Phosphorylation and Regulation of a G q/11 -coupled Receptor by Casein Kinase 1 a *

Agonist-mediated receptor phosphorylation by one or more of the members of the G-protein receptor kinase (GRK) family is an established model for G-protein-cou-pled receptor (GPCR) phosphorylation resulting in receptor desensitization. Our recent studies have, how-ever, suggested that an alternative route to GPCR phosphorylation may be an operation involving casein kinase 1 a (CK1 a ). In the current study we investigate the involvement of CK1 a in the phosphorylation of the human m3-muscarinic receptor in intact cells. We show that expression of a catalytically inactive mutant of CK1 a , designed to act in a dominant negative manner, inhibits agonist-mediated receptor phosphorylation by ; 40% in COS-7 and HEK-293 cells. Furthermore, we present evidence that a peptide corresponding to the third intracellular loop of the m3-muscarinic receptor (Ser 345 -Leu 463 ) is an inhibitor of CK1 a due to its ability to both act as a pseudo-substrate for CK1 a and form a high affinity complex with CK1 a . Expression of this peptide was able to reduce both basal and agonist-mediated m3-muscarinic receptor phosphorylation in intact cells. These results support the notion that CK1 a is able to mediate GPCR phosphorylation in an agonist-depend-ent manner and that this may provide a novel mechanism for GPCR phosphorylation. The functional role of phosphorylation was investigated using a mutant of the

It is now well established that G-protein-coupled receptor (GPCR) 1 phosphorylation is a general phenomenon that controls specific key signaling properties of receptors. Originally associated with receptor desensitization (1,2), GPCR phospho-rylation has now been implicated in a number of processes including receptor internalization (3)(4)(5)(6) and as a molecular switch that determines coupling to specific signaling pathways (7,8). The receptor-specific kinases involved are generally considered to belong to the G-protein-coupled receptor kinase (GRK) family which are characterized by their sequence homology to rhodopsin kinase (GRK-1) and that include the extensively studied ␤-adrenergic receptor kinases 1 and 2 (GRK-2 and -3, respectively) (2,9). Reconstitution experiments using purified, or partially purified, receptors have demonstrated that in addition to the ␤ 2 -adrenergic receptor a number of GPCRs including muscarinic ((10 -12), substance P (13), bradykinin B 2 (14), and adenosine A 3 receptors (15) can act as GRK substrates. Furthermore, a GRK-2 dominant negative mutant (16) has been widely employed to probe the role of endogenous GRK-2 in the regulation of GPCRs (3,(17)(18)(19). These studies, and others, have led to the proposal that the GRKs, and in particular GRK-2, have a broad receptor substrate specificity and are able to phosphorylate and regulate GPCRs coupled to both adenylyl cyclase via G s/i and those coupled to the phospholipase C pathway via G q/ 11 .
In contrast to this model of GRK-mediated phosphorylation of GPCRs, our studies on the G q/11 -coupled m3-muscarinic receptor have suggested that there may be an alternative mechanism mediating agonist-dependent receptor phosphorylation. This receptor is rapidly phosphorylated on serine following agonist addition (20) with a time course that closely correlates with receptor desensitization as measured by diminished inositol (1,4,5)-trisphosphate (Ins(1,4,5)P 3 ) and intracellular calcium responses (21). Initial characterization of the kinase involved in this phosphorylation event eliminated a role for protein kinase A, protein kinase C, and Ca 2ϩ /calmodulin-dependent protein kinase (20). Crude membranes prepared from CHO cells expressing recombinant m3-muscarinic receptor were also found to contain receptor kinase activity and that this activity was insensitive to inhibition by heparin and zinc at concentrations that were known to inhibit GRK-2 activity (22). These were the first data suggesting that the m3-muscarinic receptor was phosphorylated by a kinase that was distinct from GRK-2. By using a bacterial fusion protein of the third intracellular loop of the m3-muscarinic receptor as a pseudo-substrate for the "putative" muscarinic receptor kinase, we were able to purify, from porcine cerebellum, a 40-kDa protein kinase that in membrane reconstitution experiments was able to phosphorylate the m3-muscarinic receptor in an agonist-dependent manner (23). Amino acid sequence analysis identified this protein kinase as casein kinase 1␣ (CK1␣) (24). Importantly, the ability of CK1␣ to drive receptor phosphorylation was not restricted to the m3-muscarinic receptor since both rhodopsin and the m1-muscarinic receptor were also shown to be in vitro substrates that were phosphorylated in a stimulusdependent manner (24,25). These in vitro studies suggested that CK1␣ may act as a cellular kinase for specific GPCRs, thereby offering an alternative and distinct route to GPCR phosphorylation from that of the GRKs.
In the present study we explore this hypothesis further by using a catalytically inactive mutant of CK1␣ and a peptide corresponding to the third intracellular loop of the m3-muscarinic receptor to inhibit endogenous CK1␣ activity. These experiments provide evidence for a cellular role of CK1␣ in the phosphorylation and regulation of the m3-muscarinic receptor.
