- Email: [email protected]
The GluR5 subtype of kainate receptor regulates excitatory synaptic transmission in areas CA1 and CA3 of the rat hippocampus
Neuropharmacology 37 (1998) 1269 – 1277
The GluR5 subtype of kainate receptor regulates excitatory synaptic transmission in areas CA1 and CA3 of the rat hippocampus M. Vignes a,b, V.R.J. Clarke a, M.J. Parry a, D. Bleakman c,d, D. Lodge c,d, P.L. Ornstein d, G.L. Collingridge a,* b
a Department of Anatomy, Uni6ersity of Bristol, Uni6ersity Walk, Bristol BS8 1TD, UK Laboratoire ‘Plasticite´ Ce´re´brale’, EP 628 CNRS, Uni6ersite´ Montpellier II, Place Euge`ne Bataillon, 34095 Montpellier cedex 05, France c Lilly Research Centre Ltd., Erl Wood Manor, Windlesham GU20 6PH, UK d Eli Lilly and Company Ltd., Lilly Corporate Center, Indianapolis, IN 46285, USA
Accepted 12 August 1998
Abstract Activation of kainate receptors depresses excitatory synaptic transmission in the hippocampus. In the present study, we have utilised a GluR5 selective agonist, ATPA [(RS)-2-amino-3-(3-hydroxy-5-tert-butylisoxazol-4-yl)propanoic acid], and a GluR5 selective antagonist, LY294486 [(3SR,4aRS,6SR,8aRS)-6-({[(1H-tetrazol-5-yl)methyl]oxy}methyl)-1,2,3,4,4a,5,6,7,8,8a-decahydroisoquinoline-3-carboxylic acid], to determine whether GluR5 subunits are involved in this effect. ATPA mimicked the presynaptic depressant effects of kainate in the CA1 region of the hippocampus. It depressed reversibly AMPA (a-amino-3-hydroxy-5-methylisoxazole-4-propionic acid) receptor-mediated field excitatory postsynaptic potentials (field EPSPs) with an IC50 value of :0.60 mM. The dual-component excitatory postsynaptic current (EPSC) and the pharmacologically isolated NMDA (N-methyl-D-aspartate) receptor-mediated EPSC were depressed to a similar extent by 2 mM ATPA (61 9 7% and 58 96%, respectively). Depressions were associated with an increase in the paired-pulse facilitation ratio suggesting a presynaptic locus of action. LY294486 (20 mM) blocked the effects of 2 mM ATPA on NMDA receptor-mediated EPSCs in a reversible manner. In area CA3, 1 mM ATPA depressed reversibly mossy fibre-evoked synaptic transmission (by 82 9 10%). The effects of ATPA were not accompanied by any changes in the passive properties of CA1 or CA3 neurones. However, in experiments where K + , rather than Cs + , containing electrodes were used, a small outward current was observed. These results show that GluR5 subunits comprise or contribute to a kainate receptor that regulates excitatory synaptic transmission in both the CA1 and CA3 regions of the hippocampus. © 1998 Elsevier Science Ltd. All rights reserved. Keywords: ATPA; EPSP; Hippocampus; Kainic acid; LY294486; Mossy fibre; Presynaptic receptor; Synaptic transmission
1. Introduction It has been known for many years that kainate receptors can be distinguished from AMPA receptors (Davies et al., 1979, McLennan and Lodge, 1979, Watkins and Evans, 1981). However, the lack of selective kainate receptor agonists and antagonists has, until recently, greatly hindered the identification of the roles of kainate receptors in the vertebrate central nervous system. The use of the selective non-competitive AMPA receptor antagonists GYKI52466 (Tarnawa et al., 1990, * Corresponding author. Fax: +44 117 92916870.
