Introduction

Top Summary Introduction Materials and Methods Results Discussion References

Endothelin-1 (ET-1) has been implicated in a wide variety of physiological functions associated with the cardiovascular, endocrine, pulmonary, renal, and nervous systems. Two main receptor subtypes (ET_A and ET_B) have been cloned (Arai et al., 1990; Sakurai et al., 1990). Molecular characterization of ET receptors has revealed that they belong to the G protein-coupled receptor superfamily. The mechanisms of ET-1 action involve second messenger generation via different signal transduction processes. These include activation of phospholipase C (PLC) degrading phosphatidylinositol 4,5-bisphosphate with an increased production of inositol 1,4,5-trisphosphate (InsP₃) and diacylglycerol, inhibition or activation of adenylyl cyclase, activation of phospholipase A₂ and D (Rubanyi and Polokoff, 1994; Webb and Meek, 1997), and activation of Ca²⁺-permeable nonselective cation channel via a cyclic guanosine monophosphate signaling system (Minowa et al., 1997).

In smooth muscle cells, both ET_A and ET_B are expressed and may evoke an increase in cytosolic Ca²⁺ concentration ([Ca²⁺]_i) via different G proteins (Palacios et al., 1997; Saita et al., 1997). InsP₃ generation induced by either ET_A or ET_B receptors has been reported to be inhibited by pertussis toxin in various cell types, suggesting a coupling to PLC via G_i/o proteins. However, in rat R22-D vascular myocytes, the ET receptors are coupled to PLC via a pertussis toxin-insensitive G protein (Douglas and Ohlstein, 1997). Members of the G_q family (G_q, G₁₁, G₁₄, and G₁₅) can activate PLC-₁, -₃, and -₄ through their  subunit (Jhon et al., 1993), whereas dimers released from either pertussis toxin-sensitive or -insensitive G protein can activate the PLC-₂, ₃, and ₁ (Camps et al., 1992; Ushio-Fukai et al., 1998).

The aim of the present study was two fold: 1) to identify the ET receptor subtype responsible for the increase in [Ca²⁺]_i in rat portal vein myocytes and 2) to determine the composition of the G protein that couples the ET-1 receptor subtype to PLC. Experiments with antisense oligonucleotides designed to block synthesis of G protein subunits revealed the G protein heterotrimer, composed of ₁₁, ₃, and ₅ subunits, which coupled ET-1 receptors to elevation in [Ca²⁺]_i. Identification of the G protein subunits involved in the activation of PLC was resolved by using G scavengers. In addition, the delayed accumulation of [³H]InsP₃ induced by ET-1 was associated with the slow kinetics of the ET-1-evoked Ca²⁺ response.

	Materials and Methods

Top Summary Introduction Materials and Methods Results Discussion References

Cell Preparation. Isolated myocytes from rat portal vein were obtained by enzymatic dispersion, as described previously (Macrez-Leprêtre et al., 1994). Cells were seeded at a density of about 10³ cells/mm² on glass slides imprinted with squares for localization of injected cells, and maintained in short-term primary culture in M199 medium containing 2% fetal calf serum, 2 mM glutamine, 1 mM pyruvate, 200 U/ml penicillin, and 200 µg/ml streptomycin; they were kept in an incubator gassed with 95% air and 5% CO₂ at 37°C, and were used within 96 h.

Fluorescence Measurements. Cells were loaded by incubation in physiological solution containing either 1 µM Indo-1-acetoxymethylester (AM) or 1 µM Fura-2-acetoxymethylester for 30 min at room temperature. These cells were washed and allowed to cleave the dye to the active Indo-1 or Fura-2 for 60 min. Indo-1 or Fura-2 loading was usually uniform over the cytoplasm and compartmentalization of the dyes was never observed. Measurement of [Ca²⁺]_i in single cells and calibration curves for Indo-1 and Fura-2 were determined within cells after 1 or 4 days in primary culture, as reported previously (Macrez-Leprêtre et al., 1994; Morel et al., 1996). Briefly, [Ca²⁺]_i was calculated from mean ratios according to Grynkiewicz's formula (Grynkiewicz et al., 1985): [Ca²⁺]_i = K_d  (R  R_min)/R_max  R), where R_min and R_max are the ratios obtained in the absence of Ca²⁺ and at saturating Ca²⁺. K_d is the effective dissociation constant of the Ca²⁺-sensitive dyes.  is defined as the ratio of the signals observed at excitation wavelength 380 nm (Fura-2) or at emission wavelength 480 nm (Indo-1) in the absence and presence of saturating Ca²⁺, and takes into account corrections for the optics of the systems. Mean K_d values of 825 nM (Fura-2) and 1350 nM (Indo-1) were obtained with their respective set-up. Cells were made permeable to external Ca²⁺ using 20 µM ionomycin and R_min and R_max were obtained by either Ca²⁺ depletion (no added Ca²⁺, 10 mM EGTA) or Ca²⁺ saturation (2 mM Ca²⁺) of the physiological solution. At 1 day of culture, the mean R_min and R_max values for Fura-2 were 0.61 ± 0.02 and 6.10 ± 0.25 (n = 12), respectively. These values were not significantly different at 4 days of culture as R_min and R_max values were 0.60 ± 0.02 and 5.90 ± 0.20 (n = 12), respectively. Similarly, the mean R_min and R_max values for Indo-1 were calculated at 1 and 4 days of culture. The R_min and R_max values were 0.30 ± 0.01 and 0.32 ± 0.02 (n = 10) and 4.25 ± 0.21 and 4.18 ± 0.15 (n = 10), respectively at 1 and 4 days of culture. Indo-1- or Fura-2-loaded cells were mounted in a perfusion chamber and placed on the stage of an inverted microscope (Nikon Diaphot, Tokyo, Japan). Most experiments were carried out in the presence of 1 µM oxodipine (a light-stable dihydropyridine derivative) to inhibit voltage-dependent Ca²⁺ channels. All measurements were made at 25 ± 1°C.

