Palmitoylation of the 5-Hydroxytryptamine4a Receptor Regulates Receptor Phosphorylation, Desensitization, and -Arrestin-Mediated Endocytosis
Abteilung Neuround Sinnesphysiologie, Physiologisches Institut (E.P, M.H., K.G., D.W.R.)
Abteilung Immunologie (M.O.), Universitt Gttingen, Gttingen, Germany
Institute of Functional Genomics, Montpellier, France (A.D., F.G., G.B.)
Abstract
The mouse 5-hydroxytryptamine4a (5-HT4a) receptor is an unusual member of the G protein-coupled receptor superfamily because it possesses two separate carboxyl-terminal palmitoylation sites, which may allow the receptor to adopt different conformations in an agonist-dependent manner (J Biol Chem 277:2534eC2546, 2002). By targeted mutation of the proximal (Cys-328/329) or distal (Cys-386) palmitoylation sites, or a combination of both, we generated 5-HT4a receptor variants with distinct functional characteristics. In this study, we showed that upon 5-HT stimulation, the 5-HT4a receptor undergoes rapid (t 2 min) and dose-dependent (EC50 180 nM) phosphorylation on serine residues by a staurosporine-insensitive receptor kinase. Overexpression of GRK2 significantly reduced the receptor-promoted cAMP formation. The Cys328/329-Ser mutant, which is constitutively active in the absence of ligand, exhibited enhanced receptor phosphorylation under both basal and agonist-stimulated conditions and was more effectively desensitized and internalized via a -arrestin-2 mediated pathway compared with the wild-type 5-HT4a. In contrast, G protein activation, phosphorylation, desensitization, and internalization of the other palmitoylation-deficient receptor mutants were affected differently. These findings suggest that palmitoylation plays an important role in modulating 5-HT4a receptor functions and that G protein activation, phosphorylation, desensitization, and internalization depend on the different receptor conformations.
5-Hydroxytryptamine (5-HT or serotonin) is an important neurotransmitter and a local hormone that acts in the central nervous system as well as in various peripheral organs. It regulates cellular activities by binding to specific membrane-bound 5-HT receptors. Most of these proteins belong to the superfamily of G protein-coupled receptors (GPCR). The 5-HT4 receptor was shown to mediate several important physiological effects of 5-HT, including memory facilitation and the regulation of intestinal motility (Eglen et al., 1995; Bockaert et al., 1997). Mice deficient in 5-HT4 receptors displayed abnormal stress-induced feeding and locomotor behavior, which suggests that this receptor may serve as drug target in the treatment of eating disorders (Compan et al., 2004). Moreover, 5-HT4 receptors have been implicated in the pathogenesis of central and peripheral disorders such as neurodegenerative diseases, irritable bowel syndrome, and atrial fibrillation (Wong et al., 1996; De and Tonini, 2001; Callahan, 2002).
Murine 5-HT4 receptors are coded by a complex gene that generates four carboxyl-terminal splice variants: 5-HT4a, 5-HT4b, 5-HT4e, and 5-HT4f (Claeysen et al., 1999). Cloning of 5-HT4 homologs from human and rat tissue revealed further structural diversity of this receptor (Blondel et al., 1998). All of these receptor variants share the same sequence up to the common splicing site (Leu-358), followed by a unique C terminus. All splice variants were able to stimulate adenylyl cyclase and raise intracellular cAMP levels upon agonist stimulation (Blondel et al., 1998; Claeysen et al., 1999; Bender et al., 2000; Mialet et al., 2000a; Vilaro et al., 2002), albeit with different efficacy and potency (Mialet et al., 2000a,b). Differences may also extend to the type of G proteins to which the various splice variants are coupled. For example, we recently showed that the 5-HT4a receptor is coupled to both Gs and G13 proteins (Ponimaskin et al., 2002b), whereas the 5-HT4b isoform activates Gi and Gs proteins (Pindon et al., 2002). Moreover, different 5-HT4 receptor variants significantly differ in their sensitivity toward homologous agonist-induced desensitization (Mialet et al., 2003). Together, these results indicate that the carboxyl terminus of the 5-HT4 receptor plays an important role in G protein coupling as well as in the initiation of counter-regulatory mechanisms.
The 5-HT4 receptor contains potential sites for post-translational modifications within their cytoplasmic carboxyl-terminal domains. We showed previously that the 5-HT4a receptor is palmitoylated at two different sites, which involve a conserved cysteine pair at amino acid positions 328/329 and an unusual additional site (Cys-386) near its C terminus (Ponimaskin et al., 2002a). We showed that 5-HT4a receptor palmitoylation is dynamically regulated upon agonist stimulation of the receptor (Ponimaskin et al., 2001). Common to many GPCRs, acylation of 5-HT4 receptors is believed to anchor their C-terminal tails to the plasma membrane, thereby creating additional intracellular loops.
The role of agonist-induced receptor phosphorylation by second-messengereCactivated and GPCR kinases for regulating the receptor function and cell surface expression is well documented for many receptors (Ferguson, 2001). According to the paradigm, established mainly in the 2-adrenergic receptor system, phosphorylation on serine or threonine residues located preferentially in the receptor carboxyl terminus promotes -arrestin binding, which interferes with further G protein activation. -Arrestin also acts as an adaptor protein to couple the phosphorylated receptor with the heavy chain of clathrin, thereby initiating receptor endocytosis.
Palmitoylation has been shown to affect several aspects of GPCR function, including the efficiency and selectivity of G-protein coupling, receptor phosphorylation and desensitization, endocytosis, and transport to the plasma membrane (Qanbar and Bouvier, 2003). In the case of the 2-adrenergic receptor, it was postulated that agonist-dependent changes in receptor palmitoylation modulates access of regulatory kinases such as cAMP-dependent protein kinase (PKA) and G protein-coupled receptor kinases (GRK) to the C-terminal domain (Moffett et al., 2001). Mutation of putative palmitoylation sites was also shown to increase basal phosphorylation of the adenosine A3 receptor (Palmer and Stiles, 2000). In contrast, palmitoylation-deficient forms of the CCR5 chemokine receptor or the V1a vasopressin receptor showed significant defects in receptor phosphorylation, both at basal and agonist-stimulated levels (Hawtin et al., 2001; Kraft et al., 2001). Together, these results suggest that palmitate turnover controls access of carboxyl-terminal receptor phosphorylation sites to regulatory kinases in a differential manner.
In the present study, we demonstrate that the 5-HT4a receptor is phosphorylated in an agonist-dependent manner and examine the significance of palmitoylation for controlling receptor phosphorylation, desensitization and -arrestin-mediated endocytosis.
Materials and Methods
Materials. 32Pi label was ordered from MP Biomedicals (Irvine, CA). ECL Western blotting analysis system and peroxidase-conjugated secondary antibodies were purchased from Amersham Biosciences. Enzymes used in molecular cloning were obtained from New England Biolabs (Beverly, MA). Protein A-Sepharose CL-4B beads and 5-HT were from Sigma-Aldrich (St. Louis, MO). Forskolin, PMA, A23187 [GenBank] , Ro-20-1724, staurosporin, and H-89 were from Calbiochem (San Diego, CA). TC-100 insect cell medium, DMEM, Cellfectin, and LipofectAMINE 2000 reagents were purchased from Invitrogen (Carlsbad, CA). Cell culture dishes were from NUNC GmbH & Co. KG (Wiesbaden, Germany).
Recombinant DNA Procedures. All basic DNA procedures were performed as described by Sambrook et al. (1989). The plasmid encoding -arrestin2/GFP has been described previously (Kraft et al., 2001). The construction of recombinant baculovirus encoding for the murine 5-HT4a receptor wild-type and its acylation deficient mutants with substitutions of serine for cysteine Cys328/329-Ser, Cys386-Ser, and Cys328/329/386-Ser has been reported previously (Ponimaskin et al., 2002a). All mutants were verified by double-stranded dideoxy DNA sequencing at the level of the final plasmid.
Phosphorylation Experiments. Spodoptera frugiperda (Sf.9) cells were grown in TC-100 medium supplemented with 10% fetal calf serum and 1% penicillin/streptomycin (complete TC-100). For expression, Sf.9 cells (1.5 x 106) grown in 35-mm dishes were infected with recombinant baculovirus encoding wild-type or mutated 5-HT4a receptors at a multiplicity of infection of 10 plaque-forming units per cell. Labeling experiments were started after 48 h of incubation. One hour before 32Pi-labeling, complete TC-100 was replaced by phosphate-free TC-100 medium, and cells were labeled with [32Pi] (150 e藽i/ml) in phosphate-free TC-100 medium for the time periods indicated in the figure legends. In some experiments, 5-HT, forskolin, PMA, A23187 [GenBank] , or staurosporin were added to final concentrations, which are given in the figure legends. After labeling, cells were washed twice with ice-cold PBS and lysed in 600 e蘬 of buffer A (0.5% Nonidet P-40, 150 mM NaCl, 50 mM Tris/HCl, pH 7.9, 5 mM EDTA, 10 mM iodoacetamide, and 1 mM phenylmethylsulfonyl fluoride) containing 10 mM sodium fluoride and 10 mM disodium pyrophosphate. Insoluble material was pelleted (5 min, 20,000g), and the antibody AS9459 against the C-terminal domain of the 5-HT4a receptor (Ponimaskin et al., 2001) was added to the supernatant together with 30 e蘬 of protein A-Sepharose CL-4B. Samples were incubated under gentle rotation for 4 h. After brief centrifugation, the pellet was washed three times with ice-cold buffer A, and the immunocomplexes were released from the beads by incubation for 30 min at 37°C in nonreducing electrophoresis sample buffer (62.5 mM Tris-HCl, pH 6.8, containing 20% glycerol, 6% SDS, and 0.002% bromphenol blue). Radiolabeled polypeptides were analyzed by SDS-polyacrylamide gel electrophoresis (PAGE) on 12% acrylamide gels, visualized by autoradiography and analyzed by filmless autoradiographic analysis.
The expression of 5-HT4a receptors was analyzed in parallel by immunoblot analysis with antibodies AS9459 at a 1:1000 dilution. Detection was achieved with the enhanced chemiluminescence system (ECL+; Amersham Biosciences, Piscataway, NJ) and protein bands were visualized by fluorography using Kodak X-Omat AR films. The amount of [32Pi]phosphate incorporated into each of these mutants was then calculated by autoradiographic analysis in relation to the expression level of the various proteins as assessed by densitometry of ECL films.
Phosphoamino Acid Analysis. For the determination of phosphoamino acids, 32Pi-labeled Sf.9 cells (3 x 106 cells per 60-mm dish) expressing 5-HT4a receptor wild-type or its acylation-deficient mutants were treated for 5 min with 10 e 5-HT. Receptors were immunoprecipitated with the receptor-specific antibody AS9459 and resolved by SDS-PAGE on 12% polyacrylamide gel. After electrophoretic transfer to a polyvinylidene difluoride membrane (Immobilon; Millipore Corporation (Bedford, MA)) and autoradiography, receptors were excised and hydrolyzed in 6 N HCl for 2 h at 110°C. Hydrolysates were vacuum-dried, resolubilized in pH 1.9 buffer [formic acid/acetic acid/H2O, 10:100:1890 (v/v/v)] containing phosphoamino acid standards and spotted on a thin-layer cellulose plate. Phosphoamino acids were separated by electrophoresis (900 V/1.5 h) at pH 1.9 buffer, followed by a second electrophoresis (900 V/45 min) at pH 3.5 [pyridine/acetic acid/H2O, 10:100:1890 (v/v/v) and 0.5 mM EDTA] in the orthogonal direction. After ninhydrin staining of phosphoamino acid standards, thin-layer plates were exposed to autoradiographic screens.
Electrophysiology and Data Fitting. Electrophysiological measurements were performed in Sf.9 cells using methods described previously (Heine et al., 2002). Membrane currents were measured in the whole-cell, patch-clamp mode. We used a discontinuous single electrode voltage clamp amplifier SEC-05L (NPI Electronic, Tamm, Germany), which was connected to a computer via an ITC-16 interface (Instrutech Corp., Port Washington, NY). Pipettes had a resistance of 3 to 7 M. Data were acquired and stored by Pulse-PulseFit 8.31 software (HEKA, Lambrecht/Pfalz, Germany) and data analysis was performed with Igor-Wavemetrics software (Wavemetrics, Lake Oswego, OR). The temperature of the superfusing solution was stabilized at 25 ± 0.2°C controlled with a Peltier control unit (ESF-electronic, Gttingen, Germany). Recordings were performed in TC-100 medium without fetal calf serum. The used pipette solution contained 110 mM potassium gluconate, 1 mM CaCl2, 2 mM MgCl2, 2mMNa2ATP, 0.4 mM GTP, 10 mM HEPES, and 10 mM EGTA (pH 7.25, osmolarity was 20 mOsmol below the osmolarity of the TC-100 medium). The cells were stimulated with serotonin (5-HT) by local pressure-application via pipettes with a resistance of 2 to 4 M (PDES 2L-unit; NPI Electronic) for the periods as indicated in Fig. 4. Before repeated stimulation, the agonist was washed out. The activation time constants () of hyperpolarization-induced currents were estimated by exponential curve fitting as described previously (Heine et al., 2002).
