Camp. Biochem. Pfiysiol. Vol. 103C, No. I, Printed in Great Britain
pp.
13-17, 1992 0
0306~4492/92 IFS.00+ 0.00 1992 Pergamon Press Ltd
MINI-REVIEW A METABOTROPIC L-GLUTAMATE RECEPTOR AGONIST: PHARMACOLOGICAL DIFFERENCE BETWEEN RAT CENTRAL NEURONES AND CRAYFISH NEUROMUSCULAR JUNCTIONS H. SHINOZAKI The
andM.
ISHIDA
Tokyo Metropolitan Institute of Medical Science, 3-18-22, Honkomagome, Bunkyo-ku, Tokyo 113, Japan. (Fax: (03) 3823-2965) (Received 15 October 1991; accepted for publication 1 April 1992)
Abstract-l.
2S,3S,4S-2-(carboxycyclopropyl)glycine (L-CCG-I), a conformationally restricted glutamate analogue, is a potent metabotropic L-glutamate receptor agonist in the mammalian central nervous system. 2. Depolarizing actions of L-CCG-I and trans-( k )-I-amino-1,3+yclopentanedicarboxylic acid (trunsACPD) in the newborn rat spinal motoneurone are temperature-sensitive, and are not depressed by 3-[( & )-2-carboxypiperazin-4-yl] propyl-l-phosphonic acid (CPP) and/or 6-cyano-7-nitroquinoxaline-2,3dione (CNQX). 3. L-CCG-I and truns-ACPD induced oscillatory responses in Xenopus oocytes injected with rat brain mRNA. Oocytes with oscillatory responses to L-CCG-I and trans-ACPD showed reversal potential of about -20 mV, which was very close to the equilibrium potential of chloride ions. 4. In rat hippocampal synaptoneurosomes, L-CCG-I stimulated phosphoinositide hydrolysis in a concentration dependent manner. L-CCG-I was less potent than quisqualate but more potent than trans-ACPD. 5. At low concentrations, L-CCG-I did not cause any depolarization of newborn rat spinal motoneurones, but reduced substantially amplitudes of monosynaptic reflexes. 6. At the crayfish neuromuscular junction L-CCG-I, acting presynaptically, reduced the amplitude of excitatory junctional potentials. This action was prevented by application of picrotoxin but not pertussis toxin. The actions of trans-ACPD differ from those of either L-CCG-I or ibotenate at the crayfish neuromuscular junction. 7. L-CCG-I has a potential to provide further useful information on metabotropic L-glutamate receptor function.
quisqualate acts as an excitant in both invertebrates and vertebrates, but non-NMDA agonists in the mammalian CNS such as AMPA and 5-bromowillardiine are quite inactive when tested at the crayfish neuromuscular junction (NMJ). A competitive antagonist for non-NMDA receptors, 6-cyano-7nitroquinoxaline-2,3-dione (CNQX), does not block depolarizing responses to quisqualate and L-glutamate at this junction. NMDA receptor agonists, such as NMDA and quinolinate, do not induce any detectable depolarization of crayfish muscle fibres, and selective antagonists for NMDA receptors such as 3-[( + )-2-carboxypiperazin-4-yllpropyl-1 -phosphonic acid (CPP) and D-( - )-2-amino-5-phosphonovaleric acid (D-APV) hardly depress the L-glutaKainic acid is mate induced depolarization. significantly less potent in causing depolarization at the crayfish NMJ than in the mammalian CNS, and depresses depolarizing responses to quisqualate in a competitive manner at this junction, but significantly potentiates the depolarization induced by bathapplied L-glutamate. Ibotenate depresses excitatory junctional potentials (EJPs) at the crayfish NMJ by increasing chloride permeability of the presynaptic
L-Glutamate is considered to function as an excitatory neurotransmitter in a wide range of species. Some agonists and antagonists for excitatory amino acids have made great contributions to the identification of L-glutamate as an excitatory neurotransmitter and to the elucidation of its function (Shinozaki, 1980, 1988). In the mammalian central nervous system (CNS), receptors for amino acids have been classified into at least five groups based on their pharmacological properties (Monaghan et al., 1989). They are N-methyl-D-aspartate (NMDA), kainate, cl-amino-3-hydroxy-S-methyl-isoxazole-4-propionate (AMPA) and L-2-amino-4-phosphonobutyrate (LAP4) receptors, and in addition to these ‘ionotropic’ receptors, the existence of a ‘metabotropic’ L-glutamate receptor has been reported. This classification of excitatory amino acids in the mammalian CNS may not necessarily apply to invertebrates. Indeed there appears to be some differences in pharmacological actions of mammalian Lglutamate receptor ligands when they are tested on invertebrates (Shinozaki, 1980, 1988). For example, 13
14
H.
SHINOZAKI
membrane, that was blocked by picrotoxin (Shinozaki and Ishida, 1980), and stimulates extrajunctional receptors on locust muscle fibres to increase chloride permeability (Lea and Usherwood, 1973). Recently, ibotenate was used to characterize an L-glutamate gated chloride channel in an identified insect motor neurone (Wafford and Sattelle, 1989). Ibotenate interacts with excitatory amino acid receptors in the mammalian CNS (Johnston et al., 1968, 1974; Shinozaki and Konishi, 1970), primarily at the NMDA-preferring receptors (London and Coyle, 1979; Sharif and Roberts, 1981). The same agonist produces an initial excitation followed by a sustained depression of mammalian spinal interneurones (MacDonald and Nistri, 1978). A metabotropic L-glutamate receptor in mammalian CNS is also stimulated by ibotenate, as well as quisqualate (Monaghan et al., 1989; Sugiyama et al., 1989). Stizolobic acid is a competitive antagonist for the quisqualate-type receptor at the crayfish NMJ, and decreases EJP amplitude in a concentration dependent manner (Shinozaki and Ishida, 1988), whereas on spinal motoneurones in newborn rats it causes a significant depolarization, which is blocked by CNQX (Ishida and Shinozaki, 1988). These pharmacological differences between vertebrates and invertebrates are attractive for the elucidation of the diversity of L-glutamate receptor function. In the present paper we add another example to these cases with particular reference to the actions of a novel metabotropic L-glutamate agonist. A metabotropic L-glutamate receptor, that is linked to the stimulation of phosphoinositide (PI) hydrolysis, has recently been reported (Sladeczek et al., 1985; Nicoletti et al., 1986; Sugiyama et al., 1987, 1989; Palmer et al., 1989). This receptor subtype in the mammalian CNS is activated by excitatory amino acids such as quisqualate, ibotenate and l-aminocyclopentane-trans-1,3-dicarboxylic acid (trans-ACPD). lS,3R-ACPD appears to be a selective agonist for this receptor and others are nonselective agonists (Schoepp et al., 1990). In the crustacean NMJ, the existence of a metabotropic L-glutamate receptor has been described, which may regulate the transmitter release through the mediation of a GTP-binding protein (Miwa et al., 1990). Much less is known of its pharmacology than is the case for the mammalian CNS metabotropic receptor. A major contributory factor has been lack of potent and selective agonists and antagonists of this class of L-glutamate receptor. Recently, we discovered a new potent metabotropic L-glutamate receptor agonist that is effective on rat spinal motoneurones (Shinozaki et al., 1989b; Nakagawa et al., 1990; Ishida et al., 1990a). Here we describe its actions on the mammalian CNS and compare the pharmacology of this vertebrate receptor with that of its counterpart in invertebrates. The interactions of glutamate with its various receptors is not well understood with respect to their preference for particular conformers of the neurotransmitter molecules. It is reasonable to assume that an acyclic molecule has different conformations due to rotation, vibration, steric repulsion and so on, and is capable of binding to and activating the different types of receptors. The glutamate molecule is rela-
and M.
