Portal de Periódicos da CAPES

Negative Cooperativity of Glutamate Binding in the Dimeric Metabotropic Glutamate Receptor Subtype 1

2004; Elsevier BV; Volume: 279; Issue: 34 Linguagem: Inglês

10.1074/jbc.m404831200

1083-351X

Yoshikazu Suzuki, Eiko Moriyoshi, Daisuke Tsuchiya, Hisato Jingami,

Nicotinic Acetylcholine Receptors Study

Metabotropic glutamate receptor (mGluR) subtype 1 is a Class III G-protein-coupled receptor that is mainly expressed on the post-synaptic membrane of neuronal cells. The receptor has a large N-terminal extracellular ligand binding domain that forms a homodimer, however, the intersubunit communication of ligand binding in the dimer remains unknown. Here, using the intrinsic tryptophan fluorescence change as a probe for ligand binding events, we examined whether allosteric properties exist in the dimeric ligand binding domain of the receptor. The indole ring of the tryptophan 110, which resides on the upper surface of the ligand binding pocket, sensed the ligand binding events. From saturation binding curves, we have determined the apparent dissociation constants (K0.5) of representative agonists and antagonists for this receptor (3.8, 0.46, 40, and 0.89 μm for glutamate, quisqualate, (S)-α-methyl-4-carboxyphenylglycine ((S)-MCPG), and (+)-2-methyl-4-carboxyphenylglycine (LY367385), respectively). Calcium ions functioned as a positive modulator for agonist but not for antagonist binding (K0.5 values were 1.3, 0.21, 59, and 1.2 μm for glutamate, quisqualate, (S)-MCPG, and LY367385, respectively, in the presence of 2.0 mm calcium ion). Moreover, a Hill analysis of the saturation binding curves revealed the strong negative cooperativity of glutamate binding between each subunit in the dimeric ligand binding domain. As far as we know, this is the first direct evidence that the dimeric ligand binding domain of mGluR exhibits intersubunit cooperativity of ligand binding. Metabotropic glutamate receptor (mGluR) subtype 1 is a Class III G-protein-coupled receptor that is mainly expressed on the post-synaptic membrane of neuronal cells. The receptor has a large N-terminal extracellular ligand binding domain that forms a homodimer, however, the intersubunit communication of ligand binding in the dimer remains unknown. Here, using the intrinsic tryptophan fluorescence change as a probe for ligand binding events, we examined whether allosteric properties exist in the dimeric ligand binding domain of the receptor. The indole ring of the tryptophan 110, which resides on the upper surface of the ligand binding pocket, sensed the ligand binding events. From saturation binding curves, we have determined the apparent dissociation constants (K0.5) of representative agonists and antagonists for this receptor (3.8, 0.46, 40, and 0.89 μm for glutamate, quisqualate, (S)-α-methyl-4-carboxyphenylglycine ((S)-MCPG), and (+)-2-methyl-4-carboxyphenylglycine (LY367385), respectively). Calcium ions functioned as a positive modulator for agonist but not for antagonist binding (K0.5 values were 1.3, 0.21, 59, and 1.2 μm for glutamate, quisqualate, (S)-MCPG, and LY367385, respectively, in the presence of 2.0 mm calcium ion). Moreover, a Hill analysis of the saturation binding curves revealed the strong negative cooperativity of glutamate binding between each subunit in the dimeric ligand binding domain. As far as we know, this is the first direct evidence that the dimeric ligand binding domain of mGluR exhibits intersubunit cooperativity of ligand binding. Glutamate is a major neurotransmitter in the excitatory synapses of the central nervous system, and two types of glutamate receptors are expressed in nerve cells: one is an ionotropic glutamate receptor, and the other is a metabotropic glutamate receptor (mGluR). 