Abstract
Exposure of D1 dopamine receptors to agonists results in rapid desensitization of the receptor-stimulated accumulation of cAMP. It is believed that agonist-induced phosphorylation of the receptor plays a critical role in the processes that underlie this phenomenon. To investigate the role of agonist-induced receptor phosphorylation, a FLAG epitope was added to the amino terminus of the rat D1dopamine receptor and this construct was stably expressed in C6 glioma cells. It was found that the D1 receptor was stoichiometrically phosphorylated under basal conditions and that its phosphorylation state was increased by 2- to 3-fold upon exposure of the cells to dopamine for 10 min. The dopamine-induced receptor phosphorylation could be blocked by D1-selective antagonists but was unaffected by inhibitors of either protein kinase A or protein kinase C. The incorporation of phosphate into the receptor was rapid but transient, despite the continued presence of dopamine. A comparison of the rates of receptor phosphorylation (t1/2 < 1 min) and dopamine-induced desensitization (t1/2 ∼7 min) revealed that receptor phosphorylation was not the rate limiting step for receptor desensitization. Upon removal of dopamine, the receptor was rapidly dephosphorylated (t1/2 ∼10 min) and this was not blocked by agents (i.e., concanavalin A or hypertonic sucrose) that inhibit D1receptor internalization. Using specific inhibitors, the phosphatase involved in D1 receptor dephosphorylation was shown not to correlate with the recently identified “G protein-coupled receptor phosphatase” (Proc Natl Acad Sci USA 92:8343–8347, 1995). These results suggest that the phosphorylated D1 receptor is processed through a novel recovery pathway and that internalization is not required for receptor dephosphorylation.
Dopamine receptors (DARs) belong to the large G protein-coupled receptor superfamily and molecular cloning studies have revealed the existence of five structurally distinct subtypes (Sibley and Monsma, 1992;Civelli et al., 1993; Neve and Neve, 1997). These can be divided into two subgroups on the basis of their amino acid sequences as well as their pharmacological and transductional properties. The first subgroup comprises the D1 and D5DARs and is termed “D1-like”. When expressed in mammalian cells, activated D1-like receptors stimulate adenylyl cyclase and raise intracellular levels of cAMP (Robinson and Caron, 1997). The second DAR subgroup includes the D2, D3, and D4 receptors and is termed “D2-like”. The D2-like DARs are coupled to the inhibition of adenylyl cyclase as well as the modulation of potassium and calcium ion channels (Huff, 1997). As with other G protein-coupled receptors, DARs are subject to a wide variety of regulatory mechanisms, which can either positively or negatively modulate their expression and functional activity (Sibley and Neve, 1997).
One of the most important forms of regulatory mechanisms that modulate signaling by G protein-coupled receptors is that of agonist-induced desensitization—defined as the tendency of receptor-mediated responses to wane over time despite continued agonist stimulation. Recent studies with β-adrenergic receptor systems have suggested a general paradigm for agonist-induced desensitization (Krupnick and Benovic, 1998;Lefkowitz, 1998). This involves phosphorylation of the receptors by a member of the G protein-coupled receptor kinase (GRK) family leading to the binding of an Arrestin-like protein ultimately resulting in uncoupling of the receptor from its cognate G protein and decreased functional activity. The binding of an Arrestin molecule also promotes internalization of the receptor through clathrin-coated pits into an endosomal compartment, where it may be dephosphorylated by a G protein-coupled receptor phosphatase (Pitcher et al., 1995) and recycled to the cell surface or degraded via a lysosomal pathway. Although this desensitization paradigm has in some instances been shown to be operative for other G protein-coupled receptors, recent studies have suggested that there may be significant exceptions and widespread variations to this general scheme (Innamorati et al., 1998; Oakley et al., 1999; Vickery and von Zastrow, 1999; Walker et al., 1999; Zhang et al., 1999).
Previous investigations of agonist-induced regulation of DARs have revealed great variability among the subtypes. For instance, agonist-induced desensitization is not always observed with the D2 DAR and, in some instances, agonist occupancy of this subtype results in increased receptor expression (Sibley and Neve, 1997). In contrast, the D1 DAR has been shown to exhibit agonist-induced refractoriness in both endogenous and recombinant/heterologous cellular expression systems (reviewed inSibley and Neve, 1997; Lewis et al., 1998; Jiang and Sibley, 1999). Recent data have also provided support for a phosphorylation pathway underlying agonist-induced desensitization of the D1 DAR. For instance, studies using intracellular inhibitors of protein kinases (Zhou et al., 1991) or elimination of phosphate acceptor sites in the receptor via site-directed mutagenesis (Jiang and Sibley, 1999) have indirectly implicated a role for phosphorylation in D1 DAR desensitization. Moreover, studies involving the expression of the D1 DAR in Sf9 (Ng et al., 1994) or HEK-293 cells (Tiberi et al., 1996), have shown that the receptor undergoes agonist-induced phosphorylation and that in the HEK-293 cells, this phosphorylation is enhanced by coexpression of GRKs. Neither of these latter studies, however, directly addressed the role of receptor phosphorylation in the agonist-induced desensitization process.
We now demonstrate that the D1 DAR, when expressed in C6 glioma cells, is stoichiometrically phosphorylated in response to agonist activation and that this phosphorylation is both rapid and transient. The transient nature of this phosphorylation was investigated further; it was found that, unlike β-adrenergic receptors, internalization is not required for D1DAR receptor dephosphorylation to take place. Our current data significantly characterizes the role that phosphorylation plays in agonist-induced regulation of D1 DAR function and further suggest that the phosphorylated D1receptor is processed through a novel recovery pathway.
