Norepinephrine- and Phorbol Ester-induced Phosphorylation of a 1a -Adrenergic Receptors

Maximal adrenergic responses in Rat-1 fibroblasts expressing a 1a -adrenergic receptors are not blocked by activation of protein kinase C. In contrast, activation of protein kinase C induces the phosphorylation of a 1b - adrenoreceptors and blocks their actions. The effect of norepinephrine and phorbol esters on a 1a -adrenorecep- tor phosphorylation and coupling to G proteins were studied. Both stimuli lead to dose-dependent receptor phosphorylation. Interestingly, protein kinase C activation affected to a much lesser extent the actions of a 1a - adrenergic receptors than those of the a 1b subtype (norepinephrine elicited increases in calcium in whole cells and [ 35 S]GTP g S binding to membranes). Basal phosphorylation of a 1a -adrenergic receptors was much less than that observed with the a 1b subtype. The carboxyl terminus seems to be the main domain for receptor phosphorylation. Therefore, chimeric receptors, where the carboxyl-terminal tails a of the a subtype had a high basal phosphorylation and displayed a strong phosphorylation in response to norepinephrine and phorbol esters. Our results demon-strate that stimulation of a 1a -adrenergic receptor, or activation of protein kinase C, leads to a 1a -adrenergic receptor phosphorylation. a 1a are af- fected to a much lesser extent than a 1b -adrenoreceptors by protein kinase C activation. of the bovine a 1a ORF, with Not I and Eco RI sites. The initial constructions on pcDNA3 were used on polym- erase chain reactions with the convenient combination of oligonucleotides, and the resulting products were digested with Eco RI- Not I and ligated with pcDNA3 digested with the same enzymes. These new constructions were sequenced using an ABI Prism 310 genetic analyzer from Perkin-Elmer. Transient transfection of wild type and chimeric receptors into Cos-1 cells was performed using DEAE-dextran (40). 50–70% (seeded were transfected m plasmid each 10-cm Petri Experiments transfected pCH110 b -galactosid-ase. H]prazosin B Cos-1 1ab

Phosphorylation of G protein-coupled receptors is considered the initial step in the process of desensitization. The paradigm of desensitization is based mainly on exhaustive studies performed with the Gs-coupled ␤ 2 -ARs 1 (1)(2)(3)(4)(5). Accordingly, when agonist-occupied receptors activate heterotrimeric G proteins, the released G␤␥ complexes recruit soluble GRKs (particularly GRK-2) to the site of receptor activation (2). Receptors are phosphorylated by these enzymes and bind arrestin proteins that stabilize the uncoupled state of the receptors (3). Arrestins act as bridges that bind to clathrin molecules, initiating the internalization of phosphorylated receptors into vesicles, where specific phosphatases remove the phosphates and allow the dephosphorylated receptors to return to the cell surface, completing the cycle of activation-desensitization-resensitization (4,5). Second messenger-activated kinases, such as PKA (protein kinase A) and PKC, also promote receptor phosphorylation, eliciting heterologous desensitization by a parallel process that does not require receptor activation and in which arrestins do not seem to participate (1).
G protein-coupled receptor phosphorylation is associated with desensitization, particularly with homologous desensitization (1,6). Some heterologous stimuli lead to receptor phosphorylation and consequent desensitization. For example, ␣ 1b -ARs are phosphorylated and desensitized after their activation by agonists, as a result of a cross-talk with endothelin ET A receptors or activation of PKC by phorbol esters (7)(8)(9)(10)(11). However, this same receptor is phosphorylated without evidencing desensitization by a cross-talk with B2 bradykinin receptors (12) or when active GRK-5 are expressed (9). Few examples exist in which desensitization takes place with no detectable receptor phosphorylation; these include studies with the parathyroid hormone receptor (13) and the luteinizing hormone/ chorionic gonadotropin receptor (14).