1The abbreviations used are: mGluR, metabotropic glutamate receptor; mGluR1, mGluR subtype 1; GPCR, G-protein-coupled receptor; LBD, ligand binding domain; MCPG, α-methyl-4-carboxyphenylglycine; LY367385, (+)-2-methyl-4-carboxyphenylglycine; PEG, polyethylene glycol.1The abbreviations used are: mGluR, metabotropic glutamate receptor; mGluR1, mGluR subtype 1; GPCR, G-protein-coupled receptor; LBD, ligand binding domain; MCPG, α-methyl-4-carboxyphenylglycine; LY367385, (+)-2-methyl-4-carboxyphenylglycine; PEG, polyethylene glycol. The former is a glutamate-gated ion channel, which induces a synaptic potential upon glutamate binding, whereas the latter is a G-protein-coupled receptor (GPCR), which induces various cellular responses to glutamate stimulation, e.g. inositol triphosphate production and the subsequent elevation of intracellular calcium, or a cytoplasmic cyclic AMP concentration change caused by the modulation of adenylyl cyclase activity. Because these cellular responses modulate the degree of synaptic neurotransmission, mGluRs are believed to be involved in higher order neuronal activities such as memory, learning, and so on (1Nakanishi S. Masu M. Annu. Rev. Biophys. Biomol. Struct. 1994; 23: 319-348Crossref PubMed Scopus (421) Google Scholar, 2Hollmann M. Heinemann S. Annu. Rev. Neurosci. 1994; 17: 31-108Crossref PubMed Scopus (3658) Google Scholar). The mGluR belongs to the Class III GPCR and forms a subfamily consisting of eight subtypes (mGluR1–8) (3Masu M. Tanabe Y. Tsuchida K. Shigemoto R. Nakanishi S. Nature. 1991; 349: 760-765Crossref PubMed Scopus (989) Google Scholar, 4Tanabe Y. Masu M. Ishii T. Shigemoto R. Nakanishi S. Neuron. 1992; 8: 169-179Abstract Full Text PDF PubMed Scopus (889) Google Scholar, 5Conn P.J. Pin J.P. Annu. Rev. Pharmacol. Toxicol. 1997; 37: 205-237Crossref PubMed Scopus (2716) Google Scholar, 6Jingami H. Nakanishi S. Morikawa K. Curr. Opin. Neurobiol. 2003; 13: 271-278Crossref PubMed Scopus (101) Google Scholar). One outstanding feature of the receptor is the large extracellular ligand binding domain (LBD), which is characteristic of the Class III GPCRs. The mGluR1 LBD consists of ∼520 amino acids and forms a clamshell-like bilobate domain (LB1 and LB2) (7Kunishima N. Shimada Y. Tsuji Y. Sato T. Yamamoto M. Kumasaka T. Nakanishi S. Jingami H. Morikawa K. Nature. 2000; 407: 971-977Crossref PubMed Scopus (1098) Google Scholar, 8Tsuchiya D. Kunishima N. Kamiya N. Jingami H. Morikawa K. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 2660-2665Crossref PubMed Scopus (323) Google Scholar). Our previous biochemical and crystallographic studies demonstrated that the LBD forms a homodimer by not only an intersubunit disulfide bond but also hydrophobic interactions (7Kunishima N. Shimada Y. Tsuji Y. Sato T. Yamamoto M. Kumasaka T. Nakanishi S. Jingami H. Morikawa K. Nature. 2000; 407: 971-977Crossref PubMed Scopus (1098) Google Scholar, 8Tsuchiya D. Kunishima N. Kamiya N. Jingami H. Morikawa K. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 2660-2665Crossref PubMed Scopus (323) Google Scholar, 9Okamoto T. Sekiyama N. Otsu M. Shimada Y. Sato A. Nakanishi S. Jingami H. J. Biol. Chem. 1998; 273: 13089-13096Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar, 10Tsuji Y. Shimada Y. Takeshita T. Kajimura N. Nomura S. Sekiyama N. Otomo J. Usukura J. Nakanishi S. Jingami H. J. Biol. Chem. 2000; 275: 28144-28151Abstract Full Text Full Text PDF PubMed Scopus (119) Google Scholar). In the crystal structures, one protomer of the dimeric LBD adopts two different conformations: an open conformation and a closed conformation. A symmetric structure with both protomers in the open conformation is observed in the absence of glutamate, whereas the structure of the glutamate binding state