ISHIDA
tively flexible, and at present, there is no way to determine directly the conformation of transmitter molecules as they activate their particular receptors. The only currently available approach to this question is by chemical modification of agonists and their pharmacological characterization. In an earlier study (Shinozaki et al., 1989a, b; Ishida and Shinozaki, 1990b), we gained some insight into possible conformations of L-glutamate molecules which activated each L-glutamate receptor subtype. 2-(Carboxycyclopropyl)glycine (CCG) is a conformationallyrestricted analogue of glutamate, and the cyclopropyl group fixes the glutamate chain in an extended or folded form. CCG has eight stereoisomers theoretically, so we expected that CCG would provide useful information about the interaction between the conformation of glutamate molecules and activation of L-glutamate receptor subtypes. When the depolarizing actions of eight CCG isomers were examined for electrophysiological effects in the isolated spinal cord of the newborn rat, they demonstrated a large variety of depolarizing actions. The 2S,3$4S-isomer (LCCG-I, Fig. 1) and the 2S,3R,4R-isomer (L-CCG-II), an extended form of L-CCG, showed a clear preference for non-kainate, non-quisqualate and nonNMDA type receptors (Ishida et al., 1990a), and L-CCG-I has been concluded to be a potent metabotropic L-glutamate receptor agonist based on experimental evidence described below (Nakagawa et al., 1990; Ishida et al., 1990a). The depolarizing activity of L-CCG-I was much higher than that of L-glutamate in newborn rat spinal motoneurones, but lower than that of quisqualate, NMDA and kainate. This minimum effective concentration of L-CCG-I to cause depolarization of newborn rat spinal motoneurones was less than 10 PM, and the depolarizing activity of L-CCG-I was about two times more potent than racemic trans-ACPD. Both L-CCG-I and trans-ACPD generated somewhat longer depolarizing responses than other ionotropic excitants. When L-CCG-I was applied for an extended period, the depolarization was not maintained, but rather was decreased in the presence of the agonist (a desensitization-like phenomenon). This decrease was more rapid at higher agonist concentrations. When L-CCG-I or trans-ACPD was added to the bathing fluid repetitively at constant intervals, peak amplitudes of successive responses gradually decreased to a plateau level, while those to NMDA did not reduce, but rather increased slightly. Usually, the amplitudes of responses to L-glutamate or other ionotropic L-glutamate receptor agonists did not decline during their application in the rat spinal motoneurone. Even after the development of desensitization-like phenomenon induced by L-CCG-I, further depolarization was induced by NMDA, AMPA and kainate, but not by trans-ACPD.
“oo;$coo”
LCCGI
u~&coo”
L-CCGII Fig. 1. Chemical structure of L-CCG-I.
Metabotropic L-glutamate receptor agonists
The depolarization induced by kainate and AMPA was effectively reduced by CNQX in a concentration dependent manner, but depolarizing responses to L-CCG-I and trans-ACPD were decreased slightly even in the presence of high concentrations of CPP alone, or in combination with high concentrations of CNQX. Thus, these actions do not mimic those of other agonists such as AMPA, kainate and NMDA (Shinozaki et al., 1989b; Ishida et al., 1990a). Since L-glutamate and quisqualate are mixed-type agonists, it was reasonable to expect that they will continue to generate substantial depolarization even in the presence of high concentration of CNQX and CPP. If L-CCG-I and trans-ACPD act through activating intracellular enzyme systems, their depolarizing responses should be temperature-dependent. When the temperature of the perfusing fluid was varied from 35 to 15°C peak amplitudes and time-courses of responses varied substantially. Amplitudes of responses to L-CCG-I and trans-ACPD were markedly decreased by reducing the temperature to IS’C, while the responses to ionotropic L-glutamate receptor agonists such as L-glutamate, kainate, AMPA and quisqualate were markedly increased and those to NMDA were unchanged (Ishida et al., 1990a). LCCG-I and truns-ACPD caused a maximal depolarizing response in the temperature range of 25 to 27°C and an increase of temperature from 27 to 35°C slightly depressed their response amplitudes. Kainate, NMDA and AMPA had a relatively slow time-course of action at 15”C, whereas responses at 35°C were much faster. Although the uptake mechanisms and the open time of agonist-gated ion channels may show some temperature-dependence, the results presented here are consistent with the conclusion that the intracellular enzyme systems are related to the depolarization induced by L-CCG-I or truns-ACPD. Xenopus oocytes are of great value for electrophysiological identification of metabotropic L-glutamate receptor agonists, because the oocytes injected with poly(A) + mRNA isolated from the mammalian brain express GTP-binding protein-coupled metabotropic L-glutamate receptors, which induce a characteristic oscillatory membrane current through PI break down, which probably arises through common second messenger pathway involving IP, and CaZ+ (Gundersen et al., 1984; Sugiyama et al., 1987, 1989). Pertussis toxin blocks the stimulation of metabolism (Ui, 1984). If L-CCG-I is a GTP-binding proteincoupled metabotropic L-glutamate receptor agonist, it would be expected to show the oscillatory response in Xenopus oocytes injected with mRNA from the mammalian brain, as is the case for L-glutamate, quisqualate, ibotenate, truns-ACPD and homocysteic sulphinic acid. Several days after the injection of rat cortex mRNA, oocytes responded to L-glutamate, kainate, AMPA, quisqualate, truns-ACPD, L-CCG-I and II. Bath application of kainate and AMPA induced significant smooth inward currents in voltage-clamped oocytes. On the other hand, currents induced by L-glutamate, quisqualate, truns-ACPD, L-CCG-I and II had two components, smooth currents and oscillatory ones with a longer latency (Ishida et al., 1990a). L-CCG-I induced oscillatory responses in a manner quite similar to quisqualate or truns-ACPD, although the threshold concentration CBPIO)C,I--B
15
