Directory UMM :Data Elmu:jurnal:B:Brain Research:Vol881.Issue2.2000:

www.elsevier.com / locate / bres

3

Direct radiolabeling by [ H]quisqualic acid of group I metabotropic

glutamate receptor in rat brain synaptic membranes

a c b b b

Eiichi Hinoi , Kiyokazu Ogita , Yutaka Takeuchi , Hiroshi Ohashi , Takaharu Maruyama ,

a ,

_*

Yukio Yoneda

a

Department of Molecular Pharmacology, Kanazawa University Faculty of Pharmaceutical Sciences, 13-1 Takara-machi, Kanazawa,

Ishikawa920-0934, Japan

b

Banyu Pharmaceutical Co. Ltd., Tsukuba, Ibaraki 300-2611, Japan

c

Department of Pharmacology, Faculty of Pharmaceutical Sciences, Setsunan University, Hirakata, Osaka 573-0101, Japan Accepted 8 August 2000

Abstract

3

[ H]Quisqualic acid (QA) was synthesized and used to label metabotropic glutamate receptor (mGluR) in rat brain synaptic membranes in the presence of three different ionotropic glutamate receptor agonists at respective saturating concentrations. Of several mGluR agonists

3

tested, group I agonists were more potent in displacing [ H]QA binding than group II and group III agonists in the presence of the three

3

ionotropic agonists. [ H]QA binding was markedly inhibited by guanine nucleotide analogues in a concentration-dependent manner at a

3

concentration range of 10 nM to 1 mM. Scatchard analysis revealed that [ H]QA binding consisted of a single component with a K of_d

3

50.965.3 nM and a Bmaxof 431.6618.3 fmol / mg protein. These results suggest that [ H]QA indeed labels group I mGluR functionally coupled to GTP binding protein in rat brain synaptic membranes when determined under the experimental conditions employed.  2000 Elsevier Science B.V. All rights reserved.

Theme: Neurotransmitters, modulators, transporters, and receptors

Topic: Excitatory amino acid receptors: physiology, pharmacology and modulation

3

Keywords: Metabotropic glutamate receptor; [ H]QA binding; Group I agonists; GTP; Scatchard analysis

1. Introduction [1,19]. Both (6

)-1-aminocyclopentanetrans-1,3-dicarbox-ylic acid (trans-ACPD) and (S )-3,5-dihydrox-Membrane receptors for L-glutamic acid (Glu) are yphenylglycine ((S )-3,5-DHPG) have relatively high

affini-categorized into two major classes such as ionotropic ty for the group I mGluR subtype. Groups II and III of (iGluR) and metabotropic (mGluR) receptors according to mGluR are linked to adenylate cyclase to inhibit formation intracellular signal transduction systems. The former is of cAMP. (2S,29R,39R)-2-(29,39 -dicarboxycyclop-further divided into three groups based on selectivity to ropyl)glycine (DCG-IV) is a group II agonist [14], while N-methyl-D-aspartic (NMDA), DL-a-amino-3-hydroxy-5- L-(1)-2-amino-4-phosphonobutyric acid (L-AP4) is a

methylisoxazole-4-propionic (AMPA) and kainic (KA) group III agonist [11,15]. Although cloning studies have acids, while the latter is classified into three distinct revealed the existence of at least eight different peptides subtypes, including groups I, II and III. Of these mGluR composing these mGluR subtypes, radiolabeling of subtypes, the group I subtype is coupled to phospholipase mGluRs is rather unsuccessful in brain synaptic mem-C to stimulate hydrolysis of membrane phospholipids, with branes to date, mainly due to unavailability of radioligand

L-quisqualic acid (QA) as one of the most potent agonists specific for each subtype. 3

[ H]Glu has been used for preferential radiolabeling of mGluR in autoradiographic [6,7] and membrane binding

*Corresponding author. Tel.: 181-76-234-4471; fax: 1

81-76-234-[18] strategies. A recent study employs a novel radioligand

4471.