Transfection of COS-7 Cells. The cDNA encoding the wild-type or the mutated 5-HT4a receptors in pTracer-CMV2 plasmid, as well as the cDNA coding for -arrestin2/GFP or GRK2, were introduced into COS-7 cells by electroporation as described previously (Claeysen et al., 1996). In brief, 1 x 107 cells were trypsinized, centrifuged, resuspended in 500 e蘬 of buffer B (50 mM K2HPO4, 20 mM CH3CO2K, 20 mM KOH, and 26.7 mM MgSO4, pH 7.4) containing 2.5 to 10 e蘥 of receptor cDNA. The total amount of DNA was kept constant at 15 e蘥 per transfection using the pTracer vector. After a 15-min exposure at room temperature, 300 e蘬 of cell suspension was transferred to a 0.4-ml electroporation cuvette (Bio-Rad) and pulsed using a Gene-Pulser apparatus (1000 e蘁, 280 V). Afterward, cells were diluted in DMEM (106 cells/ml) containing 10% dialyzed fetal bovine serum (dFBS) and plated on 15-cm dishes or into 12-well clusters at the desired density.
Determination of cAMP Production in Transfected COS-7 cells. Intracellular cAMP levels were determined by measuring the conversion of the [3H]adenine nucleotide precursor [3H]ATP to [3H]cAMP. In brief, 6 h after transfection, the complete medium was exchanged for serum-free DMEM containing 2 e藽i/ml [3H]adenine to label the endogenous ATP pool. After overnight incubation, cAMP accumulation was measured as described previously (Dumuis et al., 1988).
Desensitization Analysis. COS-7 cells expressing wild-type or acylation-deficient 5-HT4a receptors were grown in serum-free DMEM and incubated with [3H]adenine for 24 h. To analyze receptor desensitization, cells were preincubated with 5-HT (1 e) for the time periods indicated in figure legends. After removal of the agonist, cells were washed three times with DMEM, and cAMP accumulation was initiated by addition of medium containing 5-HT (10 e) and phosphodiesterase inhibitor RO-20-1724 (0.1 mM). The cAMP accumulation within a 10-min stimulation period was determined. To analyze the role of GRK2 in the agonist-promoted desensitization of the 5-HT4a receptor, COS-7 cells transiently expressing either the 5-HT4a WT receptor or acylation-deficient mutants were cotransfected with or without the vector encoding GRK2 (500 ng/107 cells). The 5-HT (10 e) induced cAMP production was then directly measured during a 10-min incubation period.
Assay for [3H]GR 113808 Binding in Intact COS-7 Cells. To measure radioligand binding in intact cells, transiently transfected COS-7 cells were plated into 12-well dishes (6 x 105 cells/ml), grown for 6 h in DMEM containing 10% dFBS and then for another 20 h in serum-free medium. Experiments were performed in HEPES-buffered saline (20 mM HEPES, 150 mM NaCl, 4.2 mM KCl, 0.9 mM CaCl2, 0.5 mM MgCl2, 0.1% glucose, and 0.1% bovine serum albumin). The reaction mixture (250 e蘬) containing saturating concentrations of [3H]GR 113808 (0.4 to 0.6 nM, Kd = 0.12 nM; specific activity, 83 Ci/mmol) was added to the cells and incubated for 2 h at 4°C. Incubation was terminated by aspiration of the incubating buffer followed by washing the cells with ice-cold HEPES-buffered saline. Cells were lysed in 0.1 N NaOH, then 8 ml of scintillation liquid (PCS scintillation cocktail; Amersham Biosciences) was added and radioactivity was measured by scintillation counting. The receptor densities were estimated as described previously (Claeysen et al., 1999). An excess of 5-HT (0.5 e) was added to determine nonspecific binding. Protein concentration was determined using a Bio-Rad protein assay.
Fluorescent Microscopy. To analyze the receptor internalization, COS-7 cells growing on coverslips were cotransfected with the plasmids encoding for the YFP-tagged 5-HT4a receptor and GFP-tagged -arrestin2. At 14 to 16 h after transfection, cells were washed and starved for 24 to 26 h in serum-free DMEM. Then, cells were treated either with vehicle (H2O) or with 5-HT (1 e) for 1 h and fixed with paraformaldehyde (3% in phosphate-buffered saline) for 15 min. The cells were washed three times with phosphate-buffered saline, and coverslips were mounted in 90% (v/v) glycerol. Finally, cells were monitored under a confocal laser-scan microscope LSM510-Meta (Zeiss, Welwyn Garden City, UK) at 63x magnification by using a mode setting that allows for spectral unmixing using individual reference spectrums.
Results
Phosphorylation of the 5-HT4a Receptor. G protein-coupled receptors often undergo agonist-induced phosphorylation, which is functionally linked to receptor desensitization. We therefore studied whether the 5-HT4a receptor is phosphorylated upon agonist stimulation. Sf.9 insect cells expressing recombinant 5-HT4a receptors were incubated with 32Pi to label their ATP pool and then exposed to the agonist 5-HT (10 e, 10 min). After immunoprecipitation and SDS-PAGE, the receptor was analyzed by autoradiography as described under Materials and Methods. As shown in Fig. 1A, exposure of the receptor to 5-HT promoted the phosphorylation of a single protein band with a molecular mass of approximately 42 kDa corresponding to the predicted molecular mass of the 5-HT4 receptor. This band comigrated with the 5-HT4a protein band detected by the Western blot and was absent in noninfected, 32Pi-labeled control cells. To examine the dose dependence of 32Pi incorporation, Sf.9 cells expressing 5-HT4a receptor were incubated in the presence of varying concentrations of 5-HT. Levels of radiolabel incorporation were then quantified by autoradiographic analysis after immunoprecipitation of 5-HT4a receptors and SDS-PAGE. Figure 1B demonstrates that 5-HT induced a dose-dependent increase in the phosphorylation intensity of the receptor with an EC50 of 180 nM. Kinetic analysis revealed a rapid 5-HT-induced increase in 32Pi incorporation into the 5-HT4a receptor (t1/2 2 min), and maximal phosphorylation level was reached after 5 min (Fig. 1C).
Phosphorylation of GPCR may be controlled by multiple second messenger-activated protein kinases, including protein kinase C, PKA, and calcium/calmodulin-dependent kinase. We investigated whether these protein kinases may participate in phosphorylation of the 5-HT4a receptor. Treatment of 32Pi-labeled cells with forskolin, PMA, or with the Ca2+ ionophore A23187 [GenBank] to stimulate different kinase activities, however, did not induce increased receptor phosphorylation (Fig. 1A). Moreover, pretreatment of the receptor-expressing cells with the protein kinase inhibitor staurosporin at 1 e concentration did not reduce the agonist-promoted increase in 5-HT4a receptor phosphorylation (Fig. 1D). These results demonstrate that the second messenger-dependent kinases tested do not mediate 5-HT4a receptor phosphorylation.
Phosphorylation of Palmitoylation-Deficient 5-HT4a Receptor Mutants. We have shown recently that the 5-HT4a receptor undergoes agonist-dependent palmitoylation and also localized the palmitoylation sites by site-directed mutagenesis (Ponimaskin et al., 2001, 2002a). Because the 5-HT4a receptor undergoes agonist-induced phosphorylation, we next analyzed a possible interplay between these two post-translational receptor modifications. Sf.9 insect cells expressing either the wild-type 5-HT4a receptor or its palmitoylation-deficient mutants were loaded with 32Pi, stimulated with 5-HT, and then subjected to immunoprecipitation, SDS-PAGE, and autoradiography.
Immunoblot analysis revealed that all mutants were expressed at levels comparable with those of the 5-HT4a wild-type control (Fig. 2A), which allows for a quantitative comparison of phosphorylation levels in the different products. The amount of [32Pi]phosphate incorporated into each of these mutants was calculated by densitometry of fluorograms and normalized for expression levels of the various proteins as determined by immunoblotting. Analysis of the results obtained in 32Pi labeling experiments demonstrated that both basal and agonist-mediated phosphorylation efficiency for all acylation-deficient mutants was significantly elevated up to 2.7-fold compared with the wild-type receptor (Fig. 2, A and B). The increase in the phosphorylation efficiency over the basal level was particularly pronounced in the case of Cys328/329 and Cys386-Ser receptor mutants.
Phosphoamino Acid Analysis. The C-terminal domain of the 5-HT4a receptor contains several serine and threonine residues that represent potential phosphate acceptor sites for GRK-mediated receptor phosphorylation. Next, we defined which amino acids within the 5-HT4a receptor are modified by ligand-induced phosphorylation. We also analyzed whether a different phosphorylation site(s) may be used in the acylation-deficient mutants. Phosphoamino acid analysis revealed that agonist-mediated phosphorylation of the wild-type receptor as well as all acylation-deficient mutants occurs exclusively at serine residues (Fig. 3).
Desensitization of the 5-HT4a Receptor in Sf.9 Cells as Assessed by an Electrophysiological Assay. A possible functional implication of the palmitoylation for the 5-HT4a receptor desensitization was analyzed in the insect Sf.9 cells by using the patch-clamp technique, as described previously (Heine et al., 2002). To this end, the hyperpolarization- and cyclic nucleotide-sensitive cation-channel from Heliothis virescens (HvCNG) was coexpressed in Sf.9 cells together with the heterotrimeric Gs protein (Gs, 1 and 2 subunits) and with the 5-HT4a receptor wild-type or its acylation-deficient mutants. The recorded activation currents during a hyperpolarization to eC100 mV were fitted by a single exponential function. The resulting channel activation time constant was used as a parameter for the analysis of the channel activity.
As shown in Fig. 4, treatment of the wild-type and mutated 5-HT4a receptors with 5-HT resulted in a fast and significant decrease of the activation time constant , which reflects an increase in the intracellular cAMP levels, leading to a faster activation of the HvCNG-channel (Fig. 4). The stimulatory effect was reversible and the currents returned nearly to control values after washout of the agonist. To analyze the receptor desensitization, the agonist was applied repeatedly, and changes in the activation time constant were measured over the time. After every receptor stimulation, the agonist was thoroughly washed out from the bath before the new stimulation. Figure 4 demonstrates that repetitive application of the agonist to the 5-HT4a receptor wild-type resulted in similar activation kinetics for the channel for the following stimulation after washout. It is noteworthy that repetitive agonist stimulation of the Cys328/329-Ser mutant resulted in a reduction of the channel activation reflected in an increase of the activation time constant (Fig. 4). Analysis of the ascending slope for the minimal values (maximal channel activation during agonist application) obtained after repeated agonist applications revealed that the slope value was increased from 3.7 (n = 4) for the wild-type to 13.5 (n = 4) in the case of Cys328/329-Ser mutant (Fig. 4). This demonstrates that the channel activation after repeated stimulation of the Cys328/329-Ser mutant was reduced compared with the wild-type, reflecting a reduced number of mutated receptor that induce intracellular cAMP changes. It is noteworthy that the agonist-induced receptor desensitization did not significantly differ from that of the wild-type receptor when the Cys386-Ser and Cys328/329/386-Ser mutants were analyzed (data not shown). These data suggest that the palmitoylation state of a proximal acylation site (Cys328/329), which was previously shown to selectively modulate constitutive receptor activity (Ponimaskin et al., 2002a), may also affect 5-HT4a receptor desensitization.
Desensitization of the 5-HT4a Receptor Wild-Type and Its Acylation-Deficient Mutants Expressed in COS-7 Cells. To further assess the role of palmitoylation for the agonist-induced 5-HT4a receptor desensitization in a different cellular context, the effects of sustained agonist treatment on the signaling efficacy of the WT and acylation-deficient mutants were compared in COS-7 cells. Receptor activities were measured in transfected cells that were pretreated with or without 5-HT for different periods. Desensitization was then defined as the reduction in the efficiency of agonist to stimulate adenylyl cyclase activity after pretreatment. As illustrated in Fig. 5, the onset of agonist-promoted desensitization was accelerated for all acylation-deficient mutants and particularly for the Cys328/329-Ser construct, compared with wild-type. However, the rate of desensitization was relatively low; after 1 h preincubation, we found that the residual response to the agonist was 80 ± 8%, 72 ± 6%, 68 ± 9%, and 62 ± 7% for the WT, Cys386-Ser, Cys328/329/386-Ser, and Cys328/329-Ser, respectively. Even at the longest desensitization time studied (18 h), we could still detect residual adenylyl cyclase activity of 62 ± 6%, 50 ± 5%, 45 ± 8% and 33 ± 7% for the WT, Cys386-Ser, Cys328/329/386-Ser, and Cys328/329-Ser, respectively (Fig. 5B). This is in contrast to our previous observations in rat esophagus and colliculus neurons of mouse, where exposure of cells to selective 5-HT4 receptor agonists was accompanied by a very potent and rapid desensitization of the receptor-stimulated adenylyl cyclase response (Ansanay et al., 1992; Ronde et al., 1995). Such divergent findings may either indicate cell type-specific effects based on a differential expression of regulatory proteins or may result from overexpression of receptors in transfected cell lines.