of L-CCG-I, which was almost similar to that of L-glutamate, was 10 times greater than that of quisqualate. L-CCG-II, another extended form of L-CCG, also elicited oscillatory responses, although its potency was about 5 times less than that of L-CCG-I. A similar difference in potency was seen in newborn rat spinal neurons in which L-CCG-II was 20 times less potent than that of L-CCG-I (Shinozaki et al., 1989b). Oocytes with oscillatory responses to L-CCG-I showed reversal potentials of about -20 mV, which is very close to the equilibrium potential of chloride ions, suggesting that L-CCG-I increased chloride conductance. At the reversal potential, inward currents induced by kainate were still observed. It has been shown that oscillatory responses in Xenopus oocytes injected with rat brain mRNA are due to the metabotropic action which leads to increase in chloride conductance mediated through GTP-binding protein. Therefore, the above results strongly suggest that L-CCG-I is a potent metabotropic L-glutamate receptor agonist on rat brain L-CCG-glutamate receptor expressed in Xenopus oocytes, suggesting that metabotropic L-glutamate receptors may exist on newborn rat spinal motoneurones. Our recent experiments have shown that L-CCG-I activates PI metabolism in the rat hippocampal neurone (Nakagawa et al., 1990). The stimulation of inositide phosphates (IPs) formation elicited by LCCG-I was concentration-dependent and stimulation reached plateau levels. CNQX did not reduce the formation of IPs induced by L-CCG-I. Racemic truns-ACPD, hitherto the most potent agonist of the PI-coupled L-glutamate receptor known, was five times less active than L-CCG-I. Quisqualate was more potent than L-CCG-I or truns-ACPD. The maximal stimulation evoked by L-CCG-I was similar to that elicited by quisqualate, but combination of L-CCG-I and quisqualate yielded no evidence of additive effects. Kainate, AMPA and NMDA at concentrations up to 100 PM were quite ineffective in stimulating formation of IPs in hippocampal synaptoneurosomes. The physiological roles of metabotropic L-glutamate receptors are being investigated by several laboratories. When the spinal reflex (ventral root recordings) was evoked by electrical stimulation of dorsal roots in the newborn rat, the responses were depressed by L-CCG-I. At a concentration of 1 PM L-CCG-I, monosynaptic reflexes almost disappeared with no associated depolarization of the postsynaptic membrane, however, polysynaptic reflexes were either unaffected or slightly decreased. The threshold concentration of L-CCG-I for depressing polysynaptic reflexes was about 3 PM, a concentration which did generate a depolarization of the postsynaptic membrane. L-AP4 depressed the monosynaptic reflex in a manner quite similar to that of L-CCG-I, but its actions were slightly weaker than those of L-CCG-I. truns-ACPD suppressed both monosynaptic and polysynaptic reflexes, and this was accompanied by depolarization of the postsynaptic membrane. The depression of monosynaptic reflexes by L-CCG-I was not blocked by picrotoxin at concentrations up to 10pM (Fig. 2). At present it is still unclear whether L-CCG-I activates metabotropic L-glutamate recep-
16
H. SIUNOZAKI and M. ISHIDA
tors, L-AP4 sensitive receptors or other receptors on
the presynaptic nerve terminal of newborn rats. In our previous studies, it was shown that ibotenic acid reduced the amplitude of the EJPs at the crayfish NMJ in a concentration-dependent manner without causing any depolarization of the muscle fibre (Shinozaki and Ishida, 1980). The decrease in EJP amplitudes induced by ibotenic acid was completely blocked by picrotoxin. Ibotenic acid increased the conductance of the muscle membrane in a manner similar to that of GABA, although the potency to increase the membrane conductance was approximately 10 times less than that of GABA on a molar basis. The conductance change could be induced by ibotenate even when the glutamate receptor was almost completely desensitized by the prolonged application of high concentrations of L-glutamate. Therefore, it seems likely that ibotenate acts on the GABA receptor at the crayfish NMJ (Shinozaki and Ishida, 1980). The depression of EJPs by ibotenate is through the presynaptic events, because the quanta1 content of extracellular EJPs decreases in a degree similar to that of the depression of intracellular EJPs. L-CCG-I showed results quite similar to those of ibotenate (Fig. 3A). L-CCG-I depressed EJP amplitudes in a concentration-dependent manner and the depression of EJPs was completely blocked by picrotoxin. The capacity to depress EJPs by L-CCG-I was slightly lower than that of ibotenate. On the other hand, trans-ACPD did not decrease EJP amplitudes at a concentration of 1 mM. L-AP4 caused only a slight decrease in EJP amplitudes which is resistant to picrotoxin (Fig. 3b). To investigate the possible participation of a GTPbinding protein in the depression of EJP amplitudes induced by L-CCG-I, the effects of pertussis toxin were examined. The crayfish neuromuscular preparation was preserved for 7 hr in a solution containing pertussis toxin (10 pg/ml). However, L-CCG-I still depressed the EJP amplitude at the crayfish NMJ as before applying pertussis toxin, suggesting that the depression of EJPs by L-CCG-I was not mediated through the GTP-binding protein. Miwa et al. (1990) have recently reported the existence of a metabotropic L-glutamate receptor which induces glutamateinduced-K + -currents that are GTP-binding protein coupled in lobster presynaptic membranes. This
I-
L-CCG-I 1 PM
L-CCG-I 3 FM -
a
PICROTOXIN 1OpM
”
!” $2
I
TIlli” b 4r
:ro"S-ACPD 1 mM
z
E
L-AP4 1 mM
L-CCG_IO.Smhl
PICROTOXIN 20 ,,M
a4
2 I 1
min
Fig. 3. Depression of EJP amplitude caused by L-CCG-I and its block by picrotoxin. Data points represent EIP amplitudes. EJPs were evoked by 9 repetitive electrical stimulations (90 msec intervals) of excitatory nerves every 10 sec.
metabotropic L-glutamate receptor may regulate transmitter release at the crustacean NMJ, and kainate and quisqualate activate this receptor (Miwa et al., 1990). However, it seems unlikely that L-CCG-I acts on this metabotropic L-glutamate receptor in the presynaptic membrane, because the effects of stimulation of this GTP-binding protein-coupled metabotropic L-glutamate receptor were blocked by pertussis toxin, but not by picrotoxin, quite the opposite of the observed actions of L-CCG-I. There are some questions remaining on the sites of actions L-CCG-I; does it activate the common receptor to ibotenate? Do any other metabotropic L-glutamate receptors which are different from the GTP-binding protein-coupled receptor exist at the crayfish NMJ? Does L-CCG-I activate ionotropic L-glutamate receptors at the crayfish NMJ? In any event, L-CCG-I is expected to provide further useful information about the metabotropic functions of L-glutamate. Acknowledgements-The authors wish to thank Dr Y. Ohfune for generous gifts of CCG samples, and Dr David B. Sattelle for correcting the English of the manuscript.