E-mail address: yyoneda@anet.ne.jp (Y. Yoneda). synthesized and introduced to label group II mGluR

3

subtype in rat brain homogenates [16]. These previous were incubated with 20 nM [ H]QA in 0.5 ml 50 mM

3

findings led us to synthesize [ H]QA for radiolabeling of Tris-acetate buffer (pH7.4) at 28C for 30 min. Incubation group I mGluR subtype in brain synaptic membranes. Both was terminated by the filtration method, and radioactivity QA and Glu are not specific ligands at any subtypes of retained on the filter was measured by liquid scintillation mGluR and iGluR, while QA has higher affinity and spectrometer at a counting efficiency of 40–42% [12].

3

selectivity for group I mGluR than Glu amongst all Non-specific binding of [ H]QA was defined by the different mGluR subtypes [17,19]. In this paper, therefore, addition of QA at 0.1 mM. In order to suppress binding to

3

we have attempted to label group I mGluR in rat brain iGluR, [ H]QA binding to mGluR was usually determined synaptic membranes treated with Triton X-100 using in the presence of 1 mM AMPA, 1mM KA and 100 mM

3

[ H]QA as a radioligand. NMDA, which are all saturating concentrations to inhibit

each radioligand binding [12,13]. Binding assays were invariably carried out one by one at intervals of 20 s in

2. Materials and methods triplicate with variations of less than 10%.

2.1. Materials

3. Results

3

[Methylene- H]QA was synthesized and purified in the

laboratory of Banyu Pharmaceutical Co. Ltd. (Tsukuba, 3.1. Displacement Japan). Guanosine 59-O-(3-thiotriphosphate) (GTP-g-S),

59-guanylylimidodiphosphate (Gpp(NH)p), GTP, GDP, Among various iGluR agonists examined, QA (IC₅₀5

3

GMP, ATP and NMDA were purchased from Sigma 0.0760.03 mM) was the most potent inhibitor of [ H]QA Chemical Company (St. Louis, MO, USA). KA, AMPA, binding in a manner that QA completely diminished QA, trans-ACPD, (S )-3,5-DHPG, DCG-IV and L-AP4 binding at 0.1 mM, with progressively less potent inhibi-were all obtained from Tocris Cookson (Bristol, UK). tion by Glu (0.1160.04 mM), AMPA (1.3560.55 mM), Other chemical used were all of the highest purity com- KA (1.9060.94 mM) and NMDA (363.86151.9 mM).

3

mercially available. Unlabeled QA was more potent in displacing [ H]QA

binding in the presence of NMDA, AMPA and KA at

2.2. Membrane preparation concentrations maximally effective in displacing each

radioligand binding than their absence (data not shown). Crude synaptic membrane fractions were prepared from In the absence of added iGluR agonists, Glu displaced

3

whole brains of male Wistar rats weighing 200–250 g, [ H]QA binding in a concentration-dependent manner at a followed by extensive washing (four times) and subsequent concentration range of 1mM to 1 mM (Fig. 1). Of mGluR

3

storage at2808C as described previously [22,23]. On the agonists tested, Glu was most potent in displacing [ H]QA day of the experiment, these frozen suspensions were binding in the absence of added iGluR agonists. Binding thawed at room temperature and treated with Triton X-100 was completely inhibited by the addition of Glu at 10mM. at 0.08% to deplete endogenous ligands as much as

possible [12]. The treatment did not affect biochemical and pharmacological properties of binding of a variety of radioligands to different iGluR subclasses [12,13,22].

