK_Ca channels), E₂ effects were largely unaffected (n = 4; P < 0.05; Fig. 7B). Addition of iberiotoxin resulted in a ∼50% increase in plateau Ca²⁺ levels compared with vehicle controls (likely representing Ca²⁺ influx in response to membrane depolarization; Fig. 7B). E₂ in the presence of iberiotoxin reduced these elevated plateau levels by ∼50% compared with iberiotoxin alone (P < 0.05). The extent of E₂-induced decrease in Ca²⁺ was comparable with E₂ effects on plateau Ca²⁺ in cells not exposed to iberiotoxin.

Fig. 7.

Mechanisms underlying E₂ effects on [Ca²⁺]_i regulation in human ASM cells. A: pretreatment with the l-type Ca²⁺ channel inhibitor nifedipine (1 μM) blunted E₂ effects on [Ca²⁺]_i responses to histamine (n = 4), suggesting a partial role for this influx mechanism in E₂ effects. B: however, pretreatment with the Ca²⁺-activated K⁺ channel inhibitor iberiotoxin did not considerably change E₂ effects (but did increase the [Ca²⁺]_i response to histamine as expected due to membrane depolarization; n = 4). C: using previously published protocols (1), assessment of store-operated Ca²⁺ entry (SOCE) showed that preexposure to E₂ significantly decreased the extent of Ca²⁺ influx by this mechanism [n = 4; cyclopiazonic acid (CPA), inhibitor of sarcoplasmic reticulum (SR) Ca²⁺ reuptake; see materials and methods for protocol details]. Pretreatment with the receptor-operated channel inhibitor SKF-96365 or the Na⁺-Ca²⁺ exchange inhibitor KB-R7943 slightly decreased Ca²⁺ influx as evaluated using the SOCE protocol. Nonetheless, even in the presence of either inhibitor, E₂ significantly blunted Ca²⁺ influx, indicating a predominant effect on SOCE per se (n = 4). Values are means ± SE. *Significant effect of E₂; #significant inhibitor effect (P < 0.05). Veh, vehicle.

In parallel studies, we examined E₂ effects on store-operated Ca²⁺ entry (SOCE). We (1, 39) have previously established the presence of this Ca²⁺ influx mechanism in human ASM cells. As previously described (1, 39), extracellular Ca²⁺ was first removed by perfusion with zero Ca²⁺ HBSS. Calcium influx through voltage-gated channels was then blocked by addition of 1 μM nifedipine, and the membrane was partially depolarized using 10 mM KCl. Sarcoplasmic reticulum (SR) Ca²⁺ stores were then depleted by exposure to 10 μM cyclopiazonic acid (CPA; inhibitor of SR Ca²⁺ reuptake). Following stabilization of [Ca²⁺]_i levels, 1 nM E₂ (or vehicle) was introduced. In the continued presence of nifedipine, KCl, CPA, and E₂ (or vehicle), extracellular Ca²⁺ was rapidly reintroduced, triggering SOCE. The presence of E₂ resulted in significantly smaller SOCE-mediated Ca²⁺ influx compared with vehicle controls (n = 4; P < 0.05; Fig. 7C).

We (55) have previously demonstrated that SOCE in ASM involves transient receptor potential (TRP) channels. However, Ca²⁺ influx on reintroduction of extracellular Ca²⁺ in the protocol used may also be due to other influx mechanisms that are not TRP-dependent, particularly receptor-operated calcium entry (ROC) or influx mode sodium-calcium exchange (NCX). To determine the role of these other mechanisms in E₂ effects, the above protocol was modified by introducing either 10 μM SKF-96365 (a potent inhibitor of ROC) or 5 μM KB-R7943 (NCX inhibitor) for 15 min following CPA exposure but before reintroduction of extracellular calcium. E₂ (1 nM) was added thereafter. ROC inhibition decreased the observed influx following introduction of extracellular Ca²⁺; however, even in the presence of SKF-96365, 1 nM E₂ inhibited the influx to the same extent as it did without SKF-96365 being present (Fig. 6C; n = 4; P < 0.05). NCX inhibition slightly decreased the extent of influx, and E₂ continued to inhibit the influx to the same extent as that without KB-R7943 being present (n = 4; P < 0.05; Fig. 7C).