REFERENCES
Fig. 2. Depression of monosynaptic reflexesin the newborn rat. Spinal reflexes were induced by electrical stimulation of ~5 dorsal roots at an interval of 90 set and the amplitudes of monosynaptic reflexes were plotted in the absence and presence of L-CCG-I. Upper trace: Dose-dependent depression of monosynaptic reflexes. Lower trace: Picrotoxin failed to influence the depression of monosynaptic reflexes.
Gundersen C. B., Miledi R. and Parker I. (1984) Glutamate and kainate receptors induced by rat brain messenger RNA in Xenopus oocytes. Proc. Roy. Sot. Lond. B. 221, 127-143. Ishida M., Akagi H., Shimamoto K., Ohfune Y. and Shinozaki H. (1990a) A potent metabotropic glutamate receptor agonist: electrophysiological actions of a conformationally restricted glutamate analogue in the rat spinal cord and Xenopus oocytes. Bruin Res. 537, 311-314.
Metabotropic
L-glutam iate receptor
Ishida M. and Shinozaki H. (1988) Excitatory action of a plant extract, stizolobic acid, in the isolated rat spinal cord of the rat. Brain Res. 413, 193-197. Ishida M. and Shinozaki H. (1990b) New potent excitatory amino acids and marked potentiation of glutamate responses: pharmacology of conformationally restricted glutamate analogues. In Amino Acids: Chemistry, Biology and Medicine (Edited by Lubec G. and Rosenthal G. A.), pp. 3988409. ESCOM, Leiden. Johnston G. A. R., Curtis D. R., DeGroat W. C. and Duggan A. W. (1968) Central actions of ibotenic acid and muscimol. Biochem. Phartnacol. 17, 2488-2489. Johnston G. A. R., Curtis D. R., Davies J. and McCulloch R. M. (1974) Spinal interneurone excitation by conformationally restricted analogues of L-glutamic acid. Nature 248, 804-805. Lea T. and Usherwood P. N. R. (1973) Effects of ibotenic acid on chloride permeability of insect muscle fibres. Comp. Gen. Phariac. 4, 351-363. London E. and Covle J. T. (1979) Snecific bindine of rH-kainic acid to receptor sites in rat brain. Mol. Pharmacol. 15, 492-505. MacDonald J. F. and Nistri A. (1978) A comparison of the action of glutamate, ibotenate and other related amino acids on feline spinal interneurones. J. Physiol. 275, 449465. Miwa A., Ui M. and Kawai N. (1990) G protein is coupled to presynaptic glutamate and GABA receptors in lobster neuromuscular synapse. J. Neurophysiol. 63, 173-180. Monaghan D. T., Bridge R. J. and Cotman C. W. (1989) The excitatory amino acid receptors: the classes, pharmacology, and distinct properties in the function of the central nervous system. Ann. Rev. Pharmacol. Toxicol. 29, 365402. Nakagawa Y., Saito K., Ishihara T., Ishida M. and Shinozaki H. (1990) (2!3,3S,4S)-a-(carboxycyclopropyl)glycine is a novel agonist of metabotropic glutamate receptors. Eur. J. Pharmacol. 184, 205-206. Nicoletti F., Iadorola M. J., Wroblewski J. T. and Costa E. (1986) Excitatory amino acid recognition sites coupled with inositol phosphohpid metabolism: developmental changes and interaction with cc,-adrenoceptors. Proc. Natl. Acad. Sci. U.S.A. 83, 1931-1935. Palmer E., Monaghan D. T. and Cotman C. W. (1989) Trans-ACPD, a selective agonist of the phosphoinositidecoupled excitatory amino acid receptor. Eur. J. Pharmacol. 166, 585-587. Sharif N. A. and Roberts P. J. (1981) L-Aspartate binding sites in rat cerebellum: a comparison of the binding of
agonists
17
L-13H]aspartate and L-[3Hlglutamate to synaptic membranes. krain Res. 211,2!%303. _ Shinozaki H. (1980) The nharmacoloev of the excitatorv neuromuscular junction in the crayfiyh. Prog. Neurobioi 14, 121-155. Shinozaki H. (1988) Pharmacology of the glutamate receptor. Prog. Neurobiol. 30, 399435. Shinozaki H. and Ishida M. (1980) Inhibitory action of ibotenic acid on the crayfish neuromuscular junction. Brain Rex 198, 157-165. Shinozaki H. and Ishida M. (1988) Stizolobic acid, a competitive antagonist of the quisqualate-type receptor at the crayfish neuromuscular junction. Brain Res. 451, 353-356. Shinozaki H., Ishida M., Shimamoto K. and Ohfune Y. (1989a) A conformationally restricted analogue of Lglutamate, the threo-folded form of L-a-(carboxycyclopropyl)glycine, activates the NMDA-type receptor more markedly than NMDA in the isolated rat spinal cord, Brain Res. 480, 355-359. Shinozaki H., Ishida M., Shimamoto K. and Ohfune Y. (1989b) Potent NMDA-like actions and marked potentiation of glutamate responses of conformational variants of a glutamate analogue in the rat spinal cord, Br. J. Pharmacol. 98, 1213-1224. Shinozaki H. and Konishi S. (1970) Actions of several anthelmintics and insecticides on rat cortical neurones. Brain Res. 24, 368-371. Schoepp D., Bockaert J. and Sladeczek F. (I 990) Pharmacological and functional characteristics of metabotropic excitatory amino acid receptor. Trends Pharmacol. Sci. 11, 508-515. Sladeczek F., Pin J. P., Recasens M., Bockaert J. and Weiss S. (1985) Glutamate stimulates inositol phosphate formation in striatal neurons. Nature 317, 717-719. Sugiyama H., Ito I. and Hirono C. (1987) A new type of glutamate receptor linked to inositol phospholipid metabolism. Nature 325, 531-533. Sugiyama H., Ito I. and Watanabe M. (1989) Glutamate receptor subtypes may be classified into two major categories: a study on Xenopus oocytes injected with rat brain mRNA. Neuron 3, 129-132. Ui M. (1984) Islet-activating protein, pertussis toxin: a probe for functions of the inhibitory guanine nucleotide regulatory component of adenylate cyclase. Trends Pharmacol. Sci. 5, 277-279. Wafford K. A. and Sattelle D. B. (1989) L-Glutamate receptors on the cell body membrane of an identified insect motor neurone. J. exp. Biol. 144, 449-462.
pp.
