Activation of Endothelin ETA Receptors Induces Phosphorylation of α1b-Adrenoreceptors in Rat-1 Fibroblasts*

The effect of endothelin-1 on the phosphorylation of α1b-adrenoreceptors, transfected into rat-1 fibroblasts, was studied. Basal α1b-adrenoreceptor phosphorylation was markedly increased by endothelin-1, norepinephrine, and phorbol esters. The effect of endothelin-1 was dose dependent (EC50 ≈ 1 nm), reached its maximum 5 min after stimulation, and was inhibited by BQ-123, an antagonist selective for ETA receptors. Endothelin-1-induced α1b-adrenoreceptor phosphorylation was attenuated by staurosporine or genistein and essentially abolished when both inhibitors were used together. The effect of norepinephrine was not modified by either staurosporine or genistein alone, and it was only partially inhibited when both were used together. These data suggest the participation of protein kinase C and tyrosine kinase(s) in endothelin-1-induced receptor phosphorylation. However, phosphoaminoacid analysis revealed the presence of phosphoserine and traces of phosphothreonine, but not of phosphotyrosine, suggesting that the putative tyrosine kinase(s), activated by endothelin, could act in a step previous to receptor phosphorylation. The effect of endothelin-1 on α1b-adrenoreceptor phosphorylation was not mediated through pertussis toxin-sensitive G proteins. Calcium mobilization induced by norepinephrine was diminished by endothelin-1. Norepinephrine and endothelin-1 increased [35S]GTPγS binding to control membranes. The effect of norepinephrine was abolished in membranes obtained from cells pretreated with endothelin-1. Interestingly, genistein plus staurosporine inhibited this effect of the endothelial peptide. Endothelin-1 did not induce α1b-adrenoreceptor internalization. Our data indicate that activation of ETA receptors by endothelin-1 induces α1b-adrenoreceptor phosphorylation and alters G protein coupling.

␣ 1b -Adrenoreceptors (␣ 1B -ARs) 1 belong to the seven transmembrane domains-G protein-coupled family of receptors that modulate phosphoinositide turnover (for reviews, see Refs. [1][2][3] (uppercase (␣ 1B ) is used when referring in general to this subtype or to receptors characterized pharmacologically in cells or tissues that naturally express them, and lowercase (␣ 1b ) is used exclusively, when referring to cloned receptors, as suggested by the International Union of Pharmacology (IUPHAR) (4)). It is generally accepted that their function and cellular distribution (internalization/recycling) are, at least partially, regulated by phosphorylation/dephosphorylation processes. The roles that different serine-threonine kinases specific for G protein-coupled receptors (GRKs) and second messenger-regulated kinases (like protein kinase A and PKC) play in such events have been extensively studied but remain yet only partially known, and some aspects of their function are still controversial (for reviews, see Refs. 5 and 6). The complexity of the problem of assigning roles to specific isoforms is exemplified by the ability of messenger-regulated kinases to modulate the activity of GRKs (crosstalk) (reviewed in Ref. 7).
Phosphorylation of ␣ 1B -ARs seems to be a key regulatory step for the shutdown of their function and takes place in the presence of agonist or after the activation of second messengerregulated kinases, such as PKC (8 -12). Activation of PKC with phorbol esters blocks ␣ 1B -adrenergic actions in cells that naturally express this subtype, such as hepatocytes (13)(14)(15)(16) and DDT 1 MF2 smooth muscle cells (8). Likewise, such effect has been observed in cells transfected with this receptor subtype (9,10,17). GRK2 and GRK3 have been involved in homologous desensitization of ␣ 1b -AR in COS-7 cells cotransfected with both the receptor and the kinases (10).
On the other hand, in rat-1 fibroblasts, endothelin increases IP 3 production and activates PKC (18). Interestingly, many other signaling processes are also activated by this endothelial peptide. Thus, endothelin inhibits adenylyl cyclase through pertussis toxin-sensitive GTP-binding proteins, turning off the activity of protein kinase A (19,20). It also stimulates the phosphorylation of several proteins in tyrosine residues and activates the mitogen-activated protein kinase pathway in rat-1 fibroblasts and other cell types (19 -21). The expression of c-fos and other cellular responses induced by activation of endothelin receptors involve stimulation of c-src (22). Focal adhesion kinase is phosphorylated in cells responding to endothelin (23,24). Rho-mediated endothelin-1 stimulation of phospholipase D in rat-1 fibroblasts has also been recently reported (25).