Palmer and Lodge, 1993, Donevan and Rogawski, 1993, Zorumski et al., 1993) or GYKI53655 (Paternain et al., 1995, Wilding and Huettner, 1995, Partin and Mayer, 1996, Bleakman et al., 1996a) to prevent activation of AMPA receptors has allowed several effects to be ascribed to activation of kainate receptors. Most information has been obtained using the hippocampal slice preparation, in which it has been shown that kainate receptors mediate: (1) an inward current in CA1 and CA3 neurones (Chittajallu et al., 1996, Vignes et al., 1996, Castillo et al., 1997), (2) a synaptic current in CA3 neurones resulting from mossy fibre stimulation (Castillo et al., 1997, Vignes and Collingridge, 1997),
0028-3908/98/$ - see front matter © 1998 Elsevier Science Ltd. All rights reserved. PII: S0028-3908(98)00148-8
1270
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
(3) depression of g-aminobutyric acid (GABA)-mediated synaptic transmission (Clarke et al., 1997, Rodrı´guez-Moreno et al., 1997) and (4) depression of excitatory synaptic transmission (Chittajallu et al., 1996). The recent discovery of compounds (Bleakman et al., 1996b, Clarke et al., 1997) that are selective for the GluR5 kainate receptor subunit (Bettler et al., 1990) versus the other known subunits (GluR6, GluR7, KA-1 and KA-2; see Bettler and Mulle, 1995) led to the identification of a role for GluR5 subunits in the first three effects (Clarke et al., 1997, Vignes et al., 1997). In the present study, we have used the GluR5 selective agonist ATPA [(RS)-2-amino-3-(3-hydroxy-5-tertbutylisoxazol-4-yl)propanoic acid] and the GluR5 selective antagonist LY294486 to determine whether GluR5 subunits are involved in the presynaptic regulation of excitatory synaptic transmission in the hippocampus.
2. Materials and methods Experiments were performed on transverse hippocampal slices (400 mm) obtained from 12 – 18-dayold Wistar rats using a Vibroslice. Slices were collected and maintained in medium comprising (mM) NaCl (124), KCl (3), NaHCO3 (26), NaH2PO4 (1.25), CaCl2 (2), MgSO4 (1), D-glucose (10) (bubbled with O2/CO2 95:5). After at least 1 h of equilibration and recovery time, the slices were transferred to a recording chamber and perfused with the same medium (at a rate of approximately 2 ml min − 1). Data were collected and analysed on-line using LTP software (Anderson and Collingridge, 1997). Evoked field potentials were recorded in area CA1, at approximately 30°C, using glass microelectrodes containing 4 M NaCl following stimulation of the Schaffer collateral-commissural fibres every 30 s. Whole-cell patch-clamp recordings were obtained blind, at room temperature, using glass microelectrodes (5–7 MV; seal resistance approximately 10 GV) filled with a solution which comprised (mM) CsMeSO3 (or KMeS03) (120), NaCl (1), MgCl2 (1), 1,2-bis(o - aminophenoxy)ethane - N,N,N%,N% - tetraacetic acid (BAPTA) (10), N-(2,6-dimethylphenylcarbamoylmethyl)triethylammonium bromide (QX-314) (5), HEPES (5) (adjusted to pH 7.3) and, usually, Mg-ATP (4). Whole-cell patch-clamp recordings were obtained in the CA1 or CA3 region in the presence of 50 mM picrotoxin. Excitatory postsynaptic currents (EPSCs) were evoked by low-frequency stimulation of the Schaffer collateral-commissural and mossy fibre pathways, respectively. Access resistance was monitored continually and neurones discarded if this parameter changed by more than 20%. Unless otherwise stated, neurones were voltage-clamped at −60 mV. Compounds were applied by addition to the perfusing medium. Data are presented as mean9S.E.M and statistical significance
assessed using Student’s t-test (PB 0.05 considered significant and indicated in the figures by *). GYKI53655 [LY300168; 1-(4-aminophenyl)-3methylcarbamyl-4-methyl-7,8-methylenedioxy-3,4-dihydro-5H-2,3-benzodiazepine] and LY294486 [(3SR,4aRS,6SR,8aRS) - 6 - ({[(1H - tetrazol - 5 - yl)methyl]oxy} methyl) - 1,2,3,4,4a,5,6,7,8,8a - decahydroisoquinoline - 3carboxylic acid] were synthesised in house. Other compounds were obtained from Tocris Cookson.
3. Results Data were obtained from field and whole-cell patchclamp recordings made in the CA1 and CA3 areas of the hippocampus of 12–18-day-old Wistar rats.