is asymmetric, because one protomer adopts the closed conformation and the other adopts the open conformation. Interestingly, even in the absence of glutamate, the asymmetric open-closed conformation is observed, implying that the open and closed conformations of the protomer are in equilibrium in an aqueous solution without ligands. Glutamate binding promotes the closing motion of the ligand binding pocket, so the closed conformation should be stabilized. Thus, the glutamate-bound open conformation observed in the crystal structure is a fascinating puzzle. One attractive explanation for this is an allosteric property in the dimeric LBD: the closed conformation in one protomer would negatively affect the binding mode of the other protomer. However, such an allosteric effect on ligand binding has not been demonstrated yet for this receptor. Some allosteric properties have been previously reported for several receptors. Extensive biochemical and crystallographic studies have been performed on the bacterial dimeric aspartate receptor, and the mechanism of negative cooperativity on aspartate binding has been elucidated on the basis of the structure (11Milburn M.V. Prive G.G. Milligan D.L. Scott W.G. Yeh J. Jancarik J. Koshland Jr., D.E. Kim S.H. Science. 1991; 254: 1342-1347Crossref PubMed Scopus (376) Google Scholar, 12Yeh J.I. Biemann H.P. Prive G.G. Pandit J. Koshland Jr., D.E. Kim S.H. J. Mol. Biol. 1996; 262: 186-201Crossref PubMed Scopus (137) Google Scholar, 13Kolodziej A.F. Tan T. Koshland Jr., D.E. Biochemistry. 1996; 35: 14782-14792Crossref PubMed Scopus (37) Google Scholar, 14Koshland Jr., D.E. Curr. Opin. Struc. Biol. 1996; 6: 757-761Crossref PubMed Scopus (83) Google Scholar). Allosteric properties are also inferred in several GPCRs (15Limbird L.E. Lefkowitz R.J. J. Biol. Chem. 1976; 251: 5007-5014Abstract Full Text PDF PubMed Google Scholar, 16Pizard A. Marchetti J. Allegrini J. Alhenc-Gelas F. Rajerison R.M. J. Biol. Chem. 1998; 273: 1309-1315Abstract Full Text Full Text PDF PubMed Scopus (49) Google Scholar), however, in these cases, the observed cooperativity seems to result from oligomerization of receptors on the membrane surface. Thus, the allosteric properties in terms of subunit-subunit communication are more obscure in the GPCRs than those in the bacterial aspartate receptor. Recently, dimer formation by GPCRs has been detected, and ligand selectivity appears to be broader than previously expected (17Devi L.A. Trends Pharmacol. Sci. 2001; 22: 532-537Abstract Full Text Full Text PDF PubMed Scopus (258) Google Scholar, 18Franco R. Canals M. Marcellino D. Ferre S. Agnati L. Mallol J. Casado V. Ciruela F. Fuxe K. Lluis C. Canela E.L. Trends Biochem. Sci. 2003; 28: 238-243Abstract Full Text Full Text PDF PubMed Scopus (75) Google Scholar, 19Bouvier M. Nat. Rev. Neurosci. 2001; 2: 274-286Crossref PubMed Scopus (578) Google Scholar, 20Milligan G. J. Cell Sci. 2001; 114: 1265-1271Crossref PubMed Google Scholar). It was also reported that, in the heterodimeric γ-aminobutyric acid type B receptor, the GB2 subunit and its association with the GB1 subunit control the agonist affinity in GB1 subunit (21Liu J. Maurel D. Etzol S. Brabet I. Ansanay H. Pin J.P. Rondard P. J. Biol. Chem. 2004; 279: 15824-15830Abstract Full Text Full Text PDF PubMed Scopus (76) Google Scholar). Therefore, the cooperativity in dimeric GPCRs has become a current issue. In this context, our mGluR system provides a unique opportunity to decipher the cooperativity in the ligand binding event of Class III GPCR using a purified pre-formed dimer. To investigate whether allostery functions in ligand binding, it is essential to analyze a saturation ligand binding curve with a wide range of ligand concentrations from sub-nanomolar to millimolar. 