13-17, 1992 0
0306~4492/92 IFS.00+ 0.00 1992 Pergamon Press Ltd
MINI-REVIEW A METABOTROPIC L-GLUTAMATE RECEPTOR AGONIST: PHARMACOLOGICAL DIFFERENCE BETWEEN RAT CENTRAL NEURONES AND CRAYFISH NEUROMUSCULAR JUNCTIONS H. SHINOZAKI The
andM.
ISHIDA
Tokyo Metropolitan Institute of Medical Science, 3-18-22, Honkomagome, Bunkyo-ku, Tokyo 113, Japan. (Fax: (03) 3823-2965) (Received 15 October 1991; accepted for publication 1 April 1992)
Abstract-l.
2S,3S,4S-2-(carboxycyclopropyl)glycine (L-CCG-I), a conformationally restricted glutamate analogue, is a potent metabotropic L-glutamate receptor agonist in the mammalian central nervous system. 2. Depolarizing actions of L-CCG-I and trans-( k )-I-amino-1,3+yclopentanedicarboxylic acid (trunsACPD) in the newborn rat spinal motoneurone are temperature-sensitive, and are not depressed by 3-[( & )-2-carboxypiperazin-4-yl] propyl-l-phosphonic acid (CPP) and/or 6-cyano-7-nitroquinoxaline-2,3dione (CNQX). 3. L-CCG-I and truns-ACPD induced oscillatory responses in Xenopus oocytes injected with rat brain mRNA. Oocytes with oscillatory responses to L-CCG-I and trans-ACPD showed reversal potential of about -20 mV, which was very close to the equilibrium potential of chloride ions. 4. In rat hippocampal synaptoneurosomes, L-CCG-I stimulated phosphoinositide hydrolysis in a concentration dependent manner. L-CCG-I was less potent than quisqualate but more potent than trans-ACPD. 5. At low concentrations, L-CCG-I did not cause any depolarization of newborn rat spinal motoneurones, but reduced substantially amplitudes of monosynaptic reflexes. 6. At the crayfish neuromuscular junction L-CCG-I, acting presynaptically, reduced the amplitude of excitatory junctional potentials. This action was prevented by application of picrotoxin but not pertussis toxin. The actions of trans-ACPD differ from those of either L-CCG-I or ibotenate at the crayfish neuromuscular junction. 7. L-CCG-I has a potential to provide further useful information on metabotropic L-glutamate receptor function.
quisqualate acts as an excitant in both invertebrates and vertebrates, but non-NMDA agonists in the mammalian CNS such as AMPA and 5-bromowillardiine are quite inactive when tested at the crayfish neuromuscular junction (NMJ). A competitive antagonist for non-NMDA receptors, 6-cyano-7nitroquinoxaline-2,3-dione (CNQX), does not block depolarizing responses to quisqualate and L-glutamate at this junction. NMDA receptor agonists, such as NMDA and quinolinate, do not induce any detectable depolarization of crayfish muscle fibres, and selective antagonists for NMDA receptors such as 3-[( + )-2-carboxypiperazin-4-yllpropyl-1 -phosphonic acid (CPP) and D-( - )-2-amino-5-phosphonovaleric acid (D-APV) hardly depress the L-glutaKainic acid is mate induced depolarization. significantly less potent in causing depolarization at the crayfish NMJ than in the mammalian CNS, and depresses depolarizing responses to quisqualate in a competitive manner at this junction, but significantly potentiates the depolarization induced by bathapplied L-glutamate. Ibotenate depresses excitatory junctional potentials (EJPs) at the crayfish NMJ by increasing chloride permeability of the presynaptic
L-Glutamate is considered to function as an excitatory neurotransmitter in a wide range of species. Some agonists and antagonists for excitatory amino acids have made great contributions to the identification of L-glutamate as an excitatory neurotransmitter and to the elucidation of its function (Shinozaki, 1980, 1988). In the mammalian central nervous system (CNS), receptors for amino acids have been classified into at least five groups based on their pharmacological properties (Monaghan et al., 1989). They are N-methyl-D-aspartate (NMDA), kainate, cl-amino-3-hydroxy-S-methyl-isoxazole-4-propionate (AMPA) and L-2-amino-4-phosphonobutyrate (LAP4) receptors, and in addition to these ‘ionotropic’ receptors, the existence of a ‘metabotropic’ L-glutamate receptor has been reported. This classification of excitatory amino acids in the mammalian CNS may not necessarily apply to invertebrates. Indeed there appears to be some differences in pharmacological actions of mammalian Lglutamate receptor ligands when they are tested on invertebrates (Shinozaki, 1980, 1988). For example, 13
14
H.
SHINOZAKI
membrane, that was blocked by picrotoxin (Shinozaki and Ishida, 1980), and stimulates extrajunctional receptors on locust muscle fibres to increase chloride permeability (Lea and Usherwood, 1973). Recently, ibotenate was used to characterize an L-glutamate gated chloride channel in an identified insect motor neurone (Wafford and Sattelle, 1989). Ibotenate interacts with excitatory amino acid receptors in the mammalian CNS (Johnston et al., 1968, 1974; Shinozaki and Konishi, 1970), primarily at the NMDA-preferring receptors (London and Coyle, 1979; Sharif and Roberts, 1981). The same agonist produces an initial excitation followed by a sustained depression of mammalian spinal interneurones (MacDonald and Nistri, 1978). A metabotropic L-glutamate receptor in mammalian CNS is also stimulated by ibotenate, as well as quisqualate (Monaghan et al., 1989; Sugiyama et al., 1989). Stizolobic acid is a competitive antagonist for the quisqualate-type receptor at the crayfish NMJ, and decreases EJP amplitude in a concentration dependent manner (Shinozaki and Ishida, 1988), whereas on spinal motoneurones in newborn rats it causes a significant depolarization, which is blocked by CNQX (Ishida and Shinozaki, 1988). These pharmacological differences between vertebrates and invertebrates are attractive for the elucidation of the diversity of L-glutamate receptor function. In the present paper we add another example to these cases with particular reference to the actions of a novel metabotropic L-glutamate agonist. A metabotropic L-glutamate receptor, that is linked to the stimulation of phosphoinositide (PI) hydrolysis, has recently been reported (Sladeczek et al., 1985; Nicoletti et al., 1986; Sugiyama et al., 1987, 1989; Palmer et al., 1989). This receptor subtype in the mammalian CNS is activated by excitatory amino acids such as quisqualate, ibotenate and l-aminocyclopentane-trans-1,3-dicarboxylic acid (trans-ACPD). lS,3R-ACPD appears to be a selective agonist for this receptor and others are nonselective agonists (Schoepp et al., 1990). In the crustacean NMJ, the existence of a metabotropic L-glutamate receptor has been described, which may regulate the transmitter release through the mediation of a GTP-binding protein (Miwa et al., 1990). Much less is known of its pharmacology than is the case for the mammalian CNS metabotropic receptor. A major contributory factor has been lack of potent and selective agonists and antagonists of this class of L-glutamate receptor. Recently, we discovered a new potent metabotropic L-glutamate receptor agonist that is effective on rat spinal motoneurones (Shinozaki et al., 1989b; Nakagawa et al., 1990; Ishida et al., 1990a). Here we describe its actions on the mammalian CNS and compare the pharmacology of this vertebrate receptor with that of its counterpart in invertebrates. The interactions of glutamate with its various receptors is not well understood with respect to their preference for particular conformers of the neurotransmitter molecules. It is reasonable to assume that an acyclic molecule has different conformations due to rotation, vibration, steric repulsion and so on, and is capable of binding to and activating the different types of receptors. The glutamate molecule is rela-
and M.