Taken together, our results indicate that the acylation-deficient mutants were reproducibly more sensitive to agonist pre-exposure than the receptor wild-type. The rate of agonist-promoted receptor desensitization was greatest for the proximal acylation mutant Cys328/329-Ser. This suggests that, in agreement with our findings with receptor phosphorylation, desensitization of the 5-HT4a receptor may also be regulated by its palmitoylation state.
Role of GRK in Desensitization of the 5-HT4a Receptor. Phosphorylation of the 5-HT4a receptor is not affected by stimulation of several different second messenger-activated protein kinases (Fig. 1, A and D). In addition, we had previously demonstrated that the functional desensitization of the 5-HT4 receptors in mouse colliculus neurons is cAMP-independent and agonist-selective (Ansanay et al., 1992; Ronde et al., 1995). This suggests a possible role of GRKs in the desensitization process. To test this hypothesis, we coexpressed GRK2 with the different 5-HT4a receptor mutants and determined the amount of cAMP accumulation over a 10-min incubation period. In this experimental setting, "desensitization" is interpreted as the cumulative effect of GRK overexpression on 5-HT-induced cAMP generation compared with cells with endogenous GRK levels. As shown in Fig. 6, the basal activity of cells coexpressing the different receptors together with GRK2 was not significantly changed. In contrast, the agonist-induced cyclase response was significantly (p < 0.01) impaired in cells overexpressing GRK2 compared with the control (Fig. 6). After a 10-min stimulation period with 5-HT, the maximal stimulation of adenylate cyclase in cells overexpressing GRK2 was reduced to 28 ± 8%, 35 ± 5%, 55 ± 11%, and 51 ± 9% of control cells (without coexpression of GRK2) for the wild-type, Cys328/329-Ser, Cys386-Ser, and Cys328/329/386-Ser, respectively. A 60 min-agonist pre-exposure performed before the period of cAMP accumulation did not significantly decrease the 5-HT response (data not shown). Taken together, these results strongly suggest a role for GRK2 (or a related receptor kinase) in homologous desensitization of the 5-HT4a receptor.
Role of Palmitoylation in Agonist-Induced Internalization of the 5-HT4a Receptor. In many GPCRs, GRK-induced receptor phosphorylation plays an important role in facilitating -arrestin-mediated receptor internalization. To analyze whether the 5-HT4a receptor undergoes agonist-dependent internalization, the COS-7 cells were transiently cotransfected with YFP-tagged 5-HT4a receptor and GFP-tagged -arrestin2. Functionality of receptor-YFP construct was confirmed by ligand binding, 5-HT induced adenylate cyclase activity, and serum response element assay. Distribution of the receptor and -arrestin2 was then analyzed by confocal microscopy on LSM510-Meta microscope. Without agonist stimulation, the 5-HT4a receptor was mainly localized at the plasma membrane, whereas -arrestin2 showed homogeneous cytoplasmic staining (Fig. 7A). Treatment of transfected cells with 1 e 5-HT for 60 min induced clear receptor internalization with punctate, clustered 5-HT4a receptor-YFP structures that were partially colocalized with -arrestin2 (Fig. 7A).
Having established that phosphorylation efficiency of the 5-HT4a receptor is modulated by receptor palmitoylation, we next compared the rate of agonist-mediated internalization for palmitoylation-deficient mutants. In addition, we analyzed whether arrestin might modulate this process. COS-7 cells were transiently transfected with plasmids encoding the wild-type or mutated receptors in the presence or absence of -arrestin2 over-expression. Binding of the receptor-specific antagonist [3H]GR 113808 on intact cells was used to quantify the number of receptors at the cell surface. This assay revealed that the wild-type and all palmitoylation-deficient mutants showed similar expression levels at the cell surface under basal conditions (Table 1). Cells were then treated with the agonist, and internalization was measured up to 60 min after agonist exposure. As shown in the Fig. 7B, in the absence of recombinant -arrestin2 overexpression, the wild type and all acylation-deficient mutants underwent slow agonist-mediated internalization. We found that after 60 min of incubation with 5-HT, only 25% of the wild-type, Cys386-Ser, and Cys328/329/386-Ser mutants were internalized, whereas internalization of the proximal mutant Cys328/329-Ser was more efficient; approximately 35% of this mutant was removed from the cell surface (Fig. 7, Table 1).
Comparison of the cell surface expression of the 5-HT4a receptor WT or acylation-deficient mutants in intact COS-7 cells before and after a 60-min pre-exposure to agonist in the absence or in the presence of transiently expressed -arrestin2. Each value represents the mean ± S.E.M. from at least three independent experiments performed in triplicate. A statistically significant difference between the value obtained with -arrestin2 expression after a 60-min pre-exposure to 5-HT for the C328/329-S mutant and the WT values is indicated (, p < 0.001).
Coexpression of the 5-HT4a receptor wild-type with -arrestin2 resulted only in moderate enhancement of internalization up to 30% (Fig. 7C). In the case of the Cys386-Ser and Cys328/329/386-Ser mutants, the rate of agonist-induced internalization upon coexpression with -arrestin2 was increased up to 45% (Fig. 7C). The effect of -arrestin2 overexpression was strongest with the Cys328/329-Ser mutant, where 77% of ligand binding sites were removed from the cell surface after 60-min incubation with 5-HT (Fig. 7D, Table 1). The different effects of -arrestin2 overexpression on internalization of the receptor variants were found not to be caused by different -arrestin2 expression levels in the various transfections.
The fact that the highest rate of receptor endocytosis was obtained for the mutant Cys328/329 suggests that the lack of palmitoylation at these sites renders 5-HT4a receptors more susceptible to -arrestin2-mediated internalization, and this effect becomes more apparent under experimental conditions of -arrestin2 overexpression.
Discussion
The results presented here demonstrate that the 5-HT4a receptor is a substrate for phosphorylation by endogenous cell kinase(s) upon exposure to agonist. A prominent role for GRKs in the regulation of this receptor can be assumed because of following observations: 1) The 5-HT4a receptor is rapidly (t1/2 2 min) phosphorylated upon agonist stimulation, and the concentration dependence of receptor phosphorylation (EC50 180 nM 5-HT) strictly followed ligand occupancy. 2) Neither activation nor inhibition of PKA, the kinase activated downstream of the 5-HT4a receptor, or other second messenger-activated protein kinases, such as protein kinase C or calmodulin kinase, affected 5-HT4a receptor phosphorylation in a noticeable manner. This finding is unexpected, because the receptor's third intracellular loop as well as the carboxyl terminus contains consensus sites for phosphorylation by protein kinase C and calcium/calmodulin-dependent kinase. A close correlation between dose-dependent receptor phosphorylation and ligand binding together with the apparent lack of any effect of second messenger-activated protein kinases strongly suggests the involvement of one or more G protein-coupled receptor kinases, because they specifically phosphorylate only agonist-occupied receptors (Beemann and Hosey, 1999). GRKs from insects and mammals are evolutionarily conserved, which may explain why these kinases are capable of phosphorylating a wide range of receptor substrates (Cassill et al., 1991).
Examination of the 5-HT4a receptor C terminus reveals the existence of several potential serine or threonine phosphorylation sites. Some of these sites (e.g., Ser382 or Ser385) are highly conserved among human, rat, and murine 5-HT4a receptors. These serine residues are flanked by negatively charged amino acids, which previously were found to promote GRK-mediated phosphorylation (Onorato et al., 1991). Moreover, phosphoamino acid analysis revealed that ligand-induced phosphorylation of the 5-HT4a receptor occurs exclusively on serine residues. It remains to be determined which of the six C-terminal serine residues, or possibly other serines located on cytoplasmic loops of the receptor, constitute phospho-acceptor sites for receptor kinases.
It is well documented that many GPCRs can be modified by both palmitoylation and phosphorylation and that these two post-translational modifications are interrelated. The findings of the present study are reminiscent of earlier work on the 2-adrenergic receptor in which a Cys341-Gly palmitoylation mutant found to be hyperphosphorylated at basal levels could not be enhanced by further stimulation (Moffett et al., 1993). It was concluded that lack of palmitoylated cysteine exposes a PKA phosphorylation site juxtaposed to the palmitoylation site. A marked increase in basal phosphorylation as a result of mutating two putative cysteine palmitoylation sites was also observed in the A3 adenosine receptor (Palmer and Stiles, 2000). However, GPCRs in their depalmitoylated states are not always better targets for receptor kinases. For example, a palmitoylation-deficient variant of the CCR5 chemokine receptor is not efficiently phosphorylated and reveals a profound defect in receptor desensitization and internalization (Kraft et al., 2001). Similar defects were observed for the vasopressin V1a receptor (Hawtin et al., 2001), whereas elimination of the two palmitoylated cysteine residues in the related V2 vasopressin receptor did not change basal or agonist-stimulated phosphorylation (Sadeghi et al., 1997). Moreover, a mass spectrometric study of post-translational modifications of the bradykinin B2 receptor revealed that palmitoylation at Cys356 and phosphorylation at Tyr352 were mutually exclusive (Soskic et al., 1999). Taken together, these studies clearly show that in GPCRs, palmitoylation and phosphorylation are often interrelated, albeit in different ways.
Agonist-induced 5-HT4a receptor phosphorylation established in the present study seems to be functionally significant. We reported previously that 5-HT4 receptors expressed in colliculus neurons undergo homologous receptor desensitization (Ansanay et al., 1992). Neither activation nor inhibition of cAMP-dependent protein kinase affected receptor desensitization, whereas inhibitors of G protein-coupled receptor kinases, such as heparin or Zn2+, caused significant decrease in 5-HT-induced receptor desensitization. Similar to the present findings on 5-HT4a receptor regulation in a heterologous expression system, the 5-HT4 receptor desensitization in native tissues also proceeded in a biphasic manner and possessed comparable kinetics of the rapid phase of receptor desensitization. Further support for a role for GRKs in 5-HT4a receptor regulation derives from the observation that overexpression of GRK2 significantly enhanced 5-HT4a receptor desensitization.
A role for PKA in 5-HT4 receptor regulation was also suggested recently based on the observation that receptor-independent activation of PKA could mimic 5-HT induced desensitization of the 5-HT4d receptor isoform (Mialet et al., 2003). However, mutation of the four PKA consensus sites on this receptor did not impair PKA-induced 5-HT4d receptor desensitization. In combination with our finding that, on its own, PKA activation did not result in 5-HT4a receptor phosphorylation, this implies that PKA may phosphorylate another protein substrate downstream of this receptor. Adenylyl cyclase, which is inhibited by PKA phosphorylation (Iwami et al., 1995), is a potential substrate for such regulation.
The 5-HT4a receptor is atypical among GPCRs because of the presence of two separate C-terminal palmitoylation sites operating as potential membrane anchorage points at positions Cys-328/329 and Cys-386. Together with the observation that the palmitoylation state of the receptor is agonist-sensitive (Ponimaskin et al., 2001), this dual lipidation suggests that in a native cellular environment, the 5-HT4a receptor C terminus exists in several different conformational states that can be modulated by external stimuli. In the present work, we tested the hypothesis that regulatory cytoplasmic proteins, which bind to or phosphorylate the C-terminal domain, may interact with these receptor conformers in a differential manner. Earlier studies in which the same receptor variants were used, revealed that mutation of the proximal Cys-328/329 couplet resulted in elevated constitutive activity of this receptor (Ponimaskin et al., 2002a). In contrast, mutations of the distal Cys-386 or of all C-terminal palmitoylation sites showed no increase in the basal activity. We concluded from these studies that defective palmitoylation at the proximal membrane anchorage site induces a conformation of the 5-HT4a C terminus to facilitate receptor binding and/or activation of heterotrimeric G proteins (Ponimaskin et al., 2002a). Our present finding of elevated basal phosphorylation levels of the proximal Cys328/329-Ser mutant may be caused by its higher constitutive activity, as was reported before in other receptor systems (Pei et al., 1994). This interpretation, however, does not explain why the Cys386-Ser mutant and the completely palmitoylation-deficient receptor are also better substrates for receptor kinases. Conformational changes occurring at the C terminus during basal or agonist-stimulated depalmitoylation of the receptor possibly lead to an exposure of one or several protein kinase binding site(s). Such receptor structure seems not to be identical with the conformation that is optimal for G protein activation. Several putative serine phosphorylation sites are located near the palmitoylated C-terminal cysteine residues, which supports the assumption that palmitoylation may restrain access of these sites to the receptor kinase. Enhanced receptor phosphorylation in case of the palmitoylation-deficient mutants could also be caused by the presence of additional phospho-acceptor sites that may have been introduced by the conservative replacement of cysteine by serine residues. However, this seems unlikely, because the triple mutant Cys328/329/386-Ser was phosphorylated even less than the single or double mutants.