The ability of endothelin receptors to elicit activation of different cellular pathways allows them to influence the activity of other receptors present in the same cell. Recently, it has been elegantly demonstrated that ET-1 transmodulates positively the activity of epidermal growth factor receptors in rat-1 fibroblasts (26). In this study, it was observed that activation of ET A receptors elicited phosphorylation of epidermal growth factor (EGF) receptors and its association with Grb-2 and Shc; i.e. ET-1 induces the activation of EGF receptors in the absence of its ligand (26).
In previous studies, we demonstrated that tetradecanoylphorbol acetate (TPA) attenuates the action of ␣ 1b -ARs transfected into rat-1 fibroblasts (17). It has already been demonstrated that in these cells, activation of PKC by phorbol esters results in the phosphorylation of this transfected receptor (9,10). In the present work, the effect of activation of endothelin ET A receptors (that are endogenously expressed in rat-1 cells) on the phosphorylation of ␣ 1b -ARs was examined. Since it is known that ET-1 induces activation of tyrosine kinases in rat-1 cells, the signaling pathways that participate in such effect were explored.  (27). The carboxyl-terminal peptide of the ␣ 1b -AR was obtained from Multiple Peptide Systems. Polyvinylidene difluoride membranes were obtained from Bio-Rad.

EXPERIMENTAL PROCEDURES
Cell Lines and Culture-Rat-1 fibroblasts transfected with the hamster ␣ 1b -AR (28), generously provided to us by Drs. R. J. Lefkowitz, M. G. Caron, and L. Allen (Duke University), were cultured in glutamine-containing high glucose DMEM supplemented with 10% fetal bovine serum, 300 g of the neomycin analog per ml, G-418 sulfate, 100 g/ml streptomycin, 100 units/ml penicillin, and 0.25 g/ml amphotericin B at 37°C under a 95% air/5% CO 2 atmosphere as described previously (17). In the present experiments, cells at confluence were serum deprived in unsupplemented DMEM for 24 h.
Photoaffinity Labeling of the ␣ 1b -AR-Membranes (100 g) from rat-1 cells expressing ␣ 1b -AR (2-2.5 pmol/mg protein), prepared as described previously (17), were incubated in the dark with 6 nM aryl-125 I-labeled azidoprazosin essentially as described by Lattion et al. (9) for 1 h at room temperature; then 1 ml of 50 mM Tris-HCl, pH 7.4, 150 mM NaCl, 5 mM EDTA containing protease inhibitors (leupeptin 20 g/ml, aprotinin 20 g/ml, phenylmethylsulfonyl fluoride 100 g/ml, bacitracin 500 g/ml, and soybean trypsin inhibitor 50 g/ml) was added, and the tubes were exposed to UV light for 3 min. After this treatment, membranes were centrifuged at 12,700 ϫ g for 15 min, washed, and electrophoresed in 10% SDS-polyacrylamide gel under reducing conditions. The specificity of the labeling was determined using phentolamine or prazosin as competitors.
