J Biol Chem, Vol. 274, Issue 39, 27573-27577, September 24, 1999

A Relationship between Protein Kinase C Phosphorylation and Calmodulin Binding to the Metabotropic Glutamate Receptor Subtype 7^*

Yoshiaki Nakajima, Tomomi Yamamoto, Tatsuo Nakayama^§, and Shigetada Nakanishi^¶

From the  Department of Cell Physiology, National Institute for Physiological Sciences, Okazaki 444-8585, the ^§ Department of Biochemistry, Miyazaki Medical College, Kiyotake, Miyazaki 889-1692, and the ^¶ Department of Biological Sciences, Kyoto University Faculty of Medicine, Kyoto 606-8501, Japan

	ABSTRACT

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Metabotropic glutamate receptor subtype 7 (mGluR7) is coupled to the inhibitory cyclic AMP cascade and is selectively activated by a glutamate analogue, L-2-amino-4-phosphonobutyrate. Among L-2-amino-4-phosphonobutyrate-sensitive mGluR subtypes, mGluR7 is highly concentrated at the presynaptic terminals and is thought to play an important role in modulation of glutamatergic synaptic transmission by presynaptic inhibition of glutamate release. To gain further insight into the intracellular signaling mechanisms of mGluR7, with the aid of glutathione S-transferase fusion affinity chromatography, we attempted to identify proteins that interact with the intracellular carboxyl terminus of mGluR7. Here, we report that calmodulin (CaM) directly binds to the carboxyl terminus of mGluR7 in a Ca²⁺-dependent manner. The CaM-binding domain is located immediately following the 7th transmembrane segment. We also show that the CaM-binding domain of mGluR7 is phosphorylated by protein kinase C (PKC). This phosphorylation is inhibited by the binding of Ca²⁺/CaM to the receptor. Conversely, the Ca²⁺/CaM binding is prevented by PKC phosphorylation. Collectively, these results suggest that mGluR7 serves to cross-link the cyclic AMP, Ca²⁺, and PKC phosphorylation signal transduction cascades.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Metabotropic glutamate receptors (mGluRs)¹ belong to the class of seven transmembrane domain receptors and consist of eight different subtypes (mGluR1 to mGluR8) (1-4). They are coupled to intracellular signal transduction mechanisms via G proteins and exert their effects on second messengers and ion channels (2-4). The eight mGluR subtypes can be classified into three groups (2-4). Group 1 receptors (mGluR1 and mGluR5) are coupled to the stimulation of the inositol trisphosphate (IP₃)/Ca²⁺ signaling pathway. Group 2 (mGluR2 and mGluR3) and group 3 (mGluR4, mGluR6, mGluR7, and mGluR8) receptors are coupled to the inhibitory cyclic AMP cascade in heterologously expressing cells but differ in their agonist selectivity. Among the mGluR family, mGluR7 is the most highly conserved across different mammalian species (5) and is widely distributed throughout the central nervous system (6-9). This receptor is selectively activated by L-2-amino-4-phosphonobutyrate (L-AP4) (10, 11). L-AP4 has been shown to exert a potent presynaptic inhibition of glutamate release (12, 13). At synapses, mGluR7 is located in close proximity to synaptic vesicle release sites (9, 14, 15). Recent gene targeting analysis has indicated that mGluR7 deficiency causes a reduction in high frequency synaptic transmission, post-tetanic potentiation, and short term potentiation in the CA1 synapses of hippocampal slices (16). In behavioral analyses, these knockout mice showed a deficit in fear response and conditioned taste aversion (17). Therefore, mGluR7 has been postulated to play an important role in synaptic modulation and plasticity. However, it remains elusive whether a variety of cellular and physiological functions of mGluR7 all result from coupling to the inhibitory cyclic AMP cascade mediated by this receptor subtype.

Recent biochemical and molecular studies have indicated that protein-protein interactions play a pivotal role in regulation and signal transduction of group 1 mGluRs (18-20). The identification of molecules that interact with receptors would thus provide an important clue for understanding the receptor function. In this study, we attempted to identify proteins that interact with mGluR7 by glutathione S-transferase (GST) affinity chromatography using the intracellular carboxyl terminus of mGluR7 (ct-mGluR7). We show that a 17-kDa protein, which was identified as calmodulin (CaM), binds directly to ct-mGluR7 in a Ca²⁺-dependent manner. In addition, we show that the Ca²⁺/CaM binding is inhibited by protein kinase C (PKC)-evoked phosphorylation. Furthermore, PKC phosphorylation of ct-mGluR7 is inhibited by Ca²⁺/CaM binding.