Differential regulation within a family of receptors is frequently associated with differential susceptibility of members to be modified by phosphorylation. The actions of epinephrine and norepinephrine (NE), for example, are mediated through nine known adrenergic receptors. Subtypes of ␣ 2 -and ␤-ARs seem to be subject to desensitization according to their susceptibility as kinase substrates (15)(16)(17). The information on ␣ 1 -AR phosphorylation/desensitization is far less complete. ␣ 1 -ARs couple to the phosphoinositide turnover/calcium mobilization pathway (18,19) through G q/11 GTP-binding proteins (20). Activation of other signal transduction processes such as modulation of mitogen-activated protein kinase and phosphoinositide 3-kinase activities has also been reported (21,22). These receptors participate in a plethora of physiological actions, including autonomic neurotransmission and the control of cardiovascular, respiratory, genitourinary, and gastrointestinal functions. They modulate metabolic pathways, smoothmuscle contraction, water and electrolyte metabolism, and vascular tone. Long term effects of ␣ 1 -AR stimulation include expression of c-fos and c-jun proto-oncogenes and cell growth and proliferation (20 -24). The potential participation of ␣ 1 -ARs in several pathologic states such as hypertension (25,26) and cardiac (27,28) and prostatic (29,30) hypertrophy further emphasizes their importance. Therefore, knowledge of these receptors and their regulation is of interest from basic, clinical, and therapeutic perspectives.
␣ 1a -ARs seem to be responsible for phenylephrine-promoted hypertrophy in cultured neonatal rat myocytes (28). Other studies have reported the presence of ␣ 1a -AR as the predominant subtype present in the human prostate and have suggested the use of specific ␣ 1a -AR antagonists in the treatment of benign prostatic hypertrophy (29,30). In our opinion, further knowledge of the molecular events that regulate ␣ 1a -AR may contribute to the understanding of the pathogenesis of such conditions and potentially to new approaches to alleviate them. In this work, we show that ␣ 1a -ARs are phosphorylated as a consequence of agonist binding or PKC activation. Interestingly, these events affect ␣ 1a -ARs to a very limited extent, which is in marked contrast to the desensitization that has been observed with ␣ 1b -ARs.

EXPERIMENTAL PROCEDURES
Materials-(Ϫ)-Norepinephrine, lysophosphatidic acid, TPA, staurosporine, GTP␥S, GDP, and protease inhibitors were obtained from Sigma. Ro 31-8220 and bisindolylmaleimide I were from Calbiochem. Pertussis toxin was purified from vaccine concentrates as described (36,37). Phentolamine was a generous gift from Ciba-Geigy. DMEM, fetal bovine serum, trypsin, antibiotics, and other reagents used for cell culture were from Life Technologies, Inc. [  Cell Lines and Culture-Rat-1 fibroblasts transfected with bovine ␣ 1a -AR (32) or hamster ␣ 1b -AR (31), 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/ml neomycin analog 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 (18). The receptor densities in these cell lines were in the range of 1-1.5 pmol/mg membrane protein for the cell line expressing the ␣ 1a -AR subtype and 2-2.5 pmol/mg protein for the cell line expressing the ␣ 1b -AR subtype, as previously reported (18). Cos-1 cells were cultured in the same medium but in the absence of G-418 sulfate. For all of the experiments, confluent cells were serum-deprived overnight in unsupplemented DMEM. [ 3 H]Prazosin binding was performed as described previously (18).
[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. Experiments were performed with about 10 6 cells suspended in 3 ml of the above mentioned buffer supplemented with 1.2 mM CaCl 2 . Fluorescence measurements were carried out with an Aminco-Bowman Series 2 spectrometer with the excitation monochromator set at 340 and 380 nm, with a chopper interval of 0.5 s, and the emission monochromator set at 510 nm. [Ca 2ϩ ] i was calculated according to Grynkiewicz et al. (38) using the software provided by Aminco-Bowman; traces were directly exported to the graphs.