3.1. ATPA depresses excitatory synaptic transmission The ability of ATPA to depress field excitatory postsynaptic potentials (EPSPs), evoked in area CA1 by stimulation of Schaffer collateral-commissural fibres, was investigated by applying a single concentration of ATPA for 10 min. A range of concentrations (from 30 nM to 10 mM) was tested, each application to a slice obtained from a different rat. The threshold concentration was approximately 100 nM, the IC50 was 0.60 mM and the maximum depression (at 10 mM) was approximately 70%. Depression of the EPSP was not associated with any change in the presynaptic fibre volley. Note that, at a concentration of 2 mM ATPA, a depression of 59% is predicted from the dose–response curve (Fig. 1).
3.2. E6idence for a presynaptic locus To address the locus of action of ATPA, additional experiments were performed under voltage-clamp conditions (at − 60 mV). The ability of ATPA to depress AMPA and NMDA receptor-mediated synaptic transmission was compared by studying either dual-component EPSCs (the peak of which at − 60 mV is mediated mainly via activation of AMPA receptors) or NMDA receptor-mediated EPSCs, with AMPA receptor activation eliminated using 50 mM GYKI53655. ATPA (2 mM) depressed AMPA and NMDA receptor-mediated synaptic transmission by 619 7% (n= 4) and 5896% (n= 4), respectively. The time-course of the ATPA-induced depression is shown in Fig. 2A. The depression of synaptic transmission was not associated with any effect on the holding current in either set of experiments (Fig. 2B). However, in the absence of GYKI53655, concentrations of ATPA greater than 2 mM resulted in inward currents, presumably due to
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
1271
Fig. 1. ATPA depresses field EPSPs in area CA1. (A) Graphs of pooled experiments, with traces from representative experiments at the time-points indicated (a –c), for three separate concentrations of ATPA. Each graph plots the mean 9 S.E.M (where larger than the symbol) for three slices. The data were normalised with respect to the 20-min baseline immediately preceding application of ATPA. (B) Dose – response curve plotting peak synaptic depression versus ATPA concentration. Each point plots the mean 9S.E.M. of three slices except dose at 3 mM (n=9).
activation of AMPA receptors (Lauridsen et al., 1985, Clarke et al., 1997). The observation that ATPA depresses both AMPA and NMDA receptor-mediated synaptic responses to a similar extent is most simply explained by a
presynaptic action to inhibit the amount of Lglutamate released. Consistent with this conclusion, depression of the EPSCs was associated with an increase in the paired-pulse facilitation ratio (Fig. 2C).
1272
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
Fig. 2. ATPA depresses EPSCs in area CA1. (A) A plot of peak EPSC amplitude versus time. Each point is the mean9 S.E.M. of eight slices (four dual-component EPSCs and four NMDA receptor-mediated EPSCs). (B) Plot of the effect on holding current for the same eight slices. (C) The graph plots the paired-pulse facilitation ratio (interpulse interval of 50 ms) before and during the peak of the ATPA-induced depression. Similar results were seen for five dual-component EPSC experiments and two isolated NMDA receptor-mediated EPSC experiments and so both sets of data have been pooled (n= 7). The traces show a representative paired-pulse facilitation experiment of dual-component EPSCs. The magnitude of the facilitation is more clearly evident when the depressed first EPSC is re-scaled to the peak of the control response.
3.3. Antagonism of the effect of ATPA
3.4. ATPA depresses EPSCs in area CA3
The activity of ATPA to depress excitatory synaptic transmission in the CA1 region of the hippocampus correlates with its potency as an agonist at the GluR5 subtype of kainate receptor (Clarke et al., 1997). Consistent with an action via GluR5, the GluR5-selective antagonist LY294486 (20 mM) reversibly antagonised the effects of ATPA (2 mM) on NMDA receptor-mediated EPSCs (Fig. 3). Thus, LY294486 had no effect on NMDA receptor-mediated EPSCs per se, or on the holding current, but reversed the depression induced by ATPA to 393% (n = 4). Collectively, therefore, these data suggest that GluR5-containing kainate receptors regulate excitatory synaptic transmission in the Schaffer collateral-commissural pathway, via a presynaptic mechanism.