3H-Labeled quisqualate, which is an agonist specific for mGluR1 and -5, is widely used in the ligand binding assay. However, this conventional assay cannot yield a saturation binding curve because of the upper limit of the applicable ligand concentration. Because the fluorescence emitted from tryptophan is sensitive to the environment surrounding the indole group, the intrinsic tryptophan fluorescence can be a good probe to sense ligand binding events (22Verjovski-Almeida S. Silva J.L. J. Biol. Chem. 1981; 256: 2940-2944Abstract Full Text PDF PubMed Google Scholar, 23Nelson S.W. Iancu C.V. Choe J.Y. Honzatko R.B. Fromm H.J. Biochemistry. 2000; 39: 11100-11106Crossref PubMed Scopus (26) Google Scholar, 24Xu Y. Johnson J. Kohn H. Widger W.R. J. Biol. Chem. 2003; 278: 13719-13727Abstract Full Text Full Text PDF PubMed Scopus (18) Google Scholar, 25Lakowicz J.R. Principles of Fluorescence Spectroscopy. 2nd Ed. Plenum Publishers, New York1999: 452-486Google Scholar). In the present study, we found that the fluorescence spectrum of the intrinsic tryptophans of purified mGluR1 LBD changed upon ligand binding. The system led us to obtain saturation binding curves by titration of the tryptophan fluorescence with glutamate, a native ligand, and other non-native ligands, such as quisqualate, (S)-α-methyl-4-carboxyphenylglycine ((S)-MCPG), and (+)-2-methyl-4-carboxyphenylglycine (LY367385). These binding curves allowed us to determine the apparent binding constants of these ligands and to demonstrate the positive effect of calcium ions on agonist binding. Furthermore, Hill analyses of the titration curves revealed that negative cooperativity of glutamate binding exists between each protomer of the dimeric mGluR1 LBD. Materials—l-Quisqualate, (S)-MCPG, and LY367385 were purchased from Tocris (UK). (R)-MCPG was a gift from Dr. D. Shunter (Tocris). l-Glutamate was purchased from Nacalai Tesque (Japan). Oligonucleotide primers were obtained from Proligo (Japan). All other reagents used in the present study were of molecular or analytical grade. Construction of an Expression Vector for the FLAG-tagged LBD and Its Mutant—To obtain a C-terminal FLAG-tagged mGluR1 LBD, we performed PCR with pmGluR104 (9Okamoto T. Sekiyama N. Otsu M. Shimada Y. Sato A. Nakanishi S. Jingami H. J. Biol. Chem. 1998; 273: 13089-13096Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar) as the template. The forward primer for the PCR was designed at the N terminus of the LBD with a NotI site. The reverse primer was designed at the C terminus of the LBD (at the 1566th thymine in the mGluR1 cDNA) with the DNA sequence for the FLAG epitope (DYKDDDDK) and a stop codon followed by an XbaI site. After verification by DNA sequencing, the PCR product was cloned into the pFastBac DUAL vector (Invitrogen) using the NotI and XbaI sites. The I120A and T188A point mutations were introduced into the FLAG-tagged LBD gene by replacing the NotI/PshAI region of the wild type with the same fragment containing the mutation, which was excised from the plasmid previously used in the mutation experiments (26Sato T. Shimada Y. Nagasawa N. Nakanishi S. Jingami H. J. Biol. Chem. 2003; 278: 4314-4321Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar). The W110V mutation was introduced by site-directed mutagenesis using the QuikChange site-directed mutagenesis kit (Stratagene), with pmGluR103 (9Okamoto T. Sekiyama N. Otsu M. Shimada Y. Sato A. Nakanishi S. Jingami H. J. Biol. Chem. 1998; 273: 13089-13096Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar) as the template. After sequence verification, the NotI/PshAI fragment of this mutant was exchanged for that of the wild type on the pFastBac DUAL vector. Production of Baculoviruses for Protein Expression—Baculoviruses