Differential G-protein-coupled Receptor Phosphorylation Provides Evidence for a Signaling Bar Code

Materials

Unless otherwise stated, all biochemicals and reagents were from Sigma or from previously identified sources (5). Radioisotope [³²P]orthophosphate (specific activity 8500–9120 Ci/mmol), myo-[³H]inositol (specific activity 25 Ci/mmol), and N-[³H]methylscopolamine (specific activity 83 Ci/mmol) were from PerkinElmer Life Sciences. The anti-muscarinic receptor antibodies used in this study were generated in-house against a glutathione S-transferase (GST) bacterial fusion protein containing the full third intracellular loop of the mouse M₃-muscarinic receptor (Arg²⁵²–Thr⁴⁹¹). Where rabbit polyclonal antibodies were raised, the antibodies were first passed through a GST-affinity column to remove antibodies to GST and then purified by affinity chromatography on a column containing the receptor-fusion protein. In the case of mouse monoclonal antibodies, affinity purification against the antigen was carried out.

Phosphorylation-specific antibodies, anti-phosphoserine 384, 412, and 577, were raised against peptide sequences NSTKLPpSSDNLQ, HKLQAQKpSMDDRD, and YQQRQpSVIFHK, corresponding to amino acid residues 378–389, 405–418, and 572–583 of the mouse M₃-muscarinic receptor in which serines 384, 412, and 577 were phosphorylated. The 87-day programs, which included four immunizations, were performed by Eurogentec. The resulting antisera were purified against the immunizing peptides.

Generation of GST Fusion Constructs

Mouse M₃-muscarinic receptor third intracellular loop Arg²⁵² to Thr⁴⁹¹ and the C-terminal Lys⁵⁴⁸ to Leu⁵⁸⁹ were inserted into pGEX-2t to produce N-terminal GST fusions. Escherichia coli BL21 (DE3) IRL transformed with the fusion constructs or pGEX-2t alone was grown in LB medium containing 50 μg/ml ampicillin, 50 μg/ml chloramphenicol, and 1% w/v glucose; protein expression was induced by addition of isopropyl 1-thio-β-d-galactopyranoside to a final concentration of 200 μm.

Culture of CHO-M₃ Wild-type Stable Cell Lines

CHO cells stably expressing the wild-type M₃-muscarinic receptor were maintained in Ham's F-12 medium (Invitrogen) supplemented with 10% fetal calf serum (FCS), penicillin (50 units/ml), streptomycin (50 μg/ml), and geneticin G418 (500 μg/ml). Experiments were performed in Krebs/HEPES buffer (118 mm NaCl, 1.3 mm CaCl₂, 4.3 mm KCl, 1.17 MgSO₄, 4.17 mm NaHCO₃, 1.18 mm KH₂PO₄, 11.7 mm glucose, 10 mm HEPES (pH 7.4)) or in a modified Krebs/HEPES buffer as indicated.

Preparation and Primary Culture of Mouse Cerebellar Granule Neurons

Mouse CG neurons were prepared and cultured as described previously (5), trypsin- dissociated, and plated on poly-d-lysine-coated 6-well plates at a density of 2 × 10⁶ cells well. The neurons were maintained in Eagle's basal medium (Invitrogen) supplemented with 20 mm KCl, penicillin/streptomycin, and 10% FCS. After 48 h, cytosine arabinoside (10 μm) was added to prevent glial cell proliferation, and the culture was continued for 7–8 days. Experiments were then performed on cells that were washed and then maintained in CSS-25 buffer (120 mm NaCl, 1.8 mm CaCl₂, 25 mm KCl, 15 mm glucose, 25 mm HEPES (pH 7.4)).