Experimental Procedures
Materials.
C6 Glioma cells were purchased from American Type Culture Collection (Manassas, VA). [3H]SCH-23390 (70–71.3Ci/mmol) and [3H]cAMP (31.4 Ci/mmol), were obtained from DuPont/NEN (Boston, MA). [32P]Orthophosphate (carrier-free) was obtained from Amersham Pharmacia Biotech (Piscataway, NJ). Dopamine, forskolin, Ro-20–1724, (±)-propranolol, (±)-isoproterenol, (±)-butaclamol, and anti-D1DAR monoclonal antibody were purchased from Research Biochemicals Inc. (Natick, MA). cAMP assay kits were from Diagnostic Products Corp. (Los Angeles, CA). Cell culture media and reagents were from Life Technologies (Grand Island, NY). Fetal calf serum was purchased from Summit Biotechnology (Purchase, CO). Calcium phosphate transfection kits were from Invitrogen (San Diego, CA). Concanavalin A and MiniComplete protease inhibitor cocktail were purchased from Roche Diagnostics (Nutley, NJ). Bisindolylmaleimide-1, H-89, KT5720, CPT-cAMP, PMA, okadaic acid, and calyculin A were purchased from Calbiochem (San Diego, CA). Western-Star immunoblotting kits were supplied by Tropix (Bedford, MA). M2-affinity gel and all other reagents were purchased from Sigma (St. Louis, MO).
Cell Culture and Transfections.
C6 glioma cells were cultured in Dulbecco's modified essential medium (DMEM) supplemented with 10% fetal calf serum, 1 mM sodium pyruvate, 50 U/ml penicillin, and 50 μg/ml streptomycin. Cell cultures were grown at 37°C in 5% CO2 and 90% humidity. An amino-terminal FLAG epitope-tagged construct of the rat D1 DAR (Monsma et al., 1990) was created from pSFβ2, an expression construct containing a FLAG-tagged β2-adrenergic receptor (Guan et al., 1992). The β2-adrenergic receptor sequence was excised using NcoI and SalI and, afterNcoI/SalI digestion of the rat D1 DAR sequence, the D1 DAR was inserted in-frame 3′ to the FLAG epitope sequence to create pSFD1. The pSFD1 receptor construct (30 μg) was then cotransfected with the pMAM-neo plasmid DNA (3 μg) into C6 glioma cells using the calcium phosphate precipitation method (transfection kit; Invitrogen). In brief, cells were seeded in 150-mm2 plates and transfection was carried out after 30 to 40% confluence was achieved. DNA and 60 μl of 2 M CaCl2 were mixed in H2O in a total volume of 500 μl, which was then slowly mixed with 500 μl of HEPES-buffered saline. The reaction mixture was incubated at the room temperature for 30 min and then evenly added to the cell culture dish containing 15 ml of fresh media. After overnight incubation at 37°C, the transfection media was replaced by 25 ml of standard media. The cultures were split after another 2 to 3 days and G418 (700 μg/ml) was added to the media. G418-resistant clones were selected after 2 weeks, expanded, and further screened and characterized by a radioligand binding assay.
Radioligand Binding Assays
Cells were harvested by incubation with 5 mM EDTA in Earle's balanced salt solution (EBSS) and collected by centrifugation at 300gfor 10 min. The cells were resuspended in lysis buffer (5 mM Tris, pH 7.4 at 4°C; 5 mM MgCl2) and were disrupted using a Dounce homogenizer followed by centrifugation at 34,000g for 10 min. The resulting membrane pellet was resuspended in binding buffer (50 mM Tris, pH 7.4, 1 mM EDTA, 5 mM KCl, 1.5 mM CaCl2, 4 mM MgCl2, 120 mM NaCl). The membrane suspension (final protein concentration, 50 μg/tube) was then added to assay tubes containing [3H]SCH-23390 in a final volume of 0.5 ml. (+)-Butaclamol was added at the final concentration of 1 μM to determine nonspecific binding. The assay tubes were incubated at room temperature for 1 h and the reaction was terminated by rapid filtration through GF/C filters pretreated with 0.3% polyethylenimine. Radioactivity bound to the filters was quantified by liquid scintillation spectroscopy at a counting efficiency of 47%.
Determination of cAMP Production
C6 glioma cells were seeded into 96-well plates (50,000 cells per well) and cultured for 1 day before the experiment. To assess desensitization, the cultures were first preincubated for the indicated time periods in the absence or presence of dopamine with 0.1 mM l-ascorbic acid and 5 μM (±)-propranolol (to block endogenous β-adrenergic receptors) and in 20 mM HEPES-buffered DMEM (pH 7.4 at 37°C). Subsequently, the cells were washed four times with 200 μl of EBSS (37°C) and were further incubated with various concentrations of dopamine in a total volume of 100 μl at 37°C for 15 min in the presence of 30 μM Ro-20–1724, 100 μM l-ascorbic acid, and 5 μM (±)-propranolol. The reaction was terminated by discarding the supernatant and adding 100 μl of 3% perchloric acid per well. After incubating on ice for 30 min, 40 μl of 15% KHCO3was added to the wells and the plates were further incubated for 10 min. The plates were then centrifuged for 10 min at 1,300g and 50 μl of the supernatant from each well was subsequently transferred to a 1.2-ml tube containing 250 μl of reaction mixture (150 μl of Tris-EDTA buffer, 50 μl of cAMP binding protein, and 50 μl of [3H]cAMP). After incubation at 4°C overnight, 250 μl of charcoal-dextran mix (1%) was added to each tube followed by incubation at 4°C for 15 min then centrifugation for 15 min at 1,300g. Radioactivity in the supernatant from each tube was quantified by liquid scintillation spectroscopy at a counting efficiency of 47%. cAMP concentrations were calculated using a standard curve according to the protocol of the assay kit.