Recent pharmacological experiments suggest that kainate receptors located postsynaptically on CA3 neurones contain GluR5 subunits, since their activation by either kainate or by the high frequency stimulation of mossy fibres is inhibited by GluR5 antagonists (Vignes et al., 1997). To determine whether GluR5-containing kainate receptors may also regulate the level of excitatory neurotransmission in the CA3 region of the hippocampus, we tested the effects of ATPA (1 mM, 5 min) on mossy fibre-evoked dual-component EPSCs. In each case the mossy fibre-evoked EPSC was depressed (by 829 10%; n= 4; Fig. 4A). As in CA1, the effects of ATPA were not associated with any change in the holding current (Fig. 4B). These data suggest that GluR5-containing kainate receptors also regulate exci-
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
1273
Fig. 3. Antagonism of ATPA-induced depressions in area CA1. (A) A plot of NMDA receptor-mediated EPSC amplitude versus time to illustrate the reversible block of ATPA-induced depressions by 20 mM LY294486. The traces are NMDA receptor-mediated EPSCs obtained at the times indicated (a – g). (B) Pooled data showing NMDA receptor-mediated EPSC amplitude in the presence of 1 mM ATPA (n = 4) and, for a different set of slices, in the presence of 2 mM ATPA and ATPA plus 20 mM LY294486 (n = 4).
tatory synaptic transmission in the mossy fibre pathway. The lack of any excitatory postsynaptic effect of ATPA is surprising. First, kainate is a potent excitant of CA3 neurones, an action that is antagonised by the selective GluR5 antagonist LY294486 (Vignes et al., 1997). Second, recordings using KMeSO4-filled sharp microelectrodes show ATPA hyperpolarised CA1 neurones of the adult hippocampus (Clarke et al., 1997). In five neurones we replaced Cs + with K + in the patch electrode and found that ATPA (1 mM) induced an outward current (of 1996 pA; Fig. 4C). In a further three neurones, ATPA generated an outward current in the presence of both 0.5 mM tetrodotoxin and 500 mM (S)-MCPG [(S)-a-methyl-4-carboxyphenylglycine; data not shown]. It is unlikely, therefore, that ATPA was inducing the release, in an action dependent manner, of a neurotransmitter to activate K + channels or that ATPA was activating mGlu receptors to elicit this response.
4. Discussion The present results suggest that the presynaptic kainate receptor which regulates excitatory synaptic transmission in area CA1 of the hippocampus (Chittajallu et al., 1996) contains GluR5 subunits. In addition, a similar type of receptor regulates excitatory synaptic transmission in the mossy fibre pathway in area CA3 of the hippocampus.
4.1. A presynaptic GluR5 -containing kainate receptor in area CA1 The findings that a submaximal concentration of ATPA depressed AMPA and NMDA receptor-mediated synaptic transmission to the same extent, with no associated change in holding current, and that the paired-pulse facilitation ratio was increased, strongly suggest that the locus of action of ATPA is presynaptic. These data, therefore, confirm the existence of presynaptic kainate receptors on excitatory terminals, sug-
1274
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
Fig. 4. ATPA depresses EPSCs in area CA3. (A) The graph plots peak EPSC amplitude evoked by single shock stimulation of the mossy fibre pathway versus time. The frequency of stimulation was increased from 0.033 to 0.2 Hz to facilitate the mossy fibre response (180 9 20% of control; n=4; Salin et al., 1996). Representative traces were obtained at the time-points indicated (a – c). (B) Mean data for four neurones showing that ATPA has no effect on the holding current. (C) Mean data for five neurones showing that ATPA induces an outward current when K + is the major cation in the patch pipette.
gested on the basis of the actions of kainate and domoate in the CA1 region of the hippocampus (Chittajallu et al., 1996). Based on the effectiveness of ATPA and LY294486, it seems likely that this kainate receptor comprises or contains GluR5 subunits. Thus, the potency of ATPA at depressing EPSPs in the present experiments (IC50 0.60 mM) is similar to its potency at activating human homomeric GluR5 expressed in HEK293 cells (EC50 2.1 mM; Clarke et al., 1997). Based on binding experiments, ATPA is at least three orders of magnitude less potent at other kainate receptor subunits (GluR6, GluR7, KA2, GluR6+ KA2) and, in the present study, was highly effective at a concentration subthreshold for activating AMPA receptors (Clarke et al., 1997). LY294486 is also inactive at displacing binding to these human recombinant kainate receptors and its activity as an AMPA receptor antagonist can be excluded since all of the experiments were performed in the presence of sufficient GYKI53655 to eliminate the activation of AMPA receptors (Clarke et al., 1997). Of course, the possibility that over the same concentration ranges ATPA and
LY294486 respectively activate and inhibit some unknown receptor, or untested heteromeric assembly of kainate receptor subunits, cannot be discounted.
