SYNAF'SE 12~206-213(1992)

Multiple Metabotropic Glutamate Receptors Regulate Hippocampal Function MANISHA A. DESAI, TIMOTHY S. SMITH, AND P. JEFFREY CONN Department of Pharmacology, Emory Uniuersity School of Medicine, Atlanta, Georgia 30322

KEY WORDS

Phosphoinositide hydrolysis, Neuromodulation, ACPD, AP3, EPSP, IPSP

ABSTRACT Selective activation of metabotropic glutamate receptors with trans-lamino-1,3-cyclopentanedicarboxylicacid (trans-ACPD) stimulates phosphoinositide hydrolysis and elicits three major physiological responses in area CA1 of the hippocampus. These include direct excitation of pyramidal cells, blockade of synaptic inhibition, and decreased transmission at the Schaffer collateral-CAl pyramidal cell synapse. Physiological effects of trans-ACPD are thought to be mediated by activation of phosphoinositide hydrolysis. However, it is now clear that multiple metabotropic glutamate receptor subtypes exist, some of which are not coupled to phosphoinositide hydrolysis, Thus, we performed a series of studies aimed a t determining whether the physiological effects of trans-ACPD in the hippocampus are mediated by activation of the predominant phosphoinositide hydrolysis-linked glutamate receptor. We report that ~-2-amino-3phosphonopropionic acid (L-AP ~ ), a n antagonist of trans-ACPD-stimulated phosphoinositide hydrolysis, does not inhibit the physiological effects of trans-ACPD in area CA1 at concentrations that maximally inhibit trans-ACPD-stimulated phosphoinositide hydrolysis in this region. Furthermore, lS,3S-ACPD activates the phosphoinositide hydrolysis-linked glutamate receptor but does not reduce evoked field excitatory postsynaptic potentials (EPSPs) in area CA1. However, we report that the physiological effects of 1R,3S- and lS,SR-ACPD are consistent with the hypothesis that these effects are mediated by activation of a metabotropic glutamate receptor. Thus, our data are consistent with the hypothesis that the physiological effects of trans-ACPD in area CA1 of the hippocampus are mediated by metabotropic glutamate receptors that are distinct from the AP3-sensitive phosphoinositide hydrolysis-linked glutamate receptor. 0 1992 Wiley-Liss, Inc.

see Conn and Desai, 1991; Schoepp et al., 1990a).Four INTRODUCTION members of the metabotropic glutamate receptor famGlutamate and other excitatory amino acids (EAAs) ily have been cloned (Houamed e t al., 1991; Masu et al., are the primary excitatory neurotransmitters in the 1991; Tanabe et al., 1992). These receptors have no vertebrate central nervous system (CNS). Until resequence homology and little structural similarity with cently, it was thought that all of the actions of glutaother G-protein-linked receptors; though they do have mate were mediated by receptors that belong to a famseven putative membrane-spanning regions typical of ily of ligand-regulated cation channels. These receptors G-protein-linked receptors, are referred to as ionotropic glutamate receptors and The most well characterized G-protein-linked recepinclude the N-methyl-D-aspartate (NNDA), and tor is coupled to phospholipase C (PLC) and activation kainate/[RSl-a-amino-3-hydroxy-5-methyl-isoxazoleof phosphoinositide hydrolysis. Glutamate-stimulated propionic acid ( W A M P A ) receptor subtypes (for reviews, see Collingridge and Lester, 1989; Monaghan phosphoinositide hydrolysis has been characterized in et al., 1989. It is now clear that a distinct family of brain slices and primary cell cultures from a variety of glutamate receptors exists, members of which are brain regions (for reviews, see Conn and Desai, 1991; coupled to various signal transduction processes Schoepp et al., 1990a). When expressed in Xenopus through GTP binding proteins (G-proteins). G-proteinlinked glutamate receptors are referred to as metaboReceived April 27, 1992; accepted in revised form June 8,1992. tropic glutamate receptors to distinguish them from Address reprint requests to P. Jeffrey Conn, Department of Pharmacology, members of the ionotropic receptor family for reviews, Emory University School of Medicine, Atlanta, GA 30322.0 1992 WILEY-LISS, INC.