	EXPERIMENTAL PROCEDURES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Materials-- Materials were purchased from the following sources: bovine CaM from Sigma; mouse monoclonal antibody against CaM (anti-CaM mAb) from Upstate Biotechnology, Inc. (Lake Placid, NY); rat brain PKC from Calbiochem; catalytic subunit of bovine heart protein kinase A (PKA) from Roche Molecular Biochemicals; [-³²P]ATP from Amersham Pharmacia Biotech.

Recombinant Proteins-- ct-mGluR2 (Gln⁸²⁰-Leu⁸⁷²), ct-mGluR3 (Gln⁸²⁹-Leu⁸⁷⁹), ct-mGluR4 (His⁸⁴⁸-Ile⁹¹²), ct-mGluR6 (His⁸⁴⁰-Lys⁸⁷¹),and ct-mGluR7 (His⁸⁵¹-Ile⁹¹⁵) were amplified, using the corresponding mGluR cDNAs (10, 21, 22) as templates by polymerase chain reaction (PCR). ct-mGluR8 (His⁸⁴⁴-Ile⁹⁰⁸) was amplified with rat brain total RNA by reverse transcriptase-mediated PCR (23). Primers for PCR or reverse transcriptase-mediated PCR were designed as follows: the nucleotide sequence immediately following the 7th transmembrane segment and covering 19-21 base pairs of the downstream sequence of each individual ct-mGluR was preceded by an appropriate restriction cleavage site and used as a 5' primer. The sequence containing a stop codon and the upstream sequence of each respective ct-mGluR was followed by a restriction cleavage site and used as a 3' primer. GST fusion proteins containing different ct-mGluRs (GST-ct-mGluRs) were generated by inserting the PCR products in-frame into a multiple cloning site downstream of the GST-coding region of the pGEX4T series of vectors (Amersham Pharmacia Biotech). Truncated forms of ct-mGluR7 fused to the GST protein were similarly constructed by starting with an appropriate amino-terminal sequence in ct-mGluR7 and ending with the TAA stop codon. All ct or truncated forms of mGluR7 were designed to contain the common linker amino acid sequence between GST and the inserts. A proper in-frame insertion and the absence of any sequence errors of the PCR products were confirmed by DNA sequencing in both strands of all constructs. The GST fusion proteins were expressed in an Escherichia coli strain BL21 and purified by glutathione-Sepharose 4B beads (Amersham Pharmacia Biotech). The GST-ct-mGluR7 protein was further purified by cation exchange chromatography on a MonoS column (Amersham Pharmacia Biotech) using a gradient elution from 150 to 405 mM NaCl in a solution containing 50 mM Hepes, pH 7.0, 2 mM EDTA, and 1 mM dithiothreitol. Fractions containing the GST-ct-mGluR7 protein were combined, dialyzed against 25 mM Hepes, pH 7.4, containing 150 mM NaCl, and then concentrated with Ultrafree (Millipore, Bedford, MA).

Affinity Chromatography of Brain Extracts-- Brains from adult Sprague-Dawley rats were homogenized on ice using a glass-Teflon homogenizer (20 strokes) in Buffer A (10 mM Hepes, pH 7.5, 150 mM NaCl, 2 µg/ml pepstatin, 2 µg/ml leupeptin, 2 µg/ml aprotinin, and 1 mM phenylmethylsulfonyl fluoride) and centrifuged at 100,000 × g for 1 h at 4 °C. The supernatant was precleared by passing through Sepharose 4B beads. GST fusion proteins (100 µg) were immobilized on glutathione-Sepharose 4B beads (50 µl) and incubated with the precleared brain supernatant for 4 h at 4 °C. The beads were washed with Buffer A five times, and bound proteins were eluted by adding the sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) loading buffer. Proteins were separated by SDS-PAGE (4-20%) and stained with Coomassie Brilliant Blue R-250. The concentration of proteins was measured using the DC protein assay kit (Bio-Rad) with bovine serum albumin as a standard.

Amino Acid Sequencing-- Amino acid sequences were determined essentially according to the method of Matsudaira (24). Proteins were subjected to SDS-PAGE and electrotransferred to a polyvinylidene difluoride membrane (Schleicher & Schuell). The blotted membrane was briefly stained with Coomassie Brilliant Blue R-250. After extensive washing with water, membrane pieces containing the proteins of interest were excised and then examined with a Procise 492 gas-phase sequencer (Applied Biosystems Division, Perkin Elmer).