Immunoprecipitation of ␣ 1b -AR-Immunoprecipitation of ␣ 1b -AR was standardized using photoaffinity-labeled membranes. Rabbit antiserum against the decapeptide of the carboxyl terminus sequence of the hamster ␣ 1b -AR was obtained by immunizing rabbits with the indicated peptide coupled to keyhole limpet hemocyanin. Serum samples with a titer above 1:5000, as determined by enzyme-linked immunosorbent assay (peptide), were used. Membranes were solubilized with 1 ml of solubilization buffer (10 mM Tris-HCl, pH 7.4, 50 mM NaCl, 5 mM EDTA, 1.0% Triton X-100, 0.05% SDS, 50 mM NaF, 100 M Na 3 VO 4 , 10 mM ␤-glycerophosphate, 10 mM sodium pyrophosphate, 1 mM phosphoserine, 1 mM phosphothreonine, and the previously mentioned protease inhibitors at the concentrations indicated) for 1 h at 4°C; the extracts were centrifuged at 12,700 ϫ g for 15 min at 4°C, and the supernatants were transferred to new tubes containing 10 l of immune serum and 40 l of Sepharose-coupled protein A (1:1 (v/v) slurry). Tubes were incubated overnight at 4°C. Beads were washed five times (1 ml/each) with 50 mM Hepes, 50 mM NaH 2 PO 4 , 100 mM NaCl, pH 7.2, containing 1.0% Triton X-100, 0.05% SDS, and 100 mM NaF; once with 50 mM Tris, 0.15 M NaCl, pH 7.4; and once with 10 mM Tris, 0.15 M NaCl, pH 7.4. Washed beads were boiled in 100 l of Laemmli sample buffer containing 5% mercaptoethanol and electrophoresed in 10% SDS-polyacrylamide minigels, transferred to nitrocellulose, and exposed to X-Omat x-ray film (Eastman Kodak Co.) at Ϫ80°C with an intensifying screen. Autoradiographs were analyzed with the Collage software, version 2.0.
Phosphorylation of ␣ 1b -AR-Rat-1 fibroblasts expressing the ␣ 1b -AR were grown to confluence in 10-cm culture dishes and serum starved for 24 h. On the day of the experiment, cells were maintained in phosphatefree DMEM during 1 h and then incubated in 3 ml of the same medium containing [ 32 P]P i (0.05 mCi/ml) for 3 h at 37°C. Labeled cells were stimulated with ET-1, norepinephrine, EGF, or TPA as indicated; then they were washed twice with ice-cold phosphate-buffered saline and solubilized with 1.0 ml of ice-cold solubilization buffer. The plates were maintained for 1 h on ice; then the extracts were centrifuged at 12,700 ϫ g for 15 min, at 4°C, and the supernatants were transferred to new tubes containing 10 l of anti-␣ 1b -AR antiserum and 40 l of Sepharose-coupled protein A and immunoprecipitated as described above. The effect of staurosporine, genistein, and wortmannin on the phosphorylation of ␣ 1b -AR was assayed by preincubating these agents in labeling medium for the time and concentrations indicated in the figures. The amount of phosphorylated receptor was determined by densitometric analysis of autoradiographs or by PhosphorImager analysis. At least three independent experiments were performed for each treatment. Data were analyzed and plotted using commercial software (Prism 2.01, GraphPad Software).
PTX Treatment of the Cells-Transfected rat-1 fibroblasts, in 10-cm dishes, were incubated for 16 -24 h with or without PTX (100 ng/ml). The presence of PTX was continued throughout the phosphorylation period. Pertussis toxin-catalyzed ADP-ribosylation was carried out as described previously (29).
Phosphoaminoacid Analysis-Immunoprecipitated phosphorylated receptors were transferred to polyvinylidene difluoride membranes and hydrolyzed in 0.2 ml of 6 N HCl at 110°C for 1.5 h. Samples were then centrifuged at 12,700 ϫ g for 5 min, and the acid supernatants were evaporated to dryness in a vacuum concentrator (Speed Vac). Hydrolyzed samples were solubilized in a mixture of cold phosphoaminoacid standards (Ser(P), Thr(P), and Tyr(P), 25 nmol each) and analyzed by two dimensional thin layer cellulose electrophoresis, using 750 V for each dimension (30).