ISHIDA
tively flexible, and at present, there is no way to determine directly the conformation of transmitter molecules as they activate their particular receptors. The only currently available approach to this question is by chemical modification of agonists and their pharmacological characterization. In an earlier study (Shinozaki et al., 1989a, b; Ishida and Shinozaki, 1990b), we gained some insight into possible conformations of L-glutamate molecules which activated each L-glutamate receptor subtype. 2-(Carboxycyclopropyl)glycine (CCG) is a conformationallyrestricted analogue of glutamate, and the cyclopropyl group fixes the glutamate chain in an extended or folded form. CCG has eight stereoisomers theoretically, so we expected that CCG would provide useful information about the interaction between the conformation of glutamate molecules and activation of L-glutamate receptor subtypes. When the depolarizing actions of eight CCG isomers were examined for electrophysiological effects in the isolated spinal cord of the newborn rat, they demonstrated a large variety of depolarizing actions. The 2S,3$4S-isomer (LCCG-I, Fig. 1) and the 2S,3R,4R-isomer (L-CCG-II), an extended form of L-CCG, showed a clear preference for non-kainate, non-quisqualate and nonNMDA type receptors (Ishida et al., 1990a), and L-CCG-I has been concluded to be a potent metabotropic L-glutamate receptor agonist based on experimental evidence described below (Nakagawa et al., 1990; Ishida et al., 1990a). The depolarizing activity of L-CCG-I was much higher than that of L-glutamate in newborn rat spinal motoneurones, but lower than that of quisqualate, NMDA and kainate. This minimum effective concentration of L-CCG-I to cause depolarization of newborn rat spinal motoneurones was less than 10 PM, and the depolarizing activity of L-CCG-I was about two times more potent than racemic trans-ACPD. Both L-CCG-I and trans-ACPD generated somewhat longer depolarizing responses than other ionotropic excitants. When L-CCG-I was applied for an extended period, the depolarization was not maintained, but rather was decreased in the presence of the agonist (a desensitization-like phenomenon). This decrease was more rapid at higher agonist concentrations. When L-CCG-I or trans-ACPD was added to the bathing fluid repetitively at constant intervals, peak amplitudes of successive responses gradually decreased to a plateau level, while those to NMDA did not reduce, but rather increased slightly. Usually, the amplitudes of responses to L-glutamate or other ionotropic L-glutamate receptor agonists did not decline during their application in the rat spinal motoneurone. Even after the development of desensitization-like phenomenon induced by L-CCG-I, further depolarization was induced by NMDA, AMPA and kainate, but not by trans-ACPD.
“oo;$coo”
LCCGI
u~&coo”
L-CCGII Fig. 1. Chemical structure of L-CCG-I.
Metabotropic L-glutamate receptor agonists
The depolarization induced by kainate and AMPA was effectively reduced by CNQX in a concentration dependent manner, but depolarizing responses to L-CCG-I and trans-ACPD were decreased slightly even in the presence of high concentrations of CPP alone, or in combination with high concentrations of CNQX. Thus, these actions do not mimic those of other agonists such as AMPA, kainate and NMDA (Shinozaki et al., 1989b; Ishida et al., 1990a). Since L-glutamate and quisqualate are mixed-type agonists, it was reasonable to expect that they will continue to generate substantial depolarization even in the presence of high concentration of CNQX and CPP. If L-CCG-I and trans-ACPD act through activating intracellular enzyme systems, their depolarizing responses should be temperature-dependent. When the temperature of the perfusing fluid was varied from 35 to 15°C peak amplitudes and time-courses of responses varied substantially. Amplitudes of responses to L-CCG-I and trans-ACPD were markedly decreased by reducing the temperature to IS’C, while the responses to ionotropic L-glutamate receptor agonists such as L-glutamate, kainate, AMPA and quisqualate were markedly increased and those to NMDA were unchanged (Ishida et al., 1990a). LCCG-I and truns-ACPD caused a maximal depolarizing response in the temperature range of 25 to 27°C and an increase of temperature from 27 to 35°C slightly depressed their response amplitudes. Kainate, NMDA and AMPA had a relatively slow time-course of action at 15”C, whereas responses at 35°C were much faster. Although the uptake mechanisms and the open time of agonist-gated ion channels may show some temperature-dependence, the results presented here are consistent with the conclusion that the intracellular enzyme systems are related to the depolarization induced by L-CCG-I or truns-ACPD. Xenopus oocytes are of great value for electrophysiological identification of metabotropic L-glutamate receptor agonists, because the oocytes injected with poly(A) + mRNA isolated from the mammalian brain express GTP-binding protein-coupled metabotropic L-glutamate receptors, which induce a characteristic oscillatory membrane current through PI break down, which probably arises through common second messenger pathway involving IP, and CaZ+ (Gundersen et al., 1984; Sugiyama et al., 1987, 1989). Pertussis toxin blocks the stimulation of metabolism (Ui, 1984). If L-CCG-I is a GTP-binding proteincoupled metabotropic L-glutamate receptor agonist, it would be expected to show the oscillatory response in Xenopus oocytes injected with mRNA from the mammalian brain, as is the case for L-glutamate, quisqualate, ibotenate, truns-ACPD and homocysteic sulphinic acid. Several days after the injection of rat cortex mRNA, oocytes responded to L-glutamate, kainate, AMPA, quisqualate, truns-ACPD, L-CCG-I and II. Bath application of kainate and AMPA induced significant smooth inward currents in voltage-clamped oocytes. On the other hand, currents induced by L-glutamate, quisqualate, truns-ACPD, L-CCG-I and II had two components, smooth currents and oscillatory ones with a longer latency (Ishida et al., 1990a). L-CCG-I induced oscillatory responses in a manner quite similar to quisqualate or truns-ACPD, although the threshold concentration CBPIO)C,I--B
15