Membrane Preparation and [ 35 S]GTP␥S Binding-Cells were treated with the different agents to be tested. After stimulation, cells were washed and scraped with a rubber policeman in buffer containing: 20 mM Hepes, pH 7.5, 5 mM EDTA, 1 mM Na 3 VO 4 , 10 mM ␤-glycerophosphate, 10 mM sodium pyrophosphate, 2 mM MgCl 2 , and the previously mentioned protease inhibitors. Membranes were prepared as described (11) and resuspended in binding buffer (50 mM Tris, 0.10 M NaCl, 10 mM MgCl 2 , 1 mM EDTA, pH 7.4, containing 0.1% bovine serum albumin, 1 mM dithiothreitol, and 1 M GDP. [ 35 S]GTP␥S binding was performed as described by Wieland and Jakobs (39) with minor modifications. The binding reaction was carried out in a volume of 250 l for 30 min at 25°C in binding buffer containing 0.2 nM [ 35 S]GTP␥S. The reaction was initiated by the addition of membranes (25 g protein/ tube) and terminated by the addition of 2 ml of ice-cold 50 mM Tris, pH 7.4, containing 10 mM MgCl 2 and filtration on Whatman GF/C filters using a Brandel harvester. Filters were washed three times and dried, and radioactivity was measured with a Beckman LS 6000 SC liquid scintillation counter. Nonspecific binding, determined in the presence of 30 M cold GTP␥S, represented about 10% of total binding. Statistical analysis between comparable groups was performed using analysis of variance and Newman-Keuls analysis with the software included in the GraphPad Prism 2.01 program.
Photoaffinity Labeling of the ␣ 1a -AR and Chimeric Receptors-Membranes from Rat-1 cells expressing bovine ␣ 1a -AR or Cos-1 cells expressing chimeric receptors (see below) were obtained as described previously (11). In the dark, membranes (25 g of protein) were incubated with 6 nM [ 125 I]arylazidoprazosin essentially as described (11). After 1 h at room temperature, 1 ml of 50 mM Tris-HCl, pH 7.4, containing 150 mM NaCl, 5 mM EDTA, and 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. Open tubes were exposed to UV light for 3 min (Stratalinker, Stratagene). Membranes were centrifuged at 12,700 ϫ g for 15 min, washed, and subjected to 10% SDS-polyacrylamide gel electrophoresis containing 7 M urea under reducing conditions. The specificity of the labeling was determined using 10 M phentolamine as a competitor.

Construction of a GST-␣ 1a -AR Carboxyl-terminal Fusion Protein-
The carboxyl-terminal domain of bovine ␣ 1a -AR was amplified by polymerase chain reaction. The following oligonucleotides were used: 5Ј-AAGGATCCCAAGAGTTTAAAAAGGCC-3Ј (sense) corresponding to bases 971-988 of the bovine ␣ 1a ORF with an additional BamHI site; and 5Ј-CGGAATTCTCAGACTTCCTCCCCATTTTCACT-3Ј (antisense) corresponding to bases 1378 -1398 of the gene with an EcoRI site. The amplified fragment (432 bp) was digested with BamHI-EcoRI and ligated with pGEX2T (Pierce) digested with the same enzymes. The construction was analyzed by restriction analysis and automatic sequencing. The induced fusion protein contained the complete GST protein joined to the last 139 amino acids of the bovine ␣ 1a -AR and was used to immunize rabbits.
Immunoprecipitation of ␣ 1a -ARs and ␣ 1b -ARs-Photoaffinity-labeled membranes expressing ␣ 1a -ARs were used to standardize immunoprecipitation of the receptor. Rabbit antiserum against the described GST-␣ 1a -AR fusion protein was used. Membranes were solubilized in 1 ml of solubilization buffer (50 mM Tris-HCl, pH 7.4, 250 mM NaCl, 5 mM EGTA, 1% Triton X-100, 0.25% SDS, and the previously mentioned protease inhibitors). After 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. Tubes were incubated overnight at 4°C. Then, 40 l of Sepharose-coupled protein A, 50% (v/v) slurry were added and incubated for 1-2 h. 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% Triton X-100, 0.25% SDS, and 100 mM NaF. Washed beads were suspended in 50 l of 10% SDS and vortexed, and then 50 l of 2ϫ Laemmli sample buffer containing 5% mercaptoethanol were added. Samples were incubated for 5 min in a boiling bath and electrophoresed in 10% SDS-polyacrylamide minigels containing 7 M urea. Gels were fixed, dried, and exposed to X-OMAT x-ray film (Eastman Kodak Co.) at Ϫ80°C with an intensifying screen. Immunoprecipitation of ␣ 1b -ARs was performed with an antibody generated against the carboxyl-terminal decapeptide of these receptors (11) using the procedure described above. Analysis of immunoprecipitation was performed in a Molecular Dynamics PhosphorImager with the included ImageQuant software. Serum samples able to immunoprecipitate at least 70% of the photoaffinity-labeled receptors were used.