4.2. A presynaptic GluR5 -containing kainate receptor in area CA3 In view of the high concentration of kainate receptors in the CA3 of the hippocampus, especially within the mossy fibre termination zone (Foster et al., 1981, Monaghan and Cotman, 1982, Monaghan et al., 1985, Represa et al., 1987), and given the exquisite sensitivity of this region to the excitatory, epileptogenic and neurotoxic effects of kainate (Nadler, 1978, Robinson and Deadwyler, 1981, Sloviter and Damiano, 1981, Westbrook and Lothman, 1983; Kehl et al., 1984), we extended our analysis to the CA3 region. It is likely that we were studying the monosynaptic connection between mossy fibres and CA3 pyramidal neurones, based on the placement of the electrodes (Vignes and Collingridge, 1997) and since the pathway stimulated showed rapid and pronounced frequency facilitation, a characteristic of
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
1275
Fig. 5. A schematic representation of kainate receptor subunit composition at the mossy fibre synapse. GluR5 subunits are shaded black, other subunits (e.g. GluR6) are shaded white. One GluR5 subunit is located postsynaptically to explain the lack of activation by ATPA. The presynaptic (autoreceptor) is shown as a homomeric assembly but could be a heteromer with two or more GluR5 subunits. This hypothetical scheme assumes that more than one agonist molecule needs to bind to effectively open kainate receptors.
this projection (Salin et al., 1996). However, the presence of extensive recurrent excitation within area CA3 precludes a definitive identification of the location of the ATPA-sensitive presynaptic receptors. The present finding that ATPA depresses excitatory synaptic transmission in area CA3 supports the existence of presynaptic kainate receptors in this region (Monaghan and Cotman, 1982, Sawada et al., 1985, Represa et al., 1987) and suggests that, as in area CA1, these receptors probably comprise or contain GluR5 subunits. It has been reported that kainate does not affect the frequency of miniature EPSCs in area CA3 (Castillo et al., 1997). However, interpretation of these data is complicated by the use of kainate, which will occupy AMPA receptors mediating the miniature EPSCs, and because certain presynaptic actions do not result in an alteration in the frequency of miniature EPSCs. We conclude that GluR5-containing kainate receptors are not only involved in postsynaptic events (Vignes and Collingridge, 1997) but also in presynaptic functions in the CA3 region of the hippocampus. Further experiments will be required to determine the mechanism by which activation of GluR5 receptors regulates excitatory synaptic transmission. One possibility is that ATPA causes a depolarisation block, analogous to the action of kainate on primary afferents (Agrawal and Evans, 1986). Alternatively it may involve a metabotropic action of GluR5 receptors, as suggested by the actions of kainate on inhibitory synap-
tic transmission in the hippocampus (Rodrı´guezMoreno and Lerma, 1998).