Among the different receptor variants, the Cys328/329-Ser mutant shows high agonist-independent receptor activity. It also seems to be a preferred substrate for receptor kinases and is obviously most susceptible to ligand-induced desensitization and -arrestin-mediated internalization. These findings demonstrate that the palmitoylation status of the 5-HT4a receptor modifies its ability to interact with other downstream regulatory proteins. The receptor mutants were affected in their capacity to undergo agonist-independent receptor activation, phosphorylation, desensitization, and -arrestin-mediated internalization with different rank orders, which suggests that interactions of regulatory proteins with the palmitoylated receptor have different structural requirements. Although all receptor mutants were hyperphosphorylated, only the Cys328/329-Ser mutant was rapidly and efficiently internalized in -arrestin-overexpressing cells. Thus, it is unlikely that GRK-mediated hyperphosphorylation alone is responsible for the enhanced endocytosis. Instead, our data suggest that the Cys328/329-Ser receptor mutant, having only a single palmitoylation site distal to the membrane, represents a specific receptor conformation facilitating -arrestin-mediated endocytosis. According to the sequential multisite model of -arrestin interaction with ligand-activated receptors, high-affinity binding of arrestin to GPCR involves two separate phosphorylation-recognition and activation-recognition sites that reside within different parts of the molecule (Han et al., 2001). We did not directly assess -arrestin binding to palmitoylation-deficient receptor mutants, as was reported previously for the V2 vasopressin receptor (Charest and Bouvier, 2003). In contrast to 5-HT4a receptors, nonpalmitoylated V2 vasopressin receptor showed no difference in basal or agonist-induced phosphorylation but is much more slowly internalized than wild-type receptors (Sadeghi et al., 1997; Charest and Bouvier, 2003). Mutation of the palmitoylation site also decreased the rate and the extent of -arrestin recruitment to this receptor. Likewise, receptor palmitoylation was also found to be required for ligand-induced -arrestin recruitment to the cell membrane and endocytosis of the thyrotropin-releasing hormone receptor (Groarke et al., 2001). On the contrary, mutation of the membrane-proximal palmitoylation site facilitates -arrestin-mediated internalization of 5-HT4a receptors.
In summary, our study shows that 5-HT4a receptor palmitoylation is functionally significant, because disruption of C-terminal palmitoylation sites resulted in enhanced receptor phosphorylation, desensitization, and -arrestin-mediated endocytosis. Because both the phosphorylation and the palmitoylation status of the 5-HT4a receptor are dynamically regulated by agonist stimulation, concerted interactions of these two post-translational modifications seem to play an essential role in modulating the function of 5-HT4a receptors.
These studies were supported by the Deutsche Forschungsgemeinschaft through the DFG Research Centre for Molecular Physiology of the Brain grants PO 732/1-2 (to E.G.P.) and SFB523 TP-A10 (to M.O.).
doi:10.1124/mol.104.008748.
References
Ansanay H, Sebben M, Bockaert J, and Dumuis A (1992) Characterization of homologous 5-hydroxytryptamine 4 receptor desensitization in colliculi neurons. Mol Pharmacol 42: 808eC816.
Bender E, Pindon A, Van OI, Zhang YB, Gommeren W, Verhasselt P, Jurzak M, Leysen J, and Luyten W (2000) Structure of the human serotonin 5-HT4 receptor gene and cloning of a novel 5-HT4 splice variant. J Neurochem 74: 478eC489.
Blondel O, Gastineau M, Dahmoune Y, Langlois M, and Fischmeister R (1998) Cloning, expression and pharmacology of four human 5-hydroxytryptamine 4 receptor isoforms produced by alternative splicing in the carboxyl terminus. J Neurochem 70: 2252eC2261.
Bockaert J, Fagni L, and Dumuis A (1997) 5-HT4 receptors: an update, in Serotoninergic Neurons and 5-HT Receptors in the CNS (Baumgarten HG and Gthert M eds) pp 439eC465, Springer-Verlag, Berlin.
Beemann M and Hosey MM (1999) G protein coupled receptor kinases as modulators of G protein signalling. J Physiol (Lond) 517: 5eC23.
Callahan MJ (2002) Irritable bowel syndrome neuropharmacology. A review of approved and investigational compounds. J Clin Gastroenterol 35: S58eCS67.
Cassill JA, Whitney M, Joazeiro CA, Becker A, and Zuker CS (1991) Isolation of Drosophila genes encoding G protein-coupled receptor kinases. Proc Natl Acad Sci USA 88: 11067eC11070.
Charest PG and Bouvier M (2003) Palmitoylation of the V2 vasopressin receptor carboxyl tail enhances -arrestin recruitment leading to efficient receptor endocytosis and ERK1/2 activation. J Biol Chem 278: 41541eC41551.
Claeysen S, Sebben M, Becamel C, Bockaert J, and Dumuis A (1999) Novel Brain-specific 5-HT4 receptor splice variants show marked constitutive activity: role of the C-terminal intracellular domain. Mol Pharmacol 55: 910eC920.
Claeysen S, Sebben M, Journot L, Bockaert J, and Dumuis A (1996) Cloning, expression and pharmacology of the mouse 5-HT4L receptor. FEBS Lett 398: 19eC25.
Compan V, Zhou M, Grailhe R, Gazzara RA, Martin R, Gingrich J, Dumuis A, Brunner D, Bockaert J, and Hen R (2004) Attenuated response to stress and novelty and hypersensitivity to seizures in 5-HT4 receptor knock-out mice. J Neurosci 24: 412eC419.
De PF and Tonini M (2001) Irritable bowel syndrome: new agents targeting serotonin receptor subtypes. Drugs 61: 317eC332.
Dumuis A, Bouhelal R, Sebben M, Cory R, and Bockaert J (1988) A nonclassical 5-hydroxytryptamine receptor positively coupled with adenylate cyclase in the central nervous system. Mol Pharmacol 34: 880eC887.
Eglen RM, Wong EH, Dumuis A, and Bockaert J (1995) Central 5-HT4 receptors. Trends Pharmacol Sci 16: 391eC398.
Ferguson SS (2001) Evolving concepts in G protein-coupled receptor endocytosis: the role in receptor desensitization and signaling. Pharmacol Rev 53: 1eC24.
Groarke DA, Drmota T, Bahia DS, Evans NA, Wilson S, and Milligan G (2001) Analysis of the C-terminal tail of the rat thyrotropin-releasing hormone receptor-1 in interactions and cointernalization with -arrestin 1-green fluorescent protein. Mol Pharmacol 59: 375eC385.
Han M, Gurevich VV, Vishnivetskiy SA, Sigler PB, and Schubert C (2001) Crystal structure of beta-arrestin at 1.9 : possible mechanism of receptor binding and membrane translocation. Structure 9: 869eC880.
Hawtin SR, Tobin AB, Patel S, and Wheatley M (2001) Palmitoylation of the vasopressin V1a receptor reveals different conformational requirements for signaling, agonist-induced receptor phosphorylation and sequestration. J Biol Chem 276: 38139eC38146.
Heine M, Ponimaskin E, Bickmeyer U, and Richter DW (2002) 5-HT-receptor-induced changes of the intracellular cAMP level monitored by a hyperpolarization-activated cation channel. Pflueg Arch Eur J Physiol 443: 418eC426.
Iwami G, Kawabe J, Ebina T, Cannon PJ, Homcy CJ, and Ishikawa Y (1995) Regulation of adenylyl cyclase by protein kinase A. J Biol Chem 270: 12481eC12484.
Kraft K, Olbrich H, Majoul I, Mack M, Proudfoot A, and Oppermann M (2001) Characterization of sequence determinants within the carboxyl-terminal domain of chemokine receptor CCR5 that regulate signaling and receptor internalization. J Biol Chem 276: 34408eC34418.
Mialet J, Berque-Bestel I, Eftekhari P, Gastineau M, Giner M, Dahmoune Y, Donzeau-Gouge P, Hoebeke J, Langlois M, Sicsic S, et al. (2000a) Isolation of the serotoninergic 5-HT4(e) receptor from human heart and comparative analysis of its pharmacological profile in C6-glial and CHO cell lines. Br J Pharmacol 129: 771eC781.
Mialet J, Berque-Bestel I, Sicsic S, Langlois M, Fischmeister R, and Lezoualc'h F (2000b) Pharmacological characterization of the human 5-HT 4(d) receptor splice variant stably expressed in Chinese hamster ovary cells. Br J Pharmacol 131: 827eC835.
Mialet J, Fischmeister R, and Lezoualc`h F (2003) Characterization of human 5-HT4(d) receptor desensitization in CHO cells. Br J Pharmacol 138: 445eC452.
Moffett S, Mouillac B, Bonin H, and Bouvier M (1993) Altered phosphorylation and desensitization patterns of a human beta 2-adrenergic receptor lacking the palmitoylated Cys341. EMBO (Eur Mol Biol Organ) J 12: 349eC356.
Moffett S, Rousseau G, Lagace M, and Bouvier M (2001) The palmitoylation state of the beta2-adrenergic receptor regulates the synergistic action of cyclic AMP-dependent protein kinase and beta-adrenergic receptor kinase involved in its phosphorylation and desensitization. J Neurochem 76: 269eC279.
Onorato JJ, Palczewski K, Regan JW, Caron MG, Lefkowitz RJ, and Benovic JL (1991) Role of acidic amino acids in peptide substrates of the beta-adrenergic receptor kinase and rhodopsin kinase. Biochemistry 30: 5118eC5125.
Palmer TM and Stiles GL (2000) Identification of threonine residues controlling the agonist-dependent phosphorylation and desensitization of the rat A3 adenosine receptor. Mol Pharmacol 57: 539eC545.
Pei G, Samama P, Lohse M, Wang M, Codina J, and Lefkowitz RJ (1994) A constitutively active mutant beta 2-adrenergic receptor is constitutively desensitized and phosphorylated. Proc Natl Acad Sci USA 91: 2699eC2702.
Pindon A, van Hecke G, van Gompel P, Lesage AS, Leysen JE, and Jurzak M (2002) Differences in signal transduction of two 5-HT4 receptor splice variants: compound specificity and dual coupling with Gs- and Gi/o-proteins. Mol Pharmacol 61: 85eC96.
Ponimaskin EG, Heine M, Joubert L, Sebben M, Bickmeyer U, Richter DW, and Dumuis A (2002a) The 5-hydroxytryptamine4a receptor is palmitoylated at two different sites and acylation is critically involved in regulation of receptor constitutive activity. J Biol Chem 277: 2534eC2546.
Ponimaskin EG, Profirovic J, Vaiskunaite R, Richter DW, and Voyno-Yasenetskaya TA (2002b) 5-Hydroxytryptamine 4a receptor is coupled to the G subunit of heterotrimeric G13 protein. J Biol Chem 277: 20812eC20819.
Ponimaskin EG, Schmidt MF, Heine M, Bickmeyer U, and Richter DW (2001) 5-hydroxytryptamine 4a receptor expressed in Sf9 cells is palmitoylated in an agonist-dependent manner. Biochem J 353: 627eC634.
Qanbar R and Bouvier M (2003) Role of palmitoylation/depalmitoylation reactions in G protein-coupled receptor function. Pharmacol Ther 97: 1eC33.
Ronde P, Ansanay H, Dumuis A, Miller R, and Bockaert J (1995) Homologous desensitization of 5-hydroxytryptamine4 receptors in rat esophagus: functional and second messenger studies. J Pharmacol Exp Ther 272: 977eC983.
Sadeghi HM, Innamorati G, Dagarag M, and Birnbaumer M (1997) Palmitoylation of the V2 vasopressin receptor. Mol Pharmacol 52: 21eC29.
Sambrook J, Fritsch E, and Maniatis T (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Soskic V, Nyakatura E, Roos M, Muller-Esterl W, and Godovac-Zimmermann J (1999) Correlations in palmitoylation and multiple phosphorylation of rat bradykinin B2 receptor in Chinese hamster ovary cells. J Biol Chem 274: 8539eC8545.
Vilaro MT, Domenech T, Palacios JM, and Mengod G (2002) Cloning and characterization of a novel human 5-HT4 receptor variant that lacks the alternatively spliced carboxy terminal exon. RT-PCR distribution in human brain and periphery of multiple 5-HT4 receptor variants. Neuropharmacology 42: 60eC73.