Phosphorylation of ␣ 1a -ARs and ␣ 1b -ARs-Rat-1 fibroblasts, expressing the bovine ␣ 1a -ARs or the hamster ␣ 1b -ARs, were cultured in culture dishes (10-cm diameter). Cells reaching confluence were serum-starved for 24 h. For the experiments using only cells expressing ␣ 1a -ARs, fibroblasts were maintained in phosphate-free DMEM for 1 h and then incubated in 3 ml of the same medium containing [ 32 P]P i (0.2 mCi/ml) for 3-5 h at 37°C. In the experiments in which the phosphorylation of ␣ 1a -and ␣ 1b -ARs were compared, the amount of radioactive phosphate was 0.1 mCi/ml. Labeled cells were stimulated with NE or TPA as indicated, and then they were washed with ice-cold phosphate-buffered saline and solubilized with 1.0 ml of ice-cold solubilization buffer containing 50 mM NaF, 1 mM Na 3 VO 4 , 10 mM ␤-glycerophosphate, 10 mM sodium pyrophosphate, 1 mM p-serine, 1 mM p-threonine, and 1 mM p-tyrosine. The plates were maintained on ice for 1 h. Then the extracts were centrifuged at 12,700 ϫ g for 15 min at 4°C, and the supernatants were immunoprecipitated as described above. At least three independent experiments were performed for each treatment. Receptor phosphorylation was detected with a Molecular Dynamics PhosphorImager and quantified with ImageQuant software. Data were within the linear range of detection of the apparatus and were plotted using Prism 2.01, GraphPad software.
Chimeric ␣ 1ab -and ␣ 1ba -ARs-cDNAs for wild type bovine ␣ 1a -AR in pBluescript and hamster ␣ 1b -AR in pSP65, containing 5Ј and 3Ј nontranslated regions, were generous gifts from Dr. Robert Lefkowitz (Duke University). Chimeric ␣ 1ab -AR and ␣ 1ba -AR were constructed by interchanging their carboxyl-terminal tails. A conserved PvuII site located in the coding sequence of the seventh transmembrane segment of both cDNAs was used for the ligation of fragments. cDNAs in the original plasmids were digested with EcoRI and PvuII, and the 5Ј cDNA of the ␣ 1a -AR was ligated with the 3Ј cDNA of the ␣ 1b -AR and vice versa; chimeric cDNAs were cloned into pcDNA3 for expression in Cos-1 cells. Fragments used for the construction of chimeric receptors were obtained from digestions in original plasmids. From the ␣ 1a -AR cDNA these fragments were: (a) a 1136-bp EcoRI-PvuII fragment containing the 5Ј nontranslated sequence and the information for the aminoterminal 319 amino acids; and (b) a 1456-bp PvuII-EcoRI fragment containing the coding sequence for the 147 amino acids from the carboxyl terminus and the 3Ј nontranslated sequence. Fragments from the ␣ 1b -AR cDNA, obtained by partial digestion, were: (a) a 1036-bp EcoRI-PvuII fragment containing the 5Ј nontranslated sequence plus the information for the amino-terminal 341 amino acids; and (b) a 1069-bp PvuII-EcoRI fragment containing the information for the carboxylterminal 174 amino acids and the 3Ј nontranslated sequence. Because inefficient expression (binding sites) was obtained with plasmids that contained 3Ј and 5Ј noncoding sequences, these regions from wild type and chimeric receptors were eliminated by polymerase chain reaction. Oligonucleotides were: 5Ј-AAGATATCGAATTCATGAATCCCGATCT-GGACACC-3Ј (sense) corresponding to bases 1-21 from the hamster ␣ 1b ORF, containing additional restriction sites for EcoRV and EcoRI; 5ЈTA-GAATTCGCGGCCGCCTAAAAGTGCCCGGGTGCCAG-3Ј (antisense) corresponding to bases 1527-1545 from the hamster ␣ 1b -AR ORF, with additional NotI and EcoRI restriction sites; 5Ј-AAGATATCGAATTCAT-GGTGTTTCTCTCCGGAAAT-3Ј (sense) corresponding to bases 1-21 from the bovine ␣ 1a -AR ORF, with EcoRV and EcoRI sites; and 5Ј-TA-GAATTCGCGGCCGCTTAGACTTCCTCCCCATTTTC-3Ј (antisense) corresponding to bases 1380 -1401 of the bovine ␣ 1a ORF, with NotI and EcoRI sites. The initial constructions on pcDNA3 were used on polymerase chain reactions with the convenient combination of oligonucleotides, and the resulting products were digested with EcoRI-NotI and ligated with pcDNA3 digested with the same enzymes. These new constructions were