4.3. The ATPA paradox A perplexing observation is that ATPA had no effect on the holding current in Cs + -loaded cells. Under similar conditions, large inward currents are induced by nanomolar concentrations of kainate and these are reversibly antagonised by selective GluR5 antagonists, such as LY294486 (Vignes et al., 1997). The latter suggests that GluR5 subunits contribute to the kainate receptors which mediate postsynaptic excitation in the CA3 region of the hippocampus. Previously, ATPA has been shown to excite cells containing homomeric assemblies of GluR5 as well as DRG neurones (Clarke et al., 1997). Therefore, why did ATPA not induce inward currents in the present experiments? One radical explanation is that the postsynaptic receptor on CA3 neurones is a heteromer comprising GluR5 and other subunits in a stochiometric configuration. Activation of the GluR5 subunit in this heteromeric combination may be a necessary but not sufficient requirement for the induction of an inward current i.e. the complex needs more than ATPA for excitation to be observed. In this way, a GluR5 antagonist can prevent activation of the complex. Such a situation would result if, for example, two agonist molecules were required for full activation of the kainate receptor but GluR5 only
1276
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
contributed one subunit to the complex. Such a scheme is consistent with the presence of GluR6 subunits in the postsynaptic kainate receptor (Mulle et al., 1998) and with the high levels of expression of this and other kainate receptor mRNAs in CA3 neurones (Wisden and Seeburg, 1993). A putative scheme is shown in Fig. 5. A second mystery is why ATPA hyperpolarises CA1 neurones (Clarke et al., 1997). The present finding that, when K + replaced Cs + as the main cation in the patch pipette, outward currents were observed suggests that the hyperpolarisation is due to activation of a K + conductance. Since the ATPA-induced hyperpolarisation is antagonised by LY294486 (Clarke et al., 1997) it is likely to be due to the action of ATPA on GluR5 subunits, rather than via some other class of receptor. This raises the possibility that GluR5 receptors can selectively activate a K + conductance under certain circumstances. One possibility is that the postsynaptic GluR5-containing receptor couples to its effector via a G-protein (see Henley, 1994, Rodrı´guez-Moreno and Lerma, 1998).
5. Concluding remarks The finding that GluR5 subunits are probably involved in a presynaptic kainate receptor on excitatory terminals in both the CA1 and CA3 regions of the hippocampus adds to the growing list of kainate receptors which contain this subunit — others being postsynaptic receptors on hippocampal CA3 neurones (Vignes et al., 1997) and presynaptic receptors on hippocampal GABAergic neurones (Clarke et al., 1997). Further studies are required to establish under what conditions these putative autoreceptors can be activated synaptically.
Acknowledgements Supported by the MRC and the Wellcome Trust. We would like to thank Marvin Hansen for the synthesis of LY300168 (GYKI53655).
References Agrawal, S.G., Evans, R.H., 1986. The primary afferent depolarizing action of kainate in the rat. Br. J. Pharmacol. 87, 345–355. Anderson, W.W., Collingridge, G.L., 1997. A data acquisition program for on-line analysis of long-term potentiation and long-term depression. Soc. Neurosci. Abstr. 23, 665. Bettler, B., Boulter, J., Hermans-Borgmeyer, I., O’Shea-Greenfield, A., Deneris, E.S., Moll, C., Borgmeyer, U., Hollmann, M., Heinemann, S., 1990. Cloning of a novel glutamate receptor subunit, GluR5: expression in the nervous system during development. Neuron 5, 583 – 595.