of L-CCG-I, which was almost similar to that of L-glutamate, was 10 times greater than that of quisqualate. L-CCG-II, another extended form of L-CCG, also elicited oscillatory responses, although its potency was about 5 times less than that of L-CCG-I. A similar difference in potency was seen in newborn rat spinal neurons in which L-CCG-II was 20 times less potent than that of L-CCG-I (Shinozaki et al., 1989b). Oocytes with oscillatory responses to L-CCG-I showed reversal potentials of about -20 mV, which is very close to the equilibrium potential of chloride ions, suggesting that L-CCG-I increased chloride conductance. At the reversal potential, inward currents induced by kainate were still observed. It has been shown that oscillatory responses in Xenopus oocytes injected with rat brain mRNA are due to the metabotropic action which leads to increase in chloride conductance mediated through GTP-binding protein. Therefore, the above results strongly suggest that L-CCG-I is a potent metabotropic L-glutamate receptor agonist on rat brain L-CCG-glutamate receptor expressed in Xenopus oocytes, suggesting that metabotropic L-glutamate receptors may exist on newborn rat spinal motoneurones. Our recent experiments have shown that L-CCG-I activates PI metabolism in the rat hippocampal neurone (Nakagawa et al., 1990). The stimulation of inositide phosphates (IPs) formation elicited by LCCG-I was concentration-dependent and stimulation reached plateau levels. CNQX did not reduce the formation of IPs induced by L-CCG-I. Racemic truns-ACPD, hitherto the most potent agonist of the PI-coupled L-glutamate receptor known, was five times less active than L-CCG-I. Quisqualate was more potent than L-CCG-I or truns-ACPD. The maximal stimulation evoked by L-CCG-I was similar to that elicited by quisqualate, but combination of L-CCG-I and quisqualate yielded no evidence of additive effects. Kainate, AMPA and NMDA at concentrations up to 100 PM were quite ineffective in stimulating formation of IPs in hippocampal synaptoneurosomes. The physiological roles of metabotropic L-glutamate receptors are being investigated by several laboratories. When the spinal reflex (ventral root recordings) was evoked by electrical stimulation of dorsal roots in the newborn rat, the responses were depressed by L-CCG-I. At a concentration of 1 PM L-CCG-I, monosynaptic reflexes almost disappeared with no associated depolarization of the postsynaptic membrane, however, polysynaptic reflexes were either unaffected or slightly decreased. The threshold concentration of L-CCG-I for depressing polysynaptic reflexes was about 3 PM, a concentration which did generate a depolarization of the postsynaptic membrane. L-AP4 depressed the monosynaptic reflex in a manner quite similar to that of L-CCG-I, but its actions were slightly weaker than those of L-CCG-I. truns-ACPD suppressed both monosynaptic and polysynaptic reflexes, and this was accompanied by depolarization of the postsynaptic membrane. The depression of monosynaptic reflexes by L-CCG-I was not blocked by picrotoxin at concentrations up to 10pM (Fig. 2). At present it is still unclear whether L-CCG-I activates metabotropic L-glutamate recep-
16
H. SIUNOZAKI and M. ISHIDA
tors, L-AP4 sensitive receptors or other receptors on
the presynaptic nerve terminal of newborn rats. In our previous studies, it was shown that ibotenic acid reduced the amplitude of the EJPs at the crayfish NMJ in a concentration-dependent manner without causing any depolarization of the muscle fibre (Shinozaki and Ishida, 1980). The decrease in EJP amplitudes induced by ibotenic acid was completely blocked by picrotoxin. Ibotenic acid increased the conductance of the muscle membrane in a manner similar to that of GABA, although the potency to increase the membrane conductance was approximately 10 times less than that of GABA on a molar basis. The conductance change could be induced by ibotenate even when the glutamate receptor was almost completely desensitized by the prolonged application of high concentrations of L-glutamate. Therefore, it seems likely that ibotenate acts on the GABA receptor at the crayfish NMJ (Shinozaki and Ishida, 1980). The depression of EJPs by ibotenate is through the presynaptic events, because the quanta1 content of extracellular EJPs decreases in a degree similar to that of the depression of intracellular EJPs. L-CCG-I showed results quite similar to those of ibotenate (Fig. 3A). L-CCG-I depressed EJP amplitudes in a concentration-dependent manner and the depression of EJPs was completely blocked by picrotoxin. The capacity to depress EJPs by L-CCG-I was slightly lower than that of ibotenate. On the other hand, trans-ACPD did not decrease EJP amplitudes at a concentration of 1 mM. L-AP4 caused only a slight decrease in EJP amplitudes which is resistant to picrotoxin (Fig. 3b). To investigate the possible participation of a GTPbinding protein in the depression of EJP amplitudes induced by L-CCG-I, the effects of pertussis toxin were examined. The crayfish neuromuscular preparation was preserved for 7 hr in a solution containing pertussis toxin (10 pg/ml). However, L-CCG-I still depressed the EJP amplitude at the crayfish NMJ as before applying pertussis toxin, suggesting that the depression of EJPs by L-CCG-I was not mediated through the GTP-binding protein. Miwa et al. (1990) have recently reported the existence of a metabotropic L-glutamate receptor which induces glutamateinduced-K + -currents that are GTP-binding protein coupled in lobster presynaptic membranes. This
I-
L-CCG-I 1 PM
L-CCG-I 3 FM -
a
PICROTOXIN 1OpM
”
!” $2
I
TIlli” b 4r
:ro"S-ACPD 1 mM
z
E
L-AP4 1 mM
L-CCG_IO.Smhl
PICROTOXIN 20 ,,M
a4
2 I 1
min
Fig. 3. Depression of EJP amplitude caused by L-CCG-I and its block by picrotoxin. Data points represent EIP amplitudes. EJPs were evoked by 9 repetitive electrical stimulations (90 msec intervals) of excitatory nerves every 10 sec.
metabotropic L-glutamate receptor may regulate transmitter release at the crustacean NMJ, and kainate and quisqualate activate this receptor (Miwa et al., 1990). However, it seems unlikely that L-CCG-I acts on this metabotropic L-glutamate receptor in the presynaptic membrane, because the effects of stimulation of this GTP-binding protein-coupled metabotropic L-glutamate receptor were blocked by pertussis toxin, but not by picrotoxin, quite the opposite of the observed actions of L-CCG-I. There are some questions remaining on the sites of actions L-CCG-I; does it activate the common receptor to ibotenate? Do any other metabotropic L-glutamate receptors which are different from the GTP-binding protein-coupled receptor exist at the crayfish NMJ? Does L-CCG-I activate ionotropic L-glutamate receptors at the crayfish NMJ? In any event, L-CCG-I is expected to provide further useful information about the metabotropic functions of L-glutamate. Acknowledgements-The authors wish to thank Dr Y. Ohfune for generous gifts of CCG samples, and Dr David B. Sattelle for correcting the English of the manuscript.