Wong EH, Reynolds GP, Bonhaus DW, Hsu S, and Eglen RM (1996) Characterization of [3H]GR 113808 binding to 5-HT4 receptors in brain tissues from patients with neurodegenerative disorders. Behav Brain Res 73: 249eC252., 百拇医药(Evgeni Ponimaskin, Aline )
Abteilung Immunologie (M.O.), Universitt Gttingen, Gttingen, Germany
Institute of Functional Genomics, Montpellier, France (A.D., F.G., G.B.)
Abstract
The mouse 5-hydroxytryptamine4a (5-HT4a) receptor is an unusual member of the G protein-coupled receptor superfamily because it possesses two separate carboxyl-terminal palmitoylation sites, which may allow the receptor to adopt different conformations in an agonist-dependent manner (J Biol Chem 277:2534eC2546, 2002). By targeted mutation of the proximal (Cys-328/329) or distal (Cys-386) palmitoylation sites, or a combination of both, we generated 5-HT4a receptor variants with distinct functional characteristics. In this study, we showed that upon 5-HT stimulation, the 5-HT4a receptor undergoes rapid (t 2 min) and dose-dependent (EC50 180 nM) phosphorylation on serine residues by a staurosporine-insensitive receptor kinase. Overexpression of GRK2 significantly reduced the receptor-promoted cAMP formation. The Cys328/329-Ser mutant, which is constitutively active in the absence of ligand, exhibited enhanced receptor phosphorylation under both basal and agonist-stimulated conditions and was more effectively desensitized and internalized via a -arrestin-2 mediated pathway compared with the wild-type 5-HT4a. In contrast, G protein activation, phosphorylation, desensitization, and internalization of the other palmitoylation-deficient receptor mutants were affected differently. These findings suggest that palmitoylation plays an important role in modulating 5-HT4a receptor functions and that G protein activation, phosphorylation, desensitization, and internalization depend on the different receptor conformations.
5-Hydroxytryptamine (5-HT or serotonin) is an important neurotransmitter and a local hormone that acts in the central nervous system as well as in various peripheral organs. It regulates cellular activities by binding to specific membrane-bound 5-HT receptors. Most of these proteins belong to the superfamily of G protein-coupled receptors (GPCR). The 5-HT4 receptor was shown to mediate several important physiological effects of 5-HT, including memory facilitation and the regulation of intestinal motility (Eglen et al., 1995; Bockaert et al., 1997). Mice deficient in 5-HT4 receptors displayed abnormal stress-induced feeding and locomotor behavior, which suggests that this receptor may serve as drug target in the treatment of eating disorders (Compan et al., 2004). Moreover, 5-HT4 receptors have been implicated in the pathogenesis of central and peripheral disorders such as neurodegenerative diseases, irritable bowel syndrome, and atrial fibrillation (Wong et al., 1996; De and Tonini, 2001; Callahan, 2002).
Murine 5-HT4 receptors are coded by a complex gene that generates four carboxyl-terminal splice variants: 5-HT4a, 5-HT4b, 5-HT4e, and 5-HT4f (Claeysen et al., 1999). Cloning of 5-HT4 homologs from human and rat tissue revealed further structural diversity of this receptor (Blondel et al., 1998). All of these receptor variants share the same sequence up to the common splicing site (Leu-358), followed by a unique C terminus. All splice variants were able to stimulate adenylyl cyclase and raise intracellular cAMP levels upon agonist stimulation (Blondel et al., 1998; Claeysen et al., 1999; Bender et al., 2000; Mialet et al., 2000a; Vilaro et al., 2002), albeit with different efficacy and potency (Mialet et al., 2000a,b). Differences may also extend to the type of G proteins to which the various splice variants are coupled. For example, we recently showed that the 5-HT4a receptor is coupled to both Gs and G13 proteins (Ponimaskin et al., 2002b), whereas the 5-HT4b isoform activates Gi and Gs proteins (Pindon et al., 2002). Moreover, different 5-HT4 receptor variants significantly differ in their sensitivity toward homologous agonist-induced desensitization (Mialet et al., 2003). Together, these results indicate that the carboxyl terminus of the 5-HT4 receptor plays an important role in G protein coupling as well as in the initiation of counter-regulatory mechanisms.
The 5-HT4 receptor contains potential sites for post-translational modifications within their cytoplasmic carboxyl-terminal domains. We showed previously that the 5-HT4a receptor is palmitoylated at two different sites, which involve a conserved cysteine pair at amino acid positions 328/329 and an unusual additional site (Cys-386) near its C terminus (Ponimaskin et al., 2002a). We showed that 5-HT4a receptor palmitoylation is dynamically regulated upon agonist stimulation of the receptor (Ponimaskin et al., 2001). Common to many GPCRs, acylation of 5-HT4 receptors is believed to anchor their C-terminal tails to the plasma membrane, thereby creating additional intracellular loops.
The role of agonist-induced receptor phosphorylation by second-messengereCactivated and GPCR kinases for regulating the receptor function and cell surface expression is well documented for many receptors (Ferguson, 2001). According to the paradigm, established mainly in the 2-adrenergic receptor system, phosphorylation on serine or threonine residues located preferentially in the receptor carboxyl terminus promotes -arrestin binding, which interferes with further G protein activation. -Arrestin also acts as an adaptor protein to couple the phosphorylated receptor with the heavy chain of clathrin, thereby initiating receptor endocytosis.
Palmitoylation has been shown to affect several aspects of GPCR function, including the efficiency and selectivity of G-protein coupling, receptor phosphorylation and desensitization, endocytosis, and transport to the plasma membrane (Qanbar and Bouvier, 2003). In the case of the 2-adrenergic receptor, it was postulated that agonist-dependent changes in receptor palmitoylation modulates access of regulatory kinases such as cAMP-dependent protein kinase (PKA) and G protein-coupled receptor kinases (GRK) to the C-terminal domain (Moffett et al., 2001). Mutation of putative palmitoylation sites was also shown to increase basal phosphorylation of the adenosine A3 receptor (Palmer and Stiles, 2000). In contrast, palmitoylation-deficient forms of the CCR5 chemokine receptor or the V1a vasopressin receptor showed significant defects in receptor phosphorylation, both at basal and agonist-stimulated levels (Hawtin et al., 2001; Kraft et al., 2001). Together, these results suggest that palmitate turnover controls access of carboxyl-terminal receptor phosphorylation sites to regulatory kinases in a differential manner.
In the present study, we demonstrate that the 5-HT4a receptor is phosphorylated in an agonist-dependent manner and examine the significance of palmitoylation for controlling receptor phosphorylation, desensitization and -arrestin-mediated endocytosis.
Materials and Methods
Materials. 32Pi label was ordered from MP Biomedicals (Irvine, CA). ECL Western blotting analysis system and peroxidase-conjugated secondary antibodies were purchased from Amersham Biosciences. Enzymes used in molecular cloning were obtained from New England Biolabs (Beverly, MA). Protein A-Sepharose CL-4B beads and 5-HT were from Sigma-Aldrich (St. Louis, MO). Forskolin, PMA, A23187 [GenBank] , Ro-20-1724, staurosporin, and H-89 were from Calbiochem (San Diego, CA). TC-100 insect cell medium, DMEM, Cellfectin, and LipofectAMINE 2000 reagents were purchased from Invitrogen (Carlsbad, CA). Cell culture dishes were from NUNC GmbH & Co. KG (Wiesbaden, Germany).
Recombinant DNA Procedures. All basic DNA procedures were performed as described by Sambrook et al. (1989). The plasmid encoding -arrestin2/GFP has been described previously (Kraft et al., 2001). The construction of recombinant baculovirus encoding for the murine 5-HT4a receptor wild-type and its acylation deficient mutants with substitutions of serine for cysteine Cys328/329-Ser, Cys386-Ser, and Cys328/329/386-Ser has been reported previously (Ponimaskin et al., 2002a). All mutants were verified by double-stranded dideoxy DNA sequencing at the level of the final plasmid.
Phosphorylation Experiments. Spodoptera frugiperda (Sf.9) cells were grown in TC-100 medium supplemented with 10% fetal calf serum and 1% penicillin/streptomycin (complete TC-100). For expression, Sf.9 cells (1.5 x 106) grown in 35-mm dishes were infected with recombinant baculovirus encoding wild-type or mutated 5-HT4a receptors at a multiplicity of infection of 10 plaque-forming units per cell. Labeling experiments were started after 48 h of incubation. One hour before 32Pi-labeling, complete TC-100 was replaced by phosphate-free TC-100 medium, and cells were labeled with [32Pi] (150 e藽i/ml) in phosphate-free TC-100 medium for the time periods indicated in the figure legends. In some experiments, 5-HT, forskolin, PMA, A23187 [GenBank] , or staurosporin were added to final concentrations, which are given in the figure legends. After labeling, cells were washed twice with ice-cold PBS and lysed in 600 e蘬 of buffer A (0.5% Nonidet P-40, 150 mM NaCl, 50 mM Tris/HCl, pH 7.9, 5 mM EDTA, 10 mM iodoacetamide, and 1 mM phenylmethylsulfonyl fluoride) containing 10 mM sodium fluoride and 10 mM disodium pyrophosphate. Insoluble material was pelleted (5 min, 20,000g), and the antibody AS9459 against the C-terminal domain of the 5-HT4a receptor (Ponimaskin et al., 2001) was added to the supernatant together with 30 e蘬 of protein A-Sepharose CL-4B. Samples were incubated under gentle rotation for 4 h. After brief centrifugation, the pellet was washed three times with ice-cold buffer A, and the immunocomplexes were released from the beads by incubation for 30 min at 37°C in nonreducing electrophoresis sample buffer (62.5 mM Tris-HCl, pH 6.8, containing 20% glycerol, 6% SDS, and 0.002% bromphenol blue). Radiolabeled polypeptides were analyzed by SDS-polyacrylamide gel electrophoresis (PAGE) on 12% acrylamide gels, visualized by autoradiography and analyzed by filmless autoradiographic analysis.
The expression of 5-HT4a receptors was analyzed in parallel by immunoblot analysis with antibodies AS9459 at a 1:1000 dilution. Detection was achieved with the enhanced chemiluminescence system (ECL+; Amersham Biosciences, Piscataway, NJ) and protein bands were visualized by fluorography using Kodak X-Omat AR films. The amount of [32Pi]phosphate incorporated into each of these mutants was then calculated by autoradiographic analysis in relation to the expression level of the various proteins as assessed by densitometry of ECL films.
Phosphoamino Acid Analysis. For the determination of phosphoamino acids, 32Pi-labeled Sf.9 cells (3 x 106 cells per 60-mm dish) expressing 5-HT4a receptor wild-type or its acylation-deficient mutants were treated for 5 min with 10 e 5-HT. Receptors were immunoprecipitated with the receptor-specific antibody AS9459 and resolved by SDS-PAGE on 12% polyacrylamide gel. After electrophoretic transfer to a polyvinylidene difluoride membrane (Immobilon; Millipore Corporation (Bedford, MA)) and autoradiography, receptors were excised and hydrolyzed in 6 N HCl for 2 h at 110°C. Hydrolysates were vacuum-dried, resolubilized in pH 1.9 buffer [formic acid/acetic acid/H2O, 10:100:1890 (v/v/v)] containing phosphoamino acid standards and spotted on a thin-layer cellulose plate. Phosphoamino acids were separated by electrophoresis (900 V/1.5 h) at pH 1.9 buffer, followed by a second electrophoresis (900 V/45 min) at pH 3.5 [pyridine/acetic acid/H2O, 10:100:1890 (v/v/v) and 0.5 mM EDTA] in the orthogonal direction. After ninhydrin staining of phosphoamino acid standards, thin-layer plates were exposed to autoradiographic screens.
Electrophysiology and Data Fitting. Electrophysiological measurements were performed in Sf.9 cells using methods described previously (Heine et al., 2002). Membrane currents were measured in the whole-cell, patch-clamp mode. We used a discontinuous single electrode voltage clamp amplifier SEC-05L (NPI Electronic, Tamm, Germany), which was connected to a computer via an ITC-16 interface (Instrutech Corp., Port Washington, NY). Pipettes had a resistance of 3 to 7 M. Data were acquired and stored by Pulse-PulseFit 8.31 software (HEKA, Lambrecht/Pfalz, Germany) and data analysis was performed with Igor-Wavemetrics software (Wavemetrics, Lake Oswego, OR). The temperature of the superfusing solution was stabilized at 25 ± 0.2°C controlled with a Peltier control unit (ESF-electronic, Gttingen, Germany). Recordings were performed in TC-100 medium without fetal calf serum. The used pipette solution contained 110 mM potassium gluconate, 1 mM CaCl2, 2 mM MgCl2, 2mMNa2ATP, 0.4 mM GTP, 10 mM HEPES, and 10 mM EGTA (pH 7.25, osmolarity was 20 mOsmol below the osmolarity of the TC-100 medium). The cells were stimulated with serotonin (5-HT) by local pressure-application via pipettes with a resistance of 2 to 4 M (PDES 2L-unit; NPI Electronic) for the periods as indicated in Fig. 4. Before repeated stimulation, the agonist was washed out. The activation time constants () of hyperpolarization-induced currents were estimated by exponential curve fitting as described previously (Heine et al., 2002).