M₃-muscarinic Receptor Purification and Mass Spectrometry

For the mass spectrometry experiments, a stably transfected CHO cell line was generated that expressed a mouse M₃-muscarinic receptor HA-tagged at the C terminus. For receptor purification, 20 confluent T175 flasks were harvested and resuspended in 40 ml of Krebs/HEPES buffer and stimulated with methacholine (100 μm, 5 min). Membranes were then prepared and solubilized by addition of 5 ml of PBS containing 1% Nonidet P-40 plus a mixture of protease and phosphatase inhibitors. After centrifugation at 20,000 × g, the resulting supernatant was diluted 1:1 with PBS, and the receptor was then purified on anti-HA affinity resin (Roche Diagnostics). After extensive washing with solubilization buffer containing 0.5% Nonidet P-40, the resin was resuspended in 2× SDS-PAGE sample buffer. The sample was resolved by SDS-PAGE on 8% gels and stained with colloidal Coomassie Blue. Purified M₃-muscarinic receptors were excised from the polyacrylamide and washed three times for 15 min with 100 mm ammonium bicarbonate. Reduction and alkylation of cysteines were performed by addition of 10 mm dithiothreitol (DTT) in 50 mm ammonium bicarbonate at 60 °C for 30 min followed by addition 100 mm iodoacetamide in 50 mm ammonium bicarbonate for 30 min in the dark. Gel slices were washed three times for 5 min with 50 mm ammonium bicarbonate containing 50% acetonitrile and incubated overnight at 37 °C in 50 mm ammonium bicarbonate containing 1 μg of sequencing grade trypsin (Promega).

LC-MS/MS was carried out on each sample using a 4000 Q-Trap mass spectrometer (Applied Biosystems, Warrington, UK). Peptides resulting from in-gel digestion were loaded at a high flow rate onto a reverse-phase trapping column (0.3 mm inner diameter × 1 mm), containing 5 μm of C18 300 Å Acclaim PepMap media (Dionex, UK) and eluted through a reverse-phase capillary column (75 μm inner diameter × 150 mm) containing Symmetry C18 100 Å media (Waters) that was self-packed using a high pressure packing device (Proxeon Biosystems, Odense, Denmark). The output from the column was sprayed directly into the nanospray ion source of the 4000 Q-Trap mass spectrometer. Additional mass spectrometry experiments were performed on an LTQ Orbitrap mass spectrometer by “FingerPrints” Proteomics Facility (College of Life Sciences, University of Dundee).

The resulting spectra were searched against the UniProtKB/SwissProt data base using MASCOT software (Matrix Science Ltd.) with peptide tolerance set to 5 ppm for LTQ OrbiTrap data or 0.1 Da for 4000 Q-Trap data, and the MS/MS tolerance was set to 0.6 Da. Fixed modifications were set as carbamidomethyl cysteine with variable modifications of phosphoserine, phosphothreonine, phosphotyrosine, and oxidized methionine. The enzyme was set to trypsin/P, and up to two missed cleavages were allowed. Peptides with a Mascot score greater than 20 and where the probability (p) that the observed match was a random event was <0.05 were included in the analysis. The spectra of peptides reported as being phosphorylated were interrogated manually to confirm the precise sites of phosphorylation.

[³²P]Orthophosphate Labeling and M₃-muscarinic Receptor Immunoprecipitation

In vivo [³²P]orthophosphate labeling, receptor solubilization, and immunoprecipitation were conducted as described previously (5). In brief, CHO cells stably expressing the human M₃-muscarinic receptor were grown in 6-well plates, washed, and incubated for 1 h in KH₂PO₄-free Krebs buffer containing 100 μCi/ml [³²P]orthophosphate (PerkinElmer Life Sciences). Cells were then stimulated with 0.1 mm methacholine for 5 min and lysed in RIPA buffer (2 mm EDTA, 20 mm β-glycerophosphate, 160 mm NaCl, 1% Nonidet P-40, 0.5% deoxycholate, 10 mm Tris (pH 7.4)). M₃-muscarinic receptors were immunoprecipitated using an in-house anti-M₃-muscarinic receptor polyclonal antibody (5). Immunoprecipitated proteins were resolved by SDS-PAGE on 8% gels, transferred to nitrocellulose membrane, and visualized by autoradiography. The membrane was subsequently blocked and immunoblotted with another in-house anti-mouse M₃-muscarinic receptor monoclonal antibody for the detection of total receptors. To dephosphorylate the immunoprecipitated receptor, the immune complexes were washed three times with 10 mm Tris (pH 7.4) containing 0.25% n-octyl glucoside before treatment with calf intestinal phosphatase (20 units per reaction) for 2 h at 37 °C. Receptors were then resolved by SDS-PAGE on 8% gels and immunoblotted with phosphorylation-specific antibodies.