Generation Of F-CK1␣K46R-Wild type bovine casein kinase 1␣ that had been tagged at the N terminus with the FLAG epitope (F-CK1␣) and cloned into pcDNA-3 (Invitrogen, see Ref. 24) was used as a template for the Quikchange site-directed mutagenesis kit (Stratagene). The mutagenesis primer used was GAGGAGGTGGCAGTGCGACTA-GAATCCCAGAAGGCGAGGCATCCCCAGTTG. This created a Lys-Arg change at position 46 in the amino acid sequence of CK1␣. The resulting construct F-CK1␣K46R was contained in pcDNA-3 and also possessed a FLAG epitope tag on the N terminus. The mutation was confirmed by DNA sequencing.
Generation of the Third Intracellular Loop Peptide (3i Loop Peptide)-The sequence encoding amino acids Ser 345 -Leu 463 from the third intracellular loop of the m3-muscarinic receptor was amplified using the polymerase chain reaction primers, 5Ј primer, GGGGGTAC-CGCCACCATGTCCCTGGAGAACTCCGCCTCCTCCGAC, and 3Ј primer, GGGTCTAGACTACAGAGTGGCTTCCTTGAAGGACAGAGG, and cloned into KpnI and XbaI sites in pcDNA-3. The resulting construct was then used in transient transfections of HEK-293 cells or COS-7 cells or used to make stably expressing CHO cell lines.
Generation of the m3-Muscarinic Receptor Deletion Mutant Lys 370 -Ser 425 Deletion Mutant-The m3-muscarinic receptor coding sequence contained in pcDNA-3 was digested with HindIII and then religated. This removed the coding sequence for amino acids Lys 370 -Ser 425 inclusive but maintained the reading frame of the remaining cDNA.
Construction of Bacterial Fusion Proteins-Generation of the GST bacterial fusion proteins used in this study have been described previously (23).
Antibody Production-Production and characterization of the m3muscarinic receptor specific antiserum raised against residues Ser 345 -Leu 463 in the third intracellular loop of the m3-muscarinic receptor has been previously described (20). The pan-M antiserum was raised against a peptide (DRYFSVTRPLSYRAKRTPRC) corresponding to amino acids Asp 122 -Arg 140 of the m1-muscarinic receptor. This sequence is conserved in the muscarinic receptor family, and the resulting antiserum would be expected to cross-react with all of the muscarinic receptor subtypes. The peptide was conjugated to Keyhole Limpet hemocyanin and injected into New Zealand White rabbits using standard protocols. Characterization of the antiserum using Western blots showed that the pan-M antiserum cross-reacted with the m1 and m3 muscarinic receptors. This antibody was used in immunoprecipitations where the phosphorylation of the Lys 370 -Ser 425 deletion mutant was investigated.
The CK1␣-specific antiserum was raised against a peptide corresponding to the N terminus of bovine CK1␣ (MASSSGSKAEFIVG-GKYKLC). Characterization of the antiserum using Western blots showed that it was able to cross-react with purified recombinant bovine CK1␣ and endogenous CK1␣ present in CHO, HEK-293, COS-7 cells, and rat brain.
Transient Transfections of HEK-293, CHO, and COS-7 Cells-Cells were plated onto 12-well dishes 24 h before transfection (cells were 40 -60% confluent at the start of transfection). Cells were transfected with m3-muscarinic receptor cDNA (contained in pcDNA-3) or co-transfected with m3-muscarinic receptor plus F-CK1␣K46R or 3i loop peptide constructs. The transfection reagent used was Fugene (Roche Molecular Biochemicals) using a total DNA concentration of 0.5 g/well.
In experiments to determine inositol (1,4,5)-trisphosphate levels, HEK-293 cells were plated onto 24-well dishes, and each well was transfected with 0.25 g of DNA. Cells were used 48 -72 h after transfection.
Stable Transfection Of CHO Cells with the Third Intracellular Loop Peptide (3i Loop Peptide)-CHO-K1 cells were transfected with the 3i loop peptide construct (described above) using the Fugene method. Clones expressing the peptide were selected in G418 (200 g/ml) and screened for expression by Western blot using the m3-muscarinic receptor antiserum that was raised against this peptide (20 , was added, and incubations were continued for a further 5 min. Reactions were terminated by rapid aspiration of the drug-containing media and application of 1 ml of ice-cold solubilization buffer (10 mM Tris, 10 mM EDTA, 500 mM NaCl, 1% Nonidet P-40, 0.5% deoxycholate, pH 7.4). Samples were left on ice for 15 min and then cleared by microcentrifugation. Antiserum (0.2 g) was added, and the samples were left on ice for 60 -90 min. Immune complexes were isolated on protein A-Sepharose beads, and the beads were washed three times with TE buffer (10 mM Tris-HCl, 10 mM EDTA, pH 7.4). In the case of receptor immunoprecipitations, the protein A-Sepharose pellet was then resuspended in 1 ml of TE, and an aliquot of the protein A slurry was removed corresponding to a known quantity of receptors as determined by radioreceptor assay (see below). This ensured that for each experiment the same number of receptors from each transfection was run on the gel. Isolated immune complexes were then resolved on 8% SDS-PAGE gels in the case of muscarinic receptors or 15% gels for the 3i loop peptide. The gels were dried and subjected to autoradiography, and the level of phosphorylation was assessed with a Bio-Rad model GS 670 densitometer.