Transfection of COS-7 Cells. The cDNA encoding the wild-type or the mutated 5-HT4a receptors in pTracer-CMV2 plasmid, as well as the cDNA coding for -arrestin2/GFP or GRK2, were introduced into COS-7 cells by electroporation as described previously (Claeysen et al., 1996). In brief, 1 x 107 cells were trypsinized, centrifuged, resuspended in 500 e蘬 of buffer B (50 mM K2HPO4, 20 mM CH3CO2K, 20 mM KOH, and 26.7 mM MgSO4, pH 7.4) containing 2.5 to 10 e蘥 of receptor cDNA. The total amount of DNA was kept constant at 15 e蘥 per transfection using the pTracer vector. After a 15-min exposure at room temperature, 300 e蘬 of cell suspension was transferred to a 0.4-ml electroporation cuvette (Bio-Rad) and pulsed using a Gene-Pulser apparatus (1000 e蘁, 280 V). Afterward, cells were diluted in DMEM (106 cells/ml) containing 10% dialyzed fetal bovine serum (dFBS) and plated on 15-cm dishes or into 12-well clusters at the desired density.
Determination of cAMP Production in Transfected COS-7 cells. Intracellular cAMP levels were determined by measuring the conversion of the [3H]adenine nucleotide precursor [3H]ATP to [3H]cAMP. In brief, 6 h after transfection, the complete medium was exchanged for serum-free DMEM containing 2 e藽i/ml [3H]adenine to label the endogenous ATP pool. After overnight incubation, cAMP accumulation was measured as described previously (Dumuis et al., 1988).
Desensitization Analysis. COS-7 cells expressing wild-type or acylation-deficient 5-HT4a receptors were grown in serum-free DMEM and incubated with [3H]adenine for 24 h. To analyze receptor desensitization, cells were preincubated with 5-HT (1 e) for the time periods indicated in figure legends. After removal of the agonist, cells were washed three times with DMEM, and cAMP accumulation was initiated by addition of medium containing 5-HT (10 e) and phosphodiesterase inhibitor RO-20-1724 (0.1 mM). The cAMP accumulation within a 10-min stimulation period was determined. To analyze the role of GRK2 in the agonist-promoted desensitization of the 5-HT4a receptor, COS-7 cells transiently expressing either the 5-HT4a WT receptor or acylation-deficient mutants were cotransfected with or without the vector encoding GRK2 (500 ng/107 cells). The 5-HT (10 e) induced cAMP production was then directly measured during a 10-min incubation period.
Assay for [3H]GR 113808 Binding in Intact COS-7 Cells. To measure radioligand binding in intact cells, transiently transfected COS-7 cells were plated into 12-well dishes (6 x 105 cells/ml), grown for 6 h in DMEM containing 10% dFBS and then for another 20 h in serum-free medium. Experiments were performed in HEPES-buffered saline (20 mM HEPES, 150 mM NaCl, 4.2 mM KCl, 0.9 mM CaCl2, 0.5 mM MgCl2, 0.1% glucose, and 0.1% bovine serum albumin). The reaction mixture (250 e蘬) containing saturating concentrations of [3H]GR 113808 (0.4 to 0.6 nM, Kd = 0.12 nM; specific activity, 83 Ci/mmol) was added to the cells and incubated for 2 h at 4°C. Incubation was terminated by aspiration of the incubating buffer followed by washing the cells with ice-cold HEPES-buffered saline. Cells were lysed in 0.1 N NaOH, then 8 ml of scintillation liquid (PCS scintillation cocktail; Amersham Biosciences) was added and radioactivity was measured by scintillation counting. The receptor densities were estimated as described previously (Claeysen et al., 1999). An excess of 5-HT (0.5 e) was added to determine nonspecific binding. Protein concentration was determined using a Bio-Rad protein assay.
Fluorescent Microscopy. To analyze the receptor internalization, COS-7 cells growing on coverslips were cotransfected with the plasmids encoding for the YFP-tagged 5-HT4a receptor and GFP-tagged -arrestin2. At 14 to 16 h after transfection, cells were washed and starved for 24 to 26 h in serum-free DMEM. Then, cells were treated either with vehicle (H2O) or with 5-HT (1 e) for 1 h and fixed with paraformaldehyde (3% in phosphate-buffered saline) for 15 min. The cells were washed three times with phosphate-buffered saline, and coverslips were mounted in 90% (v/v) glycerol. Finally, cells were monitored under a confocal laser-scan microscope LSM510-Meta (Zeiss, Welwyn Garden City, UK) at 63x magnification by using a mode setting that allows for spectral unmixing using individual reference spectrums.
Results
Phosphorylation of the 5-HT4a Receptor. G protein-coupled receptors often undergo agonist-induced phosphorylation, which is functionally linked to receptor desensitization. We therefore studied whether the 5-HT4a receptor is phosphorylated upon agonist stimulation. Sf.9 insect cells expressing recombinant 5-HT4a receptors were incubated with 32Pi to label their ATP pool and then exposed to the agonist 5-HT (10 e, 10 min). After immunoprecipitation and SDS-PAGE, the receptor was analyzed by autoradiography as described under Materials and Methods. As shown in Fig. 1A, exposure of the receptor to 5-HT promoted the phosphorylation of a single protein band with a molecular mass of approximately 42 kDa corresponding to the predicted molecular mass of the 5-HT4 receptor. This band comigrated with the 5-HT4a protein band detected by the Western blot and was absent in noninfected, 32Pi-labeled control cells. To examine the dose dependence of 32Pi incorporation, Sf.9 cells expressing 5-HT4a receptor were incubated in the presence of varying concentrations of 5-HT. Levels of radiolabel incorporation were then quantified by autoradiographic analysis after immunoprecipitation of 5-HT4a receptors and SDS-PAGE. Figure 1B demonstrates that 5-HT induced a dose-dependent increase in the phosphorylation intensity of the receptor with an EC50 of 180 nM. Kinetic analysis revealed a rapid 5-HT-induced increase in 32Pi incorporation into the 5-HT4a receptor (t1/2 2 min), and maximal phosphorylation level was reached after 5 min (Fig. 1C).
Phosphorylation of GPCR may be controlled by multiple second messenger-activated protein kinases, including protein kinase C, PKA, and calcium/calmodulin-dependent kinase. We investigated whether these protein kinases may participate in phosphorylation of the 5-HT4a receptor. Treatment of 32Pi-labeled cells with forskolin, PMA, or with the Ca2+ ionophore A23187 [GenBank] to stimulate different kinase activities, however, did not induce increased receptor phosphorylation (Fig. 1A). Moreover, pretreatment of the receptor-expressing cells with the protein kinase inhibitor staurosporin at 1 e concentration did not reduce the agonist-promoted increase in 5-HT4a receptor phosphorylation (Fig. 1D). These results demonstrate that the second messenger-dependent kinases tested do not mediate 5-HT4a receptor phosphorylation.
Phosphorylation of Palmitoylation-Deficient 5-HT4a Receptor Mutants. We have shown recently that the 5-HT4a receptor undergoes agonist-dependent palmitoylation and also localized the palmitoylation sites by site-directed mutagenesis (Ponimaskin et al., 2001, 2002a). Because the 5-HT4a receptor undergoes agonist-induced phosphorylation, we next analyzed a possible interplay between these two post-translational receptor modifications. Sf.9 insect cells expressing either the wild-type 5-HT4a receptor or its palmitoylation-deficient mutants were loaded with 32Pi, stimulated with 5-HT, and then subjected to immunoprecipitation, SDS-PAGE, and autoradiography.
Immunoblot analysis revealed that all mutants were expressed at levels comparable with those of the 5-HT4a wild-type control (Fig. 2A), which allows for a quantitative comparison of phosphorylation levels in the different products. The amount of [32Pi]phosphate incorporated into each of these mutants was calculated by densitometry of fluorograms and normalized for expression levels of the various proteins as determined by immunoblotting. Analysis of the results obtained in 32Pi labeling experiments demonstrated that both basal and agonist-mediated phosphorylation efficiency for all acylation-deficient mutants was significantly elevated up to 2.7-fold compared with the wild-type receptor (Fig. 2, A and B). The increase in the phosphorylation efficiency over the basal level was particularly pronounced in the case of Cys328/329 and Cys386-Ser receptor mutants.
Phosphoamino Acid Analysis. The C-terminal domain of the 5-HT4a receptor contains several serine and threonine residues that represent potential phosphate acceptor sites for GRK-mediated receptor phosphorylation. Next, we defined which amino acids within the 5-HT4a receptor are modified by ligand-induced phosphorylation. We also analyzed whether a different phosphorylation site(s) may be used in the acylation-deficient mutants. Phosphoamino acid analysis revealed that agonist-mediated phosphorylation of the wild-type receptor as well as all acylation-deficient mutants occurs exclusively at serine residues (Fig. 3).
Desensitization of the 5-HT4a Receptor in Sf.9 Cells as Assessed by an Electrophysiological Assay. A possible functional implication of the palmitoylation for the 5-HT4a receptor desensitization was analyzed in the insect Sf.9 cells by using the patch-clamp technique, as described previously (Heine et al., 2002). To this end, the hyperpolarization- and cyclic nucleotide-sensitive cation-channel from Heliothis virescens (HvCNG) was coexpressed in Sf.9 cells together with the heterotrimeric Gs protein (Gs, 1 and 2 subunits) and with the 5-HT4a receptor wild-type or its acylation-deficient mutants. The recorded activation currents during a hyperpolarization to eC100 mV were fitted by a single exponential function. The resulting channel activation time constant was used as a parameter for the analysis of the channel activity.
As shown in Fig. 4, treatment of the wild-type and mutated 5-HT4a receptors with 5-HT resulted in a fast and significant decrease of the activation time constant , which reflects an increase in the intracellular cAMP levels, leading to a faster activation of the HvCNG-channel (Fig. 4). The stimulatory effect was reversible and the currents returned nearly to control values after washout of the agonist. To analyze the receptor desensitization, the agonist was applied repeatedly, and changes in the activation time constant were measured over the time. After every receptor stimulation, the agonist was thoroughly washed out from the bath before the new stimulation. Figure 4 demonstrates that repetitive application of the agonist to the 5-HT4a receptor wild-type resulted in similar activation kinetics for the channel for the following stimulation after washout. It is noteworthy that repetitive agonist stimulation of the Cys328/329-Ser mutant resulted in a reduction of the channel activation reflected in an increase of the activation time constant (Fig. 4). Analysis of the ascending slope for the minimal values (maximal channel activation during agonist application) obtained after repeated agonist applications revealed that the slope value was increased from 3.7 (n = 4) for the wild-type to 13.5 (n = 4) in the case of Cys328/329-Ser mutant (Fig. 4). This demonstrates that the channel activation after repeated stimulation of the Cys328/329-Ser mutant was reduced compared with the wild-type, reflecting a reduced number of mutated receptor that induce intracellular cAMP changes. It is noteworthy that the agonist-induced receptor desensitization did not significantly differ from that of the wild-type receptor when the Cys386-Ser and Cys328/329/386-Ser mutants were analyzed (data not shown). These data suggest that the palmitoylation state of a proximal acylation site (Cys328/329), which was previously shown to selectively modulate constitutive receptor activity (Ponimaskin et al., 2002a), may also affect 5-HT4a receptor desensitization.
Desensitization of the 5-HT4a Receptor Wild-Type and Its Acylation-Deficient Mutants Expressed in COS-7 Cells. To further assess the role of palmitoylation for the agonist-induced 5-HT4a receptor desensitization in a different cellular context, the effects of sustained agonist treatment on the signaling efficacy of the WT and acylation-deficient mutants were compared in COS-7 cells. Receptor activities were measured in transfected cells that were pretreated with or without 5-HT for different periods. Desensitization was then defined as the reduction in the efficiency of agonist to stimulate adenylyl cyclase activity after pretreatment. As illustrated in Fig. 5, the onset of agonist-promoted desensitization was accelerated for all acylation-deficient mutants and particularly for the Cys328/329-Ser construct, compared with wild-type. However, the rate of desensitization was relatively low; after 1 h preincubation, we found that the residual response to the agonist was 80 ± 8%, 72 ± 6%, 68 ± 9%, and 62 ± 7% for the WT, Cys386-Ser, Cys328/329/386-Ser, and Cys328/329-Ser, respectively. Even at the longest desensitization time studied (18 h), we could still detect residual adenylyl cyclase activity of 62 ± 6%, 50 ± 5%, 45 ± 8% and 33 ± 7% for the WT, Cys386-Ser, Cys328/329/386-Ser, and Cys328/329-Ser, respectively (Fig. 5B). This is in contrast to our previous observations in rat esophagus and colliculus neurons of mouse, where exposure of cells to selective 5-HT4 receptor agonists was accompanied by a very potent and rapid desensitization of the receptor-stimulated adenylyl cyclase response (Ansanay et al., 1992; Ronde et al., 1995). Such divergent findings may either indicate cell type-specific effects based on a differential expression of regulatory proteins or may result from overexpression of receptors in transfected cell lines.