HIPPOCAMPAL FUNCTION

oocytes, at least one of the metabotropic glutamate receptor clones couples to PLC in a pertussis toxin-sensitive manner. The pharmacological properties of this cloned receptor are similar to those of the phosphoinositide hydrolysis-linked glutamate receptor studied in intact neurons. The recent introduction of selective agonists of the metabotropic glutamate receptors, such as trans-lamino-1,3-cyclopentanedicarboxylic acid (trans-ACPD) (Desai and Conn, 1990; Palmer et al., 1989), has allowed determination of the physiological effects of metabotropic glutamate receptor activation. trans-ACPD has a number of actions in the mammalian CNS. These include depolarization of thalamic (Hall et al., 1979) and spinal cord (McLennan et al., 1982; McLennan and Liu, 1982) neurons, induction of membrane potential oscillations in neurons of the rat dorsolateral septa1 nucleus (Zheng and Gallagher, 1991), decreased excitatory postsynaptic potentials (EPSPs) in the striatum (Lovinger, 1991), and induction of Ca2+mobilization in cultured cerebellar neurons (Irving et al., 1990). The effects of trans-ACPD have been most thoroughly characterized in the hippocampus. trans-ACPD has three primary actions in hippocampal area CA1. These include a decrease in evoked EPSPs a t the Schaffer collateral-CAl pyramidal cell synapse (Baskys and Malenka, 1991a,b), direct excitatory effects on pyramidal cells (Desai and Conn, 1991; Stratton et al., 19891,and blockade of synaptic inhibition (Desai and Conn, 1991). The direct excitatory effects include pyramidal cell depolarization (accompanied by an increase in input resistance), blockade of spike frequency adaptation, and blockade of a slow afterhyperpolarizing potential (AHP) that follows a burst of action potentials (Desai and Conn, 1991; Stratton et al., 1989). Charpak et al. (1990) found that trans-ACPD has similar effects on pyramidal cells in area CA3. Because trans-ACPD is highly selective for the phosphoinositide hydrolysis-linked glutamate receptor relative to ionotropic glutamate receptors, actions of moderate concentrations of trans-ACPD are often assumed to be mediated by activation of phosphoinositide hydrolysis. However, the recent cloning of multiple metabotropic glutamate receptor subtypes (Tanabe et al., 1992) indicates that some of the actions of trans-ACPD may not be mediated by activation of the phosphoinositide hydrolysis-linked glutamate receptor. Furthermore, we recently found that trans-ACPD-stimulated phosphoinositide hydrolysis and trans-ACPD-induced excitation of CA1 pyramidal cells do not undergo a parallel developmental regulation, suggesting that these responses may be mediated by different receptors (Boss et al., 1992). Thus, we performed a series of studies aimed at determining whether the physiological effects of transACPD in the hippocampus are mediated by activation of the predominant phosphoinositide hydrolysis-linked glutamate receptor in this region. We provide evidence

207

that trans-ACPD-induced excitation of CA1 pyramidal cells, disinhibition, and reduction of EPSPs are each mediated by metabotropic glutamate receptors that are distinct from the AP3-sensitive phosphoinositide hydrolysis-linked glutamate receptor. MATERIALS AND METHODS Measurement of phosphoinositide hydrolysis Phosphoinositide hydrolysis was measured in crosschopped slices of whole hippocampus using the method described by Conn and Wilson (1991).A modification of this method was used to measure phosphoinositide hydrolysis in microdissected hippocampal subregions. Adult male Sprague-Dawley rats were decapitated, the brain was removed, and the hippocampi were dissected on ice. Transverse hippocampal slices (400 pm) were prepared using a McIlwain tissue chopper, suspended in Kreb's bicarbonate buffer containing glucose (KRB), and incubated for 30 minutes in a shaking water bath. All incubations were at 37°C and were under an atmosphere of 95% 0,/5% CO,. Slices were washed with warmed KRB and were transferred to test tubes containing KRB (four slices per test tube). Aliquots of KRB containing 2 pCi [3H] inositol were added to each tube to bring the volume to 225-250 pl, and the slices were incubated for 2 hours. LiCl (10 mM) and ~-2-amino-3) used) were phosphonopropionic acid ( L - A P ~(when then added to bring the final volume to 300 pl, and slices were incubated for an additional 45 minutes. KRB contained (in mM) 108 NaC1, 4.7 KC1, 2.5 CaC12, 1.2 MgSO,, 1.2 KH,P04, 25 NaHC03, and 10 glucose and was equilibrated with 95% 02/5% CO,. The reaction was stopped by placing the tubes on ice and washing slices with ice-cold KRB. Individual slices were then further microdissected on ice into areas CA1, CA3, and dentate gyrus (DG). Individual sections of each region from three microdissected slices were transferred to separate tubes containing 300 p1 KRB and 900 pl chloroform:methanol(l:2).Aqueous and organic phases were separated by adding 300 pl each of chloroform and 0.5 N HC1 and vortexing. One milliliter aliquots of the aqueous phase were added to anion exchange columns containing Dowex-1 in the formate form. Free [3Hlinositol and [3H1glyceroinositol phosphate were washed from the columns and discarded. [3H] inositol monophosphate ([3HlInsP) was eluted directly into scintillation vials, and radioactivity present in [3H]InsP was determined by liquid scintillation counting. Electrophysiological analysis of metabotropic glutamate receptor activation All experimental procedures used for the determination of physiological effects of trans-ACPD have been described previously (Desai and Conn, 19911, and were followed except for slight modifications in the slicemaking procedure. Briefly, adult male Sprague-Dawley

208

M.A. DESAI ET AL.

rats were decapitated, and brains were removed and placed in ice-cold artificial cerebrospinal fluid (ACSF) for 30 seconds. The hippocampus was dissected on ice, and transverse hippocampal slices (400 pm) were prepared using a McIlwain tissue chopper. Slices were incubated at room temperature for 1 hour before being transferred to the stage of a brain slice chamber where they were maintained fully submerged and continuously perfused with warmed ACSF (30°C). ACSF contained (in mM) 124 NaC1, 2.5 KC1, 1.3 MgSO,, 2.0 CaCl,, 1.0 NaH2P04, 26 NaHC03, and 10 glucose and was equilibrated with 95% 02/5% C02. trans-ACPD, lS,3R-ACPD, lS,3S-ACPD, L - A P ~and , D-2-amino-5-phosphonopentanoic acid (D-AP~) were obtained from Tocris Neuramin (Essex, UK). 1R,3SACPD was a generous gift from Dr. James A, Monn of Eli Lilly Research Laboratories (Indianapolis, IN).

