G protein-coupled estrogen receptor 1-mediated effects in the rat myometrium

Abstract

G protein-coupled estrogen receptor 1 (GPER), also named GPR30, has been previously identified in the female reproductive system. In this study, GPER expression was found in the female rat myometrium by reverse transcriptase-polymerase chain reaction and immunocytochemistry. Using GPER-selective ligands, we assessed the effects of the GPER activation on resting membrane potential and cytosolic Ca²⁺ concentration ([Ca²⁺]_i) in rat myometrial cells, as well as on contractility of rat uterine strips. G-1, a specific GPER agonist, induced a concentration-dependent depolarization and increase in [Ca²⁺]_i in myometrial cells. The depolarization was abolished in Na⁺-free saline. G-1-induced [Ca²⁺]_i increase was markedly decreased by nifedipine, a L-type Ca²⁺ channel blocker, by Ca²⁺-free or Na⁺-free saline. Intracellular administration of G-1 produced a faster and transitory increase in [Ca²⁺]_i, with a higher amplitude than that induced by extracellular application, supporting an intracellular localization of the functional GPER in myometrial cells. Depletion of internal Ca²⁺ stores with thapsigargin produced a robust store-activated Ca²⁺ entry; the Ca²⁺ response to G-1 was similar to the constitutive Ca²⁺ entry and did not seem to involve store-operated Ca²⁺ entry. In rat uterine strips, administration of G-1 increased the frequency and amplitude of contractions and the area under the contractility curve. The effects of G-1 on membrane potential, [Ca²⁺]_i, and uterine contractility were prevented by pretreatment with G-15, a GPER antagonist, further supporting the involvement of GPER in these responses. Taken together, our results indicate that GPER is expressed and functional in rat myometrium. GPER activation produces depolarization, elevates [Ca²⁺]_i and increases contractility in myometrial cells.

estrogens are steroid hormones with a critical role in reproduction. The effects of 17 β-estradiol, the most bioactive estrogen, are mediated by nuclear estrogen receptors (ER) α (ERα) and β (ERβ) and by the newly identified G protein-coupled estrogen receptor 1 (GPER), also named GPR30 (18, 39, 44). Several reports indicate the involvement of GPER in breast, ovarian, and uterine cancer (1, 2, 16, 17, 23, 44, 45), while the physiological role of GPER in the reproductive system is largely unknown.

In the ovary, GPER is expressed during perinatal development and mediates estrogen action on primordial follicle (47). Nongenomic actions of steroids were reported earlier in myometrium (34). However, whether or not GPER is involved in myometrial functions remains to be clarified. The present study was undertaken to investigate the effects of GPER activation on membrane potential, cytosolic Ca²⁺ concentration ([Ca²⁺]_i) and contractility on rat myometrium, using the selective GPER agonist G-1 (4) and antagonist G-15 (14).

MATERIALS AND METHODS

Primary Culture of Myometrial Cells

Primary myometrial cells were prepared from rat uterus according to the procedure described by Krizsan-Agbas et al. (26) with some modifications (13). Animal protocols were reviewed and approved by the Institutional Animal Care and Use Committee. Adult (nonpregnant) female Sprague-Dawley rats (Ace Animals, Boyertown, PA) were anesthetized with urethane (1.2 g/kg ip); uterine horns were immediately excised and immersed in cold Hanks balanced salt solution (HBSS). The myometrium was rapidly separated from endometrium and stroma and then minced into fragments (about 1 mm³). Tissues were digested for 1.5 h at 37°C by collagenase type II (600 U/ml, Sigma-Aldrich, St. Louis, MO) in DMEM/F12 culture medium, followed by trituration with a sterile Pasteur pipette and filtration through a cell strainer with 70-μm pores. After collection, the cells were washed twice by centrifugation (1,000 RPM, 10 min) and resuspended in fresh DMEM/F12 medium supplemented with 10% fetal bovine serum, 100 U/ml penicillin, and 100 μg/ml streptomycin and then plated on collagen-coated coverslips. Cultures were maintained at 37°C in a humidified atmosphere with 5% CO₂; the medium was changed every 2–3 days. Myometrial cells were used for calcium imaging experiments after 3–5 days in culture.