For phosphopeptide mapping, CHO-M3 cells were labeled with 100 μCi/ml of [³²P]orthophosphate. Stimulation and immunoprecipitation were carried out as described above. The immunoprecipitated samples of one entire 6-well plate of CHO cells were pooled and resolved by SDS-PAGE on 8% gels. Tryptic phosphopeptide maps were prepared as described previously (5). The resolved phosphopeptides were then visualized using a STORM 820 phosphorimager (GE Healthcare) and quantified using AlphaEaseFC software (Alpha Innotech).

M₃-muscarinic Receptor Immunoprecipitation from Tissue Samples

To establish the phosphorylation status of M₃-muscarinic receptors in preparations from the hippocampus, cerebral cortex, and whole pancreas, tissues from adult wild-type or adult M₃-muscarinic receptor knock-out mice on a mixed genetic background (129/SvEv × CF1) (25) were extracted into ice-cold HIB buffer (25 mm HEPES (pH 7.4), 120 mm NaCl, 5 mm KCl, 9.1 mm glucose) containing protease inhibitors and phosphatase inhibitors (Complete EDTA-free, PhosStop, Roche Diagnostics). Membranes were prepared and solubilized in buffer composed of 20 mm Tris (pH 7.4), 150 mm NaCl, 3 mm EDTA, and 1% Nonidet P-40. The receptor was immunoprecipitated with the anti-M₃-muscarinic receptor monoclonal antibody. Immune complexes were washed three times in solubilization buffer and resuspended in 2× SDS-PAGE sample buffer. Receptors were separated by SDS-PAGE on 8% gels, transferred on to nitrocellulose membranes, and immunoblotted with anti-phosphoserine 384, anti-phosphoserine 412, or anti-phosphoserine 577 antibodies. For the detection of total receptors, membranes were stripped and re-probed with an anti-mouse M₃ muscarinic receptor antibody. Western blots were digitized and quantified as described above.

Total [³H]Inositol Phosphate Accumulation Assay

To assess agonist efficacy, total [³H]inositol phosphate accumulation assay was carried out CHO-m3 cells. In brief, cells seeded at 100,000 cells/well in 24-well plates were labeled with 2.5 μCi/ml myo-[³H]inositol (PerkinElmer Life Sciences) for 24 h at 37 °C. Cells were washed twice in Krebs/HEPES buffer (pH 7.4) and incubated with 10 mm LiCl for 20 min at 37 °C. Appropriate concentrations of agonist were added for 10 min to stimulate [³H]inositol phosphate production. Reactions were terminated by rapid aspiration of the buffer and addition of 1 m ice-cold trichloroacetic acid (500 μl). Cell lysates were transferred to centrifuge tubes and incubated with 10 mm EDTA (50 μl) and 1,1,2-trichlorofluoroethane/tri-n-octylamine (1:1, v/v, 500 μl) for 15 min at room temperature. Samples were centrifuged at 20,000 × g for 4 min. An ∼400-μl aliquot from the top layer was recovered and transferred to fresh tubes containing 60 mm NaHCO₃. [³H]Inositol mono-, bis-, and trisphosphate ([³H]InsPx) fraction was recovered by anion exchange chromatography as follows: Dowex-1 (formate form) columns were regenerated with 10 ml of ammonium formate (2 m), formic acid (0.1 m) and washed thoroughly with distilled water. Samples were applied to the columns, and the columns washed with 10 ml of distilled water. Columns were then washed with ammonium formate (60 mm), sodium tetraborate (10 mm) solution. Total [³H]InsPx was eluted in 10 ml of ammonium formate (0.75 m), formic acid (0.1 m) and collected in large scintillation vials. A 5-ml aliquot from the eluate was mixed with 10 ml of SafeFluor scintillation mixture, and radioactivity was detected by liquid scintillation counting.

Muscarinic Receptor Expression Levels

The expression levels of the M₃-muscarinic receptor were determined in membranes isolated from the hippocampus, cortex, and submandibular salivary gland by subtracting the specific binding of the muscarinic antagonist N-[³H]methylscopolamine obtained from wild-type mice from that obtained from M₃-muscarinic receptor knock-out mice. N-[³H]Methylscopolamine binding was conducted as described previously (13). The M₃-muscarinic receptor expression levels in hippocampus, cortex, and submandibular gland were 258, 351, and 335 fmol/mg protein, respectively.