Quantification of m3-Muscarinic Receptor Expression-m3-Muscarinic receptor expression for each transfection was determined by incubating cells plated down onto 12-well dishes with 0.5 ml of Krebs/ HEPES buffer (as above but containing KH 2 PO 4 1.17 mM) containing a saturating concentration of muscarinic receptor antagonist [ 3 H]Nmethylscopolamine (ϳ0.5 nM) for 60 min at 37°C. Cells were washed with ice-cold Krebs/HEPES buffer (3 times), and bound [ 3 H]N-methylscopolamine was determined by liquid scintillation counting of cell extracts solubilized in solubilization buffer. Nonspecific binding was determined in the presence of 10 M atropine and was Ͻ3% of the total binding.
Mass Ins(1,4,5)P 3 Determination-Cells grown in 24-well dishes were washed with Krebs/HEPES buffer and challenged with agonist for the appropriate times. Incubations were terminated by rapid aspiration, addition of ice-cold 0.5 M trichloroacetic acid, and transfer to an ice bath. After 15 min the supernatant was removed and neutralized by addition of EDTA and Freon/tri-N-octylamine as described previously (21). Extracts were brought to pH 7 by addition of NaHCO 3 and stored at 4°C until analysis. Ins(1,4,5)P 3 mass measurements were performed using a radioreceptor assay described previously (26).
GST Fusion Protein Pull Down Assay-The hippocampus and cerebral cortex from one rat was homogenized in 15 ml of TE buffer (10 mM Tris-HCl, 2.5 mM EDTA, pH 7.4) by a 5-s pulse in a Polytron (maximum setting). A soluble brain fraction was prepared by centrifugation at 50,000 ϫ g for 15 min. The supernatant was taken and used in the pull down experiment.
GST-m3-muscarinic receptor fusion proteins (5 g) were incubated with soluble rat brain extract (50 g) for 1 h at 4°C. GST fusion protein complexes were isolated on glutathione-Sepharose beads and washed three times in TE buffer. Beads were resuspended in Laemmli buffer and resolved by 12% SDS-PAGE. The presence of CK1␣ was then determined by Western blot using the casein kinase 1␣-specific antibody.
In Vitro Kinase Assay for FLAG-tagged Casein Kinase 1␣ (F-CK1␣) and F-CK1␣K46R-HEK-293 cells were transiently transfected with either recombinant bovine FLAG-tagged F-CK1␣, FLAG-tagged F-CK1␣K46R, or vehicle. 72 h after transfection cells were lysed with 1 ml of ice-cold lysis buffer (20 mM Tris-HCl, 0.5% Nonidet P-40, 250 mM NaCl, 3 mM EDTA, 3 mM EGTA, 2 mM Na 3 VO 4 , 1 mM dithiothreitol, 1 mM phenylmethylsulfonyl fluoride, pH 7.6). The lysate was centrifuged at 21,000 ϫ g for 5 min at 4°C and the pellet discarded. To the lysate was added mouse M2 anti-FLAG antibody (0.1 g) for 1 h at 4°C followed by rabbit anti-mouse IgG (1 g) for 20 min at 4°C. The immune complexes isolated on protein A-Sepharose beads and washed twice with lysis buffer and twice with kinase buffer (10 mM Tris-HCl, 1 mM MgCl 2 , pH 7.4). The immune complex was then used in a kinase assay by resuspending the protein A beads in kinase buffer containing 20 M [␥-32 P]ATP (2.5 Ci/nmol), 5 g of ␣-casein (total volume ϭ 30 l). The reactions were allowed to proceed for 15 min at 37°C and were terminated by the addition of Laemmli buffer (10 l). Samples were resolved on a 12% SDS-PAGE gel which were then stained with Coomassie Blue to visualize ␣-casein, and an autoradiograph was obtained.