SR Ca²⁺ reuptake.

The initial protocol for E₂-treated cells (e.g., Fig. 3) was repeated, and the decay time constant of the [Ca²⁺]_i responses with or without 1 nM E₂ was evaluated by single exponential decay fit (calculated from peak [Ca²⁺]_i until return to baseline; n = 3; Fig. 8). Although the amplitude of [Ca²⁺]_i response was decreased as reported above, the rate of fall was not significantly affected by E₂ (Fig. 8).

Fig. 8.

A and B: role of SR Ca²⁺ reuptake in E₂ effects. The rate of fall of the [Ca²⁺]_i response to histamine (calculated as a decay time constant, τ, using a single exponential fit), representing Ca²⁺ reuptake as well as efflux, was largely unaffected by preexposure to E₂. Absence of plasma membrane Ca²⁺ fluxes (inhibited by 0 Ca²⁺ and La³⁺) did not affect this conclusion, suggesting lack of E₂ effect on SR Ca²⁺ reuptake.

To verify lack of E₂ effect on SR Ca²⁺ reuptake, additional experiments were performed in a separate set of cells in the absence of extracellular Ca²⁺. Following initial perfusion in HBSS, extracellular Ca²⁺ was removed by perfusion with zero Ca²⁺ HBSS, and Ca²⁺ influx was nonspecifically blocked with 1 mM LaCl₃. This resulted in a typical, small decrease in baseline [Ca²⁺]_i levels. Cells were then exposed to 1 μM histamine in the presence or absence of 1 nM E₂. Under these conditions, the [Ca²⁺]_i responses were comparable (vehicle 363 ± 40 nM; E₂ 352 ± 17 nM; n = 3; Fig. 8). In parallel experiments, E₂ effects on sarco(endo)plasmic reticulum Ca²⁺-ATPase (SERCA) reuptake were evaluated. In identical extracellular conditions described above, preexposure to 1 nM E₂ did not alter the rate of decline of [Ca²⁺]_i following histamine exposure as calculated from [Ca²⁺]_i peak to baseline value as described above (n = 3; Fig. 8).

DISCUSSION

ASM is the main regulator of airway hyperreactivity and inflammation and has become an important target with the recent increases in asthma diagnoses. Sex disparities in clinical findings as well as changes in airway function with the menstrual cycle, pregnancy, and menopause have raised the question of hormonal influences on ASM. The few studies that have looked at the effects of sex steroids on airway function are conflicting. ERs are expressed in a wide variety of tissues including vascular endothelium and smooth muscle and even the lung; however, few studies have examined ER expression specifically within ASM, especially in humans. We have shown for the first time that both full-length ERα and ERβ are present in ASM of female patients and that activation of these receptors results in reduction of agonist-induced elevations in [Ca²⁺]_i. These novel data set the foundation for examining dysregulation of ER signaling as a potential mechanism underlying sex differences in airway diseases such as asthma.

[Ca²⁺]_i Regulation in ASM

Airway tone reflects a balance between bronchoconstriction and bronchodilation. A key element of airway tone is ASM contractility, which, in turn, is determined to large extent by [Ca²⁺]_i. Elevation of [Ca²⁺]_i in ASM following bronchoconstrictor stimulation involves both SR Ca²⁺ release and plasma membrane Ca²⁺ influx (49). Subsequent decrease in [Ca²⁺]_i and maintenance of Ca²⁺ homeostasis involve SR Ca²⁺ reuptake and plasma membrane efflux. In this regard, the role of both SR and plasma membrane in [Ca²⁺]_i responses to histamine are well-recognized. Ca²⁺ influx occurs in ASM through both voltage-gated (57) and receptor-gated (38) channels. Ca²⁺ influx may also be indirectly modulated via alterations in plasma membrane potential, mediated via mechanisms such as iberiotoxin-sensitive Ca²⁺-activated K⁺ channels, although its contribution may be species-dependent (9, 50). In addition, recent studies, including our own, have demonstrated controlled Ca²⁺ influx in response to SR Ca²⁺ depletion (SOCE; Refs. 1, 42). SOCE involves members of the TRP channel including TRP canonical (TRPC) 1, 3, 4, 5, and 6 (10, 55). In addition, members of the TRP vanilloid (TRPV) family (ROC) can also contribute to Ca²⁺ influx in ASM [e.g., with changing osmolarity (12, 23)]. Furthermore, some studies suggest a role for NCX in ASM (21). Accordingly, multiple [Ca²⁺]_i regulatory targets exist in human ASM for estrogens. The results of the present study indicate that E₂ indeed inhibits several mechanisms but almost exclusively at the plasma membrane.