Membrane Current and [Ca²⁺]_i Measurements. Voltage-clamp and membrane current recordings were made with a standard patch-clamp technique using an EPC-7 patch-clamp amplifier (List, Darmstadt-Eberstadt, Germany). Patch pipettes had resistances of 3-4 M. When [Ca²⁺]_i measurements were carried out simultaneously, Indo-1 (50 µM) was added to the pipette solution in the whole-cell recording mode and [Ca²⁺]_i was estimated from the 405/480 nm fluorescence ratio using a calibration determined within cells as described previously (Morel et al., 1996). R_min was determined by loading cells with a pipette solution containing 10 mM EGTA. Mean R_min values were 0.25 ± 0.02 (n = 10) and 0.26 ± 0.03 (n = 10) at 1 and 4 days of culture, respectively. R_max was determined by hyperpolarizing the cells to 200 mV. This caused membrane breakdown and a large increase in Ca²⁺ permeability without resulting in a significant dye leakage. Mean R_max values were 4.19 ± 0.25 (n = 10) and 4.25 ± 0.30 (n = 10) at 1 and 4 days of culture, respectively. An intracellular value for K_d was determined according to Almers and Neher (1985). The 405/480-nm ratio was determined in cells loaded with a pipette solution containing 8 mM Ca²⁺ and 10 mM EGTA so that the free Ca²⁺ concentration was around 240 nM. A mean value for K_d of 1156 nM was obtained under these conditions. Patch-clamp experiments were performed at 30 ± 1°C.

Antibodies, Peptides, and Heparin. Anti-G_com antibody, G_q/₁₁ peptide, -adrenergic receptor kinase-1 (ARK₁) peptide, and heparin were added to the pipette solution to allow dialysis of the cell after a breakthrough in whole-cell recording mode for at least 7 to 8 min, a time longer than that expected theoretically for diffusion of substances in solution (Pusch and Neher, 1988).

Microinjection of Oligonucleotides. Antisense or nonsense oligonucleotides were injected into the nucleus of myocytes by a manual injection system, as described previously (Macrez-Leprêtre et al., 1997a,b). The injection solution contained 10 µM oligonucleotides in water; approximately 10 fl were injected into the nucleus. The myocytes were cultured for 72 h (the time needed to obtain the highest inhibition of protein expression) in culture medium before [Ca²⁺]_i measurements. The sequences of the anti-_q, anti-₁₁, anti-₁₂, anti-₁₃, anti-₁₄, anti-₁₅, anti-_icom, anti-₀₁, anti-_z, anti-₁, anti-₂, anti-₃, anti-₄, anti-₅, anti-₁, anti-₂, anti-₃, anti-₄, anti-₅, anti-₇, and anti-₈ antisense oligonucleotides have been published previously (Gollasch et al., 1993; Macrez-Leprêtre et al., 1997a,b). Nonsense oligonucleotides had the same composition that the corresponding antisense oligonucleotides, but in the opposite sense.

Solutions. The normal physiological solution contained (in mM): NaCl 130, KCl 5.6, MgCl₂ 1, CaCl₂ 2, glucose 11, and HEPES 10, pH 7.4 with NaOH. The basic pipette solution contained (in mM): KCl 130 and HEPES 10, pH 7.3 with NaOH. Ca²⁺-free solution was prepared by omitting CaCl₂ and by adding 0.5 mM EGTA. ET-1 and active compounds were applied to the recorded cell by pressure ejection for the period indicated on the records.

Immunocytochemistry. Three days after injection, venous myocytes were washed with PBS, fixed with 3% formaldehyde (v/v) for 30 min at room temperature, and permeabilized in PBS containing 3% fetal calf serum and 0.1 (w/v) saponin for 30 min. Cells were incubated with the same buffer containing 5% fetal calf serum, 0.1 (w/v) saponin, and the anti-G protein antibody at 1:100 dilution overnight at 4°C. Then, cells were washed in PBS containing 3% fetal calf serum and 0.1 (w/v) saponin (4 × 10 min) and incubated with goat anti-rabbit IgG conjugated to fluorescein isothiocyanate (diluted 1:200) in the same solution for 3 h at 20°C. Thereafter, cells were washed (4 × 10 min) in PBS and mounted in Vectashield (Biosys, Compiègne, France). Images of the stained cells were obtained with a confocal microscope (Bio-Rad MRC 1000, Bio-Rad Laboratories, Inc., Hemel Hempstead, UK). Only cells on the same glass slide were compared with each other by keeping acquisition parameters (gray values, exposure time, aperture, etc.) constant. Immunostaining fluorescence was estimated by gray level analysis using MPL software (Bio-Rad).