Generation of a Catalytically Inactive Mutant of CK1␣-To
investigate whether the cellular kinase responsible for m3muscarinic receptor phosphorylation was CK1␣, we tested the ability of a catalytically inactive form of CK1␣ to inhibit agonist-mediated m3-muscarinic receptor phosphorylation. Lysine 46 in bovine CK1␣ corresponds to the conserved lysine found at the ATP-binding site of all protein kinases (27). By point mutagenesis we constructed a lysine to arginine mutation at position 46 (called F-CK1␣K46R) that would be predicted to result in a catalytically inactive kinase. Expression of F-CK1␣K46R was confirmed in transiently transfected HEK-293 cells by Western blotting for the FLAG epitope that was engineered at the N terminus (Fig. 1A). Due to the epitope tag the recombinant mutant ran at a slightly higher molecular mass than the endogenous CK1␣. Hence by Western blotting with a polyclonal CK1␣ antiserum that detected both endogenous CK1␣ and recombinant mutant kinase, we estimated that F-CK1␣K46R and endogenous CK1␣ were expressed at approximately equivalent levels (Fig. 1B). Similar results were obtained for F-CK1␣K46R expressed in COS-7 cells (data not shown).
In order to determine enzymatic activity, HEK-293 cells were transiently transfected with recombinant bovine F-CK1␣ or F-CK1␣K46R, both of which were tagged at the N terminus with the FLAG epitope. In vitro kinases assays on FLAG antiserum immunoprecipitates revealed that the F-CK1␣K46R had no detectable kinase activity (Fig. 1C).
In each experiment receptor expression was determined, and the amount of receptor applied to the gel was adjusted so that equal receptor numbers were run on the gel. Co-expression of the F-CK1␣K46R with the receptor did not influence the level of m3-muscarinic receptor expression. Any differences we observed in the level of receptor expression between control and co-transfected cells (usually Ͻ30%) were probably due to experimental variations in transfection efficiencies.
A Peptide Corresponding to the Third Intracellular Loop of the m3-Muscarinic Receptor Inhibits Receptor Phosphorylation-Earlier studies from our laboratory demonstrated that a bacterial fusion protein containing a portion of the third intracellular loop of the m3-muscarinic receptor (Ser 345 -Leu 463 ) was able to inhibit CK1␣-mediated muscarinic receptor phosphorylation in membranes (Ref. 23; also see "Discussion"). Here we tested the ability of this peptide to inhibit m3-muscarinic receptor phosphorylation in intact cells.
We also developed a stable CHO cell line that expressed the 3i loop peptide constitutively. This cell line was transiently transfected with the m3-muscarinic receptor and receptor phosphorylation compared with native CHO-K1 cells transiently transfected with the receptor. In these experiments FIG. 1. Expression of the catalytically inactive F-CK1␣K46R mutant. A, HEK-293 cells were transfected with (K46R) or without (CNT) the F-CK1␣K46R construct. Cells were lysed, and an aliquot (ϳ20 g of protein) of the cell lysis was resolved on a 12% SDS-PAGE gel and immunoblotted with the M2 anti-FLAG epitope antiserum. B, cell lysis from transfected cells (ϳ20 g of protein) were resolved on a 12% gel and immunoblotted using the CK1␣-specific antiserum. C, in vitro kinase assays were used to assess the activity of F-CK1␣K46R. Cells transfected with recombinant FLAG epitope-tagged bovine F-CK1␣ (CK1␣) or F-CK1␣K46R (K46R) or non-transfected (CNT) were lysed, and the M2 anti-FLAG antiserum was used to immunoprecipitate the recombinant kinases. The immunoprecipitate was then used in an in vitro kinase assay with ␣-casein as the substrate. The proteins were resolved by 12% SDS-PAGE, and the gel was stained with Coomassie Blue to visualize the ␣-casein. Gels were then dried, and an autoradiograph was obtained. The positions of molecular mass markers are shown in kilodaltons. The experiments shown are representative of at least three experiments.
The 3i loop peptide expressed in CHO cells was itself phosphorylated, but this phosphorylation was not altered by m3muscarinic receptor stimulation (Fig. 5). The same results were obtained in COS-7 and HEK-293 cells (data not shown).
Determination of a Putative CK1␣-binding Site on the m3-Muscarinic Receptor-CK1␣ contained in a crude soluble rat brain fraction specifically associated with the muscarinic receptor portion of a glutathione S-transferase (GST) bacterial fusion protein that contains the third intracellular loop sequence Ser 345 -Leu 463 , designated Ex-m3 (Fig. 6). This interaction appeared particularly strong since washes in salt (KCl) up to a concentration of 2 M was not sufficient to disrupt binding of CK1␣ (data not shown). Deletion mutants of Ex-m3 were used to map the binding site of CK1␣ (Fig. 7A). Truncation at the Nand C-terminals of Ex-m3 did not affect the ability of the fusion protein to interact with CK1␣ present in rat brain supernatant (Fig. 7B) or recombinant bovine CK1␣ partially purified from infected sf-9 cells (data not shown). However, deletion of the region Lys 370 -Ser 425 (⌬Lys 370 -Ser 425 ) resulted in no detectable binding of CK1␣ (Fig. 7B). A smaller deletion of 18 amino acids (His 374 -Val 391 ) also resulted in a fusion protein that was unable to associate with CK1␣ (Fig. 7B). (Note, the identity of the doublet in the Ex-m3 pull down (lane 2, Fig. 7C) is likely to be CK1␣ running at its correct molecular mass (the lower band) and CK1␣ that is still associated with Ex-m3 (upper band). The doublet in lane 5 (Fig. 7C) is likely to be CK1␣ (the upper band) and an unknown protein that cross-reacts with the CK1␣ antiserum (lower band).)