Taken together, our results indicate that the acylation-deficient mutants were reproducibly more sensitive to agonist pre-exposure than the receptor wild-type. The rate of agonist-promoted receptor desensitization was greatest for the proximal acylation mutant Cys328/329-Ser. This suggests that, in agreement with our findings with receptor phosphorylation, desensitization of the 5-HT4a receptor may also be regulated by its palmitoylation state.
Role of GRK in Desensitization of the 5-HT4a Receptor. Phosphorylation of the 5-HT4a receptor is not affected by stimulation of several different second messenger-activated protein kinases (Fig. 1, A and D). In addition, we had previously demonstrated that the functional desensitization of the 5-HT4 receptors in mouse colliculus neurons is cAMP-independent and agonist-selective (Ansanay et al., 1992; Ronde et al., 1995). This suggests a possible role of GRKs in the desensitization process. To test this hypothesis, we coexpressed GRK2 with the different 5-HT4a receptor mutants and determined the amount of cAMP accumulation over a 10-min incubation period. In this experimental setting, "desensitization" is interpreted as the cumulative effect of GRK overexpression on 5-HT-induced cAMP generation compared with cells with endogenous GRK levels. As shown in Fig. 6, the basal activity of cells coexpressing the different receptors together with GRK2 was not significantly changed. In contrast, the agonist-induced cyclase response was significantly (p < 0.01) impaired in cells overexpressing GRK2 compared with the control (Fig. 6). After a 10-min stimulation period with 5-HT, the maximal stimulation of adenylate cyclase in cells overexpressing GRK2 was reduced to 28 ± 8%, 35 ± 5%, 55 ± 11%, and 51 ± 9% of control cells (without coexpression of GRK2) for the wild-type, Cys328/329-Ser, Cys386-Ser, and Cys328/329/386-Ser, respectively. A 60 min-agonist pre-exposure performed before the period of cAMP accumulation did not significantly decrease the 5-HT response (data not shown). Taken together, these results strongly suggest a role for GRK2 (or a related receptor kinase) in homologous desensitization of the 5-HT4a receptor.
Role of Palmitoylation in Agonist-Induced Internalization of the 5-HT4a Receptor. In many GPCRs, GRK-induced receptor phosphorylation plays an important role in facilitating -arrestin-mediated receptor internalization. To analyze whether the 5-HT4a receptor undergoes agonist-dependent internalization, the COS-7 cells were transiently cotransfected with YFP-tagged 5-HT4a receptor and GFP-tagged -arrestin2. Functionality of receptor-YFP construct was confirmed by ligand binding, 5-HT induced adenylate cyclase activity, and serum response element assay. Distribution of the receptor and -arrestin2 was then analyzed by confocal microscopy on LSM510-Meta microscope. Without agonist stimulation, the 5-HT4a receptor was mainly localized at the plasma membrane, whereas -arrestin2 showed homogeneous cytoplasmic staining (Fig. 7A). Treatment of transfected cells with 1 e 5-HT for 60 min induced clear receptor internalization with punctate, clustered 5-HT4a receptor-YFP structures that were partially colocalized with -arrestin2 (Fig. 7A).
Having established that phosphorylation efficiency of the 5-HT4a receptor is modulated by receptor palmitoylation, we next compared the rate of agonist-mediated internalization for palmitoylation-deficient mutants. In addition, we analyzed whether arrestin might modulate this process. COS-7 cells were transiently transfected with plasmids encoding the wild-type or mutated receptors in the presence or absence of -arrestin2 over-expression. Binding of the receptor-specific antagonist [3H]GR 113808 on intact cells was used to quantify the number of receptors at the cell surface. This assay revealed that the wild-type and all palmitoylation-deficient mutants showed similar expression levels at the cell surface under basal conditions (Table 1). Cells were then treated with the agonist, and internalization was measured up to 60 min after agonist exposure. As shown in the Fig. 7B, in the absence of recombinant -arrestin2 overexpression, the wild type and all acylation-deficient mutants underwent slow agonist-mediated internalization. We found that after 60 min of incubation with 5-HT, only 25% of the wild-type, Cys386-Ser, and Cys328/329/386-Ser mutants were internalized, whereas internalization of the proximal mutant Cys328/329-Ser was more efficient; approximately 35% of this mutant was removed from the cell surface (Fig. 7, Table 1).
Comparison of the cell surface expression of the 5-HT4a receptor WT or acylation-deficient mutants in intact COS-7 cells before and after a 60-min pre-exposure to agonist in the absence or in the presence of transiently expressed -arrestin2. Each value represents the mean ± S.E.M. from at least three independent experiments performed in triplicate. A statistically significant difference between the value obtained with -arrestin2 expression after a 60-min pre-exposure to 5-HT for the C328/329-S mutant and the WT values is indicated (, p < 0.001).
Coexpression of the 5-HT4a receptor wild-type with -arrestin2 resulted only in moderate enhancement of internalization up to 30% (Fig. 7C). In the case of the Cys386-Ser and Cys328/329/386-Ser mutants, the rate of agonist-induced internalization upon coexpression with -arrestin2 was increased up to 45% (Fig. 7C). The effect of -arrestin2 overexpression was strongest with the Cys328/329-Ser mutant, where 77% of ligand binding sites were removed from the cell surface after 60-min incubation with 5-HT (Fig. 7D, Table 1). The different effects of -arrestin2 overexpression on internalization of the receptor variants were found not to be caused by different -arrestin2 expression levels in the various transfections.
The fact that the highest rate of receptor endocytosis was obtained for the mutant Cys328/329 suggests that the lack of palmitoylation at these sites renders 5-HT4a receptors more susceptible to -arrestin2-mediated internalization, and this effect becomes more apparent under experimental conditions of -arrestin2 overexpression.
Discussion
The results presented here demonstrate that the 5-HT4a receptor is a substrate for phosphorylation by endogenous cell kinase(s) upon exposure to agonist. A prominent role for GRKs in the regulation of this receptor can be assumed because of following observations: 1) The 5-HT4a receptor is rapidly (t1/2 2 min) phosphorylated upon agonist stimulation, and the concentration dependence of receptor phosphorylation (EC50 180 nM 5-HT) strictly followed ligand occupancy. 2) Neither activation nor inhibition of PKA, the kinase activated downstream of the 5-HT4a receptor, or other second messenger-activated protein kinases, such as protein kinase C or calmodulin kinase, affected 5-HT4a receptor phosphorylation in a noticeable manner. This finding is unexpected, because the receptor's third intracellular loop as well as the carboxyl terminus contains consensus sites for phosphorylation by protein kinase C and calcium/calmodulin-dependent kinase. A close correlation between dose-dependent receptor phosphorylation and ligand binding together with the apparent lack of any effect of second messenger-activated protein kinases strongly suggests the involvement of one or more G protein-coupled receptor kinases, because they specifically phosphorylate only agonist-occupied receptors (Beemann and Hosey, 1999). GRKs from insects and mammals are evolutionarily conserved, which may explain why these kinases are capable of phosphorylating a wide range of receptor substrates (Cassill et al., 1991).
Examination of the 5-HT4a receptor C terminus reveals the existence of several potential serine or threonine phosphorylation sites. Some of these sites (e.g., Ser382 or Ser385) are highly conserved among human, rat, and murine 5-HT4a receptors. These serine residues are flanked by negatively charged amino acids, which previously were found to promote GRK-mediated phosphorylation (Onorato et al., 1991). Moreover, phosphoamino acid analysis revealed that ligand-induced phosphorylation of the 5-HT4a receptor occurs exclusively on serine residues. It remains to be determined which of the six C-terminal serine residues, or possibly other serines located on cytoplasmic loops of the receptor, constitute phospho-acceptor sites for receptor kinases.
It is well documented that many GPCRs can be modified by both palmitoylation and phosphorylation and that these two post-translational modifications are interrelated. The findings of the present study are reminiscent of earlier work on the 2-adrenergic receptor in which a Cys341-Gly palmitoylation mutant found to be hyperphosphorylated at basal levels could not be enhanced by further stimulation (Moffett et al., 1993). It was concluded that lack of palmitoylated cysteine exposes a PKA phosphorylation site juxtaposed to the palmitoylation site. A marked increase in basal phosphorylation as a result of mutating two putative cysteine palmitoylation sites was also observed in the A3 adenosine receptor (Palmer and Stiles, 2000). However, GPCRs in their depalmitoylated states are not always better targets for receptor kinases. For example, a palmitoylation-deficient variant of the CCR5 chemokine receptor is not efficiently phosphorylated and reveals a profound defect in receptor desensitization and internalization (Kraft et al., 2001). Similar defects were observed for the vasopressin V1a receptor (Hawtin et al., 2001), whereas elimination of the two palmitoylated cysteine residues in the related V2 vasopressin receptor did not change basal or agonist-stimulated phosphorylation (Sadeghi et al., 1997). Moreover, a mass spectrometric study of post-translational modifications of the bradykinin B2 receptor revealed that palmitoylation at Cys356 and phosphorylation at Tyr352 were mutually exclusive (Soskic et al., 1999). Taken together, these studies clearly show that in GPCRs, palmitoylation and phosphorylation are often interrelated, albeit in different ways.
Agonist-induced 5-HT4a receptor phosphorylation established in the present study seems to be functionally significant. We reported previously that 5-HT4 receptors expressed in colliculus neurons undergo homologous receptor desensitization (Ansanay et al., 1992). Neither activation nor inhibition of cAMP-dependent protein kinase affected receptor desensitization, whereas inhibitors of G protein-coupled receptor kinases, such as heparin or Zn2+, caused significant decrease in 5-HT-induced receptor desensitization. Similar to the present findings on 5-HT4a receptor regulation in a heterologous expression system, the 5-HT4 receptor desensitization in native tissues also proceeded in a biphasic manner and possessed comparable kinetics of the rapid phase of receptor desensitization. Further support for a role for GRKs in 5-HT4a receptor regulation derives from the observation that overexpression of GRK2 significantly enhanced 5-HT4a receptor desensitization.
A role for PKA in 5-HT4 receptor regulation was also suggested recently based on the observation that receptor-independent activation of PKA could mimic 5-HT induced desensitization of the 5-HT4d receptor isoform (Mialet et al., 2003). However, mutation of the four PKA consensus sites on this receptor did not impair PKA-induced 5-HT4d receptor desensitization. In combination with our finding that, on its own, PKA activation did not result in 5-HT4a receptor phosphorylation, this implies that PKA may phosphorylate another protein substrate downstream of this receptor. Adenylyl cyclase, which is inhibited by PKA phosphorylation (Iwami et al., 1995), is a potential substrate for such regulation.
The 5-HT4a receptor is atypical among GPCRs because of the presence of two separate C-terminal palmitoylation sites operating as potential membrane anchorage points at positions Cys-328/329 and Cys-386. Together with the observation that the palmitoylation state of the receptor is agonist-sensitive (Ponimaskin et al., 2001), this dual lipidation suggests that in a native cellular environment, the 5-HT4a receptor C terminus exists in several different conformational states that can be modulated by external stimuli. In the present work, we tested the hypothesis that regulatory cytoplasmic proteins, which bind to or phosphorylate the C-terminal domain, may interact with these receptor conformers in a differential manner. Earlier studies in which the same receptor variants were used, revealed that mutation of the proximal Cys-328/329 couplet resulted in elevated constitutive activity of this receptor (Ponimaskin et al., 2002a). In contrast, mutations of the distal Cys-386 or of all C-terminal palmitoylation sites showed no increase in the basal activity. We concluded from these studies that defective palmitoylation at the proximal membrane anchorage site induces a conformation of the 5-HT4a C terminus to facilitate receptor binding and/or activation of heterotrimeric G proteins (Ponimaskin et al., 2002a). Our present finding of elevated basal phosphorylation levels of the proximal Cys328/329-Ser mutant may be caused by its higher constitutive activity, as was reported before in other receptor systems (Pei et al., 1994). This interpretation, however, does not explain why the Cys386-Ser mutant and the completely palmitoylation-deficient receptor are also better substrates for receptor kinases. Conformational changes occurring at the C terminus during basal or agonist-stimulated depalmitoylation of the receptor possibly lead to an exposure of one or several protein kinase binding site(s). Such receptor structure seems not to be identical with the conformation that is optimal for G protein activation. Several putative serine phosphorylation sites are located near the palmitoylated C-terminal cysteine residues, which supports the assumption that palmitoylation may restrain access of these sites to the receptor kinase. Enhanced receptor phosphorylation in case of the palmitoylation-deficient mutants could also be caused by the presence of additional phospho-acceptor sites that may have been introduced by the conservative replacement of cysteine by serine residues. However, this seems unlikely, because the triple mutant Cys328/329/386-Ser was phosphorylated even less than the single or double mutants.
