Blockade of both epileptogenesis and glutamate release by (1S,3S)-ACPD, a presynaptic glutamate receptor agonist

BRAIN RESEARCH ELSEVIER

Brain Research 698 (1995) 155-162

Research report

Blockade of both epileptogenesis and glutamate release by ( 1S,3S)-ACPD, a presynaptic glutamate receptor agonist Philip J.E. Attwell a Sailesh Kaura a Georgia Sigala a Henry F. Bradford a,,, Martin J. Croucher h David E. Jane c, Jeffrey C. Watkins c ~ Department of Biochemistry, Imperial College of Science, Technology and Medicine, London SW7 2AZ, UK Department of Pharmacology, Charing Cross and Westminster Medical School London W6 8RF, UK Department of Pharmacology, School of Medical Sciences, Bristol BS8 ITD, UK Accepted 28 June 1995

Abstract

The influence of intracerebrally focally administered doses of a presynaptic metabotropic glutamate receptor agonist, (IS,3S)-ACPD, and of the post-synaptically targeted competitive NMDA receptor antagonist, D-CPPene (SDZ EAA 494), was tested on the development of amygdaloid kindling. The actions of these drugs, compared to that of D-CPP, was also tested on fully developed stage 5 amygdala kindled seizures. Both (1S,3S)-ACPD and D-CPPene dose-dependently increased the generalised seizure threshold in fully kindled animals. They showed a similar potency, with (1S,3S)-ACPD acting presynaptically and D-CPPene postsynaptically. Both drugs reversibly inhibited epileptogenesis at 10 nmol in 0.5 kd of injection vehicle, keeping the kindling stage at or below stage 2. All animals reached stage 5 after withdrawal of the 2 drugs. Whereas (1S,3S)-ACPD inhibited depolarisation-induced release of [3H]I:glutamate and [3H]D-aspartate from cortical synaptosomes (IC50 63 #M and 50 #M, respectively), D-CPPene (postsynaptically active) was without effect. These findings suggest a new approach to the development of clinically effective anticonvulsants through the development of presynaptic glutamate receptor agonists which could be administered systemically to control the extent of synaptic release of glutamate. Keywords: Presynaptic glutamate receptor; Glutamate release; Kindling; Epileptogenesis; Seizures; (1S,3S)-ACPD: ~)-CPPenc: o-(TP

1. Introduction

Competitive and non-competitive antagonists acting at the postsynaptic NMDA sub-class of excitatory amino acid (EAA) receptors (e.g. AP5, AP7, CPP, CPPene, MK-801, PCP), have been shown to exert anticonvulsant and antiepileptogenic actions in various seizure models such as cobalt-induced seizures [10], audiogenically-induced and pentylenetetrazole-induced seizures in mice [9,14], in photosensitive baboons [34], genetically epilepsy-prone rats [18,19] and in kindled rats [13,20,27,35]. In recent years the presynaptic mechanisms involved in seizure generation have become a focus of interest. The electrophysiological and other functional properties of presynaptic metabotropic

* Corresponding author. Fax: (44) (171) 225-0960. 0006-8993/95/$09.5t) © 1995 Elsevier Science B.V. All rights reserved ~SDI 0 0 0 6 - 8 q 9 3 ( 9 5 ) 0 0 8 8 6 - l

glutamate receptors have recently been studied [2,7,24,37,39,42,43] including their involvement in the kindling model of epilepsy [1]. The possible roles of metabotropic glutamate receptors in CNS function are currently the focus of intensive investigation [3,31,36,43]. Metabotropic receptors modulate the production of intracellular second messengers and belong to the family of G-protein coupled receptors [41]. These are of two classes, one positively linked to phospholipase C, mediating phosphoinositide hydrolysis and release of Ca 2+ from intracellular stores, and the other negatively linked to the cyclic AMP cascade [36]. The former have been suggested to be active mainly at postsynaptic sites and the latter at presynaptic sites [47]. In the present work the involvement of presynaptic glutamate receptors in the development of kindled epilepsy, and of full seizure expression, have been investigated. The (1S,3S) isomer of 1-aminocyclopentane- 1,3-dicarboxylate ((1S,3S)-ACPD) has been shown to have relatively selective agonist activity