sequenced using an ABI Prism 310 genetic analyzer from Perkin-Elmer. Transient transfection of wild type and chimeric receptors into Cos-1 cells was performed using DEAE-dextran (40). Cells at 50 -70% confluence (seeded the previous day) were transfected with 2 g of plasmid for each 10-cm Petri dish. Experiments with transient transfected cells were performed 48 -72 h after transfection. Transfection efficiency (60 -80%) was determined in parallel dishes transfected with pCH110 and evaluated by the activity of ␤-galactosidase. The expression levels reached by each receptor and affinity were determined by [ 3 H]prazosin binding and were as follows: Receptor phosphorylation studies with Cos-1 cells were performed as described above using 0.1 mCi/ml [ 32 P]P i . The antibody against the corresponding carboxyl terminus present in each receptor was used for immunoprecipitation, i.e. anti-␣ 1a antibody for the experiments using the wild type ␣ 1a and the chimeric ␣ 1ba -ARs and the anti-␣ 1b antibody for the wild type ␣ 1b and chimeric ␣ 1ab -ARs. No cross-immunoprecipitation was observed.

Functional Consequences of PKC Activation on ␣ 1a -AR-induced [Ca 2ϩ ] i Mobilization and G Protein
Coupling-We previously reported a differential blockade in the action of ␣ 1a -, ␣ 1b -, and ␣ 1d -ARs by phorbol esters (18). To obtain further information on the effect of PKC stimulation on ␣ 1a -AR activity, agonist-induced [Ca 2ϩ ] i mobilization and G protein coupling to this receptor were studied. It can be observed (Fig. 1, upper left panel; see also Ref. 18) that a marked [Ca 2ϩ ] i mobilization was induced by 10 M NE and that this effect was not inhibited by preincubation with 1 M TPA for 15 min. Neither did shorter incubations with 1 M TPA (1 and 5 min) affect the response to 10 M NE (data not shown). A slight displacement to the right in the concentration-response curve to NE was observed in cell pretreated with 1 M TPA for 15 min (the EC 50 changed from 130 to 240 nM) (Fig. 1, lower left panel). The repetitive addition of 10 M NE resulted in gradual decreases of [Ca 2ϩ ] i , but this effect seems to be at least partially because of calcium depletion (18). The effect of preincubation with 1 M TPA on the NE-induced [Ca 2ϩ ] i was also studied in cells expressing ␣ 1b -ARs. It was observed that TPA markedly reduced the effect of NE and increased the EC 50 from 190 nM in the controls to 1790 nM in the cells treated with the phorbol ester (Fig. 1, right panels). This finding is in agreement with previous findings and confirms that the susceptibility of these receptors to desensitization by PKC activation differs greatly (18).
To test whether agonist stimulation of ␣ 1a -ARs or activation of PKC with phorbol esters affects the coupling of these receptors to GTP-binding proteins, we determined the ability of NE to stimulate the binding of [ 35 S]GTP␥S to membranes from control, NE-stimulated, or TPA-stimulated cells. Fig. 2, left panel, shows that NE was able to increase [ 35 S]GTP␥S binding in vitro to membranes from control, NE-treated, or TPAtreated fibroblasts. The maximal stimulation levels were similar, but the treatments with NE and TPA induced a shift to the right in the concentration-response curves (EC 50 ϭ 115 nM in control membranes, 550 nM in membranes from NE-treated cells, and 700 nM in membranes from TPA-treated cells). None of these treatments modified the response to 1 M lysophosphatidic acid, used as control (data not shown). Overnight treatment of the cells with pertussis toxin blocked the effect of lysophosphatidic acid but had no influence on the action of 10 M NE (data not shown).
The same treatments applied to cells expressing ␣ 1b -ARs gave very different results. As shown in Fig. 2, right panel, cell treatment with NE or TPA markedly decreased maximal in vitro NE-stimulated [ 35 S]GTP␥S binding and also altered the EC 50 values (125 nM in control membranes, 700 nM in membranes from NE-treated cells, and 1350 nM in membranes from TPA-treated cells).