Measurement of [³H]Inositol Phosphates. Experiments on intact cells were performed as described previously (Mouillac et al., 1989; Meneton et al., 1992). Cells (0.6 × 10⁶) were incubated in 2 ml of culture medium containing 5 µCi/ml myo[2-³H]inositol (20 Ci/mmol) for 3 days leading to steady-state labeling of cellular inositol lipids (Menniti et al., 1990; Mouillac et al., 1990). Then, cells were washed with M199 without serum and myo[2-³H]inositol during 30 min. The M199 medium was removed and replaced by physiological solution with 10 mM LiCl. After a 10-min delay, ET-1 or norepinephrine (NE) was added. Inositol phosphate (InsP) formation was stopped by ice-cold perchloric acid (final concentration 5%). Acid-lysed cells were centrifuged at 2500 rpm, 4°C for 5 min. Supernatant contained InsPs and the pellet contained phosphatidylinositols and lipids. Supernatant was neutralized by HEPES 75 mM and KOH 1.5 M (pH = 6-8) and incubated 5 min on ice for the precipitation of potassium perchlorate. Neutralized supernatant was centrifuged at 2500 rpm, 4°C for 5 min. Supernatant was eluted through a poly-prep chromatography column (Bio-Rad) containing 1.6 ml anion exchange resin (DOWEX AG1-X8, formate form, 200-400 mesh). On the basis of control experiments with labeled inositol phosphate standards, free inositol was eluded with 12 ml water, glycerophosphoinositol with 12 ml of 30 mM ammonium formate (pH = 5), inositol monophosphate (InsP₁) with 12 ml of 180 mM ammonium formate/0.1 M formic acid, inositol bisphosphate (InsP₂) with 12 ml of 400 mM ammonium formate/0.1 M formic acid, and InsP₃ with 12 ml of 700 mM ammonium formate/0.1 M formic acid. The perchloric acid-precipitated pellets that contained phosphoinositides were resuspended with 1 ml chloroform-methanol-10 M HCl (200:100:1, v/v/v). These suspensions were mixed with 350 µl HCl and 350 µl chloroform and centrifuged for 5 min at 2500 rpm to separate the phases. The lower hydrophobic phase was recovered and dried in counting vials to determine radioactivity in total phosphoinositides. Recoveries of inositol phosphates and phosphoinositides were about 75%.

Chemicals and Drugs. Collagenase was obtained from Worthington Biochemical Corp. (Freehold, NJ); pronase (type E), bovine serum albumin, NE, pertussis toxin, heparin (from porcine intestinal mucosa; molecular weight, 6000), and chondroitin sulfate were purchased from Sigma (St. Louis, MO). M199 medium was obtained from Flow Laboratories (Puteaux, France). Fetal calf serum was obtained from Flobio (Courbevoie, France). Streptomycin, penicillin, glutamine, and pyruvate were purchased from Gibco (Paisley, UK). Oxodipine was a gift from Dr. Galiano (Instituto de Investigacion y Desarrollo Quimico Biologica, Madrid, Spain). Caffeine was purchased from Merck (Nogent sur Marne, France). Indo-1 and Indo-1 AM were obtained from Calbiochem (Meudon, France). AII, ET-1, BQ123, and BQ788 were purchased from Neosystem Laboratories (Strasbourg, France). Carboxyl-terminal G_q/11 peptide (LQLNLKEYNLV) and peptide corresponding to the G-binding protein domain of ARK₁ (WKKELRDAYREAQQLVQRVPKMKNKPRS) were synthesized by Genosys (Cambridge, UK). 1-(6-((17-3 methoxystra 1,3,5 (10)-trien-17-yl) amino) hexyl)-1H-pyrrole-2,5-dione (U73122) and 1-(6-((17-3 methoxystra 1,3,5 (10)-trien-17-yl) amino) hexyl)-1H-pyrrolipine-dione (U73343) were obtained from Biomol (Plymouth Meeting, PA). Anti-G_q (SC393), anti-G₁₁ (SC394), anti-G_com (SC378), anti-₃ (SC381) and anti-₅ (SC376) antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Oligonucleotides were a gift from F. Kalkbrenner (University of Berlin, Germany). Myo[2-³H]inositol (20 Ci/mmol) was purchased from NEN (Paris, France). Fluorescein isothiocyanate-conjugated goat anti-rabbit IgG was obtained from Immunotech (Marseille, France).