156

P.J.E. Attwell et al. /Brain Research 698 (1995) 155-162

at presynaptic receptors that mediate synaptic depression in the isolated hemisected spinal cord of the neonatal rat [24,28,37,44]. This compound is considered likely to activate presynaptic metabotropic glutamate receptors negatively coupled to cyclic A M P synthesis, in view of the close similarity of its synaptic depressant actions to those of (1S,3R)-ACPD and (+__)-trans-ACPD (mixture of (1S,3R) and (1R,3S) isomers) [2,7,24,28,37], and the known actions of all these forms of A C P D on cyclic A M P synthesis [8,28,46]. The presynaptic depressant and metabotropic glutamate receptor actions of (1S,3S)-ACPD are both blocked by the metabotropic glutamate antagonist o~-methyl-4-carboxyphenylglycine ( M C P G ) [21,24]. As (1S,3S)-ACPD is somewhat more selective for cyclic AMP-linked r e l a t i v e to p h o s p h o i n o s i t i d e - l i n k e d metabotropic glutamate receptors [8,37] compared with the (1S,3R)- or (+)-trans-forms of ACPD, this compound was selected for the present study. It was likewise preferred to other presynaptic depressants considered to act at metabotropic receptors negatively coupled to cyclic A M P synthesis, such as (2S,I'S,2'S)-2-(2'carboxycyclopropyl)glycine (L-CCG-I) and (2S, 1' R,2' R,3' R ) - 2 - ( 2 ' , 3 ' - d i c a r b o x y c y c l o p r o p y l ) g l y c i n e (DCG-IV), on grounds of p r e s y n a p t i c / p o s t s y n a p t i c (LCCG-I) or m e t a b o t r o p i c / i o n o t r o p i c (DCG-IV) selectivity of action [23,24,37,44]. The actions of ( I S , 3 S ) - A C P D , which has no significant activity at ionotropic glutamate receptors [44], on epileptogenesis are compared with those of the postsynaptically targeted N M D A receptor antago-

0

nists o-CPP and D-CPPene. In parallel studies, the actions of these compounds were investigated on synaptic release of glutamate from isolated synaptosomes.

2. Materials and methods

2.1. Kindling procedure Details of the surgery and kindling procedures employed can be found elsewhere [15]. Briefly, male S p r a g u e - D a w l e y rats ( 2 9 0 - 3 3 0 g), were implanted with a g u i d e - c a n n u l a / b i p o l a r electrode assembly, in the vicinity of the right basolateral amygdala under h a l o t h a n e / N 2 0 anaesthesia. The stereotaxic coordinates for the tips of the g u i d e - c a n n u l a / b i p o l a r electrode assembly unit were: AP =-0.8, L=-3.8, V=-8.8 from the skull surface (nose bar 2.5mm above the interaural line), aimed at the right basolateral amygdala. At least one week after surgery, afterdischarge thresholds (ADTs) were determined using a method of ascending limits, and animals were kindled by daily stimulation with 125% of their threshold currents. The progressive daily development of seizure activity was rated on a 5-point scale based on that of Racine [38]. After 3 consecutive fully kindled (stage 5) seizures, generalised seizure thresholds (GSTs) were accurately determined as for the ADTs. The afterdischarge duration (ADD) and motor seizure duration (MSD) of the fully kindled seizures were also monitored.

0 r~

4.

o tJ o~

/

mt

N 60

2-

1-

O-

~t

6

~

*o

~,

g

to~

~l

8

ib

i>

14

l(a

18

2'0

Stimulation number

Fig. 1. Inhibition of electrical kindling of the rat amygdala by (IS,3S)-ACPD and D-CPPene. Animals were given daily intra-amygdaloid injections of vehicle (0.5 /xl, n = 6, open circles), (1S,3S)-ACPD (10 nmol, n = 6, filled circles) or D-CPPene (10 nmol, n = 3, open diamonds) 20 min prior to applying the kindling stimulus. The evoked motor response was rated on a scale of 0-5 where: 0 = no behavioural response, 1 = immobility with facial muscle twitching, 2 =jaw myoclonus and/or head 'bobbing', 3 = 2 plus unilateral forelimb ,nyoclonus, 4 = 3 plus rearing with bilateral forelimb myoclonus, and 5 = 4 plus loss of balance resulting in repeated rearing and falling. Values shown are mean + S.E.M. 2-way ANOVA with replication shows significant variation between the treatment groups (P < 0.001). The significance of differences in seizure activity on individual days was calculated using Tukey-Kramers multiple comparisons test at the individual stimulation time points. Drug treatment was withdrawn after 10 days when all control animals had achieved three consecutive stage 5 seizures (dotted vertical line). *, * * & * * * Indicate differences verses control animals of P < 0.05. P < 0.01 and P < 0.001, respectively. There was no significant difference between the two drug-treated groups.

P.J.E. Attwell et al. / Brain Research 698 (19951 155-162

157

15o]

2.2. Action of drugs against kindled seizures Every second day, animals then received 0.5 /_tl vehicle (phosphate buffer, pH 7.4) either alone, or c o n t a i n i n g the following drugs, by focal intra-amygdaloid injection 20 m i n prior to re-estimation of GST: ( 1 S , 3 S ) - A C P D ( 0 . 0 1 1.0 nmol), D-CPPene ( 0 . 0 3 - 0 . 3 nmol), or D-CPP ( 0 . 1 - 1 . 0 nmol). The influence of each drug on GST, A D D , M S D and on gross behavioural activity observed in an open field were determined relative to control (vehicle injection) tests in the same animals. The open field was an animal box (28 × 44 cm with 12 cm high walls) in which the rat had freedom of m o v e m e n t so that any gross behavioural changes (i.e. locomotor activity, exploratory behaviour, circling, ataxia, paraplegia, etc.) could be noted.