for protein expression were obtained by following the protocol of the Bac-To-Bac baculovirus expression system (Invitrogen). Briefly, the vector DNA was transformed into DH10Bac Escherichia coli cells (Invitrogen). Then, the recombinant bacmid DNA purified from the DH10Bac cells was transfected into Sf9 insect cells, using the Cellfectin reagent (Invitrogen). After an incubation for 72 h at 27 °C, the viruses were harvested from the cell culture medium. Then, the recombinant viruses were amplified by re-infecting Sf9 cells to enhance the viral titer. Finally, we checked the viral titer by a plaque formation assay using an immobilized monolayer culture of Sf9 cells. Protein Expression and Purification—The wild-type and mutant FLAG-tagged LBD proteins were expressed by inoculating the baculoviruses into HighFive cells as previously described (9Okamoto T. Sekiyama N. Otsu M. Shimada Y. Sato A. Nakanishi S. Jingami H. J. Biol. Chem. 1998; 273: 13089-13096Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar, 10Tsuji Y. Shimada Y. Takeshita T. Kajimura N. Nomura S. Sekiyama N. Otomo J. Usukura J. Nakanishi S. Jingami H. J. Biol. Chem. 2000; 275: 28144-28151Abstract Full Text Full Text PDF PubMed Scopus (119) Google Scholar). Purification of the protein was done by taking advantage of the FLAG tag. The cell culture medium (∼500 ml) into which the target protein was sufficiently secreted was collected 4–5 days after the inoculation. After the addition of protease inhibitors (10 μg/ml, 2 μg/ml, and 0.1 mm for leupeptin, pepstatin, and phenylmethylsulfonyl fluoride, respectively), the cells were pelleted by centrifugation at 6,700 × g for 15 min at 4 °C. Then, the supernatant was directly applied to ∼1 ml of anti-FLAG M2-agarose (Sigma) packed in a disposable column (Bio-Rad). After the column was washed with a low salt buffer containing 10 mm Tris-HCl, pH 7.5, and 20 mm NaCl, the proteins bound to the beads were eluted by a high salt buffer containing the FLAG peptide at a concentration of 150 μg/ml, dissolved in 10 mm HEPES, pH 7.4, and 300 mm NaCl. The eluted fractions were collected, and the protein was concentrated and buffer-exchanged to the assay buffer (20 mm HEPES, pH 7.4, and 50 mm NaCl) using an Ultra-free centrifugation filter unit (Millipore). The protein concentration was determined by the absorbance at 280 nm with a molar extinction coefficient of 126,500 m–1cm–1 as a dimer, which was calculated from the number of tryptophans, tyrosines, and cystines in the protein (27Gill S.C. von Hippel P.H. Anal. Biochem. 1989; 182: 319-326Crossref PubMed Scopus (5034) Google Scholar). The molecular weight of the FLAG-tagged protein was 119,800 as calculated from the amino acid sequence, including the C-terminal FLAG tag. Purification of the wild-type LBD of mGluR1 (non-tagged) was done as previously described by using an immunoaffinity column conjugated with monoclonal antibodies (mG1Na-1) (9Okamoto T. Sekiyama N. Otsu M. Shimada Y. Sato A. Nakanishi S. Jingami H. J. Biol. Chem. 1998; 273: 13089-13096Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar). The molecular weight of the wild-type LBD was estimated to be 117,800 from the amino acid sequence. Steady-state Fluorescence Measurement—Steady-state tryptophan fluorescence was measured by an F-4500 spectrofluorometer (Hitachi, Japan) with an excitation wavelength of 290 nm at 20 °C using a stirring cuvette. Protein concentration was 0.67 μm as the dimer. Measurements were performed in a buffer composed of 20 mm HEPES, pH 7.4, and 50 mm NaCl with or without 2.0 mm CaCl2. Emission spectra from 300 to 400 nm were recorded. For the titration experiments, small aliquots of ligand (0.1–1.0% of total volume) were sequentially added to the cuvette, and the fluorescence