	Results

Top Summary Introduction Materials and Methods Results Discussion References

Effects of ET-1 on [Ca²⁺]_i in Vascular Myocytes. Application of 10 nM ET-1 for 30 s resulted in a delayed increase in [Ca²⁺]_i (Fig. 1A). In 20% of the cells tested (n = 30), the ET-1-induced Ca²⁺ response seemed to possess two components (Fig. 1A) with a faster component followed by a more or less sustained increase in [Ca²⁺]_i, which returned to basal [Ca²⁺]_i value within 2 to 3 min. The transient peak was obtained 35 ± 7 s (n = 9) after application of ET-1. Its mean amplitude was 135 ± 32 nM from a resting [Ca²⁺]_i of 43 ± 3 nM (n = 9) in Indo-1 AM-loaded myocytes. In 80% of the cells tested (n = 30), ET-1 initiated a slow and monophasic Ca²⁺ response (Fig. 1B) with a mean amplitude of 129 ± 25 nM (n = 21). Removal of external Ca²⁺ (in the presence of 0.5 mM EGTA) decreased the transient Ca²⁺ peak to 86 ± 12 nM (n = 30) and suppressed the sustained increase in [Ca²⁺]_i in all the types of Ca²⁺ responses (Fig. 1, A and B). External application of 1 µM oxodipine (to inhibit L-type Ca²⁺ channels) reduced the transient peak to 93 ± 15 nM (n = 30) and largely inhibited the sustained Ca²⁺ phase, but did not suppress it (Fig. 1, A and B). These results suggest that the peak increase in [Ca²⁺]_i is largely due to release of Ca²⁺ from the intracellular store. In the following experiments, ET-1-induced Ca²⁺ responses were measured either in the presence of 1 µM oxodipine for at least 5 min or in Ca²⁺-free, 0.5 mM EGTA-containing solution for 30 s.

View larger version (13K): [in this window] [in a new window]   Fig. 1.   Increase in [Ca2+]i evoked by 10 nM ET-1 in rat portal vein myocytes. The cells were loaded with Indo-1 AM and not patch-clamped. A, biphasic Ca2+ response evoked by 30-s application of 10 nM ET-1 in physiological solution containing 2 mM CaCl2 (a), becomes monophasic in Ca2+-free (0.5 mM EGTA) solution for 30 s (b), and after application of 1 µM oxodipine for 5 min (c). B, monophasic Ca2+ response evoked by 30-s application of 10 nM ET-1 in 2 mM CaCl2 solution (a), in Ca2+-free solution (0.5 mM EGTA) solution for 30 s (b), and after application of 1 µM oxodipine for 5 min (c).

View larger version (14K): [in this window] [in a new window]   Fig. 2.   Effects of ET-1 and endothelin receptor antagonists on [Ca2+]i. A, concentration-response curve to ET-1 obtained by fitting the experimental points. The Hill coefficient is equal to unity. B, inhibitions of the ET-1-stimulated [Ca2+]i changes by BQ123 () and BQ788 (). [Ca2+]i values are expressed as a percentage of the response obtained with 10 nM ET-1. Data are means ± S.E.M. for five to eleven cells. Myocytes were loaded with Indo-1 AM and not patch-clamped. External solution contained 1 µM oxodipine.

Effects of Heparin and PLC Inhibitor. Heparin blocks InsP₃ receptors (Guillemette et al., 1989) and inhibits Ca²⁺ release via these receptors in smooth muscle cells (Somlyo and Somlyo, 1994). The effect of ET-1 was studied in myocytes held at 50 mV with 1 mg/ml heparin in the pipette solution. The basal [Ca²⁺]_i level (61 ± 3 nM, n = 11) was not significantly different from that observed under control conditions (65 ± 5 nM, n = 11), but the ET-1-induced Ca²⁺ response was completely removed (Fig. 3A). This effect is specific to heparin as, when the glycosaminoglycan chondroitin sulfate (used at 1 mg/ml as a negative control; Worley et al., 1987) was added to the pipette solution instead of heparin, the Ca²⁺ response to ET-1 (102 ± 24 nM, n = 7) was similar to that obtained in control conditions (104 ± 25 nM, n = 7). U73122 has been shown to inhibit phosphatidylinositol-specific PLC in vascular myocytes without affecting the intracellular Ca²⁺ store (Macrez-Leprêtre et al., 1996). As shown in Fig. 3B, the ET-1-induced Ca²⁺ response was concentration-dependently inhibited by U73122 with an estimated IC₅₀ value of 0.25 µM. Complete inhibition was obtained with 3 µM U73122. In contrast, 3 µM 73343 (the inactive analog of U73122) had no significant effects on the ET-1-induced Ca²⁺ response (Fig. 3B). These results suggest that ET-1 activates a PLC to generate InsP₃ and the subsequent activation of InsP₃-gated Ca²⁺ release channels in the sarcoplasmic reticulum.