14o1

u) (-~

I00,

C 90CO Ul nJ © k_

80-

C

70-

o~

2.3. Epileptogenesis The actions of the drugs on the d e v e l o p m e n t of kindling was tested by injecting drug (10 n m o l / 0 . 5 ~1) or vehicle alone (0.5 /.H) intracerebrally, 20 m i n before a p p l y i n g each k i n d l i n g stimulus over the whole period, i.e. 10 days. Subsequently, the drug was w i t h d r a w n from the experimental animals, and the kindling process was c o n t i n u e d on day 11 in the absence of drug.

'O

L

-

01

'i

.......

Concentration

2.4. Synaptosome preparation S y n a p t o s o m e s were prepared from rat cerebral cortex as decribed by Hughes et al. [22]. Adult S p r a g u e - D a w l e y rats ( 2 0 0 - 2 5 0 g) were used. Table 1 Influence of D-CPPene and (1S,3S)-ACPD treatment on the number of stimulations required to reach full stage 5 kindled seizures (only achieved after drug withdrawal), and on the measurable parameters of these

'~

. . . . . . . .

i'o

(nmoJ)

Fig. 2. Influence of (1S,3S)-ACPD (open circles), t)-CPPene (filled circles) and D-CPP (diamonds), on the gcneralised seizure threshold o[ fully kindled seizures. Animals were focally pre-treated (0.5 /zl; intraamygdaloid; n = 3-5) with drug, or vehicle alone, 20 rain prior to estimation of GST in the fully kindled rats. Log(dose-response) curves were plotted for each of the antagonists, as shown, and GSTm, values were estimated graphically. * * * & " ~ * Indicate P < 0.(15, P < 0.(11, P <0.001, respectively, compared with vehicle in the same animal (Student's 2-tailed t-test for matched pairs).

seizures

Number of stimula- MSD Is) tions to achieve first stage 5 seizure Control 5.83 ± 0.60 (0.5 Izl buffer) In=6) (IS,3S)-ACPD 13.67+_1.31 * ' * (10 nmol/0.5 #11 (n = 6) D-CPPene 13.511+ 1.50 " " (10 nmol/0.5 #11 In=3)

53.00__+8.14

ADD Is)

2.5. Superfusion of 6ynaptosomes 84.33-+11.88

66.83_+15.09 131.33_+27.76

58.50_+ 14.50 131.50_+38.50

The table shows the influence of daily focal injections of (1S,3S)-ACPD, 10nmol or ~CPPene, 10 nmol 20 min prior to applying the kindling stimulus on the number of stimulations required to evoke a fully kindled stage 5 seizure. Note that full kindling was only acheived upon cessation of drug pretreatment. The parameters of the first stage 5 kindled seizure response are also compared. Significance of differences between groups was assessed using ANOVA, and where significant, Tukey-Kramers multiple comparisons post-test. & * * * Indicate P < 0.01 & P < 0.001, respectively, with respect to control mean. *

*

Isolated cerebrocortical synaptosomes were microsuperfused in an apparatus described by Hughes et al. [22]. S y n a p t o s o m e s were preincubated for 5 min at 37°C in Krebs-bicarbonate superfusion fluid of composition (mM): NaCI 140, KCI 4.9, NaHCO~ 25, M g S O 4 1.22, KH2PO,~ 1.25, CaCI~ 1.2, glucose 10, gassed with O y C O 2 (95:5% v / v ) c o n t a i n i n g 0.15 m M [-~H]L-glutamate or [3H]oaspartate. Radiolabelled synaptosomes (0.5 ml) were then placed in each of 10 chambers and then superfused at 1 m l / m i n for 15 rain to wash out excess radioactivity. Fractions were then collected at 25 s intervals (420 /zl). Each batch of s y n a p t o s o m e s was: (a) stimulated for 2 rain with 50 /zM veratridine alone IS1), then (b) treated with A C P D or D-CPPene (at concentrations specified) or saline (control) for .5 rain before stimulating a second time with veratridine at 50 /xM c o n t a i n i n g the drugs ($21. The ratios of S1 and $2 in experimental and control superfusion lines

158

P.J.E. Attwell et al. / B r a i n Research 698 (1995) 155-162

were then compared. Inhibition was expressed as this ratio in the presence and absence of the drug. To detect radioactivity in the superfusate samples, 50 tzl of each sample was added to 5 ml of Packard Ultragold fluor and counted in a Packard Tricarb 1900 /3-counter.