Membrane Preparation and [ 35 S]GTP␥S
Binding-Overnight serumdeprived confluent cultures were treated with the different agents to be tested, washed, and scraped with a rubber policeman in buffer containing 20 mM Hepes, pH 7.5, 5 mM EDTA, 100 M Na 3 VO 4 , 10 mM ␤glycerophosphate, 10 mM sodium pyrophosphate, 2 mM MgCl 2 , and the previously mentioned protease inhibitors at the concentrations indicated. Membranes were prepared according to the method of Mattingly et al. (31) and resuspended in binding buffer (50 mM Tris, 10 mM MgCl 2 , 1 mM EDTA, pH 7.5). [ 35 S]GTP␥S binding was performed as described by Wieland and Jakobs (32) with minor modifications. In brief, membranes were preincubated in binding buffer supplemented with 100 M Gpp(NH)p, 100 M GDP, and 1 mM dithiothreitol for 30 min at 30°C in a shaking water bath. After this preincubation, membranes were washed and resuspended in binding buffer without nucleotides. The binding reaction was carried out in a volume of 250 l for 30 min at 30°C in the same buffer containing 0.2 nM [ 35 S]GTP␥S. The reaction was initiated by the addition of membranes (25 g of protein/tube) and terminated by the addition of 2 ml of ice-cold binding buffer without EDTA and filtration on Whatman GF/C filters using a Brandel harvester. Nonspecific binding was determined in the presence of 100 M Gpp(NH)p and represented Ϸ10% of total binding. Filters were washed three times and dried, and radioactivity was measured with a liquid scintillation counter. Statistical analysis between comparable groups was performed using analysis of variance with Newman-Keuls analysis.
[Ca 2ϩ ] i Measurements-Confluent fibroblasts were incubated overnight in DMEM without serum and antibiotics. Cells were loaded with 5 M Fura-2/AM in Krebs-Ringer-Hepes containing 0.05% bovine serum albumin, pH 7.4, for 1 h at 37°C. Cells were detached by gentle trypsinization (17). Fluorescence measurements were carried as described, but the excitation monochromators were set at 340 and 380 nm, with a chopper interval of 0.5 s, and the emission monochromator was set at 510 nm. [Ca 2ϩ ] i was calculated according to the method of Grynkiewicz et al. (33) using the software provided by AMINCO-Bowman; traces were directly exported to the graphs.
Surface Receptors-Intact cell receptor binding assays were performed as described by Diviani et al. (10) by incubating cell monolayers grown in 24-well clusters containing 1 nM [ 3 H]prazosin. Incubation was at 4°C for 12-15 h. After binding, cells were washed three times with ice-cold phosphate-buffered saline, scraped, and counted. Phentolamine (10 M) was used to determine nonspecific binding which was Ϸ10% of total binding.

RESULTS
Photoaffinity Labeling and Immunoprecipitation of the ␣ 1b -AR-A polyclonal rabbit antibody directed against the last 10 amino acids of the hamster ␣ 1b -AR carboxyl terminus was generated and used for immunoprecipitation. To define that the receptor was immunoprecipitated, membranes where the receptor was labeled by photoaffinity cross-linking with aryl-125 I-azidoprazosin were used. Photoaffinity labeling of the rat-1 membranes resulted in a major band with a molecular weight of Ϸ85,000 (Fig. 1a), which is within the expected size for the ␣ 1b -AR (9, 10). As anticipated, the labeling was blocked with phentolamine ( Fig. 1a) and prazosin (data not shown), confirming the identity of the receptor. Other minor bands were also observed after prolonged exposure of the gels: a band of high molecular weight (Ͼ150,000), which may correspond to aggregates of receptors or oligomers (34); and two other faint bands with molecular weights of Ϸ40,000 and Ϸ25,000. These later two bands do not seem to correspond to the receptor since they were also observed when the labeling was performed in the presence of phentolamine (Fig. 1a). Fig. 1b shows that the ␣ 1b -AR is immunoprecipitated from aryl-125 I-azidoprazosin-labeled membranes; the procedure was very reproducible and most efficient when the incubation with the antibody was performed overnight at 4°C (70 -80% of the labeled receptor was immunoprecipitated under these conditions; range of 3 determinations in 3 experiments using different membrane preparations) compared with 1 h at the same temperature (Fig. 1b). Co-immunoprecipitation of the high molecular weight band and of the nonspecific band with a molecular weight of Ϸ40,000 was observed in some of the experiments. No immunoprecipitation of ␣ 1b -AR was detected when preimmune serum was used as control in the same conditions under which a maximum efficiency was achieved with immune serum (Fig. 1b). Antiserum against ␣ 1b -AR identified the receptor in Western blots of membranes obtained from ␣ 1b -AR-transfected rat-1 fibroblasts but no signal was detected in membranes from ␣ 1a -ARtransfected cells (data not shown).
