NMDA receptor complex

European Journal ol Pharmacology, 191 (1990) 477480 Elsevier

411

E!JP 20748

Short communication

Lakhbir Singh, Caroline A. Vass, John C. Hunter, Geoffrey N. Woodruff and John Hughes Parke-Davis Research Unit, Addenbrookes Hospital Site, Hills Road, Cambridge CB2 2QB. U.K. Received 6 September 1990, accepted 16 October 1990

The selective K-OpiOid receptor agonist CI-977, stereoselectively antagonised clonic seizures induced by slow i.v. infusion of N-methyl-DL-aspartate in the mouse. It was found to be more efficacious and lo-fold more potent than the competitive N-methyl-D-aspartic acid receptor antagonist CPP (3-( + )-(2-carboxypiperazin-4-yl)propyl-l-phosphonic acid). The anticonvulsant action of U-977 was antagonised by norbinaltorphimine, indicating a specific interaction with the K-receptor. The effect of U-977 but not that of CPP was also antagonised by the glycine/NMDA receptor agonist D-serine. These results provide evidence for a possible interaction between the K-receptor and the glycine/NMDA receptor.

K-Opioid receptors;

Glycine/NMDA

receptors; NMDLA

1. Introduction

It has been demonstrated that K-opioid-selective agonists of the benzeneacetamide series, such as U50488, exhibit an anticonvulsant profile against various models of experimental epilepsy (Tortella et al., 1986; Von Voigtlander et al., 1987). In particular, U50488 has been shown to block convulsions induced by the excitatory amino acid (EAA) receptor agonist N-methyl-D-aspartate (NMDA), quisqualate and kainate, an effect that would appear to be mediated through the K-receptor (Von Voigtlander et al., 1987). Indeed, although the low potency of U50488 in these models has always raised concern over selectivity, we have recently demonstrated that PD117302. a more

Correspondence to: L. Singh, Parke-Davis Research Unit, Addenbrcokes Hospital Site, Hills Road, Cambridge CB2 2QB, U.K. 0014-2999/90/$03.50

(N-methyl-DL-aspartate);

NMDLA

infusion;

Seizures

potent and selective agonist for the K-receptor, also produces protection against NMDA-induced convulsions in rats (Tortella et al., 1990). It has been proposed that a final common mechanism for this K-mediated anticonvulsant activity involves calcium (Von Voigtlander et al., 1987). While this may be the case, it has yet to be resolved whether this interaction involves physiological antagonism through two completely independent pathways or whether there is a more intimate association between the K-receptor and one or more of the EAA receptors. We now demonstrate in the present study that the latest member of our K-series of potent antinociceptive agents, CI-977 (Hunter et al., in press) is a potent anticonvulsant against N-methyl-DLaspartic acid (NMDLA)-evoked seizures in mice by what would appear to involve a specific interaction between the K-reCeptOr and the glycinemodulatory site of the NMDA receptor complex.

0 1990 Elsevier Science Publishers B.V. (Biomedical

Division)

2.1. Seizures induced by NMDLA

N-Methyl-DL-aspartic acid (NMDLA, 85 ml) was given by i.v. infusion (0.14 ml/min) into a tail vein of restrained mice (Male TO, 25-30 g, Bantin and Kingman, Hull, U.K.) until a clonic convulsion was elicited. This convulsion represented the end-point of the test and mice were killed immediately after it had occurred. The threshold dose of NMDLA required to elicit this seizure was calculated for each mouse. J.2. Drug administration The test compounds were administered i-v. 30 min before infusion of NMDLA. For antagonism studies, nor-binaltorphimine (NORBNI) and Dserine were administered i.c.v. 60 and 30 min respectively, before NMDLA as described previously (Singh et al., 1990b). 2.3. Radio&and binding studies The preparation of rat cerebral cortex homogenates and the final assay conditions were exactly as previously described (Foster and Wong, 1987; Murphy et al., 1987; Kemp et al., 1988; Boyle et al., 1990). The concentration of radioligands were: [3H]CI-977 (0.2 nM), [3H]MK-801 (2.0 nM), [“Hlglycine (25.0 nM) and [3H]CPP (10.0 nM) with non-specific binding determined as that remaining in the presence of 1 FM naloxone, 100 FM MK-801, 1 mM glycine and 1 mM L-glutamate respectively.