Reverse Transcriptase-Polymerase Chain Reaction

Total RNA from rat myometrium was isolated using TRIzol reagent according to the manufacturer's protocol (Invitrogen, Carlsbad, CA). The purity and concentration of RNA were assessed by spectrophotometrical analysis. First-strand cDNA was synthesized from 3 μg of total RNA with random hexamer primers and the SuperScript III reverse transcriptase according to the manufacturer's protocol (Invitrogen). Three microliters of 20 μl cDNA product were amplified with the following specific primers based on GPER (NM_133573) and β-actin (NM_031144) sequences: GPER: 5′-TCTGTGACCCTTTCAGCAAGTCCT-3′ (upstream), 5′-AGAAGGCTGCCAGGTTGACTATGT-3′ (downstream); β-actin: 5′-CCATTGAACACGGCATTGTCACCA-3′ (upstream), 5′-ACTCCTGCTTGCTGATCCACATCT-3′ (downstream). Amplifications were performed using Platinum Taq DNA polymerase (Invitrogen) as follows. GPER: 38 cycles of 94°C for 60 s, 58°C for 60 s, and 72°C for 60 s; β-actin: 30 cycles of 94°C for 30 s, 55°C for 45 s, and 72°C for 60 s. The polymerase chain reaction products (987 bp and 876 bp, respectively) were visualized by an ethidium bromide staining under UV light following electrophoresis on 1% agarose gel.

Immunocytochemistry

Myometrial cells on coverslips were fixed in 4% paraformaldehyde for 20 min. After being blocked with normal donkey serum (1:50 in PBS, 0.5% BSA, 0.4% Triton X-100), the cells were incubated with GPER/GPR30 antibody (1:200 dilution) as previously described (5). GPER antibody was a rabbit polyclonal antibody directed against the human COOH-terminus (MBL International, Woburn, MA). For control experiments, the antibody was incubated with a 10× excess of peptide at 4°C overnight. After several washes with PBS, cells were incubated with rabbit IgG Texas Red (1:30, Vector Laboratories, Burlingame, CA). The cell nuclei were stained by mounting with DAPI-Fluoromount-G (Southern Biotech, Birmingham, AL). Sections were examined under a confocal scanning laser microscope (Leica TCS SL) with excitation wavelengths set to 405 nm for DAPI and 561 nm for Texas Red, in the sequential mode.

Optical Imaging Using DiBAC₄(3)

Bis-oxonol (bis-[1,3-dibutylbarbituric acid] trimethineoxonol) [DiBAC₄(3)], a slow-response voltage-sensitive fluorescent dye, has been successfully used to assess relative changes in membrane potential of single cells (15). The method was similar to that previously described (8, 9). Briefly, myometrial cells were incubated for 30 min in HBSS containing 0.5 mM DiBAC₄(3). The fluorescence (excitation wavelength = 480 nm, emission wavelength = 540 nm) was continuously recorded at a rate of 10 points/min. Background values (windows of identical area placed beside the cells) were subtracted. The dye partition between the cell membrane and the cytosol is a function of membrane potential. Depolarization of the membrane leads to a sequestration of the dye into cytosol and is associated with an increase in fluorescence intensity; whereas the dye concentrates in the cell membrane during hyperpolarization, leading to a decrease of fluorescence intensity in cytosol (11). Calibration of DiBAC₄(3) fluorescence was performed using the Na⁺-K⁺ ionophore gramicidin in Na⁺-free physiological solution (11). In experiments with ionic substitution, Na⁺ was replaced with N-methyl-d-glucamine.