In Vitro Binding Assay-- GST or GST fusion proteins (2 µg each) were immobilized on glutathione-Sepharose 4B beads (20 µl). CaM (1 µg) was incubated with GST or GST fusion protein-immobilized beads in 500 µl of Buffer B (25 mM Hepes, pH 7.4, 150 mM NaCl, and 1% Triton X-100) in the presence of either 2 mM CaCl₂ or 5 mM EGTA without addition of CaCl₂ for 2 h at 4 °C. The beads were washed with the incubation buffer, and bound proteins were eluted by the SDS-PAGE loading buffer and incubated for 30 min at 60 °C. Proteins were separated by SDS-PAGE (15% gel) and bound CaM was visualized by immunoblotting with anti-CaM mAb. Chemiluminescence of bound CaM was quantitated by densitometric analysis using The Discovery Series (pdi, Huntington Station, NY). An affinity of CaM for binding to GST fusion proteins was determined by immobilizing a fixed amount of a GST fusion protein (60 pmol) on glutathione-Sepharose 4B beads, followed by incubation with increasing concentrations of CaM. Bound CaM was immunoblotted after SDS-PAGE, and amounts of the bound CaM were calculated from the standard curve with known amounts of CaM run on the same gel. Densities of immunoblotted CaM showed a linear relationship with known amounts of CaM, and each data point was obtained within the linear range of the standard curve. Saturation curves were made by fitting the data using the single site relationship B = B_max × (F/K_d)/(1 + F/K_d), where B is the amount bound, F is the amount free, K_d is the dissociation constant, and B_max represents the maximal number of binding sites. Fitting and calculation of K_d were done by using the Origin software (Microcal Software, Northampton, MA).

Phosphorylation of Fusion Proteins-- GST fusion proteins (60 pmol) were incubated with none or increasing amounts of CaM for 2 h at 4 °C. For phosphorylation of PKC or PKA, the incubation mixture (10 µl) contained 8.75 mM Hepes, pH 7.4, 52.5 mM NaCl, and 1 mM CaCl₂. Phosphorylation was started by addition of the following solution (10 µl): for PKC phosphorylation, 40 mM Tris-Cl, pH 7.5, 20 mM MgCl₂, 200 µg/ml L--phosphatidyl-L-serine, 40 µg/ml 1,2-dioleoyl-sn-glycerol, 200 µM [-³²P]ATP (50 mCi/mmol), and 9.6 milliunits of PKC; for PKA phosphorylation, 100 mM MES, pH 6.9, 20 mM MgCl₂, 2 mg/ml bovine serum albumin, 2 mM dithiothreitol, 400 µM [-³²P]ATP (50 mCi/mmol), and 10 microunits of catalytic subunit of PKA. Reactions were carried out at 30 °C for the times indicated and stopped by boiling in the SDS-PAGE loading buffer. The phosphorylated proteins were separated by SDS-PAGE (15% gel), fixed, dried, and finally exposed to an x-ray film. To determine effects of PKC phosphorylation on CaM binding, each of the fusion proteins (2 µg) was phosphorylated for 3 h at 30 °C as described above except that [-³²P]ATP was replaced with non-labeled ATP. As a control, ATP was depleted in the phosphorylation reaction.

	RESULTS

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

Identification of an mGluR7-interacting Protein-- We attempted to identify proteins that interact with mGluR7 by affinity chromatography using the ct region of mGluR7 fused to GST. GST alone or GST-ct-mGluR2 was used as control. Rat brain cytosolic fractions were prepared and loaded to glutathione-Sepharose 4B beads coated with each GST protein. After extensive washing, bound proteins were eluted by addition of the SDS-PAGE loading buffer, separated by SDS-PAGE, and stained with Coomassie Brilliant Blue R-250. A prominent band with a mobility of approximately 17 kDa was detected in the eluate from GST-ct-mGluR7 affinity beads (Fig. 1, lane 2). No such 17-kDa protein was retained with affinity beads attached with (lane 3) or without (lane 4) GST. The specificity of interaction toward ct-mGluR7 was confirmed by using GST-ct-mGluR2 which showed no detectable binding of the 17-kDa protein (lane 1). Peptide sequencing of this 17-kDa protein and subsequent data base analysis indicated that the partial sequence determined (Thr-Ile-Asp-Phe-Pro-Glu-Phe-Leu-Thr-Met-Met-Ala-Arg-Lys-Met-Lys-Asp) precisely corresponded to the sequence of rat CaM.