View larger version (13K): [in this window] [in a new window]   Fig. 3.   Effects of heparin and PLC inhibitor on ET-1-induced Ca2+ release. A, Ca2+ responses evoked by 10 nM ET-1 in control conditions (a) and in cells dialyzed with a pipette solution containing 1 mg/ml heparin (b) or chondroitin sulfate for 7 to 8 min (c). Myocytes were loaded with Indo-1 through the patch pipette and held at 50 mV. B, inhibition of the ET-1-induced Ca2+ release by U73122 () and U73123 (). [Ca2+]i values are expressed as a percentage of the response obtained with 10 nM ET-1. Data are means ± S.E.M. for seven cells. Myocytes were loaded with Indo-1 AM and not patch-clamped. External solution contained 1 µM oxodipine.

Subunit Composition of the G Protein. Previous identification of the  subunits of G proteins in rat portal vein membranes by immunoblot analysis has revealed the presence of pertussis toxin (PTX)-sensitive and PTX-insensitive G proteins (Macrez-Leprêtre et al., 1995). When cells were incubated in the presence of 0.5 µg/ml PTX for 20 h, the ET-1-induced Ca²⁺ response was not significantly affected (control: 101 ± 16 nM, n = 11; PTX-pretreated: 99 ± 20 nM, n = 11). In contrast, intracellular application of the carboxyl-terminal G_q/11 peptide inhibited in a concentration-dependent manner the ET-1-evoked increase in [Ca²⁺]_i (Fig. 4) with complete inhibition obtained at 0.5 µg/ml. To identify the heterotrimeric G protein that transduced the ET_A receptor signal, 10 µM phosphorothioate-modified antisense oligonucleotides directed against the , , and  subunits of G proteins were injected into the nucleus of vascular myocytes (Macrez-Leprêtre et al., 1997a,b). Figure 5, A and B, illustrates the time course of the efficiency of anti-_q and anti-₁₁ antisense oligonucleotides in inhibiting the NE- and ET-1-induced increases in [Ca²⁺]_i, respectively. In both cases, the highest inhibition (75-80%) was obtained 3 days after injection. A slight recovery was observed 4 days after injection. A similar delay after injection (48-72 h) was needed to obtain the highest inhibition of Ca²⁺ responses by various anti- and anti- antisense oligonucleotides. Figure 5C illustrates the efficiency and specificity of anti-₁, anti-₃, anti-₃, and anti-₅ antisense oligonucleotides in inhibiting ET-1 and AII-induced increase in [Ca²⁺]_i. Therefore, in the following experiments, the increases in [Ca²⁺]_i were measured 3 days after oligonucleotide injection, in response to successive applications of 10 nM ET-1, 10 nM AII, and 10 mM caffeine on the same cells (Fig. 6). For each experiment, the Ca²⁺ responses of antisense or nonsense oligonucleotide-injected cells located within a marked area on the glass slide were compared with those of noninjected cells outside this marked area. This procedure ensures that oligonucleotide-injected cells were always compared with control cells that were grown, treated (i.e., incubation and loading with Fura-2 AM), and analyzed under identical conditions (Macrez-Leprêtre et al., 1997a,b). The increases in [Ca²⁺]_i were measured for each cell, and mean values were calculated from all cells of each experiment. Myocytes injected with anti-₁₁, anti-₃, and anti-₅ antisense oligonucleotides showed a strong and similar inhibition (65-80%) of the ET-1-induced Ca²⁺ responses when compared with noninjected cells or cells injected with ₁₁, ₃, and ₅ nonsense oligonucleotides (Fig. 7). These results suggest that the heterotrimeric G protein composed of ₁₁, ₃, and ₅ subunits controls the Ca²⁺ response evoked by ET-1 application.

View larger version (10K): [in this window] [in a new window]   Fig. 4.   Effects of a carboxyl-terminal Gq/11 peptide on the ET-1-induced Ca2+ release. Inhibition curve obtained with increasing concentrations of Gq/11 peptide. [Ca2+]i values are expressed as a percentage of the response obtained with 10 nM ET-1. Data are means ± S.E.M. for five to nine cells. Myocytes were loaded with Indo-1 and held at 50 mV. External solution contained 1 µM oxodipine.

View larger version (22K): [in this window] [in a new window]   Fig. 5.   Time course of inhibition of mediator-induced Ca2+ responses in cells injected with antisense oligonucleotides. A, mean amplitudes of Ca2+ responses evoked by 10 µM NE in cells injected with 10 µM anti-q antisense oligonucleotides in function of time after injection. B, mean amplitudes of Ca2+ responses evoked by 10 nM ET-1 in cells injected with 10 µM anti-11 antisense oligonucleotides in function of time after injection. C, Ca2+ responses evoked by 10 nM ET-1 and 10 nM AII in the same cells 72 h after injection with 10 µM of the indicated antisense oligonucleotides. Controls correspond to Ca2+ responses in noninjected cells. Data are means ± S.E.M. for five to eight cells in (A) and (B) and for the number of cells tested indicated in parentheses in (C). Myocytes were loaded with Fura-2 AM and not patch-clamped.