intensity at 350 nm was recorded. Volume changes in the titration experiments were corrected before analyzing the data. To obtain the titration curves, we also used the peak values of the fluorescence spectra instead of the values at 350 nm, because the peak shift of the fluorescence spectrum would affect the calculated values of the Hill coefficients. Both calculations yielded similar Hill coefficient values, and, hence, in this report we used the values calculated from the fluorescence intensity at 350 nm. Transient Kinetics Measurement—Transient kinetics of the intrinsic tryptophan fluorescence change were measured with an SX18 stopped-flow spectrophotometer (Applied Photophysics, UK) at 20 °C. The excitation wavelength was 290 nm, and the emission was monitored with a 335-nm longpass optical filter. The protein and ligand were dissolved in a buffer consisting of 20 mm HEPES, pH 7.4, and 50 mm NaCl. Measurements were done with a fixed protein concentration of 0.84 μm (value after mixing). Ligand Binding Assay—The 3H-labeled ligand binding assay was performed using the polyethylene glycol (PEG) precipitation method (10Tsuji Y. Shimada Y. Takeshita T. Kajimura N. Nomura S. Sekiyama N. Otomo J. Usukura J. Nakanishi S. Jingami H. J. Biol. Chem. 2000; 275: 28144-28151Abstract Full Text Full Text PDF PubMed Scopus (119) Google Scholar) as previously described with minor modifications. Briefly, 20 nm3H-labeled quisqualate or glutamate (Amersham Biosciences) and the protein solution (1 μg of protein) were mixed in 150 μl of binding buffer composed of 40 mm HEPES, pH 7.4, and 2.5 mm CaCl2 at 4 °C for 1 h. Then, 6-kDa PEG was added to the sample to a final concentration of 15% with 3 mg/ml of γ-globulin. After vortexing and centrifugation, the precipitated material was washed twice with 1 ml of the binding buffer containing 8% of 6-kDa PEG, and then it was dissolved in 1 ml of water. After the addition of 14 ml of Clearsol II (Nacalai Tesque), the radioactivity was measured using a scintillation counter. Data Analysis of the Titration Experiments—Titration curves for glutamate and quisqualate binding were made by plotting the values of (F – F0)/ΔFmax against the ligand concentrations, where F and F0 are the fluorescence intensities in the presence and absence of ligand, respectively, and ΔFmax is the maximum value of the fluorescence change in the titration experiment. For the titration curves of (S)-MCPG and LY367385 binding, the values of (F0 – F)/ΔFmax were used because the intrinsic tryptophan fluorescence decreased upon the addition of these antagonists. The titration curves were fitted to the following equation by the KaleidaGraph software (Synergy), ΔF/ΔFmax = SnH/(Kapp + SnH), where ΔF equals F – F0 or F0 – F, S is the concentration of the ligand, Kapp is the apparent dissociation constant, and nH is the Hill coefficient. Hill coefficients were also calculated from the Hill plot, which yielded values similar to those derived from the curve fitting. With respect to Kapp, we also obtained similar values from the half-maximal value of the titration curves. This value is denoted as K0.5 in the text, and we used it to estimate the affinity of the ligand in this report. To elucidate intersubunit allostery in the dimeric ligand binding domain of mGluR1, we worked out a plan to utilize the intrinsic tryptophan fluorescence signal to detect a ligand binding event. For the first step, we constructed and purified the FLAG-tagged LBD of mGluR1 to test whether the intrinsic tryptophan fluorescence changes upon ligand binding. Purification of the FLAG-tagged Ligand Binding Domain— Because the protein from the anti-FLAG-antibody-conjugated agarose gel was eluted with the FLAG peptide at a neutral pH instead of with an acidic