␣ 1b -AR Phosphorylation-It can be observed in Fig. 2 that a labeled band with a molecular weight of Ϸ85,000 was immunoprecipitated from rat-1 cells metabolically labeled with ra-dioactive phosphate. Such basal receptor phosphorylation was markedly increased in cells treated for 5 min with 10 M norepinephrine, 10 nM ET-1, or 1 M TPA. The ET-1-stimulated ␣ 1b -AR phosphorylation took place very rapidly, reaching its maximum at 5 min (Fig. 3, a and b) and gradually decreasing afterward. The effect was dependent on the concentration of ET-1 used, with an apparent EC 50 of Ϸ1.3 nM (Fig. 4, a and b). The stimulation of ␣ 1b -AR phosphorylation by 10 nM ET-1 was inhibited in a concentration-dependent fashion by BQ-123 (Fig.  4, d and e), a selective ET A receptor antagonist (IC 50 Ϸ 235 nM, K i Ϸ25 nM). The K i obtained is within what has been observed for ET A receptors (35), which is the subtype present in rat-1 cells (36). No effect of the antagonist on the basal phosphorylation of the ␣ 1b -AR was detected (data not shown). The action of norepinephrine was also dose dependent, showing an apparent EC 50 of 25 nM; the effect of norepinephrine was of a smaller magnitude than that of ET-1 (Fig. 4, a and c) but had a very similar time course (Fig. 3, a and b).
As indicated before, some effects of ET-1 have been reported to be mediated by activation of PKC, whereas other actions seem to involve tyrosine kinases. Therefore, we investigated the participation of these enzymes in the events that lead to the phosphorylation of the ␣ 1b -AR. Preincubation of the cells with either the tyrosine kinase inhibitor, genistein, or the PKCinhibitor, staurosporine, inhibited only partially the ␣ 1b -AR phosphorylation elicited by ET-1 (Fig. 5, a, b, e, and f, and Fig.  5, c-f; respectively). None of these inhibitors blocked the ␣ 1b -AR phosphorylation induced by norepinephrine (Fig. 5, e and f). When these inhibitors were used together, they essentially abolished the phosphorylation of the ␣ 1b -AR induced by ET-1 and reduced the effect of norepinephrine (Fig. 5, e and f).
No effect of the inhibitors on the basal phosphorylation of the ␣ 1b -AR was detected (data not shown). Wortmannin (100 nM) did not affect the ␣ 1b -AR phosphorylation induced by either norepinephrine or ET-1 (Fig. 5e).
Phosphoaminoacid Analysis and Effects of PTX and EGF-To get a more clear picture of the effect of ET-1 on the phosphorylation of ␣ 1b -AR, phosphoaminoacids were analyzed. It was observed that in all the experimental conditions tested Ser(P) was the major phosphoaminoacid detected with only trace amounts of label in Thr(P). We were unable to detect Tyr(P) in any of three independent experiments performed for each experimental condition (Fig. 6).
PTX-sensitive G proteins participate in some actions of ET-1 (24). Therefore, we investigated the effect of PTX pretreatment on the phosphorylation of ␣ 1b -ARs induced by stimulation of endothelin receptors. For this purpose, cells were incubated in serum-free DMEM for 16 -24 h with 100 ng/ml PTX, and the phosphorylation of ␣ 1b -AR was assayed as before. This whole cell treatment essentially inactivated all PTX-sensitive G proteins as evidenced by the virtual absence of in vitro PTXcatalyzed ADP-ribosylation in membranes from toxin-treated cells (Fig. 7c). The stimulations of ␣ 1b -AR phosphorylation induced by norepinephrine, ET-1 and TPA were not different in PTX-treated cells as compared with control cells (Fig. 7, a and  b).
It has recently been shown that ET-1 induces tyrosine phosphorylation and activation of EGF receptors in rat-1 fibroblast (26); therefore, we tested whether the phosphorylation of ␣ 1b -AR induced by activation of ET A receptors was mediated by the EGF receptors. We were unable to observe a consistent stimulation by EGF of ␣ 1b -AR phosphorylation (data not shown).