Deletion of the Region Lys 370 -Ser 425 from the Third Intracellular Loop of the m3-Muscarinic Receptor Results in Reduced
Agonist-mediated Phosphorylation-Our previous studies have demonstrated that CK1␣-mediated phosphorylation of the bacterial fusion protein ⌬Lys 370 -Ser 425 is reduced from that of the full-length fusion protein, Ex-m3 (23). It was, therefore, decided to make the same deletion in the intact m3-muscarinic receptor with the aim to produce a receptor unable to undergo agonist-mediated phosphorylation.
Hence a stable CHO cell line was generated expressing a m3-muscarinic receptor containing a Lys 370 -Ser 425 deletion in the third intracellular loop. The clone used (clone 17) was carefully selected to have a similar receptor expression level as the control CHO cell line expressing the wild type m3-muscarinic receptor (Lys 370 -Ser 425 deletion mutant expression ϭ 782 Ϯ 67 fmol/mg protein; wild type receptor expression ϭ 908 Ϯ 124 fmol/mg protein).
Deletion of this region did not significantly (p Ͼ 0.05 Student's t test) affect agonist or antagonist binding properties of the receptor. Ser 425 deletion mutant and wild type m3-muscarinic receptor, respectively.
Since the polyclonal m3-muscarinic receptor-specific antiserum used throughout this study was raised against the region Ser 345 -Leu 463 , it was possible that the Lys 370 -Ser 425 deletion mutant may have epitopes removed that would prevent immunoprecipitation by this antiserum. For this reason an alternative antiserum was raised against a peptide conserved among the muscarinic receptor family (Asp 122-Arg 140 in the 2nd intracellular loop of the m1-muscarinic receptor). This antiserum recognized both m1-and m3-muscarinic receptors as determined by Western blot (data not shown) and was designated pan-M.
By using the pan-M antiserum in immunoprecipitation studies, it was found that the level of agonist-mediated phospho-rylation of the receptor containing the Lys 370 -Ser 425 deletion was dramatically reduced (ϳ80%) compared with wild type receptor (Fig. 8). Note that the Lys 370 -Ser 425 deletion mutant runs at ϳ90 kDa compared with the wild type muscarinic receptor that runs at ϳ110 kDa.
The pan-M antiserum did, however, cross-react with phosphoproteins other than the m3-muscarinic receptor as evident by bands at ϳ120 and ϳ82 kDa in non-transfected control cells (Fig. 8.). The identities of these proteins are not known.

Analysis of the Ins(1,4,5)P 3 Response in the Lys 370 -Ser 425 Deletion Mutant of the m3-Muscarinic
Receptor-These experiments were carried out on CHO cells stably expressing either the wild type m3-muscarinic receptor or the Lys 370 -Ser 425 deletion mutant. Analysis of the wild type m3-muscarinic receptor Ins(1,4,5)P 3 response to agonist challenge showed a characteristic peak of Ins(1,4,5)P 3 production followed by a plateau phase (Fig. 9A) consistent with previous reports (21,28). Also consistent with previous studies was the demonstration that the peak Ins(1,4,5)P 3 response could be desensitized by 41.5% following a 5-min pre-stimulation with agonist ( Fig. 9A) (21,28).
The temporal profile of the Ins(1,4,5)P 3 response on agonist stimulation of the Lys 370 -Ser 425 deletion mutant was similar to the wild type receptor (Fig. 9B). Importantly, despite the fact that agonist-mediated phosphorylation of the Lys 370 -Ser 425 deletion mutant was dramatically reduced, this receptor was FIG. 4. Expression of the 3i loop peptide reduces basal and agonist-mediated phosphorylation of the m3-muscarinic receptor. Cells expressing the m3-muscarinic receptor alone (CNT) or cotransfected with the 3i loop peptide (3i-P) were prelabeled with [ 32 P]orthophosphate and stimulated with 0.1 mM carbachol (CCh) for 5 min. m3-muscarinic receptor phosphorylation was then determined by immunoprecipitation using an m3-muscarinic receptor antiserum. An equal number of muscarinic receptors was then applied to an 8% SDS-PAGE gel, and an autoradiograph was obtained. A, a representative gel from an experiment using transiently transfected COS-7 cells. In this example m3-muscarinic receptor levels were ϳ1.7 pmol/mg protein. B, a representative gel from an experiment using transiently transfected HEK-293 cells. In this example m3-muscarinic receptor levels were ϳ1.2 pmol/mg protein. C, a representative gel from an experiment where native CHO-K1 cells (CNT) or cells that were stably expressing the 3i loop peptide (3i-P) were transiently transfected with the m3muscarinic receptor. In the experiment shown m3-muscarinic receptor levels were ϳ0. FIG. 6. Complex formation between CK1␣ and a GST fusion protein corresponding to the third intracellular loop of the m3-muscarinic receptor. Either GST (5 g) or the GST fusion protein, Ex-m3, that contains the third intracellular loop sequence Ser 345 -Leu 463 (Ex-m3, 5 g) were incubated with a soluble rat brain lysate preparation (50 g of protein). Bacterial fusion proteins were then isolated on glutathione-Sepharose beads and resolved on a 12% gel. The gel was then immunoblotted using the CK1␣-specific antiserum. Recombinant FLAG-tagged F-CK1␣ purified from infected insect sf-9 cells was used as a standard (CK1␣). Note that due to the FLAG tag at the N terminus the standard runs at a slightly higher molecular mass than the brain CK1␣. The position of molecular mass markers are shown in kilodaltons.