Among the different receptor variants, the Cys328/329-Ser mutant shows high agonist-independent receptor activity. It also seems to be a preferred substrate for receptor kinases and is obviously most susceptible to ligand-induced desensitization and -arrestin-mediated internalization. These findings demonstrate that the palmitoylation status of the 5-HT4a receptor modifies its ability to interact with other downstream regulatory proteins. The receptor mutants were affected in their capacity to undergo agonist-independent receptor activation, phosphorylation, desensitization, and -arrestin-mediated internalization with different rank orders, which suggests that interactions of regulatory proteins with the palmitoylated receptor have different structural requirements. Although all receptor mutants were hyperphosphorylated, only the Cys328/329-Ser mutant was rapidly and efficiently internalized in -arrestin-overexpressing cells. Thus, it is unlikely that GRK-mediated hyperphosphorylation alone is responsible for the enhanced endocytosis. Instead, our data suggest that the Cys328/329-Ser receptor mutant, having only a single palmitoylation site distal to the membrane, represents a specific receptor conformation facilitating -arrestin-mediated endocytosis. According to the sequential multisite model of -arrestin interaction with ligand-activated receptors, high-affinity binding of arrestin to GPCR involves two separate phosphorylation-recognition and activation-recognition sites that reside within different parts of the molecule (Han et al., 2001). We did not directly assess -arrestin binding to palmitoylation-deficient receptor mutants, as was reported previously for the V2 vasopressin receptor (Charest and Bouvier, 2003). In contrast to 5-HT4a receptors, nonpalmitoylated V2 vasopressin receptor showed no difference in basal or agonist-induced phosphorylation but is much more slowly internalized than wild-type receptors (Sadeghi et al., 1997; Charest and Bouvier, 2003). Mutation of the palmitoylation site also decreased the rate and the extent of -arrestin recruitment to this receptor. Likewise, receptor palmitoylation was also found to be required for ligand-induced -arrestin recruitment to the cell membrane and endocytosis of the thyrotropin-releasing hormone receptor (Groarke et al., 2001). On the contrary, mutation of the membrane-proximal palmitoylation site facilitates -arrestin-mediated internalization of 5-HT4a receptors.
In summary, our study shows that 5-HT4a receptor palmitoylation is functionally significant, because disruption of C-terminal palmitoylation sites resulted in enhanced receptor phosphorylation, desensitization, and -arrestin-mediated endocytosis. Because both the phosphorylation and the palmitoylation status of the 5-HT4a receptor are dynamically regulated by agonist stimulation, concerted interactions of these two post-translational modifications seem to play an essential role in modulating the function of 5-HT4a receptors.
These studies were supported by the Deutsche Forschungsgemeinschaft through the DFG Research Centre for Molecular Physiology of the Brain grants PO 732/1-2 (to E.G.P.) and SFB523 TP-A10 (to M.O.).
doi:10.1124/mol.104.008748.
References
Ansanay H, Sebben M, Bockaert J, and Dumuis A (1992) Characterization of homologous 5-hydroxytryptamine 4 receptor desensitization in colliculi neurons. Mol Pharmacol 42: 808eC816.
Bender E, Pindon A, Van OI, Zhang YB, Gommeren W, Verhasselt P, Jurzak M, Leysen J, and Luyten W (2000) Structure of the human serotonin 5-HT4 receptor gene and cloning of a novel 5-HT4 splice variant. J Neurochem 74: 478eC489.
Blondel O, Gastineau M, Dahmoune Y, Langlois M, and Fischmeister R (1998) Cloning, expression and pharmacology of four human 5-hydroxytryptamine 4 receptor isoforms produced by alternative splicing in the carboxyl terminus. J Neurochem 70: 2252eC2261.
Bockaert J, Fagni L, and Dumuis A (1997) 5-HT4 receptors: an update, in Serotoninergic Neurons and 5-HT Receptors in the CNS (Baumgarten HG and Gthert M eds) pp 439eC465, Springer-Verlag, Berlin.
Beemann M and Hosey MM (1999) G protein coupled receptor kinases as modulators of G protein signalling. J Physiol (Lond) 517: 5eC23.
Callahan MJ (2002) Irritable bowel syndrome neuropharmacology. A review of approved and investigational compounds. J Clin Gastroenterol 35: S58eCS67.
Cassill JA, Whitney M, Joazeiro CA, Becker A, and Zuker CS (1991) Isolation of Drosophila genes encoding G protein-coupled receptor kinases. Proc Natl Acad Sci USA 88: 11067eC11070.
Charest PG and Bouvier M (2003) Palmitoylation of the V2 vasopressin receptor carboxyl tail enhances -arrestin recruitment leading to efficient receptor endocytosis and ERK1/2 activation. J Biol Chem 278: 41541eC41551.
Claeysen S, Sebben M, Becamel C, Bockaert J, and Dumuis A (1999) Novel Brain-specific 5-HT4 receptor splice variants show marked constitutive activity: role of the C-terminal intracellular domain. Mol Pharmacol 55: 910eC920.
Claeysen S, Sebben M, Journot L, Bockaert J, and Dumuis A (1996) Cloning, expression and pharmacology of the mouse 5-HT4L receptor. FEBS Lett 398: 19eC25.
Compan V, Zhou M, Grailhe R, Gazzara RA, Martin R, Gingrich J, Dumuis A, Brunner D, Bockaert J, and Hen R (2004) Attenuated response to stress and novelty and hypersensitivity to seizures in 5-HT4 receptor knock-out mice. J Neurosci 24: 412eC419.
De PF and Tonini M (2001) Irritable bowel syndrome: new agents targeting serotonin receptor subtypes. Drugs 61: 317eC332.
Dumuis A, Bouhelal R, Sebben M, Cory R, and Bockaert J (1988) A nonclassical 5-hydroxytryptamine receptor positively coupled with adenylate cyclase in the central nervous system. Mol Pharmacol 34: 880eC887.
Eglen RM, Wong EH, Dumuis A, and Bockaert J (1995) Central 5-HT4 receptors. Trends Pharmacol Sci 16: 391eC398.
Ferguson SS (2001) Evolving concepts in G protein-coupled receptor endocytosis: the role in receptor desensitization and signaling. Pharmacol Rev 53: 1eC24.
Groarke DA, Drmota T, Bahia DS, Evans NA, Wilson S, and Milligan G (2001) Analysis of the C-terminal tail of the rat thyrotropin-releasing hormone receptor-1 in interactions and cointernalization with -arrestin 1-green fluorescent protein. Mol Pharmacol 59: 375eC385.
Han M, Gurevich VV, Vishnivetskiy SA, Sigler PB, and Schubert C (2001) Crystal structure of beta-arrestin at 1.9 : possible mechanism of receptor binding and membrane translocation. Structure 9: 869eC880.
Hawtin SR, Tobin AB, Patel S, and Wheatley M (2001) Palmitoylation of the vasopressin V1a receptor reveals different conformational requirements for signaling, agonist-induced receptor phosphorylation and sequestration. J Biol Chem 276: 38139eC38146.
Heine M, Ponimaskin E, Bickmeyer U, and Richter DW (2002) 5-HT-receptor-induced changes of the intracellular cAMP level monitored by a hyperpolarization-activated cation channel. Pflueg Arch Eur J Physiol 443: 418eC426.
Iwami G, Kawabe J, Ebina T, Cannon PJ, Homcy CJ, and Ishikawa Y (1995) Regulation of adenylyl cyclase by protein kinase A. J Biol Chem 270: 12481eC12484.
Kraft K, Olbrich H, Majoul I, Mack M, Proudfoot A, and Oppermann M (2001) Characterization of sequence determinants within the carboxyl-terminal domain of chemokine receptor CCR5 that regulate signaling and receptor internalization. J Biol Chem 276: 34408eC34418.
Mialet J, Berque-Bestel I, Eftekhari P, Gastineau M, Giner M, Dahmoune Y, Donzeau-Gouge P, Hoebeke J, Langlois M, Sicsic S, et al. (2000a) Isolation of the serotoninergic 5-HT4(e) receptor from human heart and comparative analysis of its pharmacological profile in C6-glial and CHO cell lines. Br J Pharmacol 129: 771eC781.
Mialet J, Berque-Bestel I, Sicsic S, Langlois M, Fischmeister R, and Lezoualc'h F (2000b) Pharmacological characterization of the human 5-HT 4(d) receptor splice variant stably expressed in Chinese hamster ovary cells. Br J Pharmacol 131: 827eC835.
Mialet J, Fischmeister R, and Lezoualc`h F (2003) Characterization of human 5-HT4(d) receptor desensitization in CHO cells. Br J Pharmacol 138: 445eC452.
Moffett S, Mouillac B, Bonin H, and Bouvier M (1993) Altered phosphorylation and desensitization patterns of a human beta 2-adrenergic receptor lacking the palmitoylated Cys341. EMBO (Eur Mol Biol Organ) J 12: 349eC356.
Moffett S, Rousseau G, Lagace M, and Bouvier M (2001) The palmitoylation state of the beta2-adrenergic receptor regulates the synergistic action of cyclic AMP-dependent protein kinase and beta-adrenergic receptor kinase involved in its phosphorylation and desensitization. J Neurochem 76: 269eC279.
Onorato JJ, Palczewski K, Regan JW, Caron MG, Lefkowitz RJ, and Benovic JL (1991) Role of acidic amino acids in peptide substrates of the beta-adrenergic receptor kinase and rhodopsin kinase. Biochemistry 30: 5118eC5125.
Palmer TM and Stiles GL (2000) Identification of threonine residues controlling the agonist-dependent phosphorylation and desensitization of the rat A3 adenosine receptor. Mol Pharmacol 57: 539eC545.
Pei G, Samama P, Lohse M, Wang M, Codina J, and Lefkowitz RJ (1994) A constitutively active mutant beta 2-adrenergic receptor is constitutively desensitized and phosphorylated. Proc Natl Acad Sci USA 91: 2699eC2702.
Pindon A, van Hecke G, van Gompel P, Lesage AS, Leysen JE, and Jurzak M (2002) Differences in signal transduction of two 5-HT4 receptor splice variants: compound specificity and dual coupling with Gs- and Gi/o-proteins. Mol Pharmacol 61: 85eC96.
Ponimaskin EG, Heine M, Joubert L, Sebben M, Bickmeyer U, Richter DW, and Dumuis A (2002a) The 5-hydroxytryptamine4a receptor is palmitoylated at two different sites and acylation is critically involved in regulation of receptor constitutive activity. J Biol Chem 277: 2534eC2546.
Ponimaskin EG, Profirovic J, Vaiskunaite R, Richter DW, and Voyno-Yasenetskaya TA (2002b) 5-Hydroxytryptamine 4a receptor is coupled to the G subunit of heterotrimeric G13 protein. J Biol Chem 277: 20812eC20819.
Ponimaskin EG, Schmidt MF, Heine M, Bickmeyer U, and Richter DW (2001) 5-hydroxytryptamine 4a receptor expressed in Sf9 cells is palmitoylated in an agonist-dependent manner. Biochem J 353: 627eC634.
Qanbar R and Bouvier M (2003) Role of palmitoylation/depalmitoylation reactions in G protein-coupled receptor function. Pharmacol Ther 97: 1eC33.
Ronde P, Ansanay H, Dumuis A, Miller R, and Bockaert J (1995) Homologous desensitization of 5-hydroxytryptamine4 receptors in rat esophagus: functional and second messenger studies. J Pharmacol Exp Ther 272: 977eC983.
Sadeghi HM, Innamorati G, Dagarag M, and Birnbaumer M (1997) Palmitoylation of the V2 vasopressin receptor. Mol Pharmacol 52: 21eC29.
Sambrook J, Fritsch E, and Maniatis T (1989) Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press, Cold Spring Harbor, NY.
Soskic V, Nyakatura E, Roos M, Muller-Esterl W, and Godovac-Zimmermann J (1999) Correlations in palmitoylation and multiple phosphorylation of rat bradykinin B2 receptor in Chinese hamster ovary cells. J Biol Chem 274: 8539eC8545.
Vilaro MT, Domenech T, Palacios JM, and Mengod G (2002) Cloning and characterization of a novel human 5-HT4 receptor variant that lacks the alternatively spliced carboxy terminal exon. RT-PCR distribution in human brain and periphery of multiple 5-HT4 receptor variants. Neuropharmacology 42: 60eC73.
Wong EH, Reynolds GP, Bonhaus DW, Hsu S, and Eglen RM (1996) Characterization of [3H]GR 113808 binding to 5-HT4 receptors in brain tissues from patients with neurodegenerative disorders. Behav Brain Res 73: 249eC252., 百拇医药(Evgeni Ponimaskin, Aline )