3

For determination of distribution profiles of [ H]QA binding in discrete brain regions, homogenate particulates were used for binding assays as membrane preparations. Each central structure was individually homogenized in 20 mM Tris-acetate buffer (pH 7.4) containing 1 mM EDTA, 1 mM EGTA, 10 mM sodium fluoride, 10 mM sodium b-glycerophosphate, 1 mM sodium orthovanadate, 10 mM sodium pyrophosphate and 1 mg / ml of different protease inhibitors [( p-amidinophenyl)methanesulfonyl fluoride, leupeptin and antipain] to avoid possible artifactual in-fluences by endogenous phosphatases and proteases during homogenization. Homogenates were centrifuged at 50 000 g for 25 min, followed by extensive washing (four times)

3

Fig. 1. Displacement of [ H]QA binding by different agonists at mGluR

and storage at 2808C as mentioned above. ₃

subtypes. Triton-treated membranes were incubated with 20 nM [ H]QA at 28C for 30 min in buffer containing mGluR agonists at different

2.3. Binding assay _{concentrations from 1mM to 1 mM in either the presence (1N /A / K) or}

absence (2N /A / K) of three ionotropic agonists. Values are the

Under these conditions, four different mGluR agonists were much weaker in displacing binding than Glu at the same concentration range. The addition of the three iGluR agonists markedly reduced the ability of Glu to displace binding, with concomitant facilitation of displacement by (S )-3,5-DHPG. In the presence of the three iGluR agonists, (S )-3,5-DHPG was almost as potent as Glu in displacing binding.

3.2. Inhibitory potencies

These experiments were repeated at least four times, and IC50 values were calculated as shown in Table 1. The group I agonists trans-ACPD and (S )-3,5-DHPG were rendered more potent in displacing binding by the addition of the aforementioned three iGluR agonists than their absence. The addition of the three iGluR agonists induced more than 15-fold potentiation of the ability of (S

)-3,5-3

DHPG to displace binding, while that of the group II Fig. 2. Saturation isotherms of [ H]QA binding in the presence of three ionotropic agonists. Triton-treated membranes were incubated with

agonist DCG-IV was not significantly affected by the

3

different concentrations of [ H]QA from 10 to 300 nM at 28C for 30 min

addition of the three iGluR agonists. The group II agonist

in the presence of three ionotropic agonists at the respective saturating

L-AP4 was ineffective in displacing binding irrespective of

concentrations. Values are the mean6S.E. of four separate measurements.

the addition of ionotropic agonists. The inclusion of three iGluR agonists led to marked exacerbation of the ability of

Glu to displace binding. Even in the presence of the three 3.4. Regional distribution iGluR agonists, Glu was the most potent displacer tested,

3

followed by (S )-3,5-DHPG, trans-ACPD, DCG-IV and [ H]QA binding to group I mGluR subtype was detected L-AP4 in a rank order of decreasing potency. in all central structures examined [(fmol / mg protein): cerebral cortex, 157.0620.5; hippocampus, 152.8631.9; striatum, 114.4619.1; hypothalamus, 181.3611.1;

mid-3.3. Saturability brain, 141.8625.0; cerebellum, 173.1621.6;

medulla-pons, 165.1610.1; spinal cord, 83.2627.9; retina, In the presence of the three added iGluR agonists, 26.765.3]. No significant binding was found in rat

3

[ H]QA binding increased in proportion to increasing peripheral tissues, such as adrenal, pituitary and pancreas.

3

concentrations of [ H]QA at a concentration range of 10 to 150 nM with saturation at concentrations above 150 nM

(Fig. 2). Scatchard analysis of these data revealed that the 3.5. GTP binding proteins maximal number of binding sites (Bmax) was 431.6618.3

3

fmol / mg protein and the dissociation constant (K ) wasd Under the conditions employed, [ H]QA binding was

50.965.3 nM, respectively. markedly inhibited by GTP and its analogues, such as

GTP-g-S and Gpp(NH)p, in a concentration dependent manner at concentrations of 10 nM to 1 mM (Table 2).