REFERENCES
Fig. 2. Depression of monosynaptic reflexesin the newborn rat. Spinal reflexes were induced by electrical stimulation of ~5 dorsal roots at an interval of 90 set and the amplitudes of monosynaptic reflexes were plotted in the absence and presence of L-CCG-I. Upper trace: Dose-dependent depression of monosynaptic reflexes. Lower trace: Picrotoxin failed to influence the depression of monosynaptic reflexes.
Gundersen C. B., Miledi R. and Parker I. (1984) Glutamate and kainate receptors induced by rat brain messenger RNA in Xenopus oocytes. Proc. Roy. Sot. Lond. B. 221, 127-143. Ishida M., Akagi H., Shimamoto K., Ohfune Y. and Shinozaki H. (1990a) A potent metabotropic glutamate receptor agonist: electrophysiological actions of a conformationally restricted glutamate analogue in the rat spinal cord and Xenopus oocytes. Bruin Res. 537, 311-314.
Metabotropic
L-glutam iate receptor
Ishida M. and Shinozaki H. (1988) Excitatory action of a plant extract, stizolobic acid, in the isolated rat spinal cord of the rat. Brain Res. 413, 193-197. Ishida M. and Shinozaki H. (1990b) New potent excitatory amino acids and marked potentiation of glutamate responses: pharmacology of conformationally restricted glutamate analogues. In Amino Acids: Chemistry, Biology and Medicine (Edited by Lubec G. and Rosenthal G. A.), pp. 3988409. ESCOM, Leiden. Johnston G. A. R., Curtis D. R., DeGroat W. C. and Duggan A. W. (1968) Central actions of ibotenic acid and muscimol. Biochem. Phartnacol. 17, 2488-2489. Johnston G. A. R., Curtis D. R., Davies J. and McCulloch R. M. (1974) Spinal interneurone excitation by conformationally restricted analogues of L-glutamic acid. Nature 248, 804-805. Lea T. and Usherwood P. N. R. (1973) Effects of ibotenic acid on chloride permeability of insect muscle fibres. Comp. Gen. Phariac. 4, 351-363. London E. and Covle J. T. (1979) Snecific bindine of rH-kainic acid to receptor sites in rat brain. Mol. Pharmacol. 15, 492-505. MacDonald J. F. and Nistri A. (1978) A comparison of the action of glutamate, ibotenate and other related amino acids on feline spinal interneurones. J. Physiol. 275, 449465. Miwa A., Ui M. and Kawai N. (1990) G protein is coupled to presynaptic glutamate and GABA receptors in lobster neuromuscular synapse. J. Neurophysiol. 63, 173-180. Monaghan D. T., Bridge R. J. and Cotman C. W. (1989) The excitatory amino acid receptors: the classes, pharmacology, and distinct properties in the function of the central nervous system. Ann. Rev. Pharmacol. Toxicol. 29, 365402. Nakagawa Y., Saito K., Ishihara T., Ishida M. and Shinozaki H. (1990) (2!3,3S,4S)-a-(carboxycyclopropyl)glycine is a novel agonist of metabotropic glutamate receptors. Eur. J. Pharmacol. 184, 205-206. Nicoletti F., Iadorola M. J., Wroblewski J. T. and Costa E. (1986) Excitatory amino acid recognition sites coupled with inositol phosphohpid metabolism: developmental changes and interaction with cc,-adrenoceptors. Proc. Natl. Acad. Sci. U.S.A. 83, 1931-1935. Palmer E., Monaghan D. T. and Cotman C. W. (1989) Trans-ACPD, a selective agonist of the phosphoinositidecoupled excitatory amino acid receptor. Eur. J. Pharmacol. 166, 585-587. Sharif N. A. and Roberts P. J. (1981) L-Aspartate binding sites in rat cerebellum: a comparison of the binding of
agonists
17
L-13H]aspartate and L-[3Hlglutamate to synaptic membranes. krain Res. 211,2!%303. _ Shinozaki H. (1980) The nharmacoloev of the excitatorv neuromuscular junction in the crayfiyh. Prog. Neurobioi 14, 121-155. Shinozaki H. (1988) Pharmacology of the glutamate receptor. Prog. Neurobiol. 30, 399435. Shinozaki H. and Ishida M. (1980) Inhibitory action of ibotenic acid on the crayfish neuromuscular junction. Brain Rex 198, 157-165. Shinozaki H. and Ishida M. (1988) Stizolobic acid, a competitive antagonist of the quisqualate-type receptor at the crayfish neuromuscular junction. Brain Res. 451, 353-356. Shinozaki H., Ishida M., Shimamoto K. and Ohfune Y. (1989a) A conformationally restricted analogue of Lglutamate, the threo-folded form of L-a-(carboxycyclopropyl)glycine, activates the NMDA-type receptor more markedly than NMDA in the isolated rat spinal cord, Brain Res. 480, 355-359. Shinozaki H., Ishida M., Shimamoto K. and Ohfune Y. (1989b) Potent NMDA-like actions and marked potentiation of glutamate responses of conformational variants of a glutamate analogue in the rat spinal cord, Br. J. Pharmacol. 98, 1213-1224. Shinozaki H. and Konishi S. (1970) Actions of several anthelmintics and insecticides on rat cortical neurones. Brain Res. 24, 368-371. Schoepp D., Bockaert J. and Sladeczek F. (I 990) Pharmacological and functional characteristics of metabotropic excitatory amino acid receptor. Trends Pharmacol. Sci. 11, 508-515. Sladeczek F., Pin J. P., Recasens M., Bockaert J. and Weiss S. (1985) Glutamate stimulates inositol phosphate formation in striatal neurons. Nature 317, 717-719. Sugiyama H., Ito I. and Hirono C. (1987) A new type of glutamate receptor linked to inositol phospholipid metabolism. Nature 325, 531-533. Sugiyama H., Ito I. and Watanabe M. (1989) Glutamate receptor subtypes may be classified into two major categories: a study on Xenopus oocytes injected with rat brain mRNA. Neuron 3, 129-132. Ui M. (1984) Islet-activating protein, pertussis toxin: a probe for functions of the inhibitory guanine nucleotide regulatory component of adenylate cyclase. Trends Pharmacol. Sci. 5, 277-279. Wafford K. A. and Sattelle D. B. (1989) L-Glutamate receptors on the cell body membrane of an identified insect motor neurone. J. exp. Biol. 144, 449-462.