RESULTS L - A P ~has previously been shown to inhibit transACPD-stimulated phosphoinositide hydrolysis in slices of whole hippocampus (Schoepp et al., 1990b). Thus, if the physiological effects of trans-ACPD are mediated by the major phosphoinositide hydrolysis-linked glutamate receptor in the hippocampus, they should be inhibited by addition of L-AP3. We began our studies by measuring the effects of L - A P ~on trans-ACPD-stimulated phosphoinositide hydrolysis in each of the major hippocampal subregions (CA1, CA3, and DG). This allowed us to verify that L - A P inhibits ~ this response in the specific subregion in which physiological studies ~ were conducted (area CA1). We found that L - A Peffectively inhibited trans-ACPD-stimulated phosphoinositide hydrolysis in each of the hippocampal subregions examined (Fig. 1.) These results are in close agreement with those from slices of whole hippocampus previously reported by Schoepp et al., (1990b). Thus, if the physiological effects of trans-ACPD in area CA1 are mediated by the phosphoinositide hydrolysis-linked glutamate receptor, they should be blocked by 1mM L-AP3.There~ the fore, we next examined the effects of 1mM L - A Pon various physiological responses to trans-ACPD. Physiological responses measured included transACPD-induced excitation of CA1 pyramidal cells, disinhibition, and reduction of evoked EPSPs at the Schaffer collateral CA1 pyramidal cell synapse. As reported previously (Desai and Conn, 1991; Stratton et al., 19891, 30-250 FM trans-ACPD induced a depolarization of CA1 pyramidal cells, reduced the slow AHP that follows a burst of action potentials (data not shown), and reduced spike frequency adaptation (Fig. 2 . ) We found that L - A P alone ~ had no effect on spike frequency adap~ ineffective as a n antagtation; furthermore, L - A Pwas onist of trans-ACPD-induced blockade of spike frequency adaptation at a concentration that maximally inhibited phosphoinositide hydrolysis (Fig. 2). Likewise, L - A Phad ~ no effect on cell membrane potential or

L-APS

-

500

r m ACPD

Q

3

n 91 > 0

400

300

r)

Q c

c

200

91 0

; 100

a

n "

CA1

CA3

DG

Fig. 1. Effect of L - A P ~on trans-ACPD-stimulated increase in phosphonositide hydrolysis in areas CA1, CA3, and DG. Phosphoinositide hydrolysis was measured in transverse hippocampal slices which were further microdissected into areas CA1, CA3, and DG. Phosphoinositide hydrolysis is expressed as percent increase in radioactivity in r3H]InsP above basal. L - A P (1 ~ mM final concentration), when used, was added 15 minutes prior to addition of 250 pM transACPD. Incubation continued for an additional 45 minutes. Each bar is the mean ( ? S.E.M.) of three experiments, each done in triplicate.

Drug

Control

250pM ACPD

-

250pM ACPD +L-AP3

7 -_r

L

Fig. 2. L - A Pdoes ~ not block trans-ACPD-induced blockade of spike frequency adaptation in CA1 pyramidal cells. Intracellular recordings were made from CA1 pyramidal cells with microelectrodes containing 3 M KC1 or 2 M KMeSO,. The voltage response to injection of prolonged (800 msec) depolarizing current was measured before and during perfusion with L-AP3,trans-ACPD, and trans-ACPD plus L-AP3. When the response to trans-ACPD in the presence of L - A Pwas ~ measured, t-AP3 was added at least 20 minutes before addition of transACPD, and baseline traces were taken in the presence of L - A P ~Rest. ing potential was maintained with constant current injection. Data shown are from different cells for each treatment and are representative of four cells for AP3, four cells for trans-ACPD, and seven cells for AP3 and ACPD. Action potentials are truncated by digitization of the data.

the slow AHP either alone or in the presence of transACPD (data not shown). In addition to its direct excitatory effects on CA1 pyramidal cells, trans-ACPD also increases excitability in area CA1 by decreasing synaptic inhibition (Desai and

HIPPOCAMPAL FUNCTION 1 mM L-AP3

2 5 0 p M ACPD

Control

250pM ACPD + L-AP3

Fig. 3. L - A P does ~ not block trans-ACPD-induced reduction of IPSPs in CA1 pyramidal cells. Superimposed traces of IPSPs elicited before and during perfusion with either L - A Por ~ truns-ACPD or both. When the response the trans-ACPD in the presence of L - A P ~was measured, L - A P ~ was added at least 20 minutes before addition of truns-ACPD, and baseline traces were taken in the presence of L-AP3. Each pair of traces is from a separate cell. Arrows mark traces elicited during perfusion with drug, Intracellular recordings were made from CA1 pyramidal cells with microelectrodes containing 2 M KMeSO,. Resting potential was maintained with constant current injection. L-AP3(n = 4)did not decrease IPSPs whereas truns-ACPD markedly decreased IPSPs (four of seven cells). L - A P ~did not inhibit trunsACPD-induced reduction of IPSPs in any of the cells examined (n = 3).