Whole-Cell Phosphorylation Assays.
One day before the experiment, cells were seeded at 1 × 106per well of a 6-well plate and cultured overnight. Cells were then washed with EBSS and incubated for 1 h in phosphate-free DMEM. Media was then removed and replaced with 2 ml of fresh media supplemented with 200 μCi/ml [32P]H3PO4. After 90 min at 37°C, the cells were then challenged with dopamine or other agents in media supplemented with 100 μM l-ascorbic acid for the times and concentrations described in the text. Cells were then transferred to ice, washed twice with ice-cold EBSS, and solubilized for 1 h at 4°C in 1 ml of solubilization buffer (50 mM HEPES, 1 mM EDTA, 10% glycerol, 1% Triton X-100, pH 7.4 at 4°C) + 150 mM NaCl supplemented with MiniComplete protease cocktail, 0.1 mM phenylmethylsulfonyl fluoride and phosphatase inhibitors (5 mM sodium pyrophosphate, 50 mM NaF). The samples were cleared by centrifugation in a Microfuge and the protein concentration was determined by bicinchoninic acid protein assay (Pierce, Rockford, IL). Equal amounts of protein were then transferred to fresh tubes with 50 μl of washed M2-affinity gel and incubated overnight with mixing at 4°C. The samples were then washed once with solubilization buffer and 500 mM NaCl, once with solubilization buffer and 150 mM NaCl, and once with Tris-EDTA, pH 7.4 at 4°C. Samples were then incubated in 2× SDS-PAGE loading buffer for 1 h at 37°C before being resolved by 8% SDS-PAGE. The gels were dried and subjected to autoradiography. To study receptor dephosphorylation, after challenge with 10 μM dopamine for 10 min, cells were washed twice with EBSS then fresh media were added and incubated at 37°C for various times as indicated. When antagonists and other agents were used, cells were preincubated with the appropriate agent for the stated times before challenge with dopamine. The agent concentrations were then maintained until the samples were processed for immunoprecipitation. All assays included cells challenged with vehicle as an internal control.
Receptor Stoichiometry Measurements.
The amount of receptor protein used in each immunoprecipitation was determined by radioligand binding and bicinchoninic acid protein assay. The specific activity of the receptor protein was calculated assuming the immunoprecipitation was 100% efficient. After developing the autoradiograph, the appropriate region of the gel was excised and radioactivity was measured by liquid scintillation counting. The specific activity of [γ-32P]ATP was determined by the method ofHawkins et al. (1983) as described previously (Carter, 1995). ATP concentrations were calculated from parallel unlabeled samples using a luciferin-luciferase assay (Sigma).
Western Blotting
C6 glioma cell membranes, or samples that had been subjected to immunoprecipitation, were resolved by 8% SDS-PAGE and transferred to nitrocellulose. Blots were blocked in 0.1% Tween PBS (TPBS) + 5% nonfat dried milk for 1 h at room temperature. The blots were then incubated overnight at 4°C with TPBS and 5% nonfat milk plus anti-D1 DAR monoclonal antibody (1:500). Blots were washed in TPBS and incubated with secondary antibody (goat-anti rat IgG-alkaline phosphatase conjugated; 1:10,000 dilution) for 1 h at room temperature. Blots were washed and visualized with the use of WesternStar chemiluminescence (Tropix, Bedford, MA).
Immunohistochemistry and Confocal Microscopy.
The C6-FD1 cells were grown on cover slips for 1 to 2 days before antibody staining. The cultures were treated with or without dopamine for 30 min and then washed three times with 1× ice-cold TBS, pH 7.4. Before staining, the cells were treated with blocking solution (3% normal goat serum, 2% horse serum, and 1% BSA). The primary antibody, anti-FLAG M2 monoclonal antibody, was diluted with 1× TBS containing 3% BSA. Live cells were incubated with the M2 antibody (30 μg/ml) at 4°C for 1 to 2 h, washed three times with ice-cold TBS, and fixed with 2% paraformaldehyde for 30 min. After three washes with TBS, the fixed cells were stained with Cy3 conjugated donkey anti-mouse IgG (1:200 dilution in TBS) for 45 min at room temperature, followed by a final three washes of TBS. The coverslips were mounted with Vectashield mounting medium and subjected to confocal microscopy. All fluorescent images were viewed under a Zeiss LSM 510 inverted confocal microscope (Zeiss, Oberkochen, Germany)
Data Analysis.
All binding assays were routinely performed in triplicate and were repeated three to four times. cAMP experiments were performed in duplicate and were repeated three to four times. Estimations of the radioligand binding parametersK D and B max, as well as the EC50 values for dopamine-stimulation of cAMP production, were calculated using the GraphPad Prizm curve-fitting program. The curves presented throughout this manuscript, representing the best fits to the data, were generated using this software program as well. The relative intensities of phosphorylated bands were determined by scanning the autoradiographs and analyzing using the software program NIH Image.