3

Table 1 _{GDP induced less potent inhibition of [ H]QA binding}

3

Effects of mGluR agonists on [ H]QA binding in either the presence or

a

absence of three ionotropic agonists

Table 2

3

Drug IC50values (mM) Effects of guanine analogues on [ H]QA binding in the presence of three

a

Trans-ACPD 710.06107.0 346.7685.5 Gpp(NH)p 18.7613.9

DCG-IV 228.8635.0 480.26140.7 GTP 139.0670.0

L-AP4 .1000 .1000 a 3

Triton-treated membranes were incubated with 20 nM [ H]QA at 28C

a 3

Triton-treated membranes were incubated with 20 nM [ H]QA at 28C for 30 min in buffer containing guanine analogues at different con-for 30 min in buffer containing mGluR agonists at different concen- centrations from 10 nM to 1 mM in the presence of three ionotropic trations from 1 to 1000mM in either the presence or absence of three agonists. Values are the mean6S.E. of four to six independent experi-ionotropic agonists. Values are the mean6S.E. of four to six independent ments. Neither GDP nor ATP significantly inhibited binding at

than GTP, while neither GMP nor ATP significantly neurons [4]. These previous findings all argue in favor of

3

affected [ H]QA binding even at 1 mM (data not shown). the idea that group I mGluR subtype is related to molecu-lar mechanisms essential for learning and memory as well as apoptosis [4] and amyotrophic lateral sclerosis [5].

3

4. Discussion Paradoxically high [ H]QA binding in medulla-pons

could be accounted for by taking into consideration the The present study deals with the first direct radiolabeling fact that regional variations were evaluated at a fixed

3

by [ H]QA of group I mGluR subtype in rat brain synaptic concentration of the radioligand in this study. Expression membranes. The data cited above are rather confirmatory of distinct mGluR subunits would lead to localization of to previous findings obtained in independent studies using group I mGluR subtype with different affinities and electrophysiological techniques, but this is undoubtedly the densities in different brain discrete structures. Both

3

first demonstration of [ H]QA binding to group I mGluR mGluR1 and mGluR5 subunits could be expressed in subtype in rat brain synaptic membranes treated with medulla-pons at a ratio different from that in other regions, Triton X-100. In addition to three different iGluR subclas- which therefore results in expression of group I mGluR ses, QA is thought to bind to group I mGluR subtype with subtype with affinities and densities different from each high selectivity and high affinity compared to other mGluR other. For this reason, the final conclusion should await groups such as group II and group III [17,19]. The addition evaluation of Bmax values in discrete brain structures with

3

of three different iGluR agonists seems effective in almost regard to regional variations of [ H]QA binding in future completely occupying iGluR subclasses and thereby dis- studies. Evaluation of mRNA as well as immunoreactive closing group I mGluR subtype available for labeling by proteins seems insufficient to prove the expression of

3

[ H]QA. Under the experimental conditions employed functional receptors with an ability to recognize agonists here, binding to group I mGluR subtype prevails over that and antagonists in particular structures. Establishment of to iGluR subtypes as well as to other mGluR groups with membrane binding assay is undoubtedly of benefit for

3

regards to [ H]QA binding. Similar radiolabeling of conventional screening and subsequent development of

3

mGluR subtype is shown with [ H]Glu binding sensitive to drugs useful for therapy and treatment of a variety of trans-ACPD in the presence of saturating concentrations of neurological and / or psychiatric disorders associated with iGluR agonists in rat forebrain membranes [24], in addi- abnormal functions of group I mGluR subtype in human tion to those studies cited in the introduction [6,7,18]. beings.