bridge, U.K. All compounds were dissolved in 0.9% NaCl and where necessary the pH was adjusted to 6.5-7.5 with HCl or NaOH. All radioligands were purchased from New England Nuclear with the exception of [3H]CI-977 which was prepared for Parke-Davis by Amersham International.

3. Results 3. I. Anticonvulsant

effects

The i.v. administration of CPP (2.3-37.0 pmol/kg) or CI-977 (0.07-11.5 pmol/kg) 30 min beforehand, dose dependently antagonised NMDLA-induced clonic seizures with ED5,, (dose required to produce 50% of the maximal increase in the MCD of NMDLA; 95% confidence limits in parantheses) of 3.16 (2.19-4.56) and 0.42 (O.ll1.63) pmol/kg respectively.’ The maximum increase in the MCD of NMDLA produced by CPP and CI-977 was approximately 2.1- and 2.9-fold respectively (fig. 1). In contrast, PD-129289 the

500 ?

2

400 -

6i g B

300 -

4 s

200 -

‘s !

100 -

6

2.4. Drugs 0

L-Glutamate, L-glycine, D-serine, N-methylDL-aspartic acid (NMDLA) and naloxone were obtained from Sigma (Poole, Dorset, U.K.). Norbinaltorphimine (NORBNI), CPP (3-( + )-(2carboxypiperazin-4-yl)propyl-1-phosphonic acid) and MK-801 (dizocilpine) were purchased from Research Biochemicals Inc. (MA, U.S.A.). CI-977 and PD-129289 were synthesised by the Department of Medicinal Chemistry, Parke-Davis, Cam-

0.1

1.0

10.0

100.0

Dose (vmol/kg, i.v.) Fig. 1. The effect of (X-977, PD-129289 and CPP on NMDLA-induced seizures. (X-977 (0). PD-129289 (A) and i.v. 30 min before infusion of ) were administered NMDLA. The threshold dose of NMDLA to induce clonic seizures was determined. Results are shown as the mean convulsive dose (MCD) of NMDLA of 8-10 animals per group. Vertical bars show + S.E.M. Significant differences from vehicle are shown by asterisks ( * P < 0.05, * * P -K0.01). determined by Dunnett’s t-test after significant ANOVA.

479

corresponding ( + )-enantiomer to CI-977, up to 7.0 pmol/kg was found to be inactive (fig. 1). 3.2. Effect of NORBNI on the anticonvulsant action of u-977

The administration of NORBNI 30 min before a submaximal dose of CI-977 (2.0 pmol/kg i.v.) dose dependently (12.5-100 nmol i.c.v.) antagonised the anticonvulsant action of CI-977: the highest dose of NORBNI almost completely abolished the effect of (X-977 (fig. 2A). NORBNJ (100 nmol per mouse) administered alone had no effect on the MCD of NMDLA (120.7 f 12.0 mg/kg, c.f. control = 100.0 f 12.2 mg/kg, P > 0.05, paired t-test).

300

‘I

A.

_

Veh.