Results
Immunoprecipitation of Phosphorylated D1 Dopamine Receptors.
To characterize phosphorylation of the D1 DAR in intact cells, we engineered an epitope-tagged construct of the receptor by placing the FLAG peptide sequence (Guan et al. 1992) at its amino terminus. This allows for purification of the receptor from cellular homogenates via immunoprecipitation using antibodies directed to the FLAG sequence. We stably transfected this construct in C6 glioma cells and selected a clonal line (C6-FD1) which expressed about 1.3 pmol/mg of protein of receptor binding activity. Preliminary characterization of these cells indicated that the FLAG-tagged D1 receptor exhibited normal ligand binding characteristics and stimulated cAMP levels in a fashion identical to that of wild-type receptors (data not shown).
Before attempting immunoprecipitation of metabolically labeled D1 receptors, we first determined the apparent molecular mass of the D1 DAR using a monoclonal antibody directed to the D1 receptor (Levey et al., 1993) to immunoblot membranes from transfected C6 glioma cells (Fig. 1A). A broad band was observed in membranes from C6-FD1 cells that had an apparent molecular mass of 55 to 60 kDa (Fig. 1A, lane 2). This band was not present in membranes from untransfected cells (Fig. 1A, lane 1). The D1 DAR has a predicted molecular mass of about 49 kDa and contains two consensus sites for N-linked glycosylation (Monsma et al., 1990). Thus, the apparent larger molecular mass is consistent with a glycosylated receptor protein.
We next examined immunoprecipitation of the D1DAR from the transfected C6 cells using anti-FLAG antisera conjugated to an affinity gel as described under Experimental Procedures. The precipitated samples were then analyzed by SDS-PAGE and immunoblotted using the monoclonal anti-D1DAR antibody. A broad band with an apparent molecular mass of 55 to 60 kDa was observed (Fig. 1B, lane 1). This was similar to that observed in membranes from C6-FD1 cells (Fig. 1A, lane 2). Again, this band was specific to the transfected cells. As an aside, we found that, whereas the anti-FLAG antibodies were useful for immunopreciptation, they were not useful for immunoblotting. Conversely, whereas the D1 DAR monoclonal antibody was useful for immunoblotting, it was not useful for immunopreciptation. Consequently, all further immunoprecipitations were performed with the affinity gel-conjugated anti-FLAG antibodies. When C6-FD1cells were pretreated with 10 μM dopamine for 10 min before immunoprecipitation, a notable decrease in the mobility of the D1 DAR was observed (Fig. 1B, lane 2) and the receptor exhibited an apparent molecular mass of 60 to 66 kDa. Although this mobility shift was not examined intensively in this study, it should be noted that a decrease in SDS-PAGE mobility is commonly observed with phosphorylated proteins.
To directly examine the phosphorylation status of the D1 DAR, the C6 glioma cells were metabolically labeled with [32P]H3PO2to isotopically label the intracellular ATP pool followed by immunopurification of the receptor. Figure2A shows an autoradiogram of immunoprecipitates from metabolically labeled untransfected and transfected cells. As can be seen in Fig. 2A, lane 2, there is a phosphorylated protein with a molecular mass of about 55 to 60 kDa, which was not observed in untransfected cells (Fig. 2A, lane 1). This protein band corresponds with the D1 DAR, which was identified via the immunoblotting experiments shown in Fig. 1. When the C6-FD1 cells were pretreated with 10 μM dopamine for 10 min, the phosphorylation state of the D1 DAR was increased by 2- to 3-fold and the mobility of the receptor was decreased (Fig. 2A, lane 3). Figure 2B shows the results of densitometric scanning of multiple experiments of the type shown in Fig. 2A. As can be seen, treatment of the cells with 10 μM dopamine for 10 min results in a 2.5-fold increase in the phosphorylation of the D1 DAR.
To assess the stoichiometry of D1 DAR phosphorylation, we determined the specific activity of the cellular [γ-32P]ATP by the method of Hawkins et al. (1983) as described previously (Carter, 1995). Using this approach, we estimated a stoichiometry of 1 mol of phosphate/mol of receptor in the basal state with an increase to 2.7 mol of phosphate/mol of receptor after challenging the cells with 10 μM dopamine for 10 min. These values are likely to be somewhat under-estimated because our calculation assumes complete immunoprecipitation of the solubilized receptors, which is probably not the case. Nonetheless, this analysis establishes that the D1 receptor isstoichiometrically phosphorylated in the C6 glioma cells and that its phosphate content is increased by 2- to 3-fold upon agonist activation.
The pharmacology of the D1 DAR phosphorylation process was next determined (Fig. 3). When the β-adrenergic agonist isoproterenol was substituted for dopamine, no increase in D1 receptor phosphorylation was observed. It should be noted that C6 cells express functional β-adrenergic receptors. This demonstrates that receptor occupancy by dopamine is required to observe stimulation of D1 DAR phosphorylation. This was demonstrated further in that the dopamine-stimulated increase in receptor phosphorylation could be inhibited by the D1-selective antagonist SCH-23390 and by the nonselective dopamine antagonist (+)-butaclamol. The inactive isomer (−)-butaclamol did not inhibit dopamine-stimulated receptor phosphorylation. The agonist-induced D1 DAR phosphorylation in the C6 cells thus seems to be homologous in nature and pharmacologically specific.