There was, however, a dramatic difference in the magnitude of the peak Ins(1,4,5)P 3 responses between the Lys 370 -Ser 425 deletion mutant and wild type m3-muscarinic receptor (Fig.  9C). The control wild type receptor peak response was 249.7 Ϯ 5.4 pmol of Ins(1,4,5)P 3 /mg protein compared with 517.3 Ϯ 72.0 pmol of Ins(1,4,5)P 3 /mg protein (n ϭ 3 Ϯ S.E.) in the case of the Lys 370 -Ser 425 deletion mutant. This increased in Ins(1,4,5)P 3 production observed on stimulation of the mutant receptor was restricted to the peak response since the plateau responses for the mutant and wild type receptors were similar (Fig. 9C). Interestingly, despite the greater responsiveness of the Lys 370 -Ser 425 deletion mutant in terms of the magnitude of the Ins(1,4,5)P 3 response, there was no significant difference (p Ͼ 0.05 Student's t test) in the potency of the carbachol-mediated Ins(1,4,5)P 3 elevation between the wild type receptor and the Lys 370 -Ser 425 deletion mutant (EC 50 values for wild type and Lys 370 -Ser 425 deletion mutant receptors were 7.14 Ϯ 3.2 and 9.71 Ϯ 1.9 M (n ϭ 3, ϮS.E.), respectively).
The ability of the Lys 370 -Ser 425 deletion mutant to show increased stimulation of inositol phosphate production was also tested in transiently transfected HEK-293 cells. Such experiments are free from the potential clonal artifacts of the stably transfected cell lines. In these experiments the Lys 370 -Ser 425 deletion mutant peak (10 s) Ins(1,4,5)P 3 response was 24.3 Ϯ 0.4% (n ϭ 3) greater than the wild type receptor response (data not shown). The fact that the increased responsiveness of the Lys 370 -Ser 425 deletion mutant was not as great as that observed in the stable transfections may be due to the different experimental protocol and cell lines; however, the trend is the same, namely the Lys 370 -Ser 425 deletion mutant appears to generate a greater Ins(1,4,5)P 3 response than the wild type receptor.
Interestingly, the peak Ins(1,4,5)P 3 response in cells cotransfected with the m3-muscarinic receptor and the F-CK1␣K46R mutant was larger (by ϳ25%) than the peak response of cells transfected with the m3-muscarinic receptor alone (Fig. 10). It therefore appears that transient transfection

m3-Muscarinic Receptor Phosphorylation by CK1␣
of F-CK1␣K46R resulting in decreased receptor phosphorylation has no effect on the ability of the inositol phosphate response to be desensitized but does result in an increase the magnitude of the Ins(1,4,5)P 3 response. This is consistent with the data from the stably transfected CHO cells where the Lys 370 -Ser 425 deletion mutant, which showed reduced receptor phosphorylation, had an increased inositol phosphate response when compared with wild type receptors (see above). DISCUSSION The present study presents in vivo evidence that CK1␣ is responsible, at least in part, for agonist-mediated receptor phosphorylation of the m3-muscarinic receptor expressed in transfected cell lines. Furthermore, phosphorylation appears not to mediate receptor desensitization but instead may control the magnitude of the peak Ins(1,4,5)P 3 response. This represents a radical departure from the widely accepted model for GPCR phosphorylation and desensitization mediated by the GRKs.