0

200

3 .A _ 2 2

100 0

12.5

25.0

50.0

Cl-977 (2.0 pmollkg)+NORBNI

5 9 ;

300

8 =

200

1 Veh. 0

0.01

0.1

The administration

of D-serine immediately before a submaximal dose of (X-977 (2.0 pmol/kg) dose dependently (0.03-0.3 pmol i.c.v.) antagonised the increase in MCD of NMDLA produced by CI-977: the highest dose of D-serine completely antagonised the anticonvulsant action of CI-977 (fig. 2B). In contrast, lo-fold higher doses of Dserine failed to antagonise a similar submaximal anticonvulsant effect of CPP (10.0 pmol/kg fig. 2B). 3.4. Radioligand binding studies Glycine and D-serine, at concentrations

up to 10 PM, had no detectable affinity for t3H] (X-977 labelled fc-opioid binding sites. The Ki value for PD-129289 was 422 f 63 nM (n = 4). Conversely, unlabelled U-977 had negligable affinity (Ki > 10 PM) for [3H]CPP, [3H]glycine and [3H]MK-SOl sites. In addition, CI-977 (l-10.0 PM) had no demonstrable effect on [3H]MK-801 binding stimulated by either L-glutamate (0.0310.0 PM) or glycine (0.1-30.0 PM). PD-129289 was also ineffective up to 10 PM in these models.

(nmol, i.c.v.)

4. Discussion

T

B.

01

100.0

3.3. Effect of D-serine on the anticonvulsant action of CZ-977 and CPP

1.0

10.0

D-serine (pmol, i.c.v.) Fig. 2. The effect of (A) NORBNI and (B) D-serine on the anticonvulsant action of CI-977. (A) NORBNI was administered i.c.v. 30 min before U-977 (2.0 pmol/kg iv.) and (B) D-serine was administered i.c.v. immediately before CI-977 ( 2.0 pmol/kg i.v.) or CPP ( , 10.0pmol/kg iv.). NMDLA was infused 30 min later. The threshold dose of NMDLA to induce clonic seizures was determined. Results are shown as the mean convulsive dose (MCD) of NMDLA of 8-10 animals per group. Vertical bars show +S.E.M. Significant differences from the group given either U-977 or CPP alone are shown by asterisks ( * P c 0.05, * * P < O.Ol), determined by Dunnett’s t-test after significant ANOVA.

CI-977 is an exceptionally potent and selective antinociceptive agent for the fc-opioid receptor (Hunter et al., in press). The present study has now demonstrated that it is also an effective anticonvulsant against NMDLA-induced clonic seizures. Indeed, in this respect, CI-977 was found to be more efficacious and approximately lo-fold more potent than the competitive NMDA receptor antagonist CPP. The anticonvulsant action of CI-977 in this model would appear to be mediated through a specific interaction with the K-receptor. Thus, in addition to the unique selectivity of CI-977 for the K-receptor (Hunter et al., in press), antagonism of the NMDLA-induced seizures was enantiomer specific with PD-129289, the corresponding ( + )enantiomer of CI-977, which demonstrated considerably lower affinity for the K-receptor than CI-977 (Hunter et al., in press), being ineffective

up to veq high doses. In addition, the effect of -gn was antagonised by the selective K-receptor antagonist NQRBNI at doses which alone failed the seizure susceptibility of mice to owever! while the above observations would suggest a simple. direct drug-receptor-related interaction. the ability of a glycine-NMDA receptor agonist Dserine to antagonise the anticonvulsant effect of Ci-977 would suggest a more complex series of events. The doses of D-serine used here have been shown to antagonise the anticonvulsant action of several other antagonists of the glycine modulatory site (Singh et al., 1990a,b). Furthermore. D-serine has been shown to be selective for the glycine modulatory site of the NMDA receptor complex (Singh et al., 199Ob). Consistent with such an action was the failure of D-serine in the present study to influence the anticonvulsant actions of CPP and diazeparn (unpublished observations) in the NMDLA-seizure model. Certainly the lack of inhibition of [ 3H]CI-977 binding would appear to eliminate any action of D-serine at the rc-receptor. The present data therefore suggest that modulation of calcium as a final common mechanism (Von Voigtlander et al., 1987) might not be due to a non-specific physiological antagonism but rather the consequence of a specific association between the tc-receptor and the glycine/NMDA receptor complex. However, U-977 does not compete for either [3H]CPP- or [3H]TCP-labelled sites (Hunter et al., in press) and, in the present study, it failed to modulate L-glutamate and glycine stimulated [‘HIMK-801 binding to the ion channel of the NMDA receptor or to inhibit strychnine-insensitive [ 3H]glycine binding. A possible explanation may be that the interaction between the K-receptor and the NMDA receptor complex occurs only in certain discrete brain areas which would be below the sensitivity for detection by homogenate binding assays. It is also possible that the action of (2-977 might involve inhibition of (amino acid) release from presynaptic neurones. In conclusion, the results presented here indicate that the anticonvulsant action of CI-977 involves stimulation of the K-receptor which decreases the NMDA receptor complex-mediated