Characterization of Protein Kinases Involved in D1Dopamine Receptor Phosphorylation.
We next examined the contribution of the second messenger protein kinases, PKA and PKC, in dopamine-stimulated phosphorylation of the D1DAR. C6-FD1 cells were incubated with various kinase inhibitors or activators before dopamine challenge as described under Experimental Procedures. Neither PKA nor PKC seem to be largely involved in the dopamine-stimulated increase in receptor phosphorylation (Fig. 4). This was demonstrated by the relative lack of effect of the PKC inhibitor bisindolylmaleimide-1 or the PKA inhibitors H-89 and KT5720 on dopamine-stimulated D1 DAR phosphorylation (Fig.4). Furthermore, direct activation of PKA using forskolin or CPT-cAMP, both in the presence of the phosphodiesterase inhibitor Ro-20–1724, does not stimulate D1 DAR phosphorylation (Fig.4). Interestingly, it was observed that direct activation of PKC using the phorbol ester PMA resulted in a small increase in the phosphorylation state of the D1 DAR (Fig. 4). This response could be blocked with the inclusion of bisindolylmaleimide-1, but not when H-89 was used, further indicating a PKC-mediated response (Fig. 4). Overall, these results suggest that the dopamine-stimulated receptor phosphorylation probably occurs for the most part via a GRK-mediated pathway.
Relationship of Receptor Phosphorylation to Dopamine-Induced Desensitization.
To investigate the role that receptor phosphorylation plays in agonist-induced desensitization, we initially examined the dose-response relationships for these processes (Fig.5). The dopamine-stimulated receptor phosphorylation was found to be dose-dependent, as demonstrated using C6 cells that were challenged with increasing concentrations of dopamine for 10 min (Fig. 5, A and B). It was observed that 10 μM dopamine was a maximally effective concentration and that the EC50 for this response was 200 to 300 nM. The desensitization of the cAMP accumulation response by dopamine was investigated by preincubating the cells with increasing concentrations of dopamine for 10 min. The cells were then extensively washed and rechallenged with a single dose of dopamine (10 μM). It was observed that dopamine exhibited an EC50 value of 150 to 200 nM for inducing desensitization and that 10 μM was a maximally effective dose (Fig. 5C). Thus, the agonist-induced receptor phosphorylation and desensitization responses exhibit similar dose-response relationships.
We next wished to compare the time courses of dopamine-stimulated receptor phosphorylation and desensitization. The time course for receptor phosphorylation is shown in Fig.6. It was observed that receptor phosphorylation was rapid reaching a peak within 1 to 3 min after agonist exposure. The extent of receptor phosphorylation was then seen to decrease until, after about 60 min, it approached the background level observed in the absence of dopamine (basal). In contrast, the time course of agonist-induced desensitization of the cAMP response was slower and was characterized by both a decrease in maximum response and an increase in the EC50 value for dopamine (Fig.7A). More detailed time course data, which examined desensitization of the maximum response, showed that the rate of desensitization exhibited a t1/2 of about 7 min with maximum desensitization not being achieved until about 30 min (Fig. 7B). The agonist-induced receptor phosphorylation thus seems to significantly precede the desensitization response (compare Figs. 6and 7).
Characterization of Receptor Dephosphorylation and Resensitization.
We next wished to examine the rates of receptor dephosphorylation and resensitization of the cAMP response, especially given that the agonist-induced receptor phosphorylation seemed transient in nature (Fig. 6). In an initial series of experiments, whole-cell phosphorylation assays were performed on C6-FD1 cells that had been incubated with 10 μM dopamine for 10 min. Cells were then extensively washed with EBSS and further incubated in the absence of dopamine in fresh medium for various periods of time before being subjected to immunoprecipitation. It was observed that, after the removal of dopamine, the agonist-stimulated receptor phosphorylation was rapidly reversed with a t1/2 of 5 to 10 min and was back to basal levels within 30 min (Fig. 8A and B). A similar experimental design was used to investigate the resensitization of the D1 DAR that had previously been desensitized with dopamine. C6-FD1 cells were challenged with 10 μM dopamine for 60 min then washed extensively and incubated in fresh buffer without dopamine for increasing periods. After this, cAMP accumulation assays were performed using a 10 μM test dose of dopamine (Fig. 8C). Interestingly, in contrast to the rapid rate of receptor dephosphorylation, resensitization of the D1 DAR-mediated cAMP response occurred slowly, not returning to control levels until after 5 to 6 h in culture (Fig. 8C).
Given the results in Fig. 8, we decided to investigate the process of receptor dephosphorylation in greater detail. Our initial hypothesis was that the phosphorylated D1 DAR might be processed in a manner similar to that of the β2-adrenergic receptor. Detailed studies have shown that, after receptor phosphorylation, the β2-adrenergic receptor undergoes rapid sequestration via β-Arrestin- and clathrin-mediated endocytosis followed by dephosphorylation within an acidified vesicular compartment by a novel protein phosphatase termed “GRP” for G protein-coupled receptor phosphatase (Pitcher et al., 1995). After dephosphorylation, the receptor can recycle to the plasma membrane and resensitization is achieved. It has been found that if β2-adrenergic receptor endocytosis is blocked, then receptor dephosphorylation is likewise inhibited and receptor resensitization does not occur (Yu et al., 1993; Pippig et al., 1995;Krueger et al., 1997; Zhang et al., 1997).