View larger version (60K): [in this window] [in a new window]   Fig. 1.   SDS-PAGE analysis showing specific binding of ct-mGluR7 to CaM in brain extracts. Rat brain cytosolic extracts were incubated with glutathione-Sepharose 4B beads attached with GST or GST-ct-mGluRs. Proteins retained by these beads were eluted, separated by SDS-PAGE (4-20% gradient gel), and stained with Coomassie Brilliant Blue R-250. The following GST or GST fusion proteins were immobilized on beads: lane 1, GST-ct-mGluR2; lane 2, GST-ct-mGluR7; lane 3, GST alone; lane 4, without any immobilized GST proteins. Lane M, molecular mass markers (kDa).

Characterization of CaM Binding to ct-mGluR7-- Next, we examined the specificity of interaction between CaM and the ct regions of group 2 and group 3 mGluRs using GST fusion protein affinity chromatography. The GST proteins fused to the ct regions of group 2 and group 3 mGluRs were immobilized on glutathione-Sepharose 4B beads and tested for their ability to retain bovine CaM (Fig. 2, upper). CaM interacted exclusively with ct-mGluR7. In contrast, no detectable interaction was observed with group 2 ct-mGluRs and other members of group 3 ct-mGluRs as well as GST alone. Furthermore, the interaction between CaM and ct-mGluR7 was completely abolished by replacing 2 mM CaCl₂ with 5 mM EGTA in the binding solution. Staining of the SDS-polyacrylamide gel with Coomassie Brilliant Blue R-250 confirmed that comparable amounts of GST or GST fusion proteins were present in these experiments (Fig. 2, lower). The results indicate that among group 2 and group 3 mGluR subtypes, CaM interacts specifically with ct-mGluR7 in a Ca²⁺-dependent manner.

View larger version (31K): [in this window] [in a new window]   Fig. 2.   Selective Ca2+-dependent binding of CaM to ct-mGluR7. The whole ct regions of group 2 and group 3 mGluRs as indicated by numbers were fused to GST. The resultant GST fusion proteins or GST alone were immobilized on glutathione-Sepharose 4B beads and tested for their ability to bind to CaM in the presence of either 2 mM CaCl2 or 5 mM EGTA without addition of CaCl2. One-fifth of bound proteins was separated by SDS-PAGE, and bound CaM was detected by immunoblotting with anti-CaM mAb in the upper panel. CaM (100 ng) was run as a control (control lane). Sizes of molecular mass markers (kDa) are shown on the left. The lower panel shows a Coomassie Brilliant Blue R-250 staining of the GST or GST fusion proteins immobilized on glutathione-Sepharose 4B beads.

To define a CaM-interacting domain in ct-mGluR7, we constructed a series of GST fusion proteins possessing different truncation forms of ct-mGluR7 (Fig. 3A). The truncated forms containing at least residues Val⁸⁵⁶-Leu⁸⁹² (Tr1-mGluR7 and Tr4-mGluR7) showed a strong Ca²⁺-dependent interaction with CaM comparable to that of ct-mGluR7 (Fig. 3B, upper). The presence and absence of Cys⁸⁹³, which is implicated as a possible palmitoylation site (25), had no effect on Ca²⁺/CaM binding. In contrast, removal of a cluster of basic residues at the amino-terminal region of ct-mGluR7 drastically reduced Ca²⁺/CaM binding (Tr2-mGluR7 and Tr3-mGluR7). In these experiments, comparable amounts of GST fusion proteins were confirmed by Coomassie Brilliant Blue R-250 staining (Fig. 3B, lower).

View larger version (45K): [in this window] [in a new window]   Fig. 3.   Identification of the CaM-binding domain in ct-mGluR7. A, amino acid sequences of ct-mGluR7, ct-mGluR4, and ct-mGluR8 and various truncated forms (Tr1-Tr4) of ct-mGluR7. A cluster of basic amino acids at the amino-terminal region (), 4 serine residues showing possible PKC phosphorylation sites (), and a cysteine residue indicating a putative palmitoylation site (*) are marked above the ct-mGluR7 sequence. Amino acids identical between mGluR4, mGluR7, and mGluR8 are indicated by line connections. B, in the upper panel, binding of CaM to GST, GST-ct-mGluR7, and four different GST-Tr-mGluRs was analyzed in the presence of either 2 mM CaCl2 or 5 mM EGTA. In the lower panel, the amounts of GST or GST fusion proteins immobilized on beads are shown. For other explanations, see Fig. 2 legend.