3. Results

3.1. Action of drugs on epileptogenesis When (1S,3S)-ACPD or D-CPPene (both at 10 nmol/0.5 /~1) was injected intracerebrally at a site close to the electrical-kindling site in the amygdala 20 m]n before applying the lsec kindling electrical pulse, both compounds inhibited kindling (Fig. 1). It can be seen that both drugs were equally effective at this dose, with kindling not proceeding on average beyond Racine stage 2 focal seizure activity. Animals receiving vehicle alone proceeded to Racine stage 5 (generalised seizure activity) over the same period as usual. When the drug treatment was withdrawn but the kindling procedure continued, these previously drug treated animals proceeded to Racine stage 5 at a similar rate to the control animals (Fig. 1). After drug withdrawal and achievement of kindled stage 5 there were no significant differences between the three experimental groups in the MSD and the ADD values for the first stage 5 seizure (see Table 1). However, the total number of stimulations required to reach fully kindled stage 5 seizures did show significant differences, with the (1S,3S)-ACPD treated group and the D-CPPene treated group both varying

significantly from the control, vehicle treated group (Table

1). 3.2. Action of drugs on stage 5 seizures The mean ADT estimated in the group of fully kindled animals was 186.9 + 21.9 /~A (mean +__S.E.M., n = 5) with the first fully kindled seizure developed after 12 __+ 1 daily stimulations. The following seizure parameters were estimated in control (vehicle injection) tests in the fully kindled animals: GST, 208.5 ___12.1 /~A; MSD, 64.8 + 4.4 s; ADD, 86.5 + 5.1 s. The action of (1S,3S)-ACPD, D-CPP or D-CPPene against fully kindled seizures were tested by intracerebral injection. All of these compounds dose-dependently increased GST in the fully kindled animals (Fig. 2). The order of anticonvulsant potencies based on GST~0o values (dose of drug required to increase GST by 100% (in nmol)) was as follows: D-CPPene (0.5)> (1S,3S)-ACPD (0.6) > D-CPP (1.1). There were no significant changes in ADD. However, a significant reduction in the MSD for the fully kindled seizure was found for a dose of 1 nmol (1S,3S)-ACPD (control = 88.25 + 3.92 s and 1 nmol (1S,3S)-ACPD = 71.50 +__2.50 s ( P < 0.01) a decrease of 19%). There were no marked changes in gross behavioural activity, as assessed in an open-field, following any dose of the drugs.

3.3. Action of drugs on synaptic release of glutamate When cerebrocortical synaptosomes were microsuperfused with Krebs-buffer containing the presynaptic

50-

100"

40-

80.

D

30-

60.

~6

(-

20-

¢-

(-

40.

C

iO-

°.o

20.

t.'s

2;o

log [(IS,3S)-ACPB] (faN)

21~

°.o

1.'5

21o

log [(IS,3S)-ACPD] (laM)

Fig. 3. (a,b) Log(dose-response) curves for the inhibition of evoked [3H]glutamate release (a) and [3H]D-aspartate release (b) from preloaded cortical synaptosomes by (1S,3S)-ACPD as determined by microsuperperfusion. Sigmoid curves were fitted to the data using non-linear regression (GraphPad Inplot, GraphPad Software). Values are the mean + S.E.M. (n = 3, measured in duplicate).

P.J.E. Attwell et al. / Brain Research 698 (1905) 155-162

8070-

159

matergic synapses is a key step in the processes causing electrical kindling.

I

4.2. Action of drugs on seizures

60-

4030C

20-

i

10ACPD

(lOOp.M)

i'Jl 50uM

IOOuM

--I--

I

I

200uM

--D-CPPene

Fig. 4. Percent inhibition of evoked [3H]D-aspartate release from preloaded cortical synaptosomes by D-CPPene compared to 100 # M (1S,3S)-ACPD as determined by microsuperperfusion. Values are the mean+S.E.M. (n = 3, measured in duplicate).

metabotropic glutamate receptor agonist (1S,3S)-ACPD (25-200 /zM), the depolarisation-evoked release of preloaded [3H]L-glutamate was inhibited by about 40% with an IC5o of 63/xM (Fig. 3a). Similar results were obtained when the evoked release of preloaded [3H]oaspartate was studied (Fig. 3b) with an IC50 value for (1S,3S)-ACPD inhibiting evoked [3H]D-aspartate release of 50 /xM, however, in this case, the total amount of inhibition was greater at about 90%. In contrast the postsynaptic NMDA glutamate receptor antagonist D-CPPene (50-200 /xM) was without significant effect on evoked [3H]D-aspartate release from synaptosomes (Fig. 4).