View larger version (21K): [in this window] [in a new window]   Fig. 6.   Increases in [Ca2+]i evoked by ET-1, AII and caffeine (Caf) in myocytes injected with 10 µM anti-sense oligonucleotides directed against the mRNAs of Gq and G11 proteins. Ca2+ responses were obtained in the same cells with successive applications of 10 nM ET-1, 10 nM AII, and 10 mM caffeine, separated by a 3-min interval, in noninjected control cells (A) and in cells injected with 10 µM anti-q (B), or anti-11 (C) antisense oligonucleotides. Cells were used 3 days after nuclear injection of antisense oligonucleotides, were loaded with Fura-2 AM, and were not patch-clamped.

View larger version (49K): [in this window] [in a new window]   Fig. 7.   ET-1-induced Ca2+ release in myocytes injected with 10 µM antisense oligonucleotides directed against the mRNAs of , , and  subunits of G proteins. A, effects of anti-q, anti-11, anti-12, anti-13, anti-14, anti-15, anti-icom, anti-01, anti-z, and of 11 nonsense oligonucleotides on Ca2+ release evoked by 10 nM ET-1. B, effects of anti-1, anti-2, anti-3, anti-4, anti-5, and of 3 nonsense oligonucleotides on ET-1-induced Ca2+ release. C, effects of anti-1, anti-2, anti-3, anti-4, anti-5, anti-7, anti-8, and of 5 nonsense oligonucleotides on ET-1-induced Ca2+ release. Columns show means ± S.E.M. in noninjected cells (open columns) and in oligonucleotide-injected cells (hatched columns). Numbers in parentheses indicate the number of cells tested. , values significantly different from those obtained in noninjected cells (P < .05). Cells were loaded with Fura-2 AM and not patch-clamped. External solution contained 1 µM oxodipine.

View larger version (51K): [in this window] [in a new window]   Fig. 8.   Specific inhibition of G3, G5, and G11 protein expression in cells injected with anti-, anti-, and anti- antisense oligonucleotides. A: Top, myocytes stained 3 days after injection with rabbit anti-3 antibody. Visualization was obtained by staining with fluorescein isothiocyanate-conjugated goat anti-rabbit IgG (1:200). Cells injected with anti-3 or anti-5 antisense oligonucleotides were compared with control (noninjected cells) on the same glass slide. Bottom, myocytes stained with rabbit anti-5 antibody. Cells injected with anti-5 or anti-3 antisense oligonucleotides were compared with control (noninjected cells). B, diagram depicting G11 expression in cells stained with anti-11, anti-q, anti-3, and anti-5 antibody. Cells injected with anti-11 and anti-q antisense oligonucleotides were compared with control (noninjected cells). Open column (noninjected cells) and hatched columns (antisense oligonucleotide-injected cells) show means ± S.E.M. with the number of cells tested in parentheses. Immunofluorescence was expressed in arbitrary units (AU). , values significantly different from those obtained under control conditions (P < .05).

G₁₁-Dependent ET-1-Induced Increase in [Ca²⁺]_i To determine which G protein subunit (₁₁ or ) was involved in effector activation, an anti-_com antibody raised to the carboxyl terminus of the  subunit was dialyzed into the cell for 7 to 8 min, as reported previously (Macrez et al., 1997). Intracellular applications of 10 to 20 µg/ml anti-_com antibody had no effect on the ET-1-induced increase in [Ca²⁺]_i (Fig. 9A) but removed the Ca²⁺ responses evoked by 10 nM AII (Fig. 9B). In a second set of experiments, a peptide corresponding to a fragment of ARK₁ was dialyzed into the cells for 7 to 8 min. Carboxyl-terminal fragments of ARK₁ have been used to bind G subunits and to block activation of effectors (Nair et al., 1995; Stehno-Bittel et al., 1995). Intracellular application of the peptide corresponding to the G binding domain of ARK₁ had no significant effects on the ET-1-induced increase in [Ca²⁺]_i (Fig. 9A), but removed the AII-induced increase in [Ca²⁺]_i (Fig. 9B). This is in agreement with previous data showing that angiotensin AT₁ receptor uses the dimers of G₁₃ to transduce the signal leading to activation of Ca²⁺ channels and increase in [Ca²⁺]_i (Macrez et al., 1997). Taken together, these results suggest that the ET_A receptors transduce their signal to PLC through G₁₁, and not through G subunits.

View larger version (27K): [in this window] [in a new window]   Fig. 9.   Effects of anti-com antibody and ARK1 peptide on ET-1- and AII-induced Ca2+ responses. A, Ca2+ responses evoked by 10 nM ET-1 in cells held at 50 mV in which antibody or peptide were dialyzed intracellularly for 7 to 8 min with Indo-1 (hatched columns) compared to control cells (open columns). Data show means ± S.E.M. with the number of experiments in parentheses. External solution contained 1 µM oxodipine. B, Ca2+ responses evoked by 10 nM AII in cells held at 50 mV in which antibodies or peptide were dialyzed intracellularly for 7 to 8 min (hatched columns) compared to control cells (open columns). Data show means ± S.E.M. with the number of experiments in parentheses. , Values significantly different from those obtained under control conditions (P < .05).