Photoaffinity Labeling and Immunoprecipitation of the ␣ 1a -AR-To study potential receptor phosphorylation, we proceeded to isolate ␣ 1a -ARs by immunoprecipitation. An anti-␣ 1a -AR polyclonal rabbit antibody directed against a fusion protein of GST and the whole carboxyl-terminal tail of the ␣ 1a -AR (amino acids 327 to 466) was used for immunoprecipitation. Membranes in which the receptor was labeled by photoaffinity cross-linking with [ 125 I]arylazidoprazosin were employed to define the best experimental conditions for receptor immunoprecipitation. Photoaffinity labeling of membranes from ␣ 1a -AR-transfected Rat-1 fibroblasts resulted in a major band with a molecular mass of ϳ60 kDa (Fig. 3), which is within the expected size for ␣ 1a -ARs (41). As expected, the labeling was blocked with phentolamine, confirming the identity of the receptor. Other two faint bands with a molecular mass of ϳ45 and ϳ25 kDa were also observed in some experiments. The ϳ45-kDa band does not seem to correspond to the receptor, because it was also observed when the labeling was performed in the presence of phentolamine. However, the ϳ 25-kDa band could correspond to a degradation product of the receptor, because its labeling was competed with phentolamine ( Fig. 3).
About 70 -80% of the labeled ␣ 1a -ARs were immunopre-cipitated from [ 125 I]arylazidoprazosin-labeled membranes; the ϳ60-kDa band, representing the whole receptor, was the main labeled protein immunoprecipitated (Fig. 3). Longer time exposure of the autoradiograms revealed the presence of a band of ϳ120 kDa that was also immunoprecipitated and may represent receptor dimers or aggregates with other proteins. Preimmune serum did not immunoprecipitate ␣ 1a -ARs (Fig. 3). Rabbit antisera against synthetic decapeptides corresponding to the amino and carboxyl termini of the receptor were unable to immunoprecipitate more than 5% of labeled receptors, even when they were effective in recognizing the cognate peptide by enzyme-linked immunosorbent assay (results not shown). The antisera were less effective against the amino-terminal decapeptide, limiting the possible analysis of receptors truncated at the carboxyl-terminal tail. ␣ 1a -AR Phosphorylation-It can be observed in Fig. 4, left panel, that a labeled band with a molecular mass of ϳ60 kDa was immunoprecipitated from Rat-1 cells metabolically labeled with radioactive phosphate. Such a faint band represents basal receptor phosphorylation, and it was markedly increased in cells treated for 5 min with 10 M NE or 1 M TPA. We compared the phosphorylation of ␣ 1a -and ␣ 1b -ARs under identical labeling conditions (0.1 mCi of [ 32 P]P i for 3 h) in simultaneous experiments, and as shown in Fig. 4, right panel, the phosphorylation observed was much more intense in the ␣ 1b -AR.
NE-stimulated ␣ 1a -AR phosphorylation took place very rapidly, reaching its maximum at 1 min (Fig. 5, upper panel) and rapidly decreasing afterward; but it remained ϳ2-3-fold above basal level for 1 h. The effect was dependent on the concentration of NE used (EC 50 ϳ 500 nM, Fig. 6). The action of TPA was also concentration-dependent (EC 50 ϳ 25 nM, Fig. 6) but had a very different time course, with the effect of TPA being much slower (t1 ⁄2 ϳ 10 min) and sustained for 1 h (Fig. 5).
Effect of Protein Kinase C Inhibitors on the ␣ 1a -AR Phosphorylations Induced by Norepinephrine and TPA-Preincubation for 30 min with 300 nM staurosporine, 300 nM Ro 31-8220, or 1 M bisindolylmaleimide I markedly decreased but did not abolish the effects of NE and TPA (Fig. 7).