4. Discussion

4.1. Epileptogenesis In common with previous publications from this laboratory on the anti-epileptogenic properties of post-synaptic NMDA receptor antagonists such as o-CPP [13], o-CPPene was found in the present study to prevent the development of electrically kindled epilepsy. In addition, we now report that the presynaptic metabotropic glutamate receptor agonist (1S,3S)-ACPD also inhibits the development of electrically induced kindling when intracerebrally injected at 10 nmol doses in 0.5 /xl of vehicle. This agent has been proposed to reduce synaptic release of excitatory transmitter (presumably glutamate), as judged by electrophysiological data [24,28,37,44]. The present results are taken to support the hypothesis that glutamate release in gluta-

The increases in GST in fully kindled animals caused by intracerebral injection of various postsynaptic NMDAreceptor antagonists such as AP5, AP7 and CPP have previously been reported [15,26,27]. The data presented in this current paper shows that D-CPPene (also a postsynaptic NMDA-receptor antagonist) and (1S,3S)-ACPD, a presynaptic metabotropic glutamate receptor agonist, have similar actions at relatively low doses. In common with the above mentioned compounds, these too may be candidates for development as clinically useful anticonvulsants (see below). In quantative terms, administration of lnmol of (1S,3S)-ACPD decreased motor seizure duration in the fully kindled animals by 19%.

4.3. Presynaptic release of glutamate Electrophysiological investigations have strongly indicated that (1S,3S)-ACPD reduces or prevents transmitter release from primary afferent terminals in the neonatal rat spinal cord [24,28,37,44], the transmitter at these terminals most likely being glutamate. The present paper describes the inhibitory effect of (1S,3S)-ACPD on glutamate release from cerebrocortical synaptosomes studied by microsuperfusion in vitro. The ICs,~ for this effect was 63 ~M. Whilst not being a highly potent action against the relatively severe depolarization induced by 50 /xM veratridine, this inhibitory effect can be seen as being physiologically relevant since physiologically activated presynaptic release mechanisms are likely to be more easily inhibited. Thus synaptically-released glutamate is likely to reach at least 50-100 ~M in the synaptic cleft [29]. Its agonist action on presynaptic metabotropic glutamate receptors would therefore result in a reduction in its presynaptic release. The lack of a similar effect following o-CPPene administration is consistent with a recent report showing a 10-fold weaker activity of the o-compound compared with the L-isomer of CPPene in inhibiting K + evoked EAA release in vivo as measured by microdialysis. The o-isomer only showed activity at 250 /zM [30]. As release of both preloaded [3H]L-glutamate and [3H]o-aspartate was inhibited under the influence of (1S,3S)-ACPD it seems likely that it was [3H]L-glutamate itself and not its metabolites whose release was being monitored. The computer-predicted maximum extent of this inhibition was different, i.e. 44% for L-glutamate and 96% for D-aspartate. This difference may reflect the metabolism of [3H]L-glutamate to other products [4]. Thus, both k-glutamate and D-aspartate enter the neurones by the same carrier, but D-aspartate is not further metabolised. It is, however, released following neuronal depolarisation.

160

P.J.E. Attwell et al. /Brain Research 698 (1995) 155-162

4.4. Presynaptic receptor type

Although no results have yet been reported on the action of (1S,3S)-ACPD on cloned receptors, it seems highly probable that the present results reflect the activation by this compound of metabotropic glutamate receptor types 2 a n d / o r 3 [33,46]. This interpretation assumes that (1S,3S)-ACPD acts in a similar fashion to (1S,3R)- and t r a n s - A C P D on such cyclic AMP-coupled receptors, as suggested by the similar actions of these compounds on cyclic AMP systems in guinea pig cerebrocortical slices [8,28]. 4.5. Glutamate as an excitatory f a c t o r in kindling and seizures

The results reported here add support to the case identifying glutamate-mediated excitation as the major influence in the development of kindling and the initiation of seizures [5,6]. The present data demonstrate that suppression of depolarisation-induced glutamate release via stimulation of presynaptic metabotropic glutamate receptors blocks: (a) development of electrically-induced kindling; and (b) the initiation of seizures in fully-kindled animals. Substantial inhibition (40-90%) of depolarisation-induced glutamate release from isolated pre-synaptic terminals (synaptosomes), coupled with previous demonstration of suppression of synaptic glutamate release in vivo (see above) supports the notion that glutamate release is the central factor being modulated by (1S,3S)-ACPD. This laboratory reported the potent anticonvulsant action of another pre-synaptically targetted drug, namely DL-AP4, against myoclonic epilepsy induced in rat cerebral cortex by cobalt implantation [10]. In this case the drug was also administered intracerebrally, but by direct superfusion of the superficial cobalt-induced epileptic focus, meninges and dural membrane having been removed. Both contralateral myoclonic jerks and cerebrocortical epileptiform EEG were suppressed by DL-AP4 at 55 /,M and higher concentrations. Monaghan et al. [31] have summarised the evidence for the presynaptic action of L-AP4. There is a great deal of other evidence which supports a central role for glutamate in generating epileptic hyperactivity. This includes enhanced glutamate levels and reduced GABA levels in interstitial fluid of kindled amygdala, with amplified glutamate release during seizures, as measured by microdialysis [25]. Glutamate [11,12,32] and NMDA [16,17] will chemically kindle the amygdala, and this is blocked by NMDA-receptor antagonists. Moreover there is transference between this chemical kindling by glutamate, NMDA and electrical kindling, strongly suggesting that the latter involves electrically-evoked release of the excitatory glutamate molecule, and its subsequent post-synaptic action at the NMDA receptor during the kindling stimulus (see also [17,25]). This is confirmed by the blockade of both the electrical and the chemical kin-