View larger version (16K): [in this window] [in a new window]   Fig. 10.   Increase in [Ca2+]i evoked by NE and ET-1 in myocytes injected with anti-q or anti-11 antisense oligonucleotides. A, in Ca2+-free (0.5 mM EGTA)-containing solution, Ca2+ releases obtained in the same noninjected cells (after 3 days of primary culture) in response to 10 µM NE or 10 nM ET-1. B, Ca2+ releases evoked by NE and ET-1 in the same cells, injected with 10 µM anti-q antisense oligonucleotides. C, Ca2+ releases evoked by NE and ET-1 in the same cells, injected with 10 µM anti-11 antisense oligonucleotides. Myocytes were loaded with Fura-2 AM and not patch-clamped. Similar results were obtained in nine to fourteen cells.

Time Course of Inositol Phosphate Accumulation Induced by ET-1 and NE. Because the time courses of the Ca²⁺ responses evoked by ET-1 and NE in control conditions were largely different, we examined the time course of inositol phosphate accumulation induced by the two mediators in intact portal vein myocytes. With 10 mM LiCl, the unstimulated accumulations of total InsP and InsP₃ were 1136 ± 101 and 185 ± 15 dpm/10⁶ cells (n = 3), respectively. After application of ET-1 (10 nM) or NE (10 µM) for 2 min, the total InsP-associated radioactivity increased against time and reached 2287 ± 130 dpm/10⁶ cells and 2305 ± 145 dpm/10⁶ cells, respectively (n = 3; Fig. 11A). These increases in InsP radioactivity were associated with a decrease in phosphoinositide radioactivity from 53316 ± 1850 dpm/10⁶ cells in control to 43742 ± 1570 dpm/10⁶ cells in the presence of the mediators for 2 min (n = 4). When focusing on the InsP₃ production (Fig. 11B), NE produced a rapid and transient stimulation of InsP₃ accumulation (~175% at 15 s) followed by a maintained stimulation (~50% at 2 min). In contrast, stimulation of InsP₃ accumulation induced by ET-1 increased progressively, reaching ~40% at 15 s and ~130% at 2 min (Fig. 11B). The sequential generation of inositol phosphates in the order of InsP₃, InsP₂, and InsP₁ (not shown) provides evidence that both ET-1 and NE mediate the stimulation of a PLC activity that degrades phosphatidylinositol bisphosphate. The profile of InsP₃ accumulation induced by ET-1 and NE appears to support the different time courses of the Ca²⁺ responses evoked by the two mediators.

View larger version (43K): [in this window] [in a new window]   Fig. 11.   Time course of NE and ET-1-induced accumulation of inositol phosphates in rat portal vein myocytes preincubated for 3 days with [3H]inositol. The myocytes were incubated for 10 min with 10 mM LiCl before addition of 10 µM NE or 10 nM ET-1. Stimulations were stopped at times indicated, and individual [3H]inositol phosphates were separated. A, total InsPs accumulation induced by NE and ET-1. Results are expressed in dpm/106 cells and are the means ± S.E.M. of three independent experiments, each done in duplicate. B, InsP3 accumulation induced by NE and ET-1. In the presence of LiCl (basal), the mean total [3H] radioactivity was 98280 ± 3350 dpm/106 cells (n = 3). The percentage of radioactivity present in free inositol, total phosphoinositide, and total inositol phosphate pools was 44.59 ± 1.55%, 54.25 ± 1.52% and 1.16 ± 0.11%, respectively (n = 3).

	Discussion

Top Summary Introduction Materials and Methods Results Discussion References

Our results show that, in rat portal vein myocytes, the G protein heterotrimer G₁₁₃₅ is coupled to ET_A receptors and activates a transduction pathway leading to Ca²⁺ release from the intracellular store. This conclusion is based on experiments using oligonucleotides to block expression of G protein subunits, and a synthetic peptide corresponding to the carboxyl terminus of G_q/11 subunits to disrupt the ET_A receptor-evoked activation of G proteins. Use of antibodies raised against the carboxyl-terminus of G subunits and of a synthetic peptide corresponding to the carboxyl terminus of ARK₁ fragments has allowed us to identify the G protein subunits interacting with the effector.

The ET-1-induced increase in [Ca²⁺]_i displayed a transient and a more or less sustained Ca²⁺ phase. Although reduced in amplitude, the transient Ca²⁺ phase persists after the removal of external Ca²⁺ or in the presence of oxodipine (an L-type Ca²⁺ channel antagonist), and is considered to be the result of a Ca²⁺ release from the intracellular store. The sustained Ca²⁺ phase depends on transmembrane Ca²⁺ influx, as it is removed in Ca²⁺-free solution and reduced in the presence of oxodipine. These results are in good agreement with previous data obtained in other smooth muscles (Minowa et al., 1997; Saita et al., 1997). Although both ET_A and ET_B receptors have been identified in various smooth muscles including portal vein (Saita et al., 1997; Mickley et al., 1997; Wang et al., 1997), our data are consistent with the idea that only the ET_A receptor subtype is involved in intracellular Ca²⁺ release; BQ123 (an antagonist of ET_A receptors) inhibits, in a concentration-dependent manner, the ET-1-induced Ca²⁺ release, whereas BQ788 (an antagonist of ET_B receptors) is ineffective.