Calcium Imaging

Intracellular Ca²⁺ measurements were performed as previously described (5, 7, 10, 13). Briefly, cells were incubated with 5 μM fura-2 AM (Invitrogen) in HBSS at room temperature for 45 min, in the dark, washed three times with dye-free buffer, and then incubated for another 45 min to allow for complete deesterification of the dye. Coverslips were subsequently mounted in a custom-designed bath on the stage of an Eclipse TE 2000-U Nikon microscope equipped with a Roper Scientific CCD camera (Optical Apparatus, Ardmore, PA) or on a stage of a Nikon Eclipse TiE microscope equipped with a Perfect Focus System and a Photometrics CoolSnap HQ2 CCD camera (Optical Apparatus). Fura-2 AM fluorescence (emission = 510 nm), following alternate excitation at 340 and 380 nm, was acquired at a frequency of 0.33 Hz from regions of interest located on cytoplasm. Data were acquired and analyzed using a Metafluor or NIS-Elements AR 3.1 software (Optical Apparatus). The ratio of the fluorescence signals (340/380 nm) was converted to Ca²⁺ concentrations (21). In Ca²⁺-free experiments, CaCl₂ was omitted and 2.5 mM EGTA was added.

Intracellular Microinjection Of Uterine Smooth Muscle Cells

Injections were performed using Femtotips II, InjectMan NI2, and FemtoJet systems (Eppendorf) as previously reported (7, 10, 13). Pipettes were back filled with an intracellular solution composed of 110 mM KCl, 10 mM NaCl, and 20 mM HEPES (pH 7.2) (22) with or without G-1. The injection time was 0.4 s at 60 hPa with a compensation pressure of 20 hPa to maintain the microinjected volume to <1% of cell volume, as measured by microinjection of a fluorescent compound (Fura-2 free acid) (22). The intracellular concentration of G-1 was determined based on the G-1 concentration in the pipette and the volume of injection.

Tissue Preparation and Tension Measurement

Uterine strips were prepared from nonpregnant Sprague-Dawley female rats according to the procedures described by Shintani et al. (42). The middle part of the uterine horn was excised in a longitudinal direction and cut into strips measuring 2 mm in width and 5 mm in length. All measurements were performed at 37°C, in Krebs solution with the following composition (in mM): 127 NaCl, 1.9 KCl, 1.2 KH₂PO₄, 2.4 CaCl₂, 1.3 MgCl₂, 26 NaHCO₃, and 5 glucose, saturated with 95% O₂-5% CO₂.

The uterine strips were mounted vertically in a 5-ml organ bath using a force transducer (ML T0201/RAD; ADInstruments, Colorado Springs, CO) coupled to a Quad Bridge Amplifier (ADInstruments). Contractions were recorded using a PowerLab system and Chart 5 software (ADInstruments). The resting load was adjusted to ∼0.5 g to obtain a maximal response to 40 mm K⁺. The contractile response was analyzed by using three parameters: 1) area under the contractility curve was determined as the integrated force from the start of the induced contraction up to 10 min following the application of the compound; 2) frequency of myometrial contractions was determined as the number of contractions occurring during a 10-min period before and after the application of the compound; and 3) amplitude of myometrial contractions was defined as the distance between the peak and the initial baseline of the contractions in a period of 10 min before and after the application of the compound.

Statistical Analysis

One-way ANOVA or two-way ANOVA were used; a P value of <0.05 was considered statistically significant.

RESULTS

GPER Expression and Immunoreactivity in Rat Myometrium

GPER mRNA was identified using reverse transcriptase-polymerase chain reaction assay as a 987 bp product in rat myometrium; β-actin (876 bp) was the internal control (Fig. 1A). Using immunocytochemical method and a GPER-specific antibody, GPER was detected in myometrial cells as red immunofluorescence in the cytosol (Fig. 1, B–D). Very weak or no immunoreactivity was present in cells incubated with the GPER antibody preabsorbed with the GPER blocking peptide, supporting the specificity of the immunostaining (Fig. 1, E–G).

Fig. 1.

G protein-coupled estrogen receptor (GPER) expression and immunoreactivity in rat myometrium. A: rat G protein estrogen receptor (GPER) mRNA was identified in rat myometrium by reverse transcriptase-polymerase chain reaction as a 987 bp band; housekeeping gene β-actin was used as an internal control. B–D: GPER immunoreactivity (red) was detected in rat myometrial cells using an antibody against human COOH terminus; cell nuclei were identified with DAPI (blue), E and F: no immunoreactivity was present in cells incubated with the antibody preabsorbed with the blocking peptide. Scale bar, 20 μm.