Photoaffinity Labeling, Immunoprecipitation, and Phosphorylation of Chimeric ␣ 1a -ARs-To compare the phosphorylation characteristics between ␣ 1a -and ␣ 1b -Ars, both subtypes were independently transfected into Cos-1 cells. Phosphorylation of these receptors was similar to that observed in Rat-1 fibroblasts (see Fig. 4), i.e. ␣ 1a -AR phosphorylation was much less intense than that of the ␣ 1b -ARs, but in both cases NE and TPA were able to increase it (Fig. 8, left panel). The phosphorylation sites of the ␣ 1b -AR have been mapped to its carboxyl-terminal tail (10,42). Therefore, to determine the influence of the carboxyl-terminal tail in the phosphorylation characteristics of ␣ 1a -and ␣ 1b -ARs, chimeric receptors with exchanged tails were constructed. Interestingly, the origin of the carboxyl-terminal tail determined the phosphorylation characteristics of these receptors. Whereas wild-type ␣ 1a -ARs showed little basal phos-phorylation, chimeric receptors in which the carboxyl-terminal tail of ␣ 1a AR was substituted by its correspondent region from ␣ 1b AR showed marked basal and stimulated phosphorylations. The opposite was detected in ␣ 1b -ARs wearing the ␣ 1a -AR carboxyl-terminal tail (Fig. 8, right panel). DISCUSSION We previously reported that the maximally stimulated activity of ␣ 1a -ARs was unaffected by phorbol ester-mediated activation of PKC in whole cells (18). In the present experiments, measuring [Ca 2ϩ ] i , we observed a barely detectable decrease in sensitivity to NE, with no change in maximal effect, in cells treated with TPA. However, a more clear shift to the right, also with no change in the maximum, was observed when ␣ 1a -ARstimulated [ 35 S]GTP␥S binding was studied. This finding is in marked contrast with what was observed in cells expressing ␣ 1b -ARs in both parameters. These data indicate that PKC activation induced a marked desensitization of the ␣ 1b subtype and affected ␣ 1a -ARs to a much lesser extent. We provide data indicating that even when activation of PKC does not lead to clear receptor desensitization in whole cells, TPA and NE induce phosphorylation of ␣ 1a -ARs. When the phosphorylation data are compared with those obtained with cells expressing the ␣ 1b -Ars, the differences are very dramatic. First, the basal phosphorylation of the receptors is markedly different; ␣ 1b -ARs basal labeling is very strong, whereas that of ␣ 1a -ARs was observed with great difficulty. In fact, we had to increase the amount of radioactive phosphate used during the labeling period by 4-fold (as compared with the amount used in our previous studies with ␣ 1b -ARs (11,12)) to observe clearly the basal labeling. The efficiencies of immunoprecipitation of photoaffinity-labeled receptors were very similar (compare the present data with those shown in Ref. 11). However, the density of receptors in the line of Rat-1 fibroblasts that express the ␣ 1b subtype (ϳ2 pmol/mg membrane protein) (11,18) is greater than that found in cells expressing the ␣ 1a subtype (1-1.5 pmol/mg membrane protein). Nevertheless, this does not seem to explain the difference in basal receptor labeling. In fact, the data with chimeric receptors strongly suggest that the properties (sequence/conformation) of carboxyl-terminal tails determine the susceptibility of these receptors to be phosphorylated. This finding is consistent with the localization of the phosphorylation site of the hamster ␣ 1b -ARs in the carboxyl terminus (42).
Second, although the effects of NE and TPA are strong when compared with the weak basal ␣ 1a -AR labeling (i.e. ϳ 5-and ϳ 10-fold increases), the intensity of the signal is clearly less than that observed in the ␣ 1b -ARs. This was also evidenced in the studies with the chimeric receptors and again emphasizes the importance of the carboxyl-terminal tails. It is also clear that the time courses of the ␣ 1a -AR phosphorylations induced by NE and TPA differ markedly. NE induced a very rapid receptor phosphorylation that also diminished very fast. In contrast, TPA induced a relatively slow effect on receptor phosphorylation that was maintained for up to 60 min. These contrasting time courses suggested that phorbol ester-sensitive PKC isoforms might not be involved in the rapid effect of NE. However, the use of PKC inhibitors strongly suggested that PKC could be involved in agonist-induced ␣ 1a -AR phosphorylation. Nevertheless, the PKC inhibitors did not block all of the agonist-mediated effect. Current ideas suggest that GRKs play a major role in homologous desensitization of several G protein-coupled receptors, including ␣ 1b -ARs (2,3,6,9). It is therefore very likely that the NE-induced ␣ 1a -AR receptor phosphorylation could involve some isoform(s) of GRK and PKC, but this possibility remains to be defined experimentally.