Antisense oligonucleotides have been used previously in these myocytes to identify the G protein heterotrimers selectively coupling ₁-adrenoceptors and angiotensin AT₁ receptors to increase in [Ca²⁺]_i (Macrez-Leprêtre et al., 1997a,b). The present results indicate that anti-₁₁ oligonucleotide injection into the nucleus of myocytes selectively inhibited the ET_A-induced Ca²⁺ release by ~75%, a value similar to that obtained with injection of anti-₃ or anti-₅ antisense oligonucleotides. These inhibitions are correlated with reductions in the expression of the corresponding G protein subunits immunostained by anti-₁₁, anti-₃, and anti-₅ antibodies. The fact that the ET_A-induced Ca²⁺ release was selectively inhibited after blockade of each subunit expression (₁₁, ₃, or ₅) supports the concept that activation of G protein by ET-1 requires the presence of all three ₁₁, ₃, and ₅ subunits. In addition, our results indicate that the subunits (₃₅) associated with the ₁₁ subunit in the ET-1-induced Ca²⁺ release differ from those associated with both _q and ₁₃ subunits in the transduction pathways activated by NE and AII (Macrez-Leprêtre et al., 1997a,b).

Antisense oligonucleotide blockade alone cannot distinguish which parts of the G protein are required for interaction with the ET_A receptor and the effector. Intracellular application of the carboxyl-terminal _q/11 peptide inhibits, in a concentration-dependent manner, the ET_A receptor-induced Ca²⁺ release, indicating that the receptor binds to the extreme carboxyl terminus of the ₁₁ subunit to promote dissociation of the heterotrimeric G protein. The observation that ET_A receptor-induced Ca²⁺ release is not affected by an anti-_com antibody or by a synthetic peptide derived from a ARK₁ fragment that selectively binds to the G subunits (Nair et al., 1995; Stehno-Bittel et al., 1995) supports the idea that the stimulation of the effector occurs via the  subunit of the G₁₁ protein. In addition, ET-1 triggers a sequential generation of inositol phosphates in the order of InsP₃, InsP₂, and InsP₁, indicating that ET_A receptor activation leads to the stimulation of a PLC-degrading phosphatidylinositol bisphosphate. Although the PLC- isoforms mediating Ca²⁺ release from the intracellular store have not been identified in rat portal vein myocytes, one may speculate that the G₁₁ subunit predominantly activated a PLC- to hydrolyze phosphatidylinositol bisphosphate. The slow kinetics of ET_A-induced Ca²⁺ release reaching a peak within 30 to 40 s can be associated with the delayed accumulation of InsP₃. This is in contrast with the fast activation of PLC by G_q subunit in response to activation of ₁-adrenoceptors, which produces a fast increase in InsP₃ (15 s) associated with a fast and large Ca²⁺ response. Although G_q and G₁₁ have been reported to be similarly efficient in vitro and in overexpression experiments, with regard to stimulation of various PLC- isoforms (Wu et al., 1993), our results indicate that the G₁₁-activated Ca²⁺ responses evoked by ET-1 are smaller and slower than the G_q-activated Ca²⁺ responses evoked by NE in the same cells. However, the G₁₁-mediated Ca²⁺ responses have similar amplitude and time course when activated by ET-1 or NE (after inhibition of the G_q subunit). Several explanations may arise: 1) G₁₁ protein may be distributed homogeneously in the plasma membrane but in a limited amount, so that its activation leads to a delayed stimulation of PLC-. In contrast, G_q proteins have been reported to be clustered in membrane domains associated with cytoskeletal proteins including F-actin (Ibarrondo et al., 1995; Cornea et al., 1998). If ET-1 receptors are localized outside these G_q-rich membrane domains, this may also explain why ET-1-induced Ca²⁺ release is specifically transduced through G₁₁ and not through G_q; 2) G₁₁ subunits may regulate the activity of intermediate effector proteins involved in the transduction of the ET_A-induced signal; 3) It has been recently shown that the ₅ subunit is preferentially linked with membrane components rather than with F-actin, in contrast to ₃ and ₁₂ subunits, which show a specific association with F-actin (Ueda et al., 1997). It is, however, unlikely that the G₅ subunit is responsible for the delayed effect of G₁₁ subunit on PLC activation and Ca²⁺ release because different G subunits (₂ or ₃) are dissociated from the NE-activated G₁₁ protein; and 4) Finally, it is possible that the rate of dissociation of the activated G₁₁₃₅ protein may be much slower than that of the activated G_q13 protein or that a more effective ₁₁-selective guanine nucleotide-regulatory protein (RGS-type) leads to a very fast reassociation of ₁₁ subunits with ₃₅ subunits. In vitro experiments are needed to evaluate these different possibilities.

In conclusion, we show that, in portal vein myocytes, the ET_A receptor binds to a specific G₁₁₃₅ heterotrimer, leading to a slow accumulation of InsP₃ and the subsequent delayed Ca²⁺ release from the intracellular store.