Immunocytochemistry of Salivary Submandibular Glands

The submandibular glands from wild-type mice or M₃-muscarinic receptor gene knock-out mice were dissected and fixed in 10% neutral buffered formalin. The fixed tissue were embedded in paraffin and processed on a Shandon Citadel 2000 where serial sections (5 μm) were cut. Sections where dewaxed in xylene, dehydrated through alcohol, and then washed with distilled water. The sections were then treated for microwave antigen retrieval in a Sanyo Shower Wave 1000-watt microwave, at 700 watts for 20 min in a 0.01 m citrate buffer (pH 6.0). Sections were then permeabilized with Triton X-100, and the M₃-muscarinic receptor was stained with either anti-M₃ muscarinic receptor antibodies or receptor phospho-specific antibodies used at a dilution of 1:200 using methods described previously (5, 14).

Data Analysis

Data presented are means ± S.E. of at least three determinations, and statistically significance differences were determined using analysis of variance followed by the Bonferroni post test. Significance was accepted at p < 0.05.

Phosphorylation Profile of the M₃-muscarinic Receptor in Three Cell Types

The phosphorylation profile of the mouse M₃-muscarinic receptor expressed in Chinese hamster ovary (CHO) cells as a recombinant protein or as an endogenously expressed protein in the mouse insulinoma cell line, MIN6, and mouse primary CG neurons was investigated. The receptor was immunoprecipitated from [³²P]orthophosphate-labeled cells and digested with trypsin. The tryptic phosphopeptides were then resolved by two-dimensional chromatography, and an autoradiograph was obtained to provide a phosphopeptide map (Fig. 1A).

The phosphopeptide map from the three cell types revealed a quartet of phosphopeptides labeled 1–4 that served to orient the map (Fig. 1B). These peptides were seen in all three cell types as were phosphopeptides labeled A and X (Fig. 1B). In contrast, there were a number of phosphopeptides that were observed in only one of the three cell types. Thus, phosphopeptides labeled with three asterisks were only observed in receptors derived from CHO cells (Fig. 1B). Four phosphopeptides labeled with two asterisks were only observed in CG neurons, and the phosphopeptide labeled with two plus signs was only observed in insulinoma cells. There were some phosphopeptides that were shared just between two cell types as illustrated by phosphopeptides Y and B that are shared between CHO cells and CG neurons.

These studies demonstrated that although there were common features in the phosphorylation profiles of the M₃-muscarinic receptor in the three cell types investigated, there were also specific differences that make the overall phosphorylation profile of the M₃-muscarinic receptor expressed in each of the cell types unique.

Mass Spectrometry Identification of Phospho-acceptor Sites and Generation of Phospho-specific Antibodies

Although the phosphopeptide maps described above illustrate differential receptor phosphorylation, these studies do not reveal the precise sites of phosphorylation. To do this, we conducted a mass spectrometric analysis of the phospho-acceptor sites on the M₃-muscarinic receptor expressed in CHO cells. These studies revealed 13 serine and 2 threonine phospho-acceptor sites in the third intracellular loop and one serine (Ser⁵⁷⁷) in the C-terminal tail (Fig. 2).

Phospho-specific antibodies were raised to phosphorylated Ser³⁸⁴ and Ser⁴¹² in the third intracellular loop and Ser⁵⁷⁷ in the C-terminal tail. These were chosen because they represented phosphorylated residues that were not within a cluster of phosphorylation sites (which would have made specific antibody generation difficult) and that were sites determined to have high antigenicity. The antiserum from each of the immunizations was subjected to affinity purification that separated nonphospho-specific antibodies (structural antibodies) and phospho-specific antibodies. To test the specificity of the purified antiserum, the nonphosphorylated epitopes present in bacterial glutathione S-transferase (GST) fusion proteins containing either the third intracellular loop of the receptor or the C-terminal tail were probed in Western blots with the structural antibody and the phospho-specific receptor antibodies (Fig. 3, A–C). In these experiment, the structural antibodies reacted with the appropriate fusion protein. The phospho-specific antibody fraction did not recognize the GST fusion proteins (Fig. 3, B and C). The phospho-specific antibodies did, however, identify the intact M₃-muscarinic receptor immunoprecipitated from transfected CHO cells (Fig. 3D). This immunoreactivity was lost following de-phosphorylation with calf intestinal phosphatase (Fig. 3D).