Our first approach was to examine the effect of inhibiting internalization of the receptor on the dephosphorylation process. Previously, treatment of cells with the plant lectin concanavalin A or hypertonic sucrose was shown to inhibit D1receptor internalization (Ng et al., 1995; Ariano et al., 1997); however, this has not yet been demonstrated using C6 glioma cells. We thus performed the experiment shown in Fig.9. In this study, we used antibodies directed against the FLAG epitope on the D1 DAR to immunostain intact C6-FD1 cells and visualize the receptor using confocal fluorescence microscopy. Because the FLAG epitope is located on an extracellular region of the receptor and the cells are intact and not permeabilized, the observed fluorescence is extracellular in nature. In the absence of dopamine, intense fluorescence is observed around the cellular exterior suggesting that the receptor is predominantly localized to the plasma membrane (Fig.9). Under basal conditions, concanavalin A or hypertonic sucrose treatments do not seem to affect the D1 DAR distribution (Fig. 9). Under control conditions, treatment of the cells with 50 μM dopamine for 30 min induces a nearly complete loss of cellular fluorescence, suggesting that internalization of the receptor has taken place (Fig. 9). In contrast, the extracellular fluorescence is retained after dopamine treatment if the cells have been co-treated with either concanavalin A or hypertonic sucrose (Fig. 9). In the latter cases, however, the fluorescence is more punctate in appearance, suggesting an agonist-induced clustering of the receptors, but no internalization.
Based on the results in Fig. 9, we next treated the cells with concanavalin A or hypertonic sucrose to see what effect this would have on the dephosphorylation of the receptor (Fig.10). The design for these experiments was similar to that used in Fig. 8 except that a fixed recovery period of 20 min was used after removal of dopamine. As can be seen, after a 20-min recovery period, the level of receptor phosphorylation had returned to near basal levels in the control treatment group, similar to that previously observed in Fig. 8. Surprisingly, pretreatment of the cells with hypertonic sucrose or concanavalin A did not have any effect on the D1 receptor dephosphorylation after agonist removal (Fig. 10). These results suggest that receptor internalization is not required for D1 receptor dephosphorylation to occur and that the phosphatase responsible for this action may reside in a different cellular compartment than the previously identified G protein-coupled receptor phosphatase.
To further characterize the dephosphorylation pathway of the D1 receptor, we examined the effects of two protein phosphatase inhibitors, okadaic acid and calyculin A, which have been shown to inhibit the G protein-coupled receptor phosphatase and/or resensitization of the β2-adrenergic receptor (Pitcher et al., 1995; Pippig et al., 1995; Krueger et al. 1997). As indicated in Fig. 11, however, neither of these inhibitors blocked D1receptor dephosphorylation and, after 20 min of recovery, all treatment groups exhibited receptor phosphorylation states that were near basal levels. These results, although preliminary in nature, would seem to suggest that the previously identified “G protein-coupled receptor phosphatase” (Pitcher et al., 1995) is not involved in D1 receptor dephosphorylation.
Discussion
In this report, we have investigated the role that phosphorylation plays in regulating D1 DAR function. Previous investigations have provided support for a phosphorylation pathway underlying agonist-induced desensitization of the D1 DAR. Studies using intracellular inhibitors of protein kinases (Zhou et al. 1991) or elimination of phosphate acceptor sites in the receptor protein (Jiang and Sibley, 1999) have implicated a role for phosphorylation in D1 DAR desensitization. Also, studies involving the expression of the D1 DAR in Sf9 (Ng et al., 1994) or HEK-293 cells (Tiberi et al., 1996), have shown that the D1receptor is phosphorylated and that, in HEK-293 cells, this phosphorylation is enhanced by coexpression of GRKs. Here we demonstrate that the D1 DAR, when expressed in C6 glioma cells, is phosphorylated in response to agonist activation and that this phosphorylation is dose-dependent, rapid and transient. Our estimation of phosphate/receptor stoichiometries suggest that, under basal conditions, there is at least 1 mol of phosphate incorporated into the D1 DAR and that the receptor phosphate content increases by up to 3-fold upon maximal dopamine activation. Although the exact number and location of phosphate acceptor sites on the D1 DAR is unclear at present, the observation that the agonist-induced phosphorylation is stoichiometric in nature lends support to the hypothesis that this is a physiologically relevant process.
Our initial characterization of D1 DAR phosphorylation in C6 glioma cells revealed this to be strictly dependent upon agonist occupancy/activation of the receptor. The increase in receptor phosphorylation was not mimicked by antagonist occupancy, although D1-selective antagonists could block the phosphorylation increase observed with dopamine treatment. Moreover, treatment of the C6 cells with a β-adrenergic agonist had no effect on D1 DAR phosphorylation, despite the fact that this treatment results in increased intracellular levels of cAMP and activation of PKA. This suggests an absence of heterologous “cross talk” in the C6 cells and that the agonist-induced D1 DAR phosphorylation is homologous in nature, at least for relatively short periods of agonist exposure.