Most evidence associates phosphorylation of G protein-coupled receptors to desensitization. Certainly one of the most surprising findings in the present work was that, despite clearly observing TPA-induced ␣ 1a -AR phosphorylation, there was great difficulty in detecting receptor desensitization in whole cells. It is possible that receptor reserve and signal amplification steps could have contributed to such difficulty. However, even in the experiments on adrenergically mediated [ 35 S]GTP␥S binding to membranes, treating cells with NE or TPA did not decrease the maximal binding but only reduced receptor sensitivity to NE. The difference from what is observed with ␣ 1b -ARs is very clear. NE, TPA, and activation of endothelin ET A receptors elicit a strong phosphorylation of ␣ 1b -ARs that correlates with desensitization (7)(8)(9)(10)(11). However, even in the case of ␣ 1b -ARs, not all stimuli that result in phosphorylation of these receptors desensitize them. For example, activation of B2 bradykinin receptors, or coexpression of GRK-5, increases the phosphorylation of ␣ 1b -ARs without any clear negative effect on their function (9,12).
On the other hand, recent examples have indicated that G protein-coupled receptors can be desensitized without being phosphorylated (13,14). However, in this family of receptors, no information is available regarding phosphorylation events that do not result in desensitization. Current ideas suggest that in a family of receptors, diversity could often be explained in terms of differential distribution and/or differential regulation. In the case the ARs, members of the ␤ and ␣ 2 families show variation in their abilities to be desensitized, and in these cases, that variation coincides to differential phosphorylation (15)(16)(17). In the case of ␣ 1 -Ars, there are clear differences between ␣ 1a and ␣ 1b -ARs in their ability to be desensitized and susceptibility to phosphorylation.
The resistance to inactivation may have physiological significance. Persistent activity of ␣ 1a -ARs has been detected in several cellular contexts in which these receptors are endogenously expressed. An elegant study by Rokosh et al. (28) showed that the three members of the ␣ 1 -AR family are coexpressed in neonatal rat cardiac myocytes. Interestingly, chronic stimulation with phenylephrine induced down-regulation of ␣ 1b -ARs and ␣ 1d -ARs, whereas the content of ␣ 1a -ARs increased and correlated with the cellular hypertrophy elicited by phenylephrine (28). Certainly, this complex phenomenon may involve various processes. Numerous studies have revealed the participation of ␣ 1a -ARs in the etiology of benign prostatic hypertrophy (29,30,43). Some indications of vascular tone modulation by ␣ 1a -ARs have also been reported (25). However, whereas the participation of ␣ 1a -ARs in cellular events related to hypertrophy and vascular tone is unquestionable, the intracellular events that regulate these situations are still under investigation. When the different ␣ 1 -AR subtypes were expressed in Rat-1 fibroblasts, ␣ 1a -ARs showed better coupling efficiency for NE-induced [Ca 2ϩ ] i mobilization and [ 3 H]inositol phosphate production than the ␣ 1b or ␣ 1d -AR subtype (18). Recent reports indicate that even if the phosphoinositide turnover/calcium mobilization pathway is a major signal transduction mechanism employed by ␣ 1 -ARs, other signaling pathways, including activation of phosphoinositide 3-kinase and mitogen-activated protein kinase are stimulated by these receptors (22). Interestingly, differences in the intracellular processes that lead to phosphoinositide 3-kinase activation by ␣ 1a -AR and ␣ 1b -AR subtypes have been observed, and the ␣ 1d -AR subtype seems to be unable to elicit activation of this kinase (22). It is currently unknown to what extent these differences are related to susceptibility to desensitization/ phosphorylation.
In summary, our data indicate that: 1) receptor activation or stimulation of TPA-sensitive PKC isoforms induces phosphorylation of ␣ 1a -ARs, 2) NE-induced receptor phosphorylation takes place much faster than that induced by TPA, 3) ␣ 1a -ARs are much less sensitive (or more resistant) to desensitization and are phosphorylated to a lesser extent than ␣ 1b -ARs, and 4) chimeric ␣ 1a-/␣ 1b -AR receptors provided further evidence on the importance of the carboxyl-terminal tails for receptor phosphorylation.