Estrogen Signaling

In humans, E₂ is the most potent ER agonist. Two full-length ERs of similar estrogen binding affinities are of significance: ERα and ERβ both belonging to the nuclear receptor family of transcription factors (34). The role of “truncated” or shorter ERα isoforms (ERα-46 and ERα-36) that lack activation factor AF-1 is not clear (19). Accordingly, in the present study, we focused on the full-length ERs.

ERs are expressed in a wide variety of tissues such as vascular endothelium and smooth muscle. However, few studies have examined ER expression within the airway, especially in humans. ERα and ERβ are expressed in human bronchial epithelial cells (22), however, expression in ASM or innervation in vivo is not known. Again, based on perimenstrual fluctuations in asthma symptoms, ER expression is likely. In a recent study using gene and protein profiling techniques, Catley et al. (7) detected ERβ in human ASM cells in vitro exposed to the cytokine IL-1β. The present study is the first to directly demonstrate ER expression within ASM of bronchi of female patients. Although we did not limit selection of patients by age, hormone status, or airway disease, our findings of significant expression of both ERα and ERβ along with truncated isoforms suggest that ERs are likely expressed across these different groups. Growing ASM cells for 48–72 h also maintained full-length ER expression, consistent with the findings of Catley et al. (7).

The well-known transcriptional (genomic) response to estrogens is complex and cell-specific, depending (at least) on estrogen concentration and duration of exposure, coregulatory proteins, promoters in estrogen-responsive genes, active genes, and ERβ-ERα interactions (19). Given these complexities and the relative lack of knowledge on estrogen signaling in the airway, we did not address the issue of prolonged estrogen exposure in ASM exposure or genomic signaling.

In other tissues, rapid effects of E₂ (seconds to minutes) have been demonstrated (30, 36). It is not entirely clear whether nongenomic ER signaling involves the same intracellular/nuclear receptors as in genomic effects, but the observed effects are dependent on E₂ concentration. Accordingly, in the present study, we initially selected a range of E₂ concentrations that spans the physiological range (1–10 nM during menses).

Estrogen, ERs, and effects on smooth muscle.

In the vasculature, acute estrogen exposure leads to blunting of agonist-induced [Ca²⁺]_i responses (34). In comparison, relatively little is known about estrogen effects on [Ca²⁺]_i in the airway. ERα-deficient mice exhibit increased airway hyperresponsiveness (6), and hormone replacement in ovariectomized animals attenuates allergen-induced asthma (13), suggesting that estrogens contribute to bronchodilation (and/or decreased inflammation), consistent with our hypothesis. However, the role of estrogen signaling may be species- and model-specific, and the underlying mechanisms need further investigation. A single study in mouse tracheal and bronchial rings sensitized by serum from asthmatic humans demonstrated reductions in carbachol-induced contractions by 30-min exposure to 100 nM E₂ (14). Our study demonstrates that even 1 nM E₂ can acutely and substantially reduce [Ca²⁺]_i responses of human ASM cells to agonists such as histamine. These novel data strongly support the idea of acute, nongenomic bronchodilatory effects of estrogens in the human airway.

Varying concentrations of E₂ produced a nonclassic dose-response curve in human ASM for [Ca²⁺]_i effects. Although the underlying mechanisms are not clear, other studies have noted the antagonizing effects of ERβ activation on estrogen effects, mostly in the context of breast cancer, ERα-dependent transcription, and ERα degradation (19, 56). Given the likely complex role of ERα vs. ERβ signaling, specific signaling mechanisms were not examined. Regardless, what is important to note is that significant reductions in [Ca²⁺]_i were observed in the physiological range of E₂ concentrations.