G-1 Depolarizes Myometrial Cells

The resting membrane potential of rat myometrial cells, monitored by the voltage-sensitive dye DiBAC₄(3), was −51 ± 1.8 mV (n = 74). This value is similar to that measured from uterine strips by classical electrophysiological techniques (27). Extracellular administration of G-1 (5 × 10⁻⁸ M, 5 × 10⁻⁷ M, and 5 × 10⁻⁶ M) induced a concentration-dependent membrane depolarization with a mean amplitude of 3.2 ± 0.9 (n = 12), 14.6 ± 1.3 (n = 19), and 19.2 ± 1.7 mV (n = 7), respectively (Fig. 2). The depolarization induced by G-1 (5 × 10⁻⁷ M) was prevented by pretreatment with G-15 (5 × 10⁻⁶ M, 15 min) (n = 14) (Fig. 2). G-1-induced depolarization was abolished when the cells were incubated in a Na⁺-free solution (n = 16), indicating the involvement of Na⁺ influx in G-1-induced myometrial depolarization (Fig. 2).

Fig. 2.

GPER-mediated depolarization of rat myometrial cells. A: administration of specific GPER agonist G-1 (5 × 10⁻⁷ M) produced a depolarization (solid line) that was abolished by pretreatment with the GPER antagonist G-15 (5 × 10⁻⁶ M, 15 min, dotted line) or in Na⁺-free saline (dashed line). The changes in voltage were assessed in cells loaded with DiBAC₄(3) [bis-oxonol (bis-[1,3-dibutylbarbituric acid] trimethineoxonol)]. B: G-1 (5 × 10⁻⁸ M, 5 × 10⁻⁷ M, and 5 × 10⁻⁶ M) induced a concentration-dependent membrane depolarization by 3.2 ± 0.9 (n = 12 cells), 14.6 ± 1.3 (n = 19 cells), and 19.2 ± 1.7 mV (n = 7 cells), respectively; pretreatment with G-15 (5 × 10⁻⁶ M) or Na⁺-free saline reduced the response to G-1 to 1.8 ± 0.3 mV (n = 14 cells) and 0.9 ± 0.3 mV (n = 16 cells), respectively. The amplitude of depolarization has been measured on the plateau of the response 7–8 min after the administration of the compound. *Significant difference compared with the resting membrane potential; **significant difference compared with the effect of G-1 (5 × 10⁻⁷ M) (P < 0.05).

GPER Activation Increases [Ca²⁺]_i in Myometrial Cells

Extracellular administration of G-1 elevates cytosolic Ca²⁺ in a concentration-dependent manner.

The mean basal [Ca²⁺]_i of rat myometrial cells was 68 ± 1.7 nM (n = 167 cells). Extracellular administration of G-1 (5 × 10⁻⁸ M, 5 × 10⁻⁷ M, and 5 × 10⁻⁶ M) induced a concentration-dependent increase in [Ca²⁺]_i with a mean amplitude of 137 ± 1.6 nM (n = 9 cells), 783 ± 4.2 nM (n = 26 cells), and 1,421 ± 7.3 nM (n = 12 cells), respectively (Fig. 3, A and E). G-1-induced increase in [Ca²⁺]_i occurred slowly, starting 1–2 min after administration of G-1, and reached a plateau after 6–7 min; a representative example is shown in Fig. 3A. The increase in [Ca²⁺]_i produced by G-1 (5 × 10⁻⁷ M) was largely abolished (59 ± 4 nM) by pretreatment with G-15 (5 × 10⁻⁶ M, 15 min), a GPER antagonist (10) (Fig. 3, A and E). In the presence of nifedipine, a L-type Ca²⁺ channel blocker (10⁻⁶ M, 10 min pretreatment), application of G-1 (5 × 10⁻⁷ M) increased [Ca²⁺]_i only by 152 ± 2.6 nM (n = 18 cells) (Fig. 3, B and E). The effect was significantly reduced compared with G-1 alone (P < 0.5), indicating the involvement of Ca²⁺ entry through plasmalemmal Ca²⁺ channels. Indeed, in Ca²⁺-free saline, administration of G-1 (5 × 10⁻⁷ M) elevates [Ca²⁺]_i only by 74 ± 0.9 nM (n = 11 cells) (Fig. 3, C and E). Similarly, in Na⁺-free saline, G-1 (5 × 10⁻⁷ M) increases [Ca²⁺]_i only by 81 ± 3 nM (n = 15 cells) (Fig. 3, D and E).