Conn, 1991). To determine if this effect of trans-ACPD is mediated by the AP3-sensitive phosphoinositide hydrolysis-linked receptor, we measured the effect of L - A P on ~ truns-ACPD-induced reduction of inhibitory postsynaptic potentials (IPSPs)in CA1 pyramidal cells. L-AP3alone had no effect on IPSPs, whereas 250 pM truns-ACPD caused a marked reduction of IPSP amplitude (Fig. 3.) As with truns-ACPD-induced excitation of CA1 pyramidal cells, truns-ACPD-induced reduction of ~ 3). IPSPs was not inhibited by 1mM L - A P (Fig. truns-ACPD has also been reported to reduce EPSPs elicited by stimulation of Schaffer collateral afferents to CA1 pyramidal cells (Baskys and Malenka, 1991a,b). To determine if L-AP3 inhibits this effect of trunsACPD, we measured extracellular field EPSPs elicited by stimulation of Schaffer collateral afferents and recorded from the dendritic region of CA1 pyramidal cells. We found that 250 pM truns-ACPD decreased the initial slope of field EPSPs and that, in common with the other physiological responses examined, this effect ~ 4). Although 1mM was not blocked by 1mM L - A P (Fig. L-AP~ alone had no effect on the initial slope of the EPSP in most slices examined, it did decrease EPSPs in two of nine slices. This effect of L - A P alone ~ made it difficult to interpret the finding that L - A P did ~ not inhibit ACPD-induced depression of field EPSPs. We next used the various isomers of ACPD to further characterize the pharmacological properties of the receptors that mediate the different physiological effects of truns-ACPD. truns-ACPD is a racemic mixture of 1S,3R- and 1R,3S- stereoisomers. To date, all of the metabotropic glutamate receptor subtypes that have been characterized are selectively activated by relatively low concentrations of lS,3R-ACPD but not lR,3S-ACPD. For instance, 10-100 pM lS,3R-ACPD increases phosphoinositide hydrolysis (Schoepp et al., 1991) and cyclic adenosine monophosphate (CAMP)accumulation (Winder and Conn, 1992) and inhibits forskolin-stimulated CAMPaccumulation (Schoepp et al.,

209

250eM +L-APB

*

Drug

7

Fig. 4. L - A P does ~ not block trans-ACPD-induced reduction of field EPSPs at the Schaffer collateral-CAl pyramidal cell synapse. Field EPSPs were evoked by stimulation of Schaffer collateral afferents and recorded from the dendritic region of CA1 pyramidal cells. Each pair of traces is from a separate slice. L - A P had ~ no effect on the initial slope of extracellular field EPSPs in seven of nine slices, although it did reduce EPSPs in the remaining two slices (not shown). truns-ACPD (250 pM) markedly decreased the initial slope of field EPSPs in 8 of 12 slices. L - A P did ~ not inhibit ACPD-induced reduction of EPSPs in any of five cells examined. When the response to truns-ACPD in the presence of L - A P was ~ measured, L - A P ~was added at least 20 minutes before addition of truns-ACPD, and baseline traces were taken in the presence of L-AP3.

1992a) in hippocampal slices. In contrast, comparable concentrations of lR,3S-ACPD do not elicit any of these responses. Similarly, previously characterized behavioral responses to metabotropic glutamate receptor activation are selectively elicited by 1S,3R-, but not lR,3S-ACPD (Sacaan et al., 1991; Sacaan and Schoepp, 1992; Schoepp et al., 1992b). The effects of the isomers of cis-ACPD have been less well characterized. lR,3R-ACPD is a potent NMDA receptor agonist (Curry et al., 1988; McLennan and Curry, 1988), and is therefore not useful for studying metabotropic glutamate receptor function. In contrast, lS,3S-ACPD has no discernable effect on ionotropic glutamate receptors (Sunter et al., 1991). However, we have reported that lS,3S-ACPD has equal potency and efficacy with lS,SR-ACPD at stimulating increases in CAMP accumulation (Winder and Conn, 1992). Nonetheless, the effects of lS,SS-ACPD on phosphoinositide hydrolysis and other metabotropic glutamate receptormediated responses have not been determined. Therefore, we first determined the effect of lS,3S-ACPD on phosphoinositide hydrolysis. lS,3S-ACPD induced a concentration-dependent increase in phosphoinositide hydrolysis in cross-chopped hippocampal slices (Fig. 5). The EC50 of this response was 105 pM which is approximately 2.5 times the previously reported EC50 of lS,3R-ACPD at stimulating phosphoinositide hydrolysis in the same preparation (Schoepp et al., 1991).The maximum response to lS,3S-ACPD was similar to that elicited by lS,3R-ACPD (data not shown). These data, combined with the previous report that 1S,3S-ACPD stimulates cyclic AMP accumulation, suggest that

M.A. DESAI ET AL.

210

Control

ACPD

1000

a

> 0 a

25OpM 1R3S

800 600

((J 4-

C

a 2 a

a

400 200

1 OOpM 1S3R

0

i

4.5

4

3.5

L

3

-Log [ l S,SS-ACPD] (M) Fig. 5. lS,3S-ACPD elicits a concentration-dependent increase in phosphoinositide hydrolysis in rat hippocampal slices. Phosphoinositide hydrolysis was measured in cross-chopped hippocampal slices. Phosphoinositide hydrolysis is expressed as percent increase in radioactivity in [3H]InsPabove basal. Data shown are the mean (5S.E.M.) of four experiments, each done in triplicate.