Determination of Cell Type-specific M₃-muscarinic Receptor Phosphorylation Using Receptor Phospho-specific Antibodies

The phosphorylation status of the M₃-muscarinic receptor at residues 384, 412, and 577 was investigated on receptors expressed in CHO cells, CG neurons, and MIN6 insulinoma cells (Fig. 4). Receptors derived from CHO cells showed high basal levels of phosphorylation on Ser³⁸⁴ that fell by 82% following receptor stimulation with the full agonist methacholine. In contrast, agonist stimulation increased phosphorylation on Ser³⁸⁴ by 28% in receptors expressed in CG neurons (Fig. 4A). In MIN6 cells, the receptor was basally phosphorylated on Ser³⁸⁴ and was not changed following agonist stimulation (Fig. 4B). These data demonstrate that phosphorylation at Ser³⁸⁴ was regulated differentially in the three cell types.

Analysis of the phosphorylation status of the receptor at Ser⁵⁷⁷ also showed the importance of cell background. In CHO cells and CG neurons, Ser⁵⁷⁷ phosphorylation was increased following agonist stimulation by 61 and 50%, respectively (Fig. 4, A and B). Interestingly there was no evidence of basal or agonist-mediated phosphorylation at Ser⁵⁷⁷ from receptors expressed in MIN6 cells (Fig. 4B).

Similarly, there was clear agonist-mediated phosphorylation at Ser⁴¹² from receptors derived from CHO cells and CG neurons. However, there was no evidence for Ser⁴¹² phosphorylation on receptors expressed in MIN6 cells (Fig. 4, A and B).

Analysis of in Vivo M₃-muscarinic Receptor Phosphorylation

To evaluate differential phosphorylation of the M₃-muscarinic receptor in vivo, membranes were prepared from the hippocampus, cerebral cortex, and pancreas from either wild-type mice or from mice where the M₃-muscarinic receptor gene had been knocked out. The membranes were then solubilized; the M₃-muscarinic receptor was immunoprecipitated with a receptor-specific monoclonal antibody, and the immunoprecipitate was probed in Western blots with receptor phospho-specific antibodies. In these experiments, the M₃-muscarinic receptor expressed in the hippocampus was determined to be phosphorylated equally on Ser³¹⁸ and Ser⁴¹² but was phosphorylated to a lesser extent on serine⁵⁷⁷ (Fig. 5A). The phosphorylation profile for the receptor appeared different in the cerebral cortex where the receptor was equally phosphorylated on Ser³¹⁸ and Ser⁵⁷⁷ but less phosphorylated on Ser⁴¹² (Fig. 5A). The receptor showed a further unique phosphorylation profile in the pancreas where the receptor was generally poorly phosphorylated, showing a weak signal for phosphorylation on Ser³¹⁸ and Ser⁴¹² but no detectable phosphorylation of Ser⁵⁷⁷ (Fig. 5A).

The molecular weight of the M₃-muscarinic receptor immunoprecipitated from the pancreas (Fig. 5A) and from MIN6 insulinoma cells (Fig. 4B) was observed to be greater than that in the brain and CHO cells. The reason for this is likely to be differential glycosylation because the receptor was poorly phosphorylated in pancreatic cell/tissue types; hence, phosphorylation was not likely to be a major contributory factor in the apparently high molecular weight of the receptor in these experiments.

We also investigated the phosphorylation status of the M₃-muscarinic receptor in salivary submandibular glands. However, we were not able to detect the receptor in Western blots from this tissue probably because the receptor was subjected to proteolysis. We were, however, able to detect the receptor by immunocytochemistry where the M₃-muscarinic receptor was determined to be expressed in the excretory cells as well as in the cells of the excretory ducts (Fig. 5B). In these experiments weak staining for the receptor phosphorylated on Ser³⁸⁴ and Ser⁴¹² was detected, although there was stronger staining for phosphorylation on Ser⁵⁷⁷ (Fig. 5B).