Functional Consequences-To obtain information on the functional consequences of ET-1-induced ␣ 1b -AR phosphorylation, the effect of ET-1 on the ability of norepinephrine to increase [Ca 2ϩ ] i was studied. It can be observed that ET-1 increased [Ca 2ϩ ] i and dose-dependently inhibited the effect of norepinephrine on this parameter (Fig. 8). However, this effect was not selective for the ␣ 1b -adrenergic action since after the initial challenge with ET-1, its own action (second challenge) was also decreased; this was particularly clear when high concentrations of ET-1 were used in the first challenge (data not shown).
Similarly complex was the study of the effect of ET-1 on ␣ 1b -AR-mediated IP 3 production. ET-1 stimulated IP 3 production, and even after several washes the levels of this second messenger remained markedly elevated, making it difficult to evaluate the ␣ 1b -adrenergic action (data not shown).
[ 35 S]GTP␥S binding was studied to determine whether activation of ET A receptors induces any effect on ␣ 1b -AR-G protein coupling. In membranes from control fibroblasts, norepinephrine and ET-1 were clearly able to increase [ 35 S]GTP␥S binding in vitro (p Ͻ 0.01 and p Ͻ 0.001 versus their basal, respectively) ( Fig. 9, top). When membranes from norepinephrine-treated fibroblasts were used, the catecholamine was no longer able to stimulate the binding of the radiolabeled nucleotide, and ET-1 induced a small, marginal effect (Fig. 9, top). In membranes from ET-1-treated cells, neither norepinephrine nor ET-1 induced any increase of [ 35 S]GTP␥S binding (Fig. 9, top).
Since the combination of genistein and staurosporine was able to inhibit ET-1-mediated ␣ 1b -AR phosphorylation in whole cells, the in vivo effect of these protein kinase inhibitors on agonist-stimulated [ 35 S]GTP␥S binding to membranes was tested. These protein kinase inhibitors by themselves did not block the in vitro effects of norepinephrine (p Ͻ 0.001 versus its basal) and ET-1 (p Ͻ 0.001 versus its basal) on this parameter in control membranes (Fig. 9, bottom). In membranes from cells incubated with the protein kinase inhibitors and norepinephrine, the catecholamine (p Ͻ 0.05 versus its basal) and ET-1 (p Ͻ 0.001 versus its basal) increased [ 35 S]GTP␥S binding (Fig.  9, bottom). However, the effects were clearly smaller than those observed in membranes from cells incubated without norepinephrine. In membranes from cells incubated with the protein kinase inhibitors and ET-1, norepinephrine (p Ͻ 0.001 versus its basal), but not ET-1, increased [ 35 S]GTP␥S binding in vitro (Fig. 9, bottom).
Finally, we studied comparatively the effect of norepinephrine and ET-1 on surface ␣ 1b -AR. It can be observed that norepinephrine induced a small decrease in surface receptors but, in contrast, ET-1 did not produce any decrease (Fig. 10). DISCUSSION In the present work, the phosphorylation of ␣ 1b -AR, expressed by stable transfection in rat-1 fibroblasts, was studied. Our results clearly show that activation of ET A receptors, endogenously expressed in these cells, leads to phosphorylation of ␣ 1b -ARs. The effect of ET-1 was of greater magnitude than that of norepinephrine (homologous), suggesting that a strong process of rapid heterologous regulation takes place in these cells. This effect of ET-1 seems to involve the participation of serine/ threonine and tyrosine kinases, since it is partially blocked by staurosporine and genistein, and it is almost completely inhibited when these inhibitors acted together. The temporal course for the ␣ 1b -AR phosphorylation induced by ET-1 stimulation is similar to the time course detected for the translocation of PKC-␦ and PKC-⑀ isoforms in neonatal ventricular myocytes stimulated with ET-1 (37). The isoforms of PKC detected in rat-1 fibroblasts (38) and in neonatal ventricular myocytes (37) are the same (i.e. ␣, ␦, ⑀, and isoforms), which suggests the possibility that ␦ and ⑀ PKC-isoforms could be involved in the phosphorylation of ␣ 1b -ARs induced by activation of ET A receptors in rat-1 cells. Moreover, the experiments in neonatal ventricular myocytes showed that the translocation of ␦ and ⑀ PKC isoforms is stronger in the cells activated with ET-1 than in those in which phenylephrine was used (37). Coincidentally, in the present study the action of ET-1 was also stronger than that of norepinephrine. These data further suggest that the effect of ET-1 on the phosphorylation state of ␣ 1b -ARs could be mediated, at least in part, by the activation of PKC. Consistent with our present findings, it has been reported that TPA elicits a stronger phosphorylation of ␣ 1b -ARs than an homologous stimulus (9).