1OOpM

1S,3S

Fig. 6. Blockade of spike frequency adaptation is elicited by 1S,3Rand lS,3S-ACPD, but not by lR,3S-ACPD. 1R-3s-ACPD (100-250 pM) did not block spike frequency adaptation in four of four cells, whereas lS,BR-ACPD (30-125 p M ) (n = 10) and lS,3S-ACPD (100125 pM) (n = 8 ) blocked spike frequency adaptation in all cells examined. Intracellular recordings were made from CAI pyramidal cells with microelectrodes containing 3 M KC1 or 2 M KMeSO, as in Figure 2. The voltage response to injection of prolonged (800 msec) depolarizing current was measured before and during perfusion with isomers of ACPD. Resting potential was maintained with constant current injection. Each pair of traces is from a separate cell. Action potentials are truncated by digitization of the data.

lS,SS-ACPD activates at least some metabotropic glutamate receptor subtypes. We next examined the effects of the various isomers of ACPD on each of the physiological effects described in area CA1 of the hippocampus. The effects of 1S,3SACPD were measured in the presence of 20 p M D - A P ~ to eliminate the effects of possible contamination by Control ACPD lR,3R-ACPD (Sunter et al., 1991). In some cases, D - A P ~ was also included in experiments measuring the response to lS,3R-ACPD. Consistent with previous studies (Desai and Conn, 1991; Salt and Eaton, 1991; Stratton et al., 1990), we found that physiological responses to lS,3R-ACPD were identical in the presence o r absence of D - A P ~Thus, . we combined data from experiments with and without D-AP5. Consistent with the loopM 1 S.3R pharmacology of the various metabotropic glutamate receptors characterized to date, we found that 30-125 p M lS,3R-ACPD blocked spike frequency adaptation in all CA1 pyramidal cells examined, whereas 30-250 pM lR,3S-ACPD did not reduce spike frequency adaptation Fig. 7. 1S,3R- and 1S,3S-, but not lR,3S-ACPD reduce the ampliin any of the cells examined (Fig. 6). High concentra- tude of intracellular IPSPs. 100-125 pM 1S,3R-ACPD (five of seven tions (500 pM) of lR,3S-ACPD did induce a slight block cells) and 100-125 pM lS,3S-ACPD (five of seven cells) reduced the of spike frequency adaptation in two of three cells. How- amplitude of IPSPs by greater than 25%, whereas higher concentrations of lR,BS-ACPD (250-500 pM) had no effect on IPSP amplitude in ever, the magnitude of this effect was much less than any offive cells. Intracellular recordings were made from CAI pyramithat seen with 30-125 pM lS,3R-ACPD. lS,3S-ACPD dal cells with microelectrodes containing 2 M KMeSO, as in Figure 3. (100-125 p M ) also blocked spike frequency adaptation Resting potential was maintained with constant current injection. Each pair of traces is from a separate cell. in all CA1 pyramidal cells examined (Fig. 6). Pyramidal cell depolarization and blockade of the slow AHP were also seen with 1S,3R- and lS,3S-ACPD, but not with lR,BS-ACPD (data not shown). Consistent with the isomer selectivity seen for the We found that 100-125 pM lS,3R-ACPD caused a reduction of IPSP amplitude similar to that seen with other physiological responses, lS,3R-ACPD (100-125 pM), but not lR,3S-ACPD (250-500 pM) reduced the trans-ACPD. In contrast, 250-500 p M lR,3S-ACPD did not decrease IPSP amplitude in any of the cells initial slope of evoked EPSPs a t the Schaffer collateralexamined. 1S,3S-ACPD (100-125 p M ) had an effect on CA1 pyramidal cell synapse. Interestingly, this effect IPSPs that was identical to that of lS,3R-ACPD or was not mimicked by lS,3S-ACPD (100-125 p M ) (Fig. 8). trans-ACPD (Fig. 7).

?/---

HIPPOCAMPAL FUNCTION

ACPD

Control 1R,3S

b--

1OOpM 1S,3R

ii/__

250uM

1

1OOpM 1s,3s

Fig. 8. Neither 1S,3S- nor lR,SS-ACPD mimics lS,3R-ACPD-induced reduction of extracellular field EPSPs at the Schaffer collateralCA1 pyramidal cell synapse. Field EPSPs were evoked by stimulation of Schaffer collateral afferents and recorded from the dendritic region of CAI pyramidal cells as in Figure 4.1R,SS-ACPD (250-500 p M )had no effect on the initial slope of field EPSPs in four of four slices, whereas 100-125 pM lS,SR-ACPD caused a marked reduction in the initial slope of the EPSP in 7 of 10 slices. lS,3S-ACPD (100-125 pM) did not mimic this effect of lS,3R-ACPD (five of six slices). Each pair of traces is from a separate slice.