Our investigation of the biochemical pathway of D1 DAR phosphorylation indicates that this process is not predominantly mediated by either PKA or PKC. This was suggested by the relative lack of effect of intracellular activators of PKA or PKC on D1 DAR phosphorylation or by PKA/PKC inhibitors on agonist-induced receptor phosphorylation. It should be noted, however, that cellular treatment with PMA, which directly activates PKC, resulted in a slight (∼50%) but consistent increase in D1 DAR phosphorylation. Interestingly, we have previously observed that PMA treatment of C6 glioma cells results in partial desensitization of the D1 DAR-mediated cAMP response (Jiang and Sibley, 1997). This is unlikely to be related to the D1agonist-induced effect; however, because D1 DAR stimulation is not linked to activation of the PKC system and the PKC inhibitor bisindolylmaleimide-1 had no effect on agonist-induced receptor phosphorylation. Although the default hypothesis is that the observed D1 DAR phosphorylation is predominantly mediated by one or more GRKs, additional experimentation will be required to directly demonstrate GRK involvement in this biochemical process.
The lack of effect of PKA activators and inhibitors on the D1 DAR phosphorylation was especially interesting given that, in other cellular systems, intracellular activation of PKA has been shown to result in desensitization of D1mediated responses (Bates et al., 1991; Black et al., 1994). Also, Zhou et al. (1991) have shown that inhibition of PKA could partially attenuate agonist-induced desensitization of the D1 DAR. In addition, we have recently observed that mutagenesis of Thr-268, which resides within a PKA recognition motif in the rat D1 DAR, results in an attenuation of the rate, but not extent, of agonist-induced receptor desensitization (Jiang and Sibley, 1999). These effects of PKA, however, are not universally observed; a recent study using monkey D1 DAR expressed in C6 glioma cells reported no effect of treatment with cAMP analogs on D1 DAR responsiveness (Lewis et al., 1998). One possible explanation for these disparate results is that PKA may not directly phosphorylate the D1 DAR and that the observed effects of cAMP analogs (Bates et al., 1991; Black et al., 1994) are caused by phosphorylation of down-stream components. In this scenario, Thr-268 may be a substrate for a kinase other than PKA. Another possibility is that PKA does directly phosphorylate the D1 DAR (on Thr-268 or elsewhere) but that agonist occupancy of the receptor is absolutely required and this is not mimicked by cAMP analogs. One would then predict that treatment of the cells with PKA inhibitors should diminish the dopamine-induced phosphorylation, but this was, in fact, not observed (Fig. 4). However, if the number of residues phosphorylated by GRKs greatly exceeds those/that phosphorylated by PKA, then it might be difficult to detect a small reduction of phosphate content in the receptor through PKA inhibition. Obviously, final resolution of this issue will require the complete delineation of all the phosphorylation sites on the D1 DAR using mutagenesis and other approaches.
To further investigate the role that phosphorylation plays in agonist-induced desensitization, we compared the dose-response and time course relationships for these two regulatory events. It was found that the potency (EC50 ∼200 nM) of dopamine for inducing receptor phosphorylation was identical with that for inducing desensitization suggesting that these events are tightly linked. Interestingly, the potency of dopamine for inducing phosphorylation/desensitization was about 10-fold less than that for stimulating cAMP accumulation in the C6 glioma cells (EC50 = ∼20 nM; see Fig. 7). This suggests that the phosphorylation/desensitization processes are less correlated with cAMP generation and are more highly correlated with the degree of receptor occupancy by agonists, which agrees well with previous findings on GRK-mediated phosphorylation reactions (Krupnick and Benovic, 1998; Lefkowitz, 1998).
A comparison of the rate of receptor phosphorylation with the rate of desensitization clearly indicates that phosphorylation of the D1 DAR precedes desensitization. In our experiments, agonist-induced receptor phosphorylation was near maximal at the earliest measurable time point (1 min) whereas the t1/2 for attenuation of the D1 DAR mediated cAMP response was approximately 7 to 10 min. Clearly, phosphorylation of the D1 DAR is not the rate-limiting step in its desensitization. Presumably, subsequent events, such as the binding of an Arrestin protein to the phosphorylated receptor and/or the physical translocation of the receptor from its cognate G protein must be rate limiting. Surprisingly, the dopamine-induced receptor phosphorylation was transient in nature despite the continued presence of the agonist. Subsequent to maximal receptor phosphorylation (at 1–3 min of agonist exposure), the phosphate content of the D1 DAR gradually declines to basal levels by 60 min with a t1/2 of ∼20 to 30 min. We investigated this in more detail using wash-out experiments and found that, after maximal phosphorylation and agonist removal, the phosphate content of the receptor returned to basal levels with a t1/2 of about 10 min. Interestingly, the time course for receptordephosphorylation seems to be similar to, or lag slightly behind, that for the on-set of desensitization. Surprisingly, after agonist washout, the dopamine-stimulated cAMP response recovers slowly, not attaining control levels of activity until after several hours of washout. This suggests that, despite its rapid dephosphorylation, the D1 DAR does not recycle and recover in a rapid, simple fashion in C6 glioma cells.
Recent studies have shown that, after receptor phosphorylation, the β2-adrenergic receptor undergoes rapid endocytosis followed by dephosphorylation within an acidified vesicular compartment by a GRP that exhibits characteristics of the PP2A family of protein phosphatases (Pitcher et al., 1995). Given these observations, we predicted that inhibition of receptor internalization should prevent the rapid dephosphorylation of the D1 DAR. In contrast, exposure of the cells to the plant lectin concanavalin A or hypertonic sucrose, treatments which were shown to block D1 DAR internalization, did not block the rapid dephosphorylation of the D1DAR. This observation indicates that the D1 DAR need not be internalized to be dephosphorylated. To further investigate the nature of the dephosphorylation reaction, we tested the effects of okadaic acid, a potent inhibitor of the GRP, and the phosphatase inhibitor calyculin A, both of which have been shown to inhibit the dephosphorylation of the β2-adrenergic receptor (Pitcher et al., 1995; Pippig et al., 1995; Krueger et al., 1997). Neither of these agents had any effect on the D1DAR dephosphorylation process suggesting the existence of a novel pathway which does not involve the recently identified GRP.