Fig. 3.

GPER-mediated cytosolic Ca²⁺ concentration ([Ca²⁺]_i) increase in rat myometrial cells. A: G-1 (5 × 10⁻⁷ M) induced a robust increase in [Ca²⁺]_i (solid line) in fura-2 AM-loaded myometrial cells, which was markedly diminished by pretreatment with G-15 (5 × 10⁻⁶ M, 15 min), nifedipine (10⁻⁶ M, 10 min pretreatment) (B), Ca²⁺-free saline (C), or Na⁺-free saline (D) (n = 12–25 cells for each condition). E: extracellular administration of G-1 (5 × 10⁻⁸ M, 5 × 10⁻⁷ M, and 5 × 10⁻⁶ M) induced a concentration-dependent increase in [Ca²⁺]_i by (n = 47 cells); the amplitude of [Ca²⁺]_i was measured on the plateau of the response, 6–7 min after the administration of the agonist. *Significant difference compared with basal [Ca²⁺]_i; **significant difference compared with the effect of G-1 (5 × 10⁻⁷ M) in regular saline (P < 0.05).

We also assessed the participation of store-operated calcium entry (SOCE) in the response to G-1. Addition of 2 mM Ca²⁺ after 10 min in Ca²⁺-free saline elicited a constitutive Ca²⁺ entry with an amplitude of 625 ± 44 nM (n = 23 cells); in the same conditions, G-1 (5 × 10⁻⁷ M) produced an increase in [Ca²⁺]_i by 593 ± 36 nM (n = 31 cells), which was not statistically different from the constitutive Ca²⁺ entry (Fig. 4A). In myometrial cells, treatment with thapsigargin (1 μM), an inhibitor of sarcoplasmic/endoplasmic reticulum Ca²⁺ ATPase, followed by addition of Ca²⁺ produced a robust SOCE seen as an increase in [Ca²⁺]_i by 1,479 ± 82 nM (n = 17 cells). The amplitude of SOCE was not significantly affected by treatment with nifedipine (10⁻⁶ M); the increase in [Ca²⁺]_i was 1,545 ± 138 nM (n = 12 cells) after nifedipine versus 1,479 ± 82 nM in the absence of nifedipine (Fig. 4).

Fig. 4.

Store-operated calcium entry is not involved in G-1 mediated increase in [Ca²⁺]_i. A: examples of actual Ca²⁺ responses produced in myometrial cells when Ca²⁺ was added to the saline after 10 min in Ca²⁺-free saline (constitutive Ca²⁺ entry; black trace), after administration of G-1 (5 × 10⁻⁷ M, G-1-induced Ca²⁺ entry; red trace) or treatment with thapsigargin (TG, 10⁻⁶ M; store-operated Ca²⁺ entry) in the absence (green trace) or presence (blue trace) of nifedipine (10⁻⁶ M). B: comparison of the amplitude of Ca²⁺ response produced by different conditions: G-1 increased [Ca²⁺]_i by 593 ± 36 (n = 31 cells), which was not statistically different from the constitutive Ca²⁺ entry (625 ± 44 nM) (n = 23 cells), while much lower than the store-operated Ca² entry (1,479 ± 82 nM; n = 17 cells); the response was not sensitive to nifedipine (1,545 ± 138 nM: n = 12 cells). *Significant difference compared with basal [Ca²⁺]_i; **significant difference compared with the constitutive or G-1-induced Ca²⁺ entry (P < 0.05).