DISCUSSION Our data suggest that the major physiological effects of trans-ACPD that have been described in area CA1 of the hippocampus are mediated by a metabotropic glutamate receptor that is distinct from the predominant phosphoinositide hydrolysis-linked metabotropic glutamate receptor in this structure. The conclusion that ACPD-induced disinhibition and excitation of CA1 pyramidal cells is not mediated by the phosphoinositide hydrolysis-linked glutamate receptor is based on the finding that L - A P effectively ~ inhibits trans-ACPD-induced phosphoinositide hydrolysis in hippocampal area CA1, but it does not inhibit these physiological responses to trans-ACPD. These data are consistent with the previous reports that D , L - A P 3 does not inhibit transACPD-induced direct excitatory effects in CA1 pyramidal cells (Stratton et al., 1990), and that L - A Pdoes ~ not inhibit trans-ACPD-induced direct excitatory effects in CA3 pyramidal cells (Charpak and Gahwiler, 1991). The present data extend previous findings to ACPDinduced disinhibition and reduction of field EPSPs in hippocampal area CA1. However, our present data and previous reports (Schoepp et al., 1990b) suggest that L - A P serves ~ as a weak partial agonist a t the phosphoinositide hydrolysis-linked glutamate receptor. Thus, L - A P could ~ fail to inhibit the response to trans-ACPD because of its partial agonist action. If this were the case, L - A P should ~ mimic (or partially mimic) the effect ~ of trans-ACPD when added alone. However, L - A P had no effect on synaptic inhibition or pyramidal cell excitability when added in the absence of trans-ACPD. These data are consistent with the hypothesis that these effects of trans-ACPD are not mediated by activation of the major phosphoinositide hydrolysis-linked

211

glutamate receptor in area CA1. Also consistent with this hypothesis, we previously found that the direct excitatory effects of trans-ACPD and trans-ACPD-stimulated phosphoinositide hydrolysis do not undergo parallel developmental regulation (Boss et al., 1992). Glutamate- and ACPD-stimulated phosphoinositide hydrolysis are many times greater in slices from 6-7day-old rats than from adults. In contrast, the effects of ACPD and glutamate on spike frequency adaptation and AHPs are similar throughout development. Although these data are consistent with our hypothesis that trans-ACPD-induced direct excitation of pyramidal cells are mediated by a metabotropic glutamate receptor that is distinct from the phosphoinositide hydrolysis-linked receptor, it should be noted that a slight phosphoinositide hydrolysis response to ACPD may remain in the presence of L - A P ~and , the lack of parallel developmental regulation of the two responses is indirect evidence of mediation by different receptors. Thus, we cannot definitively rule out a role of phosphoinositide hydrolysis in mediating the responses examined. We also found that L-AP3fails to inhibit trans-ACPDinduced decrease in evoked EPSPs at the Schaffer collateral-CAl pyramidal cell synapse. However, in contrast with its lack of effect on synaptic disinhibition and excitability of CA1 pyramidal cells, L-AP3 decreased evoked EPSPs in two of nine slices. It is possible that this is due to the weak partial agonist effects of L - A P ~ on phosphoinositide hydrolysis. However, if L-AP3is a partial agonist at the receptor th t mediates ACPD-induced reduction of EPSPs, it should occlude the effect of ACPD when added prior to ACPD application. Yet, in each slice examined, baseline measurements were made in the presence of L - A P for ~ at least 20 minutes prior to addition of trans-ACPD. Under these conditions, trans-ACPD markedly reduced EPSPs in all slices examined. Furthermore, we found that 1S,3SACPD stimulates phosphoinositide hydrolysis with similar efficacy to lS,3R-ACPD but did not reduce evoked EPSPs. Taken together, these data suggest that ACPD-induced reduction of EPSPs may also be mediated by a receptor that is distinct from the phosphoinositide hydrolysis-linked glutamate receptor. Although the physiological effects of trans-ACPD investigated in these studies are not mediated by the AP3-sensitive phosphoinositide hydrolysis-linked glutamate receptor, our data clearly suggest that these effects are mediated by metabotropic glutamate receptors. Thus, in common with all responses to metabotropic glutamate receptor activation studied to date (Irving et al., 1990; Sacaan et al., 1991; Sacaan and Schoepp, 1992; Schoepp et al., 1991, 1992a,b; Winder and Conn, 1992), each of these physiological responses is elicited by 1S,3R-, but not lR,3S-ACPD. Also, we report that lS,3S-ACPD mimics lS,3R-ACPD-induced disinhibition and the excitatory effects of lS,3R-ACPD on CA1 pyramidal cells. This, coupled with our findings

212

M.A. DESAI ET AL.