or alkaline buffer, the protein damage was minimized. As analyzed by SDS-PAGE followed by silver staining, the purified FLAG-tagged LBD was observed as almost a single band (Fig. 1A). It was previously reported that under the non-reduced conditions the two protomers of the mGluR1 LBD are cross-linked to form a dimer through an interprotomer disulfide bond even under the denaturing conditions on an SDS-polyacrylamide gel (9Okamoto T. Sekiyama N. Otsu M. Shimada Y. Sato A. Nakanishi S. Jingami H. J. Biol. Chem. 1998; 273: 13089-13096Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar, 10Tsuji Y. Shimada Y. Takeshita T. Kajimura N. Nomura S. Sekiyama N. Otomo J. Usukura J. Nakanishi S. Jingami H. J. Biol. Chem. 2000; 275: 28144-28151Abstract Full Text Full Text PDF PubMed Scopus (119) Google Scholar). Like the nontagged wild-type LBD, the band of the FLAG-tagged LBD under the reduced conditions (Fig. 1A, +DTT) was shifted to a position corresponding to twice the molecular weight under the non-reduced conditions (Fig. 1A, –DTT), indicating that the FLAG-tagged protein maintained the ability to form an interprotomer disulfide bond as demonstrated for the non-tagged LBD (9Okamoto T. Sekiyama N. Otsu M. Shimada Y. Sato A. Nakanishi S. Jingami H. J. Biol. Chem. 1998; 273: 13089-13096Abstract Full Text Full Text PDF PubMed Scopus (141) Google Scholar). Next, we investigated the ligand binding ability of the FLAG-tagged LBD using a 3H-labeled quisqualate ([3H]quisqualate) by the previously described PEG-precipitation method (10Tsuji Y. Shimada Y. Takeshita T. Kajimura N. Nomura S. Sekiyama N. Otomo J. Usukura J. Nakanishi S. Jingami H. J. Biol. Chem. 2000; 275: 28144-28151Abstract Full Text Full Text PDF PubMed Scopus (119) Google Scholar). The final concentration of [3H]quisqualate was 20 nm. As shown in Fig. 1B, the FLAG-tagged LBD bound the [3H]quisqualate at the same level as that of the non-tagged LBD. This indicates that the additional eight amino acids at the C terminus of the protein do not influence the ligand binding capacity. From these data, we concluded that the FLAG tag at the C terminus did not perturb the activities of the protein, and therefore, we utilized the FLAG-tagged LBD for the following experiments described below. Intrinsic Tryptophan Fluorescence Changes of the FLAG-tagged LBD Induced by Ligand Binding—We examined whether the intrinsic tryptophan fluorescence changed upon the addition of ligand. In the absence of ligand, the intrinsic tryptophans of the FLAG-tagged LBD exhibited an emission spectrum with a peak at ∼345 nm when excited at 290 nm (Fig. 2A, dotted line). Upon the addition of excess glutamate (1.0 mm final concentration), a native agonist for mGluR, the fluorescence spectrum changed in a manner such that the fluorescence intensity was enhanced by about 18% and the fluorescence maximum was slightly shifted by ∼1 nm toward a longer wavelength (red shift). Upon the addition of excess quisqualate (100 μm at final concentration), a non-native strong agonist for group I mGluRs, the emission spectrum was enhanced, but amplitude of the spectral change was less than that observed for glutamate addition (Fig. 2C). This result probably reflects some environmental differences between the glutamate and quisqualate binding states around the tryptophans that contribute toward sensing the ligand binding. As a control for these observations, we measured the emission spectrum of the T188A mutant. It has been demonstrated that this mutation dramatically reduced the binding affinity of agonists and that no cellular responses were detected in HEK293 cells expressing the full-length mGluR1 carrying the same mutation (26Sato T. Shimada Y. Nagasawa N. Nakanishi S. Jingami H. J. Biol. Chem. 2003; 278: 