The affinity of Ca²⁺-dependent CaM binding to ct-mGluR7 and Tr4-mGluR7 was determined by incubating a fixed amount of the GST fusion proteins immobilized onto glutathione-Sepharose 4B beads with increasing concentrations of CaM in the presence of 2 mM CaCl₂ (Fig. 4, A and B). Analysis of saturation curves of bound CaM showed virtually identical values of dissociation constant (K_d) of CaM binding for ct-mGluR7 and Tr4-mGluR7, 38.9 ± 8.3 nM for ct-mGluR7 and 45.5 ± 13.6 nM for Tr4-mGluR7 (mean ± S.D., n = 2). The results indicate that the segment consisting of Val⁸⁵⁶-Leu⁸⁹² is sufficient for interaction between CaM and ct-mGluR7.

View larger version (12K): [in this window] [in a new window]   Fig. 4.   Determination of affinities of CaM binding to ct-mGluR7 and Tr4-mGluR7. GST-ct-mGluR7 (A) and GST-Tr4-mGluR7 (B) (60 pmol each) were immobilized on glutathione-Sepharose 4B beads and incubated with increasing concentrations of CaM in the presence of 2 mM CaCl2. Amounts of bound CaM were determined from densitometric analysis of immunoblotting with anti-CaM mAb after SDS-PAGE. Data points and bars show mean ± S.D., respectively, in representative experiments done in duplicate.

Competition between PKC Phosphorylation and CaM Binding of mGluR7 in Vitro-- The CaM-binding domain in ct-mGluR7 contains consensus sequences for both PKC phosphorylation, (Ser/Thr)-X-(Arg/Lys), and PKA phosphorylation, (Arg/Lys)-(Arg/Lys)-X-(Ser/Thr) (26). We examined whether ct-mGluR7 serves as a phosphorylation substrate for PKC or PKA and, if so, whether phosphorylation of these kinases and CaM binding are mutually affected by each other. We first examined the effects of CaM binding on PKC phosphorylation by incubating a fixed amount of either GST-ct-mGluR7 or GST-Tr4-mGluR7 with PKC and [-³²P]ATP in the absence and presence of increasing concentrations of CaM. Time courses and extents of phosphorylation of ct-mGluR7 or Tr4-mGluR7 were determined by autoradiography of a ³²P-labeled product run on an SDS-polyacrylamide gel (Fig. 5). In the absence of CaM, both ct-mGluR7 and Tr4-mGluR7 were rapidly phosphorylated in a time-dependent manner with about one-third of the proteins estimated to be phosphorylated at 1 h in both cases. Neither GST nor CaM was phosphorylated by PKC. Furthermore, there was no obvious difference in the time course and extent of PKC phosphorylation between ct-mGluR7 (Fig. 5A) and Tr4-mGluR7 (Fig. 5B). We also examined possible PKA phosphorylation of GST-ct-mGluR7 by incubating with PKA and [-³²P]ATP. Regardless of the presence and absence of Ca²⁺/CaM, GST-ct-mGluR7 was not appreciably phosphorylated by PKA (data not shown). These results show that the CaM-binding domain is a major site of PKC phosphorylation within the ct-mGluR7 sequence. In contrast, this site does not serve as a target for PKA phosphorylation.

View larger version (28K): [in this window] [in a new window]   Fig. 5.   Inhibitory effect of Ca2+/CaM binding on PKC phosphorylation of GST-ct-mGluR7 and GST-Tr4-mGluR7. A fixed amount of GST-ct-mGluR7 (A) and GST-Tr4-mGluR7 (B) (60 pmol each) was incubated with PKC and [-32P]ATP in the absence and presence of increasing amounts of CaM (18, 60, and 180 pmol) as indicated with molar ratios between the GST fusion proteins and CaM. The reaction was terminated at the indicated times (min), and the reaction product was subjected to SDS-PAGE, followed by autoradiography of a 32P-labeled product. GST alone was also subjected to the phosphorylation reaction in the absence of CaM and run on SDS-PAGE (lane GST). Molecular sizes (kDa) of marker proteins are shown on the left.