dling processes by NMDA-receptor antagonists [5] (see above). 4.6. Clinical anticoneulsants

The data reported here and by others on animal models of epilepsy, including kindling, provide a valuable insight into the mechanisms generating epileptic tissue, and identifies compounds that can block their progress. Whether such compounds can be developed as new clinically useful anticonvulsants to treat patients is dependent on: (a) absence of strong toxic properties; and (b) their ability to pass the blood-brain barrier when given systemically [26,27]. Clinical trials with the potent postsynaptic NMDA receptor antagonist D-CPPene have revealed side effects such as sedation, poor concentration, depression and related states [45]. Also, D-CPPene did not show efficacy against human complex partial seizures. Similar unacceptable contra-indications were reported for clinical trials with the postsynaptic NMDA receptor antagonists CGS19755 (4-phosphonomethyl-2-piperidine carboxylic acid) used in an attempt to reduce glutamate-induced excitotoxic damage in patients with stroke [40]. Perhaps, new systemically active compounds derived from (1S,3S)-ACPD could be developed which act as agonists at the presynaptic metabotropic glutamate receptor and, in this novel way, act as anticonvulsants and/or neuroprotective agents by blocking the development and expression of epileptic hyperactivity.

Acknowledgements We should like to thank Dr. Paul Herrling (Sandoz Research Institute, Bern) for the gift of D-CPPene (SDZ EAA 494). This work was supported by a grant from the Wellcome Trust and by Eli Lilly.

References [1] Akiyama,K., Daigen,A, Yamada, N., ltoh, T., Kohira, l., Ujike, H. and Otsuki, S., Longlasting enhancementof metabotropicexcitatory amino acid receptor-mediatedpolyphosphoinositidehydrolysisin the amygdala/piriform cortex of deep prepiriformcortical kindled rats, Brain Res., 569 (1992) 71-77. [2] Baskys, A. and Malenka, R.C., Trans-ACPD depresses synaptic transmission in the hippocampus, Eur. J. Pharmacol., 193 (1991) 131-132. [3] Baskys, A., Metabotropicreceptorsand 'slow' excitatoryactions of glutamate agonists in the hippocampus,Trends Neurosci., 15 (1992) 92-96. [4] Bradford,H.F., Chemical Neurobiology, Chapter 4, W.H. Freeman, New York, 1986. [5] Bradford,H.F., Glutamate, GABAand epilepsy, Prog. Neurobiol., (1995) in press. [6] Bradford,H.F. and Peterson, D.W., Current views of the pathobiochemistry of epilepsy, MoL Aspects Med., 9 (1987) 119-172. [7] Calabresi, P., Mercuri, N.B. and Bernardi, G., Activation of