Previous work by Montgomery et al. (36) and Levin (30) in vascular smooth muscle have tested the efficacy of selective ER agonists on relaxation compared with the nonspecific E₂. In their work, the ERα agonist 4,4′,4″-(4-propyl-[¹H]-pyrazole-1,3,5-triyl) trisphenol (PPT) was used. (R,R)-THC was selected as the ERα-specific agonist in our study because it also acts as an ERβ antagonist and has similar binding affinity for ERα as PPT. We found that even low concentrations of ERα-specific agonist produced substantial reduction in [Ca²⁺]_i, mimicking effects of the nonspecific E₂, whereas an ERβ agonist was less effective. These data also complement the findings of Montgomery et al. (36) that PPT was more effective than E₂ and DPN at relaxing mesenteric arteries, whereas DPN was less effective than both PPT and E₂. Therefore, it is possible that although both isoforms are present in human ASM, nongenomic effects may be mediated via only ERα.

There are very little data on estrogen effects on Ca²⁺ influx in the airway. The present study found that E₂ effects on [Ca²⁺]_i were blunted by inhibiting l-type channels (nifedipine). In isolated mouse tracheal or bronchial rings, Dimitropoulou et al. (14) found that estrogen effects are partly inhibited by blocking BK_Ca channels (similar to vasculature; Ref. 48). However, in human ASM, we were unable to detect any effect of iberiotoxin on estrogen-induced reduction of [Ca²⁺]_i. There are no data on estrogen modulation of SOCE in the airway. During these studies, we noted that on introduction of E₂ when [Ca²⁺]_i levels were elevated by the SERCA inhibitor CPA, there was an abrupt reduction in [Ca²⁺]_i, raising the possibility of accelerated NCX-mediated efflux. This effect was abolished in KB-R7943-treated cells, however, KB-R7943 had little effect on E₂ modulation of SOCE-mediated influx per se, suggesting that even if NCX was present in human ASM, estrogens may not specifically inhibit this mechanism.

As mentioned above, SOCE-mediated Ca²⁺ influx occurs in ASM and involves TRPC channels. Although estrogens do not appear to modulate TRP channel expression (17), there is no information on interactions between ER activation and TRP (specifically TRPC) channels in terms of [Ca²⁺]_i regulation. In the present study, pretreatment with SKF-96365, a potent inhibitor of receptor-operated channels, did not affect the observed E₂ effect on SOCE-mediated Ca²⁺ influx, indicating that the effect of estrogens is specifically on SOCE rather than other influx mechanisms not involving TRPC.

Studies in vascular smooth muscle (27) have found no evidence for estrogen effects on SR Ca²⁺ reuptake, a potential mechanism for reducing [Ca²⁺]_i. Ovariectomy or prolonged changes in estrogen levels do alter SERCA expression but not activity (4). In the present study, we also did not find any observable impact of acute E₂ exposure on SR Ca²⁺ reuptake.

Physiological and Clinical Relevance

The present study demonstrates that in human ASM, functional ERs exist and that estrogens acutely (likely in a nongenomic fashion) reduce agonist-induced [Ca²⁺]_i levels, akin to the well-known effects in the vasculature. Such estrogen effects appear to involve inhibition of Ca²⁺ influx via different mechanisms. The relevance of these findings lie in the potential modulation of airway tone by ER signaling, or its dysregulation in airway disease in women. In this regard, further study is required to understand how estrogens inhibit Ca²⁺ influx. In other tissues, estrogens can increase cAMP (54), which produces bronchodilation by targeting plasma membrane mechanisms including influx channels. Whether estrogens modulate cAMP or other pathways involved in bronchodilation is a potentially clinically relevant question that will be examined in future studies.

GRANTS

This study was supported by National Heart, Lung, and Blood Institute Grants HL-088029 (Y. S. Prakash) and HL-090595 (C. M. Pabelick) and by the Flight Attendant Medical Research Institute (FAMRI; Y. S. Prakash). E. A. Townsend is a graduate student in Y. S. Prakash's laboratory and is supported by the Mayo Graduate School.

DISCLOSURES

No conflicts of interest are declared by the author(s).