Intracellular microinjection of G-1 elevates [Ca²⁺]_i.

Intracellular microinjection of control buffer induced a small elevation in [Ca²⁺]_i with a mean amplitude of 72 ± 1.6 nM in rat myometrial cells (n = 10 cells) (Fig. 5, A, F, and G). Microinjection of G-1 (5 × 10⁻⁷ M final concentration inside the cell) produced a robust and transitory increase in [Ca²⁺]_i by 983 ± 8 nM (n = 10 cells) (Fig. 5, B, F, and G). The effect occurred faster and had a higher amplitude than that induced by extracellular application (seconds vs. minutes) (P < 0.05) (for comparison see Fig. 5F and Fig. 3A). Pretreatment with G-15 (5 × 10⁻⁶ M, 15 min) prevented the effects of intracellularly injected G-1 (Fig. 5, C, F, and G) (n = 11 cells); the small increase in [Ca²⁺]_i produced by G-1 administration in the presence of the antagonist was not significantly different from that in control injection. Pretreatment with nifedipine (10⁻⁶ M, 10 min) markedly reduced the G-1-induced increase in [Ca²⁺]_i to 102 ± 8 nM (Fig. 5, D, F, and G) (n = 8 cells). When G-1 (5 × 10⁻⁵ M in the pipette) was applied extracellularly by Femtotips, in the close vicinity (1–2 μm) of the cell membrane induced a slow increase in [Ca²⁺]_i by 103 + 2.6 nM with a plateau at 5–8 min after administration (n = 6 cells) (Fig. 5, E, H, and I).

Fig. 5.

Intracellular microinjection of G-1 increases [Ca²⁺]_i in rat myometrial cells. A–D: representative examples of changes in fura-2 fluorescence ratio (340/380 nm) before, during, and 4 min after injection; the fluorescence ratio scale (0–3) is illustrated in each panel and magnified in A; the arrows indicate the injection site. A: microinjection of control buffer induced a minimal change in fluorescence ratio. B: microinjection of G-1 (5 × 10⁻⁷ M final concentration inside the cell) induced a marked increase in fluorescence ratio that was prevented by pretreatment with G-15 (5 × 10⁻⁶ M) (C). D: microinjection of G-1 (5 × 10⁻⁷ M) in cells pretreated with nifedipine, a L-type Ca²⁺ channel blocker, produced a minimal change in fluorescence ratio. E: representative examples of changes in fura-2 AM fluorescence ratio (340/380 nm) before, at 4 min, and 8 min after extracellular administration of G-1 by Femtotips in the vicinity of the membrane; the fluorescence ratio scale (0–4) is illustrated in each panel. F: representative tracings of Ca²⁺ responses induced by microinjection of control buffer (left, red trace), G-1 (left, black trace), G-1 in cells pretreated with G-15 (middle trace), or nifedipine (right). G: comparison of changes in [Ca²⁺]_i during injection of control buffer (72 ± 1.6 nM, n = 10 cells), G-1 (983 ± 8 nM; n = 10 cells), and G-1 in the presence of G-15 (102 ± 8; n = 11 cells) or nifedipine (163 ± 9 nM; n = 8 cells). *Significant difference compared with the injection of control buffer; **significant difference compared with the effect of G-1 alone (P < 0.05). H: averaged changes in [Ca²⁺]_i (n = 6 cells) induced by extracellular injection of G-1. I: G-1 induced a slow increase in [Ca²⁺]_i by 103 ± 2.6 nM at 5–8 min after administration.