Guanine nucleotides are shown to modulate binding of a variety of radioligands to different metabotropic receptors coupled to GTP binding proteins [8,10]. The present data

Acknowledgements on selective inhibition by particular guanine analogues are

well consistent with previous studies on functional

cou-This work was supported in part by Grants-in-Aid for pling between mGluR and GTP binding proteins [1,9]. In

Scientific Research to Y.Y. from the Ministry of Education,

3

other words, membrane receptors radiolabeled by [ H]QA

Science, Sports and Cultures, Japan. are likely coupled to GTP binding proteins in rat brain

synaptic membranes, when determined in the presence of saturating concentrations of NMDA, AMPA and KA. The

successful radiolabeling enables us to evaluate mechanisms References underlying modulation of the affinity for agonists and

antagonists of group I mGluR subtype following associa- [1] T. Abe, H. Sugihara, H. Nawa, R. Shigemoto, R. Mizuno, S. Nakanishi, Molecular characterization of a novel metabotropic

tion and / or dissociation between subunits of the trimeric ₂₁

glutamate receptor mGluR5 coupled to inositol phosphate / Ca

GTP binding proteins.

signal transduction, J. Biol. Chem. 267 (1992) 13361–13368.

Group I mGluR subtype is widely distributed in the _{[2] A. Aiba, C. Chen, K. Herrup, C. Rosenmund, F.C. Stevens, S.} central nervous system [1,20]. Especially, mGluR1 subunit Tonegawa, Reduced hippocampal long-term potentiation and

con-is highly expressed in the CA3 pyramidal cells in hip- text-specific deficit in associative learning in mGluR1 mutant mice, Cell 79 (1994) 365–375.

pocampus and Purkinje cells in cerebellum [1].

Context-[3] A. Aiba, M. Kano, C. Chen, E.M. Stanton, D.G. Fox, K. Herrup,

specific associative learning is impaired with a

concomi-A.T. Zwingman, S. Tonegawa, Deficient cerebellar long-term

de-tant reduction of long-term potentiation in the hippocampal _{pression and impaired motor learning in mGluR1 mutant mice, Cell} CA3 subfield in mGluR1 mutant mice [2]. In these mGluR 79 (1994) 377–388.

mutant mice, long-term depression is deficient in cere- [4] J.W. Allen, B.A. Eldadah, A.I. Faden, Beta-amyloid-induced apop-tosis of cerebellar granule cells and cortical neurons: exacerbation

bellum with impaired motor learning [3]. Moreover,

by selective inhibition of group I metabotropic glutamate receptors,

antibodies against mGluR1 completely block long-term

Neuropharmacology 38 (1999) 1242–1252.

depression in cultured Purkinje cells [21]. b-Amyloid- _{[5] J.M. Anneser, G.D. Borasio, A. Berthele, W. Zieglgansberger, T.R.} induced apoptosis is exacerbated by selective inhibition of Tolle, Differential expression of group I metabotropic glutamate

possible implications for the pathogenesis of amyotrophic lateral sensitive metabotropic glutamate receptors, mGluR4 and mGluR7, sclerosis, Neurobiol. Dis. 6 (1999) 140–147. in the rat brain, J. Comp. Neurol. 360 (1995) 555–570.

3

[6] J.J. Cha, R.L. Makowiec, J.B. Penney, A.B. Young, L-[ H]glutamate [16] P.L. Ornstein, M.B. Arnold, T.J. Bleisch, R.A. Wright, W.J. Wheeler,

3

labels the metabotropic excitatory amino acid receptor in rodent D.D. Schoepp, [ H]LY341495, a highly potent, selective and novel brain, Neurosci. Lett. 113 (1990) 78–83. radioligand for labeling group II metabotropic glutamate receptor, [7] J.J. Cha, C.M. Kosinski, J.A. Kerner, S.A. Alsdorf, L. Mangiarini, Bioorg. Med. Chem. Lett. 8 (1998) 1919–1922.

S.W. Davies, J.B. Penney, G.P. Bates, A.B. Young, Altered brain [17] R.H. Porter, P.J. Roberts, D.E. Jane, J.C. Watkins, (S )-Homoquis-neurotransmitter receptors in transgenic mice expressing a portion of qualate: a potent agonist at the glutamate metabotropic receptor, Br. an abnormal human Huntington disease gene, Proc. Natl. Acad. Sci. J. Pharmacol. 106 (1992) 509–510.