Several studies have demonstrated the cross-regulation between receptors coupled to different signal transduction pathways. For example, activation of adenylyl cyclase counterregu- lates phosphoinositide turnover through phosphorylation of phospholipase C by protein kinase A (39), actions of ␤-adrenergic receptors can be inhibited by insulin, and this effect seems to involve ␤-adrenergic receptor tyrosine phosphorylation (40,41). Similarly, as already mentioned, G-protein-coupled receptors (ET A receptors) can influence the activity of tyrosine kinase receptors (EGF receptors) (26). Cross-regulation between receptors coupled to the same transduction pathway (particularly receptors coupled to phosphoinositide turnover and intracellular calcium mobilization) has been difficult to substantiate due to the labile characteristics of the mediators involved and their pools and to the recovery period required to elicit maximum calcium responses. The direct effect on the receptors has been studied to a lesser extent, but some examples have already suggested cross-regulation at this level. For example, it has been shown that ET-1 induces internalization of thrombin receptors via a PKC-mediated process (42). In addition, crossregulation between several members of lymphocyte chemotactic receptors involving phosphorylation and the possible participation of PKC has recently been shown (43).
Endothelin seems to be able to cross-regulate the activity of coexpressed receptors in different cell systems, and in some cases PKC activation seems to participate. In neuroblastomaglioma hybrid cells, ET-1 induces partial heterologous desensitization of the calcium response elicited by ATP and has no influence on the response elicited by bradykinin (44). In rat urinary bladder, the endothelial peptide potentiates the contractions evoked by exogenous ATP in a selective way since it did not modify the contractions induced by acetylcholine, 5hydroxytryptamine, prostaglandin F 2␣ or bradykinin (45). In cardiac myocytes, ET-1 enhances the calcium transients triggered by ATP, and this effect was sensitive to inhibitors of protein kinase C; it also inhibits a K ϩ current by means of a PKC-dependent mechanism (46).
The participation of serine/threonine and tyrosine kinases in the ET-1-induced phosphorylation of ␣ 1b -AR is supported by the partial inhibition achieved by staurosporine and genistein. The partial blockade obtained by each inhibitor suggest that such kinases acted by independent pathways in the phosphorylation of the receptor, and neither seemed to be mediated by PTX-sensitive G proteins. Phosphoaminoacid analysis revealed mainly the presence of Ser(P) and trace amounts of Thr(P) in basal and stimulated conditions, coincident with previous reports for the norepinephrine-and phorbol ester-stimulated phosphorylation of ␣ 1b -AR in DDT 1 MF-2 cells (11) and in vitro PKC-mediated phosphorylation of the receptor (12). The absence of Tyr(P) in the phosphoaminoacid analysis suggests that a putative genistein-sensitive kinase could act in a step previous to the phosphorylation of the receptor. However, we cannot discard the possibility that the detection of this phosphoaminoacid could have remained under the limits of detection of our assay. Norepinephrine-induced ␣ 1b -AR phosphorylation was not inhibited by staurosporine or genistein, and only a partial blockade was observed when both inhibitors were used together. Current ideas suggest that GRKs play a major role in homologous desensitization of ␣ 1b -ARs (10). At this point, it is not clear why the combination of staurosporine and genistein partially inhibits norepinephrine-induced ␣ 1b -AR phosphorylation. Further experiments will be required to address this point. However, the effect of the inhibitors is also observed on the ability of norepinephrine to stimulate [ 35 S]GTP␥S binding in membranes from cells pretreated with the catecholamine. Thus, there seems to be a relationship between these two parameters.