that 1S73S-ACPDstimulates phosphoinositide hydrolyREFERENCES sis (present studies) and CAMPaccumulation (Winder Baskys, A., and Malenka, R.C. (1991a)Trans-ACPD depresses synapand Conn, 1992) suggests that lS,3S-ACPD is a n agotic transmission in the hippocampus. Eur. J. Pharmacol., 193:131132. nist at some metabotropic glutamate receptors. How- Baskys, A., and Malenka, R.C. (1991b)Agonists at metabotropic glutaever, as discussed above, lS,3S-ACPD does not reduce mate receptors presynaptically inhibit EPSCs in neonatal rat hippocampus. J . Physiol., 444:687-701. evoked field EPSPs at the Schaffer collateral-CAl pyraBoss, V., Desai, M.A., Smith, T.S., and Conn, P.J. (1992) Trans-ACPDmidal cell synapse, suggesting that lS,3S-ACPD may induced phosphoinositide hydrolysis and modulation of hippocampal pyramidal cell excitability do not undergo parallel developserve as a n agonist at some, but not all, metabotropic mental regulation. Brain Res. (in press). glutamate receptors. Charpak, S., and Gahwiler, B.H. (1991) Glutamate mediates a slow ~ not inhibit the major physiological synaptic response in hippocampal slice cultures. Proc. R. SOC. Lond. Since L - A P does [Bioll, 243:221-226. responses described in hippocampal area CA1, the Charpak, S., Gahwiler, B.H., Do, K.Q., and Knopfel, T. (1990) Potasquestion arises as to the physiological roles of the phossium conductances in hippocampal neurons blocked by excitatory amino-acid transmitters. Nature, 347:765-767. phoinositide hydrolysis-linked AP3-sensitive glutaCollingridge, G.L., and Lester, R.A.J. (1989) Excitatory amino acid mate receptor. It is possible that this receptor plays a receptors in the vertebrate central nervous system. Pharmacol. Rev., 40:143-210. role in various forms of synaptic plasticity, including Conn, P.J., and Desai, M.A. (1991) Pharmacology and physiology of long-term potentiation (LTP). For instance, transmetabotropic glutamate receptors in mammalian central nervous system. Drug Dev. Res., 24:207-229. ACPD enhances tetanus-induced LTP in area CA1 P.J., and Wilson, K.M. (1991) Modifications to phosphinositide (McGuinness et al., 1991; Otani and Ben-Ari, 1991)and Conn, signalling. in: Molecular Neurobiology: A Practical Approach. J . induces LTP a t the mossy fiber-CA3 pyramidal cell synChad and H. Wheal, eds. IRL Press, Oxford, pp. 115-133. K., Peet, M.J., Magnuson, D.S.K., and McLennan, H. (1988) apse (Ito and Sugiyama, 1991). Furthermore, D,L -A P ~ Curry, Synthesis, resolution and absolute configuration of the isomers of inhibits LTP at the mossy fiber-CA3 synapse (Ito and the neuronal excitant l-amino-1,3-cyclopentanedicarboxylic acid. J . Med. Chem., 31:864-867. Sugiyama, 19911, and at the Schaffer collateral-CAl Desai, M.A., and Conn, P.J. (1990) Selective activation of phosphoisynapse (Izumi et al., 1991). 1AP3 also partially inhibnositide hydrolysis by a rigid analogue of glutamate. Neurosci. Lett., 109:157-162. its certain behavioral responses to ACPD that may inM.A., and Conn, P.J. (1991) Excitatoryeffects ofACPD receptor volve synaptic plasticity (Sacaan et al., 1991; Sacaan Desai, activation in the hippocampus are mediated by direct effects on pyramidal cells and blockade of synaptic inhibition. J. Neurophysand Schoepp, 1992). These ~-AP3-sensitiveresponses iol., 66:40-52. are candidates for responses that are mediated by the Hall, J.G., Hicks, T.P., McLennan, H., Richardson, T.L., and Wheal, AP3-sensitive phosphoinositide hydrolysis-linked gluH.V. (1979)The excitation ofmammalian central neurones by amino acids. J . Physiol., 286:29-39. tamate receptor. However, it is possible that L-AJ?~ also Houamed, K.M.. Kuiiuer. J.L.. Gilbert. T.L.. Haldeman. B.A.. OHara. blocks other yet uncharacterized metabotropic glutaP.J., Mulvihill, E.R., k m e r s , W., and Hagen, F.S. (1991)'Cloning; expression, and gene structure of a G protein-coupled glutamate mate receptors that are not coupled to phosphoinositide receptor from rat brain. Science 252:131&1321. hydrolysis. Irving, A.J., Schofield, J.G., Watkins, J.C., Sunter, D.C., and Collingridge, G.L. (1990) lS,3R-ACPD stimulates and L-AP3blocks Ca2 Previous studies indicate that stimulation of metabomobilization in rat cerebellar neurons. Eur. J. Pharmacol., 186:363tropic glutamate receptors results in activation of at 365. least four signal transduction mechanisms. These in- Ito, I., and Sugiyama, H. (1991) Roles of glutamate receptors in longterm potentiation at hippocampal mossy fiber synapses. NeuroReclude an increase in phosphoinositide hydrolysis (Desai port 2:333-336. and Conn, 19901, a decrease in a calcium conductance Izumi, Y., Clifford, D.B., and Zorumski, C.F. (1991) 2-Amino-3phosphonopropionate blocks the induction and maintenance of long(Lester and Jahr, 19901, inhibition of adenylate cyclase term potentiation in rat hippocampal slices. Neurosci. Lett., (Schoepp et al., 1992a; Tanabe et al., 1992), and a n 122~187-190. increase in CAMP accumulation (Winder and Conn, Lester, R.A.J., and Jahr, C.E. (1990) Quisqualate receptor-mediated depression of calcium currents in hippocampal neurons. Neuron, 1992). I t is likely that activation of each of these effector 4:741-749. systems contributes to the net physiological effects of Lovinger, D.M. (1991) Trans-l-aminocyclopentane-1,3-dicarboxylic acid (t-ACPD) decreases synaptic excitation in rat striatal slices trans-ACPD on hippocampal neurons. In future studthrough a presynaptic action. Neurosci. Lett., 129:17-21. ies, it will be important to characterize the metabotro- Masu, M., Tanabe, Y., Tsuchida, K., Shigemoto, R., and Nakanishi, S. (1991) Sequence and expression of a metabotropic glutamate receppic glutamate receptors that mediate the various phystor. Nature, 349:760-765. iological effects described in the present study and to McGuinness, N., Anwyl, R., and Rowan, M. (1991) Trans-ACPD enhances long-term potentiation in the hippocampus. Eur. J. Pharmaidentify the signal transduction mechanisms employed col., 197:231-232. by these receptors. Furthermore, it will be important to McLennan, H., and Curry, K. (1988) The actions of cyclopentane analogues of glutamic acid at binding sites for kainic and glutamic determine whether the receptors that mediate these acids. Exp. Brain Res., 72:436438. responses correspond to any of the recently cloned me- McLennan, H., Hicks, T.P., and Liu, J.R. (1982) On the configuration of the receptors for excitatory amino acids. Neuropharmacology, tabotropic glutamate receptor subtypes. +

ACKNOWLEDGMENTS This work was supported by a National Institutes of Health grant NS-28405-01 (P.J.C.), and a predoctoral fellowship from Lilly Research Laboratories (M.A.D.).