Hence, the phosphorylation profile of the M₃-muscarinic receptor was different in each of the four tissues. This difference is illustrated as a bar code in Fig. 5C.

Differential Regulation of Phosphorylation by Different Muscarinic Receptor Agonists

The studies above provide evidence that the phosphorylation status of the M₃-muscarinic receptor is dependent on the cell and tissue type in which the receptor is expressed. We extended this analysis by investigating if different agonists working at the receptor expressed in the same cell type could mediate different receptor phosphorylation profiles. To this end, the phosphorylation profile of the human M₃-muscarinic receptor expressed in CHO-m3 cells following stimulation by the agonists methacholine, pilocarpine, and arecoline was investigated. Previous studies had established that methacholine acts as a full agonist at muscarinic receptors and pilocarpine and arecoline as partial agonists (15, 16).

In this study, assays of inositol phosphate accumulation (InsPx) established methacholine as a full agonist at the M₃-muscarinic receptor and arecoline and pilocarpine as partial agonists (Fig. 6A). Consistent with previous studies (15, 16), the rank order of potency and maximal response (efficacy) for the InsPx response was methacholine > arecoline > pilocarpine (Fig. 6A). In phosphorylation assays, all three agonists stimulated an increase in total receptor phosphorylation with the same rank order of efficacy as that observed in the InsPx assay (Fig. 6B).

To determine the rank order of efficacy of the agonists at individual phosphorylation sites, phosphopeptide maps were generated following stimulation with the three agonists (Fig. 7, A–E). In these experiments, the relative intensity of four spots (labeled 1–4, Fig. 7) on the phosphopeptide map were compared. Spots 1 and 3 were significantly increased over basal when cells when stimulated with methacholine (Fig. 7, B and D). No significant changes in the phosphorylation of spot 1 or 3 were observed following stimulation with either arecoline or pilocarpine (Fig. 7B). Interestingly, spot 2 was highly phosphorylated in the basal state but became de-phosphorylated following methacholine stimulation (Fig. 7C). De-phosphorylation of spot 2 was also observed, but to a lesser extent, following stimulation with arecoline (Fig. 7C). No significant de-phosphorylation was observed in spot 2 with pilocarpine (Fig. 7C). Hence, the changes in phosphorylation observed for spots 1–3 more or less followed the expected rank order of efficacy for the three agonists. This, however, was in contrast to the phosphorylation of spot 4. Here, all three agonists increased phosphorylation at spot 4 with equivalent efficacy (Fig. 7E).

To further investigate this phenomenon, phospho-specific antibodies to Ser³⁸⁴, Ser⁴¹², and Ser⁵⁷⁷ were used to probe the effects of the three agonists on specific phosphorylation events on the M₃-muscarinic receptor in CHO-m3 cells. The receptor showed high basal levels of Ser³⁸⁴ phosphorylation, which were significantly decreased by methacholine treatment (Fig. 8A). Here, arecoline and pilocarpine also decreased phosphorylation at Ser³⁸⁴ with the following rank order of efficacy: methacholine = arecoline > pilocarpine (Fig. 8A). A similar decrease in the phosphorylation status of spot 2 in Fig. 7 was observed suggesting that spot 2 corresponds to the phosphorylation site Ser³⁸⁴.

Ser⁴¹² phosphorylation was relatively low in the basal state and was increased equivalently by all three agonists (Fig. 8B). This is the same as the effects of the three agonists on the phosphorylation status of spot 4 in Fig. 7, suggesting that spot 4 corresponds to Ser⁴¹².

Finally, phosphorylation of Ser⁵⁷⁷ was increased by all three agonists. The rank order of efficacy was methacholine > arecoline > pilocarpine.

Thus, analysis of the phosphorylation status of the M₃-muscarinic receptor with phospho-specific antibodies indicated that of the phosphorylation sites investigated only agonist-mediated phosphorylation at Ser⁵⁷⁷ followed the expected rank order of efficacy (i.e. methacholine > arecoline > pilocarpine). The rank order of efficacy for the changes in phosphorylation at the other two sites (Ser³⁸⁴ and Ser⁴¹²) did not conform to the expected order indicating that agonists were able to direct phosphorylation preferentially at these sites.