Based on all of these observations, we would like to propose the following hypothetical pathway for D1 DAR phosphorylation/dephosphorylation in C6 glioma cells. Agonist occupancy of the D1 DAR results in its rapid (seconds to minutes) phosphorylation by one or more GRKs. Phosphorylation of the D1 DAR by PKA may also occur, although this is not a predominant reaction. After receptor phosphorylation, the D1 DAR associates with an Arrestin protein that serves to uncouple the receptor and target it for internalization. The Arrestin-binding/receptor-translocation processes take place within minutes and represent the rate-limiting steps for D1 DAR desensitization. Thus, although D1 receptor phosphorylation is necessary, it is not sufficient for desensitization to take place. Subsequently, through a novel pathway or mechanism, the D1 DAR is dephosphorylated during the translocation process before its removal from the cell surface and entry into internal endosomal compartments. The phosphatase responsible for this reaction is not the recently identified GRP but may be a known or novel protein phosphatase and is most likely associated with the plasma membrane as opposed to intracellular domains. Once internalized, the D1DAR may recycle to the cell surface, but this is not a rapid event and represents the rate-limiting step for the receptor resensitization process. Although not investigated in the present study, prolonged (hr) agonist exposure may result in an additional down-regulation of D1 DAR expression.
Recently, in an elegant series of studies (Zhang et al., 1997 and 1999;Oakley et al., 1999; Walker et al., 1999; Zhang et al., 1997 and 1999), Caron and colleagues have suggested that, for several G protein-coupled receptors, the rate-limiting step for receptor resensitization is their dephosphorylation. Moreover, it was suggested that the rate of receptor dephosphorylation is dependent on the rate of dissociation of an Arrestin protein from the phosphorylated receptor (Oakley et al., 1999). Thus, the β2-adrenergic receptor, which resensitizes quickly, exhibits rapid dissociation of β-Arrestin such that β-Arrestin is not internalized along with the receptor in HEK-293 cells (Oakley et al., 1999). Conversely, the V2 vasopressin receptor, which re-sensitizes slowly, exhibits delayed dissociation of β-Arrestin such that the receptor-β-Arrestin complex is cointernalized (Oakley et al., 1999). Interestingly, the D1 DAR was shown to exhibit rapid β-Arrestin dissociation; however, neither the phosphorylation status of the D1 DAR nor its rate of resensitization was examined in this previous study (Zhang et al., 1999). Obviously, given our current data, the rate of D1 DAR re-sensitization in C6 cells seems to be more correlated with its rate of endosomal trafficking and recycling than with its rate of dephosphorylation. This suggests that the model of Caron and colleagues (Oakley et al., 1999) may not be applicable to all G protein-coupled receptors or to all cell types. An important issue that we are currently attempting to address is to examine directly the intracellular trafficking of the D1 DAR in response to agonist treatment of the C6 cells to follow the receptor's fate and re-cycling once it undergoes internalization.
A review of the literature indicates that several other G protein-coupled receptors exhibit similar phenomena as we have observed here for the D1 DAR. For instance, both the bradykinin B2 receptor expressed in human fibroblasts (Blaukat et al., 1996) and the vasopressin V1a receptor expressed in HEK-293 cells (Innamorati et al., 1998) exhibit rapid but transient phosphorylation in response to agonist treatment. Interestingly, as we have observed here with the D1 DAR, okadaic acid was reported not to affect the dephosphorylation of the V1a vasopressin receptor (Innamorati et al., 1998). Similarly, the CCK receptor has been reported to undergo transient agonist-induced phosphorylation as well as a complex internalization process in acinar or CHO cells (Klueppelberg et al., 1991; Roettger et al., 1995). It was further suggested that the CCK receptor may be dephosphorylated in a smooth vesicular compartment adjacent to the plasma membrane. All of these observations suggest that our current findings with the D1 DAR may have wide-spread applicability to many G protein-coupled receptors and different cell types.
Footnotes
- Received July 14, 2000.
- Accepted October 27, 2000.
-
Send reprint requests to: Dr. David R. Sibley, Experimental Therapeutics Branch, NINDS/NIH, Building 10, Room 5C108, 10 Center Drive, MSC 1406, Bethesda, MD 20892-1406. E-mail: sibley{at}helix.nih.gov
Abbreviations
- DAR
- dopamine receptor
- GRK
- G protein-coupled receptor kinase
- HEK
- human embryonic kidney
- PMA
- phorbol 12-myristate 13-acetate
- CPT-cAMP
- 8-(4-chlorophenylthio)-cAMP
- DMEM
- Dulbecco's modified essential medium
- EBSS
- Earle's balanced salt solution
- PAGE
- polyacrylamide gel electrophoresis
- TPBS
- Tween PBS
- BSA
- bovine serum albumin
- TBS
- Tris-buffered saline
- PKA
- protein kinase A
- PKC
- protein kinase C
- GRP
- G protein-coupled receptor phosphatase
- DA
- dopamine
- The American Society for Pharmacology and Experimental Therapeutics