P.J.E. Attwell et a L / Brain Research 698 119951 155-162

quisqualate metabotropic receptors reduces glutamate and GABA mediated synaptic potentials in the rat striatum, Neurosci. Lett., 139 (1992) 41-44. [8] Cartmell, J., Curtis, A.R., Kemp, J.A., Kendall, D.A. and Alexander, S.P.H., Subtypes of metabotropic excitatory amino acid receptor distinguished by stereoisomers of the rigid glutamate analogue, 1-aminocyclopentane-l,3-dicarboxylate, Neurosci. Lett., 153 (1993) 107-110. [9] Chapman, A.G., Graham, J.L. and Meldrum, B.S., Potent oral anticonvulsant activity of CPP and CPPene in DBA/2, Eur. J. Pharmacol., 178 11990) 97-99. [10] Coutinho-Netto, J., Abdul-Ghani, A.S., Collins, J.F. and Bradford, H.F., Is glutamate a trigger factor in epileptic hyperactivity? Epilepsia, 22(3) (1981) 289-296. [I 1] Croucher, M.J. and Bradford, H.F., Kindling of full limbic seizures by repeated microinjections of excitatory amino acids into the rat amygdala, Brain Res., 501 (1989) 58-65. [12] Croucher, M.J. and Bradford, H.F., NMDA receptor blockade inhibits glutamate-induced kindling of the rat amygdala, Brain Res., 506 (19911) 349-352. [13] Croucher. M.J., Bradford, H.F., Sunter, D.C. and Watkins, J.C., Inhibition of the development of electrical kindling of the prepyriform cortex by daily focal injections of excitatory amino acid antagonists, Eur. J. Pharmacol., 152 (1988) 29-38. [14] Croucher, M.J., Collins, J.F. and Meldrum, B.S., Anticonvulsan! action of excitatory amino acid antagonists, Science, 216 (19821 899-901. [15] Croucher, M.J., Cotterell, K.L. and Bradford, H.F., Competitive NMDA receptor antagonists raise electrically kindled generalised seizure thresholds, Neurochem. Res., 17 11992) 409-413. [16] Croucher, M.J., Cotterell, K.L. and Bradford, H.F., NMDA receptor and epileptogenesis, Seizure, 1 (Suppl. A) (1992) P3/03. [17] Croucher, M.J., Cotterell, K.L. and Bradford, H.F., Amygdaloid kindling by repeated focal NMDA administration: comparison with electrical kindling, Eur. J. Pharmacol., (1995) in press. [18] De-Sarro, G. and De-Sarro, A., Anticonvulsant activity of competitive antagonists of NMDA receptor in genetically epilepsy-prone rats, Eur. J. Pharmacol., 215 (1992) 221-229. [19] De-Sarro, G. and De-Sarro, A., Anticonvulsant properties of noncompetative antagonists of the NMDA receptor in genetically epilepsy-prone rats in comparison with CPPene, Neuropharmacology, 32 11993) 51-58. [20] Diirmuller, N., Craggs, M. and Meldrum, B.S., The effect of the non-NMDA receptor antagonist GYKI 52466 and NBQX and the competitive NMDA receptor antagonist D-CPPene on the development of amygdala kindling and on amygdala-kindled seizures. b,;oilepsy Reds., 17 (1994) 167-174. [21] Hayashi, Y., Sekiyama, N., Nakanishi, S., Jane, D.E., Sunter, D.C., Birse, E.F., Udvarhelyi, P.M. and Watkins, J.C., Analysis of agonist and antagonist activities of phenylglycine derivatives for different cloned metabotropic glutamate receptor sub-types, J. Neurosci., 14 119941 3370 3377. [22] Hughes, P.D., Foley, P., Bradford, H.F., Ghatei, M., Khandanian. N., Bloom, S.R. and Wu, J.Y., The differential release of amino acids and neuropeptides from purified subpopulations of mammalian GABAergic and cholinergic cerebrocortical synaptosomes, Neurochem. Res., 18 (1993) 393-4011. [23] lshida, M., Saitoh, T., Shimamoto, K., Ohfune, Y. and Shinozaki. H., A m~vel metabotropic glutamate receptor agonist: marked depression of monosynaptic excitation in the newborn rat isolated spinal cord, Br. J. Pharmacol., 109 (1993) 1169-1177. [24] Jane, D.E., Jones, P.L.St.J., Pook, P.C.-K,, Tse, H.-W. and Watkins, J.C. Actions of two new antagonists showing selectivity for different sub-types o! metabotropic glutamate receptor in the neonatal rat spinal cord, Br. J. Pharmacol., 112 (1994) 809-816. [25] Kaura, S., Bradford, H.F., Young, A.M.J., Croucher, M.J. and Hughes, P.D., The effect of amygdaloid kindling on the content and