4314-4321Abstract Full Text Full Text PDF PubMed Scopus (52) Google Scholar). As shown in Fig. 2E, the emission spectrum of the mutant did not exhibit any noticeable change upon the addition of excess glutamate. Although we also examined the spectrum upon the addition of quisqualate, no effects were observed as well (data not shown). Therefore, the enhanced emission spectra observed for the wild-type FLAG-tagged LBD resulted from agonist binding to the protein. Fig. 2B shows fluorescence spectra in the presence and absence of (S)-MCPG (100 μm final concentration), an antagonist for mGluR1. In contrast to the results with the agonists, the fluorescence intensity of the intrinsic tryptophans decreased and the fluorescence maxima shifted toward a shorter wavelength (blue shift) upon the addition of (S)-MCPG. These changes were opposite to those observed for agonist binding. A similar spectral change was observed upon the addition of an excess of LY367385 (100 μm final concentration), another antagonist (Fig. 2D), suggesting that, in the ligand binding states of the two antagonists, the environments around tryptophans involved in the fluorescence change were similar to each other. On the other hand, no spectral change was observed upon the addition of (R)-MCPG (Fig. 2F), a stereoisomer of (S)-MCPG that does not bind to the mGluR1 LBD (8Tsuchiya D. Kunishima N. Kamiya N. Jingami H. Morikawa K. Proc. Natl. Acad. Sci. U. S. A. 2002; 99: 2660-2665Crossref PubMed Scopus (323) Google Scholar). This result clearly indicates that the observed spectral changes are due to the binding of the antagonists. We measured the rate constants of ligand binding to the LBD with a stopped-flow apparatus. As shown in Fig. 3A, the tryptophan fluorescence was abruptly enhanced upon mixing with glutamate (0.5 mm after mixing), and it almost reached a plateau within 1 s. The time course of the fluorescence intensity change was roughly fitted to a single exponential curve. From the fitting curve, the observed rate constant (kobs) of glutamate binding to the LBD was estimated to be 22.2 s–1. On the other hand, upon mixing with (S)-MCPG (0.5 mm after mixing), the intrinsic tryptophan fluorescence was suddenly quenched and completely reached a plateau within 1 s (Fig. 3B). The time course of the fluorescence change was fitted well to a single exponential curve, and the observed rate constant was estimated to be 38.6 s–1. It should be noted that these rate constants of ligand binding are apparent rates measured only at the saturating concentration of ligands. They do not represent actual association rate constants of ligand. Identification of the Tryptophan Residue Contributing to the Spectral Change of the Intrinsic Tryptophan Fluorescence— Tryptophan fluorescence in a protein is generally influenced by the surrounding environment. Water molecules, attacking the indole moiety of tryptophan, and polar side chains and peptide bonds in close proximity of the indole ring are major causes of tryptophan fluorescence quenching (28Chen Y. Barkley M.D. Biochemistry. 1998; 37: 9976-9982Crossref PubMed Scopus (747) Google Scholar). There are seven tryptophan residues in one protomer (Trp-110, Trp-224, Trp-320, Trp-367, Trp-372, Trp-468, and Trp-500). To predict which tryptophan residue contributes to the observed spectral changes, we examined the environment around each tryptophan side chain using the atomic models. We first considered solvent accessibility of each tryptophan. To quantify it, we calculated the solvent-accessible area of each of the tryptophan side chains (indole rings) for the two distinct conformations of the protomer, i.e. the open conformation and the closed conformation, by the program SURFACE (29Lee B. Richards F.M. J. Mol. Biol. 1971; 55: 37