21:549-554. McLennan, H., and Liu, J. 11982) The action of six antagonists of the excitatory amino acids on neurones ofthe rat spinal cord. Exp. Brain Res., 45:151-156. Monaghan, D.T.. Bridges. R.J.. and Cotman. C.W. (1989) The excitatoriamino acid receptors: their classes, pharmacology, and distinct properties in the function of the central nervous system. Ann. Rev. Pharmacol. Toxicol., 29:365-402.

213

HIPPOCAMPAL FUNCTION Otani, S., and Ben-Ari, Y. (1991) Metabotropic receptor-mediated long-term potentiation in rat hippocampal slices. Eur. J . PharmaCO~., 205:325-326. Palmer, E., Monaghan, D.T., and Cotman, C.W. (1989) Trans-ACPD, a selective agonist of the phosphoinositide-coupledexcitatory amino acid receptor. Eur. J . Pharmacol., 166:585-587. Sacaan, A.I., Monn, J.A., and Schoepp, D.D. (1991) Intrastriatal injection of a selective metabotropic excitatory amino acid receptor agonist induces contralateral turning in the rat. J . Pharmacol. Exp. Ther., 259:1366-1370. Sacaan, A.I., and Schoepp, D.D. (1992) Activation of hippocampal metabotropic excitatory amino acid receptors leads to seizures and neuronal damage. Neurosci. Lett., 139:77-82. Salt, T.E., and Eaton, S.A. (1991) Excitatory actions of the metabotropic excitatory amino acid receptor agonist, trans-(+)-1-amino-cyclopentane-l,3-dicarboxylate (t-ACPD), on rat thalamic neurons in vivo. Eur. J . Neurosci., 3:1104-1111. Schoepp, D., Bockaert, J., and Sladeczek, F. (1990a) Pharmacological and functional characteristics of metabotropic excitatory amino acid receptors. Trends Pharmacol. Sci., 11:508-515. Schoepp, D.D., Johnson, B.G., and Monn, J.M. (1992a) Inhibition of cyclic AMP formation by a selective metabotropic glutamate receptor agonist. J . Neurochem., 58:1184-1186. Schoepp, D.D., Johnson, B.G., Sacaan, A.I., True, R.A., and Monn, J.M. (1992b) In vitro and in vivo pharmacology of 1S,3R- and 1R,3SACPD: evidence for a role of metabotropic glutamate receptors in striatal motor function. Mol. Neuropharmacol. (in press). Schoepp, D.D., Johnson, B.G., Smith, E.C.R., and McQuaid, L.A. (1990b) Stereoselectivity and mode of inhibition of phosphoinosit-

ide-coupled excitatory amino acid receptors by 2-amino-3phosphonopropionic acid. Mol. Pharmacol., 38:222-228. Schoepp, D.D., Johnson, B.G., True, R.A., and Monn, J.A. (1991) Comparison of (1S,3R)-1-aminocyclopentane-1,3-dicarboxylicacid (1S,3R-ACPD)- and lR,3S-ACPD-stimulated brain phosphoinositide hydrolysis. Eur. J . Pharmacol. 207:351-353. Stratton. K.R.. Worlev. P.F.. and Baraban. J.M. (1989) Excitation of hippocampal neurins by 'stimulation of glutamate QP receptors. Eur. J . Pharmacol., 173:235-237. Stratton, K.R., Worley, P.F., and Baraban, J.M. (1990) Pharmacological characterization of phosphoinositide-linked glutamate recepcor excitation of hippocampal neurons. Eur. J. Pharmacol., 186:357361. Sunter, D.C., Edgar, G.E., Pook, P.C-K., Howard, J.A.K., Udvarhelyi, P.M., and Watkins, J.C. I1991)Actions of the four isomers 011-aminocyclopentane-l,3-decarboxylate(ACPD) in the heniisected isolated spinal cord of the neonatal rat. Br. J. Pharmacol., 104. I'anabe, Y., Maw, M., Ishii. T., Shigemoto, H., and Nakanishi, S. (1992) A family of metaboiropic glutamate receptors. Neuron., 8:169-179. Winder, D.G., and Conn, P.J. (1992)Activation of metabotropic glutamate receptors in the hippocampus increases cyclic A M P accumulation. J. Neurochem., 59:375-378. Zheng, F., and Gallagher, J.P. (1991I Truns-ACPI) (truns-I),L,-lamino-l,3-cyclopentanedicarboxylic acid) elicited oscillation of membrane potentials in rat dorsolateral septa1 nucleus neurons recorded intracellularly in vitro. Neurosci. Lett., 125:147-150. ,

~~~~~~~