excitatory effects. The exact mechanism by which this interaction between the K-receptor and the NMDA receptor complex occurs remains unclear at the moment. However, such an interaction may explain the anticonvulsant and neuroprotective properties of K-receptor agonists.

References Boyle, S.J., K.G. Meecham, J.C. Hunter and J. Hughes, 1990, [ ‘HI-CI-977: A highly selective ligand for the kappa-opioid receptor in both guinea-pig and rat forebrain, Mol. Neuropharmacol. 1. 23. Foster. A.C. and E.H.F. Wong, 1987, The novel anticonvulsant MK-801 binds to the activated state of the N-methyl-Daspartate receptor in the rat brain, Br. J. Pharmacol. 91, 403. Hunter. J.C., G.E. Leighton, K.G. Meecham, S. Boyle, D.C. Horwell, D.C. Rees and J. Hughes, Cl-977, a novel and selective agonist for the kappa-opioid receptor. Br. J. Pharmacol. (in press). Kemp, J.A., A.C. Foster, P.D. Leeson, T. Priestley, R. Tridgett and L.L. Iversen, 1988, 7-Chlorokynurenic acid is a selective antagonist at the glycine modulatory site of the Nmethyl-D-aspartate receptor of the complex, Proc. Nat]. Acad. Sci. U.S.A. 85, 6547. Murphy, D.E.. J. Schneider, C. Boehm, J. Lehmann, J. and M. Williams, 1987, Binding of [3H]3-(2-carboxypiperazin-4yl)propyl-l-phosphonic acid to rat brain membranes: a selective, high affinity ligand for N-methyl-D-aspartate receptors, J. Pharmacol. Exp. Ther. 240, 778. Singh, L.. A.E. Donald, A.C. Foster, P.H. Hutson, L.L. Iversen, S.D. Iversen, J.A. Kemp, P.D. Leeson, G.R. Marshall, R.J. Oles, T. Priestley, L. Thorn, M.D. Tricklebank, C.A. Vass and B.J. Williams, 199Oa, Enantiomers of HA-966 (3-amino-1-hydroxypyrrolid-2-one) exhibit distinct central nervous system effects: (+)-HA-966 is a selective glycine/N-methyl-D-aspartate receptor antagonist, but ( - )-HA-966 is a potent gamma-butyrolactone-like sedative, Proc. Natl. Acad. Sci. U.S.A. 87, 347. Sir&, L.. R.J. Oles and M.D. Tricklebank, 1990b, Modulation of seizure susceptibility in the mouse by the strychnine-insensitive glycine recognition site of the NMDA receptor/ion channel complex, Br. J. Pharmacol. 99, 285. Tortella, F.C., L. Robles. E. Echevarria, J.C. Hunter and J. Hughes, 1990, PDlI7302, a selective non-peptide opioid kappa agonist, protects against NMDA and maximal electroshock convulsions in rats, Life Sci. 46, 1. Tortella, F.C.. L. Robles and J.H. Holaday, 1986, U50, 488, a highly selective kappa opioid: anticonvulsant profile in rats, J. Pharmacol. Exp. Ther. 237, 49. Von Voigtlander. P.F., E.D. Hall, M. Camacho Ochoa, R.A. Lewis and H.J. Triezenberg, 1987, U-54494A: A unique anticonvulsant related to kappa opioid agonists, J. Pharmacol. Exp. Ther. 243, 542.