USA 95 (1998) 6480–6485. [18] D.D. Schoepp, R.A. True, 1S,3R-ACPD-sensitive (metabotropic)

3

[8] B.B. Hoffman, R.J. Lefkowitz, Radioligand binding studies of [ H]glutamate receptor binding in membranes, Neurosci. Lett. 145 adrenergic receptors: new insights into molecular and physiological (1992) 100–104.

regulation, Ann. Rev. Pharmacol. Toxicol. 20 (1980) 581–608. [19] M.K. Schulte, R.J. Roon, D.J. Chalmers, D.C. Sunter, J.F. Koerner,

1

[9] J. Kehoe, Glutamate activates a K conductance increase in aplysia Utilization of the resolved L-isomer of 2-amino-6-phosphonohex-neurons that appears to be independent of G proteins, Neuron 13 anoic acid (L-AP6) as a selective agonist for a quisqualate-sen-(1994) 691–702. sitized site in hippocampal CA1 pyramidal neurons, Brain Res. 649 [10] M.E. Maguire, E.M. Ross, A.G. Gilman, Beta-adrenergic receptor: (1994) 203–207.

ligand binding properties and the interaction with adenyl cyclase, [20] R. Shigemoto, S. Nakanishi, N. Mizuno, Distribution of the mRNA Adv. Cyc. Nuc. Res. 8 (1983) 1–83. for a metabotropic glutamate receptor (mGluR1) in the central [11] Y. Nakajima, H. Iwakabe, C. Akazawa, H. Nawa, R. Shigemoto, N. nervous system: an in situ hybridization study in adult and

develop-Mizuno, S. Nakanishi, Molecular characterization of a novel retinal ing rat, J. Comp. Neurol. 322 (1992) 121–135.

metabotropic glutamate receptor mGluR6 with a high agonist [21] R. Shigemoto, T. Abe, S. Nomura, S. Nakanishi, T. Hirano, selectivity forL-2-amino-4-phosphonobutyrate, J. Biol. Chem. 268 Antibodies inactivating mGluR1 metabotropic glutamate receptor (1993) 11868–11873. block long-term depression in cultured Purkinje cells, Neuron 12 [12] K. Ogita, Y. Yoneda, Disclosure by Triton X-100 of NMDA- (1994) 1245–1255.

3

sensitive [ H]glutamate binding sites in brain synaptic membranes, [22] M. Shuto, K. Ogita, T. Minami, H. Maeda, Y. Yoneda, Inhibition of

3

Biochem. Biophys. Res. Comm. 143 (1988) 510–517. [ H]MK-801 binding by ferrous (II) but not ferric (III) ions in a [13] K. Ogita, T. Sakamoto, Da Han, Y. Azuma, Y. Yoneda, Discrimina- manner different from that by sodium nitroprusside (II) in rat brain

tion by added ions of ligands at ionotropic excitatory amino acid synaptic membranes, J. Neurochem. 69 (1997) 744–752.

receptors insensitive to N-methyl-D-aspartate in rat brain using [23] F. Stastny, E. Hinoi, K. Ogita, Y. Yoneda, Ferrous iron modulates

3

membrane binding techniques, Neurochem. Int. 23 (1994) 379–388. quinolinate-mediated [ H]-MK-801 binding to rat brain synaptic [14] H. Ohishi, R. Ogawa-Meguro, R. Shigemoto, T. Kaneko, S. membranes in the presence of glycine and spermidine, Neurosci.

Nakanishi, N. Mizuno, Immunohistochemical localization of Lett. 262 (1999) 105–108.

metabotropic glutamate receptors, mGluR2 and mGluR3, in rat [24] R.A. Wright, J.W. McDonald, D.D. Schoepp, Distribution and cerebellar cortex, Neuron 13 (1994) 55–66. ontogeny of 1S,3R-1-aminocyclopentane-1,3-dicarboxylic

acid-3