Importantly, phosphorylation of both ct-mGluR7 and Tr4-mGluR7 was progressively inhibited by adding increasing amounts of CaM to the reaction mixture. At the molar ratio of 1:1 between the PKC substrates and CaM, phosphorylation of both ct-mGluR7 and Tr4-mGluR7 was almost completely inhibited by Ca²⁺/CaM. Furthermore, no obvious difference in the sensitivity of inhibition of PKC phosphorylation by Ca²⁺/CaM binding was observed between the two PKC substrates. This finding is consistent with the results above, which show a similar affinity of these mGluR7 segments for Ca²⁺/CaM binding. The results indicate that the interaction of Ca²⁺/CaM with the ct domain of mGluR7 prevents phosphorylation by PKC.

Finally, we examined the effect of PKC phosphorylation on Ca²⁺/CaM binding to ct-mGluR7 (Fig. 6). GST-ct-mGluR7 or GST-Tr4-mGluR7 was incubated with PKC in the presence and absence of ATP. The resultant GST fusion proteins were coupled to glutathione-Sepharose 4B beads and tested for their ability to bind to CaM in a Ca²⁺-dependent manner. Non-phosphorylated GST fusion proteins could bind to Ca²⁺/CaM, but once phosphorylated, they lost their ability to bind to Ca²⁺/CaM. The results indicate that Ca²⁺/CaM binding is inhibited by PKC phosphorylation.

View larger version (36K): [in this window] [in a new window]   Fig. 6.   Inhibitory effect of PKC phosphorylation on Ca2+/CaM binding to ct-mGluR7. GST-ct-mGluR7, GST-Tr4-mGluR7, and GST alone (2 µg each) were subjected to PKC phosphorylation reaction for 3 h in the presence and absence of ATP. The resultant GST fusion proteins or GST alone were immobilized on glutathione-Sepharose 4B beads and tested for their ability to bind to CaM in the presence of either 2 mM CaCl2 or 5 mM EGTA. The lower panel shows a Coomassie Brilliant Blue R-250 staining, indicating no difference in retention of the phosphorylated and non-phosphorylated proteins on the glutathione beads.

	DISCUSSION

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES

In this study, in vitro analysis indicates that CaM directly binds to ct-mGluR7 in a Ca²⁺-dependent fashion. Among the mGluR family, mGluR5 has been shown to interact with Ca²⁺/CaM at two distinct sites of ct-mGluR5 with different affinities (27). Consistent with the sequence homology between mGluR1 and mGluR5, we have found that Ca²⁺/CaM also binds to ct-mGluR1.² Interestingly, the group 3 mGluR subtypes (mGluR4, mGluR7, and mGluR8) have highly homologous ct tails, but no significant interaction was observed between CaM and mGluR4 or mGluR8. Therefore, the binding of Ca²⁺/CaM is specific to group 1 mGluRs and mGluR7 within the mGluR family and thus is not relevant to intracellular second messengers involving these receptor subtypes. The CaM-binding domain of ct-mGluR7 diverges from those of group 1 mGluRs but possesses several structural characteristics of CaM binding (28) as follows: 1) one face possessing basic and polar residues in an -helical wheel representation, whereas the other face contains a stretch of hydrophobic amino acids; 2) a cluster of basic amino acids at the amino-terminal portion, followed by a sequence that contains relatively few basic amino acids but often possesses a consensus sequence for phosphorylation (Fig. 3). Interestingly, some of these features are shared by mGluR4 and mGluR8. However, these two subtypes show no appreciable interaction with Ca²⁺/CaM. Therefore, amino acid differences in the middle of the CaM-binding sequence of mGluR7 as compared with mGluR4 and mGluR8 (Fig. 3) seem to be critical for their ability to bind to Ca²⁺/CaM.

The CaM-binding domain of mGluR7 has been located within a segment immediately downstream of the 7th transmembrane segment and upstream of Cys⁸⁹³ in ct-mGluR7. This cysteine residue is commonly present in the ct of group 3 mGluRs (mGluR4, mGluR7, and mGluR8) (Fig. 3) and has been reported to be a putative palmitoylation site, for example in mGluR4 (25). Such palmitoylation is thought to be involved in the formation of the 4th intracellular loop structure. For mGluR5, one of the two CaM-binding domains resides at the corresponding portion immediately following the 7th transmembrane segment (27). This domain of mGluRs, together with their 2nd cytoplasmic loop, has been shown to serve as a G protein-interacting domain (29, 30). Therefore, the overlap of a CaM-binding domain and the G protein-interacting domain suggests an interesting possibility that CaM binding to specific mGluR subtypes may also affect a G protein coupling efficacy.