161

release of amino acids in vivo and in vitro from the amygdaloid complex: in vivo and in vitro studies, J. Neurochem., 65 (19951 1240-1249. [26] Kaura, S., Croucher, M.J., Bradford, H.F., Pook, P.c.K., Sunter. D.C. and Watkins, J.C., A role for NMDA receptor sub-types in kindling model of epilepsy, Can. J. Physiol. PharmacoL, 72 (1994) p13.13.32. [27] Kaura, S., Croucher, M.J., Bradford, H.F., Suntcr, D.C. and Watkins, J.C., Relative anticonvulsant potencies of CPP and some novel analogues in the kindled rat, Biochem. Soc. Trans., 21 (19931 419S. [28] Kemp, M., Robert, P.J., Pook, P.C.-K., Jane, D.E., Jones, A.W., Jones, P.L.St.J., Sunter, D.C., Udvarhelyi, P.M. and Watkins, J.('., Antagonism of presynaptically mediated depressant responses and cyclic AMP-coupled metabotropie glutamate receptors, Eur..1. Pharmacol. - - Mol. Pharmacol. Sect., 266 11994) 187-192. [29] Kuffier, S.W. and Yoshikami, D., The number of transmittcr molecules in a quantum: an estimate from iontophorctic application of acetylcholinc at the neuromuscular synapse, .1. Physiol., 251 11975) 465 482. [30] Millan, M.H., Chapman, A.G. and Meldrum, B.S., Contrasting effects of D- and b(E)-4-(3-phosphono-2-propenyl)piperazine-2carboxylie acid as anticonvulsants and as inhibitors of potassiumevoked increases in hippocampal extraeellular glutamate and aspartate levels in freely moving rats, J. Neurochem., 62 (1994) 217-222. [31] Monaghan, D.T., Bridges, R.J. and Cotman, C.W., The excitatory amino acid receptors: their classes, pharmacology, and distinct properties in the function of the central nervous system, Annu. Rer. Pharmacol. Toxicol., 29 (1989) 365-402. [32] Mori, N. and Wada, J.A., Bidirectional translcr between kindling induced by excitatory amino acids and electrical stimulation, Brain Res., 425 (1987) 45-48. [33] Nakanishi, S., Molecular diversity of glutamate receptors and implications for brain function, Science, 258 (19q2) 597 6113. [34] Patel, S., Chapman, A.G., Graham, J.L., Mcldrum, B.S. and Frcy, P., Anticonvulsant activity of the NMDA antagonists, r)-( )4-(3 phosphonopropyl)piperazine-2-carboxylic acid (D-CPP) and D( -)(E)-4-(3-phosphonoprop-2-cnyl)piperazine-2-carboxylic acid (l)-CPPene) in a rodent and a primate model of reflex epilepsy, Epilepsy Res., 7 (19901 3-111. [35] Peterson, D.W., Collins, J.F. and Bradford, H.F.. Anticonvulsant action of amino acid antagonists against kindled hippocampal seizures, Brain Res., 311 (1984) 176- 180. [36] Pin, J.-P. and Duvoisin. R., Neurotransmittcr receptors. I. The metabotropic glutamate receptors: structure and functions, Neuropharmacology, 34 (1995) 1-26. [37] Pook, P.C.K., Sunter, D.C., Udvarhelyi, P.M. and Watkins. J.('., Evidence for presynaptic depression of monosynaptic excitation in neonatal rat motoneurones by ~1S,3S)- and ~IS.3R)-ACPD, £~-p. Physiol., 77 (1992) 529-532. [38] Racine, R.J., Modification of seizure activity by electrical stimulation. II. Motor seizure, Electroencephalog, r. Clin. Neurophywiol., 32 ( 19721 281-294. [39] Rainnie, D.G. and Shinnick-Gallagher, P., Trans-ACPD and I-APB presynaptically inhibit excitatory glutamatergie transmission in the basolateral amygdala (BLA), Neurosci. Lett., 139 (19921 87-91. [40] Rogawski, M.A. and Porter, R.J., Antiepilcptic drugs: pharmacological mechanisms and clinical efficacy with consideration of promising developmental stage compounds, Pharmacol. Ret'.. 42 (1991)) 223-286. [41] Schoepp, D., Bockaert, J. and Sladeczck, F., Pharmacological and functional characteristics of metabotropic excitatory amino acid receptors, Trends Pharmacol. Sci., 11 1199111 508-515. [42] Schoepp, D.D., Johnson, B.G., Pharmacology of metabotropic glutamate receptor inhibition of cyclic AMP formation in the adult rat hippocampus, Neurochem. Int., 22 (19031 277-283. [43] Schoepp, D.D., Johnson, B.G., True, R.A. and Monn, J.A., Comparison of ( 1S,3R)- 1-aminocyclopentane- 1,3-dicarboxylic acid ( 1S,3R-

162

P.J.E. Attwell et al. / B r a i n Research 698 (1995) 155-162

ACPD)- and lR,3S-ACPD-stimulated brain phosphoinositide hydrolysis, Eur. J. Pharmacol., 207 (1991) 351-353. [44] Sunter, D.C., Edgar, G.E., Pook, P.C.K., Howard, J.A.K., Udvarhelyi, P.M. and Watkins, J.C., Actions of the four isomers of laminocyclopentane-l,3-dicarboxylate (ACPD) in the hemisected isolated spinal cord of the neonatal rat, Br. J. Pharmacol., 104 (1991) 377P. [45] Sveinbjornsdottir, S., Sander, J.W., Upton, D., Thompson, P.J., Patsalos, P.N., Hirt, D., Emre, M., Lowe, D. and Duncan, J.S., The

excitatory amino acid antagonist D-CPPene (SDZ EAA-494) in patients with epilepsy, Epilepsy Res., 16 (1993) 165-174. [46] Tanabe, Y., Nomura, A., Masu, M., Shigemoto, R., Mizuno, N. and Nakanishi, S., Signal transduction, pharmacological properties, and expression patterns of two rat metabotropic glutamate receptors, mGluR3 and mGluR4, J. Neurosci., 13 (1993) 1372-1378. [47] Watkins, J.C. and Collingridge, G.L., Phenylglycine derivatives as metabotropic glutamate receptor antagonists, Trends in Pharmacol. Sci., 15 (1994) 333-342.