Previous in vitro studies from our laboratory have established that CK1␣ is able to mediate stimulus-dependent phosphorylation of a number of GPCRs (24,25) suggesting that CK1␣ may represent a pathway, distinct from the GRKs, for receptor phosphorylation in intact cells. In the present study we have taken advantage of the fact that, like many GPCRs, the m3-muscarinic receptor is phosphorylated in an agonist-dependent manner by endogenous receptor kinases expressed in a number of commonly used cell lines (i.e. CHO, COS-7, and HEK-293 cells). To investigate whether CK1␣ was one of these endogenous receptor kinases, we tested the ability of a catalytically inactive mutant of CK1␣ to inhibit m3-muscarinic receptor phosphorylation. By mutating a conserved lysine residue (Lys 46 ) to an arginine (F-CK1␣K46R) in sub-domain II of the catalytic domain of CK1␣, known to be essential for phosphotransfer (27), the catalytic activity of the kinase was lost. Analogous mutations in other protein kinases, in particular GRK-2, have been used to generate dominant negative mutants (16). Furthermore, mutation of this conserved lysine in CK1⑀ has been shown to result in a dominant negative mutant able to block endogenous CKI⑀-mediated phosphorylation of dishevelled in Xenopus oocytes (29). We predicted, therefore, that if endogenous CK1␣ was responsible for m3-muscarinic receptor phosphorylation then expression of the F-CK1␣K46R mutant would inhibit receptor phosphorylation by acting in a dominant negative manner. We show here that F-CK1␣K46R when expressed in cells at approximately equivalent levels to the endogenous CK1␣ resulted in a dramatic reduction in the level of agonist-mediated m3-muscarinic receptor phosphorylation, suggesting that at least a proportion of the receptor phosphorylation was mediated by endogenous CK1␣.
We aimed to confirm this finding by raising an alternative inhibitor to CK1␣. Previous studies have shown that a peptide corresponding to the third intracellular loop of the m3-muscarinic receptor (Ser 345 -Leu 463 ) contained in a GST bacterial fusion protein was able to inhibit m3-muscarinic receptor phosphorylation in membranes (23). This inhibitory property was attributed to the ability of the peptides to act as a pseudosubstrate for CK1␣ and therefore compete with the receptor for endogenous CK1␣ present in the membrane preparation. Additionally, in the present study we demonstrate that this peptide forms a high affinity complex with CK1␣, and this may also contribute to its inhibitory properties. In order to investi- gate the involvement of CK1␣ in the phosphorylation of m3muscarinic receptors in intact cells, we tested the ability of the third intracellular loop peptide Ser 345 -Leu 463 (3i loop peptide) to inhibit receptor phosphorylation in transfected cell lines. Expression of the 3i loop peptide resulted in a dramatic reduction in both basal and agonist-mediated m3-muscarinic receptor phosphorylation. The peptide itself became phosphorylated suggesting that, as in the membrane experiments, the peptide was acting as a pseudo-substrate for CK1␣. Although it is possible that the 3i loop peptide may be inhibiting receptor phosphorylation by interacting with kinases other than CK1␣, the cellular data presented here is consistent with the earlier in vitro data and suggests that m3-muscarinic receptor phosphorylation in intact cells is mediated, at least in part, by CK1␣.
Interestingly, the m3-muscarinic receptor is not the only GPCR found to be phosphorylated by casein kinase I in intact cells. In a recent study, stimulus-dependent phosphorylation of the ␣-factor pheromone receptor of Saccharomyces cerevisiae (Ste2p) was also reported to be mediated by casein kinase I, and this was associated with control of receptor ubiquitination and endocytosis (30).
The fact that agonist-mediated phosphorylation of the m3muscarinic receptor in the present study was not completely inhibited by either F-CK1␣K46R or the 3i loop peptide indicates that either the inhibitors are not expressed in sufficiently high concentrations or that phosphorylation is mediated by kinases in addition to CK1␣. In the case of the GRK-2 dominant negative mutant, in vitro studies have demonstrated that a 10-fold molar excess of the dominant negative mutant was required to inhibit GRK-2-mediated ␤ 2 -adrenergic receptor phosphorylation by 60% (in the presence of G-protein ␤␥-subunits) (16). In the present study we were only able to achieve a level of F-CK1␣K46R expression that was approximately equivalent to endogenous CK1␣. This level of expression may have been insufficient to inhibit completely the endogenous kinase activity.
An alternative, and attractive, proposition is that CK1␣ is one of a number of kinases responsible for agonist-mediated phosphorylation. There is, for example, evidence for the involvement of the GRKs in muscarinic receptor phosphorylation. The adenylyl cyclase-coupled m2-muscarinic receptor was one of the first receptors to be shown to be an in vitro substrate for the GRKs (31). Phosphorylation at sites on the third intracellular loop is thought to mediate m2-muscarinic receptor internalization (6,32). m3-muscarinic receptors contained in urea-treated membranes have previously been shown to be phosphorylated by purified GRK-2 and -3 but not GRK-5 and -6 (11). A peptide corresponding to the third intracellular loop of the m3-muscarinic receptor has also been shown to be a substrate for GRK-2 in in vitro studies (33). Furthermore, the purified m1-muscarinic receptor is phosphorylated in an agonistdependent manner following reconstitution with GRK-2 (12). These in vitro reconstitution studies indicate that certain GRKs have the potential to phosphorylate members of the muscarinic receptor family coupled via both G i and G q/11 Gproteins, although direct evidence that this is the case in cells has yet to be presented. Hence, placed in context with previous studies our data suggest the intriguing possibility that the m3-muscarinic receptor may be phosphorylated by CK1␣ and possibly one or more of the GRKs. Ongoing studies in our laboratory mapping the phosphorylation sites on the m3-muscarinic receptor may reveal phospho-acceptor sites that cannot be assigned to CK1␣ and thereby indicate the involvement of other receptor kinases.