In this study, using in vitro analysis we have demonstrated that ct-mGluR7 is phosphorylated by PKC but not by PKA. Furthermore, CaM binding and PKC phosphorylation are mutually exclusive. There are several precedents that have shown a similar relationship between CaM binding and phosphorylation by PKC. For example, in mGluR5, PKC phosphorylation is inhibited by Ca²⁺/CaM binding to the receptor, and conversely this binding is prevented by PKC phosphorylation (27). In the N-methyl-D-aspartate (NMDA) glutamate receptors, binding of Ca²⁺/CaM disrupts an interaction between NMDA receptors and the cytoskeletal -actinin-2 (31, 32). Additionally, the CaM binding and protein phosphorylation in NMDA receptors have been reported to be mutually exclusive (33). In both cases, the intracellular Ca²⁺ concentration ([Ca²⁺]_i) is raised by activation of these receptors, either by the enhancement of Ca²⁺ influx through a receptor-channel complex or by mobilization of intracellular Ca²⁺ stores via IP₃ production. These examples therefore differ from the signaling mechanism of mGluR7, which is coupled to the inhibitory cyclic AMP cascade and has no ability to stimulate IP₃ formation or influence Ca²⁺ influx directly (10). In this sense, the reciprocal regulation between CaM binding and PKC phosphorylation of mGluR7 is more reminiscent to the regulation of myristoylated alanine-rich protein kinase C substrate (MARCKS) by the Ca²⁺/CaM binding and protein phosphorylation. MARCKS is a major cellular substrate of PKC, with PKC phosphorylation inhibiting CaM binding and CaM binding preventing PKC phosphorylation. The brain MARCKS is highly concentrated at the presynaptic junction and is thought to be phosphorylated by PKC through an increase in [Ca²⁺]_i triggered by neuronal cell stimulation (34). It has recently been reported that activation of PKC suppresses the ability of group 3 mGluRs to inhibit transmission at glutamatergic synapses (13). Furthermore, mGluR7 is largely located at the presynaptic terminals. Therefore, it is conceivable that mGluR7 is under dual regulation by PKC and CaM at the presynaptic terminals, when neuronal cells are stimulated and increase a Ca²⁺ influx through activated Ca²⁺ channels. Thus, our observations suggest that mGluR7 serves to cross-link the cyclic AMP, Ca²⁺, and PKC phosphorylation signal transduction cascades. In closing, this process may play an important role in modulating synaptic transmission in concert with the function of CaM which recognizes changes in [Ca²⁺]_i.

	FOOTNOTES

^* This work was supported in part by research grants from the Ministry of Education, Science and Culture of Japan, the Sankyo Foundation, the Yamanouchi Foundation, and the Biomolecular Engineering Research Institute.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

To whom correspondence should be addressed: Dept. of Biological Sciences, Kyoto University Faculty of Medicine, Yoshida, Sakyo-ku, Kyoto 606-8501, Japan. Tel.: 81-75-753-4437; Fax: 81-75-753-4404; E-mail: snakanis@phy.med.kyoto-u.ac.jp.

² K. Ishikawa, Y. Nakajima, and S. Nakanishi, unpublished observations.

	ABBREVIATIONS

The abbreviations used are: mGluR, metabotropic glutamate receptor; IP₃, inositol trisphosphate; L-AP4, L-2-amino-4-phosphonobutyrate; ct, carboxyl terminus; GST, glutathione S-transferase; kDa, kilodalton; CaM, calmodulin; PKC, protein kinase C; anti-CaM mAb, mouse monoclonal antibody against CaM; PKA, protein kinase A; PCR, polymerase chain reaction; PAGE, polyacrylamide gel electrophoresis; [Ca²⁺]_i, intracellular Ca²⁺ concentration; K_d, dissociation constant; MARCKS, myristoylated alanine-rich protein kinase C substrate; NMDA, Nmethyl-D-aspartate; MES, 2-(N-morpholino)ethanesulfonic acid.

	REFERENCES

TOP ABSTRACT INTRODUCTION EXPERIMENTAL PROCEDURES RESULTS DISCUSSION REFERENCES