G-1 Increases Uterine Strips Contractility

In basal conditions, the contractility of rat uterine strips was characterized by the following parameters: 1) the area under the contractility curve was 96 ± 3.2 g/10 min, 2) the frequency of myometrial contractions was 5.4 ± 0.8 contractions/10 min, and 3) the amplitude of myometrial contractions was 1.06 + 0.09 g. Administration of G-1 (5 × 10⁻⁷ M) increased the area under the contractility curve by 862 ± 4.1%, the frequency of myometrial contractions by 426 ± 3.4% and the amplitude of contractions by 52 ± 0.7% (n = 6 experiments) compared with basal conditions. In addition, a transitory increase in basal tone was also observed. A representative trace is shown in Fig. 6A. Pretreatment of the uterine strips with G-15 (5 × 10⁻⁶ M, 10 min) or nifedipine (10⁻⁶ M) abolished the effect of G-1 on contractility (Fig. 6, B and C). In addition, nifedipine also abolished the basal contractions (Fig. 6C).

Fig. 6.

G-1 increases uterine contractility. A–C: representative examples of contractility recordings in rat uterine strips. A: administration of G-1 (5 × 10⁻⁷ M) increased the frequency and amplitude of uterine contractions. B: pretreatment with G-15 (5 × 10⁻⁶ M, 10 min) prevented the effect of G-1. C: nifedipine (10⁻⁶ M) abolished the basal contractions and prevented the effect of G-1 on uterine strips contractility.

DISCUSSION

GPER has been identified in several tissues including uterus (25, 31). Recent evidence indicate that the receptor is involved in endometrial pathology (24, 43, 45, 46), while the role of GPER in myometrium is largely unexplored. Recently, GPER has been proposed to act as a negative regulator of ERα-dependent uterine growth in response to estradiol (20). However, GPER-deficient mice did not seem to have an impaired uterine function (32). In this study, we examined the role of GPER in nonpregnant rat myometrium, by assessing the effect of selective GPER ligands, previously characterized (4, 14), on [Ca²⁺]_i, membrane potential, and uterine contractility.

First, we detected the presence of GPER expression in nonpregnant rat myometrium using reverse transcriptase-polymerase chain reaction and immunocytochemistry. GPER immunoreactivity was identified intracellularly, as reported in several other cell types (12, 33, 39). Next, using a slow-response voltage-sensitive dye, we found that administration of G-1, a GPER-selective agonist (4), produced a concentration-dependent depolarization of rat myometrial cells. G-1-induced depolarization was abolished in a Na⁺-free saline, indicating the involvement of Na⁺ influx in the response. In myometrial cells, Na⁺ influx occurs trough voltage-gated Na⁺ channels and nonselective cation channels from transient receptor potential channels family (TRPC). All TRPCs except for TRPC3 are expressed in nonpregnant rat myometrium (3), while the human myometrium lacks TRPC2 (50). The specific Na²⁺ influx pathway(s) associated with GPER activation remains to be clarified.

In myometrium, similar to other excitable tissues, depolarization due to various stimuli leads to an opening of plasmalemmal voltage-sensitive Ca²⁺ channels, followed by an increase in cytoplasmatic Ca²⁺ concentration and contraction (49). With respect to the concentrations of GPER ligands used, notable differences were reported in different cellular systems and between endogenously expressing and GPER-transfected cells. Several recent publications, including ours (2, 5, 28), indicate that in primary cells, G-1 activates GPER in micromolar range. In transfected cells, which overexpress the receptor, lower concentrations of agonists may be sufficient to activate the receptor (4).

Our calcium imaging experiments indicate that extracellular administration of G-1 induced a delayed and sustained increase in [Ca²⁺]_i. Since Ca²⁺ entry in rat uterus occurs almost entirely via L-type Ca²⁺ channels (49), we examined the [Ca²⁺]_i response to G-1 in the presence of nifedipine, a L-type Ca²⁺ channel blocker. Pretreatment with nifedipine markedly decreased G-1-induced increase in [Ca²⁺]_i, supporting the involvement of L-type Ca²⁺ channels in the response. In Ca²⁺-free saline, the response to G-1 was markedly reduced, indicating a limited or lack of participation of sarcoplasmic reticulum. SOCE is an important Ca²⁺ influx pathway in many nonexcitable and some excitable cells (36). We examined an eventual participation of SOCE in the response to G-1. Depletion of the internal Ca²⁺ stores by treatment with thapsigargin, an inhibitor of sarcoplasmic-endoplasmic reticulum Ca²⁺-ATPase, evoked a robust SOCE, which was not affected by treatment with nifedipine. G-1-induced [Ca²⁺]_i response was similar to the constitutive Ca²⁺ entry and did not seem to involve a SOCE mechanism. Previously, we reported that GPER activation by G-1 produced an increase in [Ca²⁺]_i in neurons (5) and breast cancer cells (2). Similarly, other reports indicate that GPER activation enhanced [Ca²⁺]_i in cells endogenously expressing or transfected with the receptor (4, 35 ,39).