The functional consequences of receptor phosphorylation were investigated in a receptor where the region Lys 370 -Ser 425 in the third intracellular loop had been deleted. Compared with the wild type receptor this mutant showed an ϳ80% decrease in agonist-mediated receptor phosphorylation. The region deleted included the putative CK1␣-binding site, identified in this study to reside in the domain His 374 -Val 391 . Although the reduction in the ability of the Lys 370 -Ser 425 deletion mutant to undergo agonist-mediated receptor phosphorylation might be explained by the loss of the putative CK1␣-binding site, it must be noted that this deletion also removed eight serine residues that may act as potential phospho-acceptor sites. Furthermore, the results reported here are consistent with our previous studies using bacterial fusion proteins where we showed that deletion of either Lys 370 -Ser 425 or His 374 -Val 391 from a bacterial fusion protein expressing the majority of the third intracellular loop (i.e. Ser 345 -Leu 463 , termed Ex-m3) resulted in a dramatic reduction in the level of phosphorylation mediated by purified CK1␣ (23). Functional analysis of the wild type m3-muscarinic receptor Ins(1,4,5)P 3 response revealed a characteristic peak/plateau response to agonist stimulation where the peak response at 5-10 s was desensitized by a pre-stimulation with agonist. Previously we have speculated that due to the rapid time course of agonist-mediated m3-muscarinic receptor phosphorylation, and the large weight of evidence in the literature linking receptor phosphorylation to GPCR desensitization, that phosphorylation was the mechanism underlying the desensitization of the peak Ins(1,4,5)P 3 response (20,25,34). To our surprise the Lys 370 -Ser 425 deletion mutant receptor showed a peak/plateau Ins(1,4,5)P 3 response with a temporal profile very similar to the wild type receptor. Furthermore, pre-stimulation of the Lys 370 -Ser 425 deletion mutant resulted in desensitization of the peak Ins(1,4,5)P 3 response. These data suggest that agonist-mediated phosphorylation of the m3-muscarinic receptor was not involved in the desensitization of the peak Ins(1,4,5)P 3 response. It is, however, possible that the small amount of phosphorylation that remains in the Lys 370 -Ser 425 deletion mutant receptor, possibly mediated by one or more of the GRKs (see above), may be sufficient to induce desensitization of the peak Ins(1,4,5)P 3 response.
Recent studies have identified a number of GPCRs that undergo phosphorylation-independent desensitization. For example, heterologous desensitization of the formyl-methionylleucyl-phenylalanine and the bradykinin B 2 receptors are not associated with receptor phosphorylation (35)(36)(37). A similar lack of correlation between receptor phosphorylation and desensitization has also been reported for the chemoattractant receptor of Dictyostelium where removal of the phospho-acceptor sites does not affect desensitization of the adenylyl and guanylyl cyclase responses (38). The possibility that desensitization of the m3-muscarinic receptor is a result of mechanisms downstream of the receptor is currently being pursued in our laboratory.
In the present study the Lys 370 -Ser 425 deletion mutant gave a more robust Ins(1,4,5)P 3 response when compared with the wild type receptor (2-3 fold). This was not due to changes in agonist affinities for the receptor nor changes in the efficacy of the agonist but may reflect an enhanced coupling of the receptor to G q/11 . An increase in the magnitude of the Ins(1,4,5)P 3 response was also seen in cells where m3-muscarinic receptor phosphorylation was reduced by co-transfection with F-CK1␣K46R, suggesting that receptor phosphorylation mediated by CK1␣ was able to control the magnitude of the Ins(1,4,5)P 3 response. An enhanced inositol phosphate signal has previously been reported for truncation mutants of the platelet-activating factor (39) and neurokinin-2 receptors (40)

m3-Muscarinic Receptor Phosphorylation by CK1␣
where putative phospho-acceptor sites had been removed. In these cases receptor-stimulated phospholipase C activity appeared to be "up-regulated" by removing phospho-acceptor sites resulting in augmented signaling. It appears, therefore, that receptor phosphorylation may play a role in controlling the magnitude of the inositol phosphate response mediated by m3muscarinic receptors and possibly other phospholipase C-coupled GPCRs.
In summary we show that inhibition of endogenously expressed CK1␣ reduces agonist-mediated phosphorylation of the m3-muscarinic receptor. These data support our earlier studies suggesting that CK1␣ is a receptor kinase involved in GPCR phosphorylation. Furthermore, by using a receptor mutant that shows reduced agonist-mediated phosphorylation, we show that phosphorylation is not associated with desensitization of the peak Ins(1,4,5)P 3 response but may instead be involved in controlling the magnitude of the inositol phosphate response.