Our functional data indicate that activation of ET A receptors alters the ability of norepinephrine to increase [Ca 2ϩ ] i . These experiments are complicated since the response to a new addition of ET-1 was also decreased, and the strong effect of ET-1 on this parameter may even lead to depletion of the intracellular calcium stores. However, even in these experiments, it is clear that low concentrations of endothelin-1 that induced small increases in [Ca 2ϩ ] i (compare, e.g., 0.1 nM ET-1 with 10 nM ET-1) markedly inhibited the effect of norepinephrine. Such action cannot be attributed to calcium depletion. Moreover, the [ 35 S]GTP␥S binding data strongly suggest that activation of ET A receptors alters the functional interaction of ␣ 1b -AR with G proteins. Such alteration was avoided when staurosporine and genistein were present during the incubation of the cells with ET-1. Again, there seems to be a relationship between receptor phosphorylation and catecholamine-stimulated [ 35 S]GTP␥S binding.
In rat-1 fibroblasts, participation of cytosolic tyrosine kinases in the signaling pathways activated by endothelin is supported by the observation that in v-src-transformed cells, the accumulation of IP 3 and calcium induced by endothelin was dramatically amplified; in contrast, thrombin-dependent responses were only slightly reduced (31). Focal adhesion kinase is subjected to tyrosine phosphorylation in rat-1 fibroblasts stimulated with endothelin, and this effect was attenuated when the activity of PKC was down-regulated or inhibited. Besides, pp42 and pp44 forms of mitogen-activated protein kinase are also phosphorylated in residues of tyrosine, and this effect is reduced by pretreatment with PTX, whereas the mentioned tyrosine phosphorylation of focal adhesion kinase remained intact (24). On the other hand, rat-6 fibroblasts, overexpressing PKC-␤1, showed an increased calcium signaling in the presence of endothelin; in this case, an increase in the number of ET A receptors was detected (47). It is possible that GRKs could be activated by ET A receptors; however, GRKs are reported to phosphorylate only receptors in the agonist-occupied state; the participation of these kinases in the heterologous phosphorylation of ␣ 1b -ARs seems unlikely. However, we have no evidence on the activation of GRKs after ET-1 stimulation, and we cannot discard the possibility of their participation. Recently, a role for Rho in mediating endothelin and phorbol esters stimulation of phospholipase D has been determined in rat-1 fibroblasts (25). Besides, phospholipase D seems to be regulated by PKC and protein kinase D in these cells (48), and GRK2 is also modulated by PKC, opening the possibility of regulation of ␣ 1b -AR by different kinases activated in cascade after the stimulation of ET-1 receptors.
Our studies on surface receptors also gave some interesting results. In rat-1 fibroblasts, norepinephrine induced a relatively modest and slow decrease in surface receptors. This is consistent with data obtained by another group using the same cell type (9) but differs from observations for these receptors in other cells (49,50). Cell differences may exist in the rates and regulation of receptor internalization and recycling. Nevertheless, in our experiments a dissociation between receptor phosphorylation and internalization is evidenced; i.e. activation of ET A receptors markedly stimulated ␣ 1b -AR phosphorylation but did not induce clear loss of surface receptor binding. The possibility that phosphorylation of specific serine/threonine residues of the ␣ 1b -AR could be important in receptor internalization cannot be ruled out. In fact, it has been observed that TPA induces ␣ 1b -adrenoreceptor internalization (49,50) and that catecholamine-stimulated internalization is inhibited by staurosporine (49).
In summary, our data indicate that 1) activation of endothelin ET A receptors induces the phosphorylation of ␣ 1b -ARs, 2) receptor phosphorylation takes place mainly in serine residues, 3) PKC and tyrosine kinases seem to participate in such effect, and 4) ET-1-induced ␣ 1b -AR phosphorylation alters receptor-G protein coupling. ET A and ␣ 1B -ARs coexist in many cells, where it could be anticipated that the activation of ET-1 receptors may regulate the actions of ␣ 1B -ARs.