GPCR are active structural proteins located usually in the cell membrane. However, increasing evidence indicates that angiotensin II receptors are active not only when stimulated from outside but also from inside the cells (6, 37, 38). We recently reported that intracellular angiotensin II activates myometrial cells via an intracrine mechanism (13). Also, we characterized the intracellular activation of cannabinoid CB₁ receptors by anandamide (10). With respect to GPER, the subcellular localization has yet to be firmly established; both intracellular and plasma membrane localizations have been reported (16, 33, 39, 41). Initially, GPER was shown to be uniquely localized to the endoplasmic reticulum in cells transiently or constitutively expressing GPER (39). In contrast, other studies using morphological and functional assays showed that GPER is expressed only in the plasma membrane of GPER-transfected cells (16, 29, 44) or of CA2 pyramidal neurons of the rat hippocampus (19). In myometrial cells, G-1 administered intracellularly by microinjection produced a faster and transitory increase in [Ca²⁺]_i, with a higher amplitude than that induced by extracellular administration of G-1 in the same final concentration. Since G-1 is a membrane permeant compound, we also applied G-1 on the extracellular side of the membrane by microinjection, using Femtotips. In this case, we noted a small and delayed increase in [Ca²⁺]_i, whose shape and kinetics were similar to that produced by bath application of G-1. We attribute the differences in kinetics and amplitude of responses to intracellular versus extracellular administration of the agonist to the location of GPER in myometrial cells. Our calcium imaging and immunocytochemistry results support an intracellular location of functional GPER. This is in agreement with a recent report which found that GPER is localized intracellularly in other tissues (12). The response was abolished by pretreatment with G-15, a GPER antagonist, indicating that this response is receptor specific.

Increase of [Ca²⁺]_i and stimulation of myosin light chain kinase (MLCK) via Ca²⁺-calmodulin is essential for myometrial contraction. To identify the effect of GPER activation on uterine contraction, the next series of experiments were carried out on rat uterine strips. Administration of G-1 caused a robust stimulant effect on uterine strip contractility by increasing the frequency and amplitude of uterine contractions. Similarly, estrogen has been reported to stimulate uterine contractility in perfused nonpregnant human or swine uterus (30, 40) or in artificially inseminated sows (48). Taken together, our results support the involvement of GPER activation in myometrial depolarization and contraction. Similarly, a very recent report, published when the present study was under review, indicates that GPER activation is associated with increased contractility of the human myometrium and the receptor is most likely to be involved in myometrium physiology during pregnancy (28).

GRANTS

This study was supported by National Institutes of Health Grants HL-90804 (to E. Brailoiu) and 5R01CA127731-03 (to J. B. Arterburn and T. I. Oprea). A. A. Tica and O. S. Tica were funded by Sunset Molecular Discovery, Santa Fe, NM.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: A. A. Tica, E. C. Dun, O. S. Tica, X. Gao, and E. Brailoiu performed experiments; A. A. Tica, E. C. Dun, and O. S. Tica analyzed data; A. A. Tica, X. Gao, and G. C. Brailoiu prepared figures; A. A. Tica and G. C. Brailoiu drafted manuscript; A. A. Tica, E. C. Dun, O. S. Tica, X. Gao, J. B. Arterburn, G. C. Brailoiu, T. I. Oprea, and E. Brailoiu approved final version of manuscript; J. B. Arterburn, G. C. Brailoiu, T. I. Oprea, and E. Brailoiu edited and revised manuscript; E. Brailoiu conception and design of research; E. Brailoiu interpreted results of experiments.