Bettler, B., Mulle, C., 1995. Neurotransmitter receptors II. AMPA and kainate receptors. Neuropharmacology 34, 123 – 139. Bleakman, D., Ballyk, B.A., Schoepp, D.D., Palmer, A.J., Bath, C.P., Sharpe, E.F., Woolley, M.L., Bufton, H.R., Kamboj, R.K., Tarnawa, I., Lodge, D., 1996a. Activity of 2,3-benzodiazepines at native rat and recombinant human glutamate receptors in vitro: stereospecificity and selectivity profiles. Neuropharmacology 35, 1689 – 1702. Bleakman, D., Schoepp, D.D., Ballyk, B., Bufton, H., Sharpe, E.F., Thomas, K., Ornstein, P.L., Kamboj, R.K., 1996b. Pharmacological discrimination of GluR5 and GluR6 kainate receptor subtypes by (3S,4aR,6R,8aR)-6-[2-(1(2)H-tetrazole-5-yl)ethyl]decahydroisoquinoline-3 carboxylic acid. Mol. Pharmacol. 49, 581– 585. Castillo, P.E., Malenka, R.C., Nicoll, R.A., 1997. Kainate receptors mediate a slow postsynaptic current in hippocampal CA3 neurons. Nature 388, 182 – 188. Chittajallu, R., Vignes, M., Dev, K.K., Barnes, J.M., Collingridge, G.L., Henley, J.M., 1996. Regulation of glutamate release by presynaptic kainate receptors in the hippocampus. Nature 379, 78 – 81. Clarke, V.R.J., Ballyk, B.A., Hoo, K.H., Mandelzys, A., Pellizzari, A., Bath, C.P., Thomas, J., Sharpe, E.F., Davies, C.H., Ornstein, P.L., Schoepp, D.D., Kamboj, R.K., Collingridge, G.L., Lodge, D., Bleakman, D., 1997. A hippocampal GluR5 kainate receptor regulating inhibitory synaptic transmission. Nature 389, 599–603. Davies, J., Evans, R.H., Francis, A.A., Watkins, J.C., 1979. Excitatory amino acid receptors and synaptic excitation in the mammalian central nervous system. J. Physiol. Paris 75, 641–654. Donevan, S.D., Rogawski, M.A., 1993. GYKI 52466, a 2,3-benzodiazepine, is a highly selective, noncompetitive antagonist of AMPA/kainate receptor responses. Neuron 10, 51 – 59. Foster, A.C., Mena, E.E., Monaghan, D.T., Cotman, C.W., 1981. Synaptic localization of kainic acid binding sites. Nature 289, 73 – 75. Henley, J.M., 1994. Kainate-binding proteins: phylogeny, structures and possible functions. Trends Pharmacol. Sci. 15, 182 –190. Kehl, S.J., McLennan, H., Collingridge, G.L., 1984. Effects of folic and kainic acids on synaptic responses of hippocampal neurones. Neuroscience 11, 111 – 124. Lauridsen, J., Honore´, T., Krogsgaard-Larson, P., 1985. Ibotenic acid analogues. Synthesis, molecular flexibility, and in vitro activity of agonists and antagonists at central glutamic acid receptors. J. Med. Chem. 28, 668 – 672. McLennan, H., Lodge, D., 1979. The antagonism of amino acid-induced excitation of spinal neurones in the cat. Brain Res. 169, 83 – 90. Monaghan, D.T., Cotman, C.W., 1982. The distribution of [3H]kainic acid binding sites in rat CNS as determined by autoradiography. Brain Res. 252, 91 – 100. Monaghan, D.T., Yao, D., Cotman, C.W. L, 1985. L-[3H]Glutamate binding to kainate-, NMDA-, and AMPA-sensitive binding sites: an autoradiographic analysis. Brain Res. 340, 378 – 383. Mulle, C., Sailer, A., Pe´rez-Otan˜o, I., Dickinson-Anson, H., Castillo, P.E., Bureau, I., Maron, C., Gage, F.H., Mann, J.R., Bettler, B., Heinemann, S.F., 1998. Altered synaptic physiology and reduced susceptibility to kainate-induced seizures in GluR6-deficient mice. Nature 392, 601 – 605. Nadler, J., 1978. Intraventricular kainic acid preferentially destroys hippocampal pyramidal cells. Nature 271, 676 – 677. Palmer, A.J., Lodge, D., 1993. Cyclothiazide reverses AMPA receptor antagonism of the 2,3-benzodiazepine, GYKI 53655. Eur. J. Pharmacol. 244, 193 – 194. Partin, K.M., Mayer, M.A., 1996. Negative allosteric modulation of wild-type and mutant AMPA receptors by GYKI 53655. Mol. Pharmacol. 49, 142 – 148.
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277 Paternain, A.V., Morales, M., Lerma, J., 1995. Selective antagonism of AMPA receptors unmasks kainate receptor-mediated responses in hippocampal neurons. Neuron 14, 185–189. Represa, A., Tremblay, E., Ben-Ari, Y., 1987. Kainate binding sites in hippocampal mossy fibres: localisation and plasticity. Neuroscience 20, 739 – 748. Robinson, J.H., Deadwyler, S.A., 1981. Kainic acid produces depolarization of CA3 pyramidal cells in the in vitro hippocampal slice. Brain Res. 221, 117–127. Rodrı´guez-Moreno, A., Herreras, O., Lerma, J., 1997. Kainate receptors presynaptically downregulate GABAergic inhibition in the rat hippocampus. Neuron 19, 893–901. Rodrı´guez-Moreno, A., Lerma, J., 1998. Kainate receptor modulation of GABA release involves a metabotropic function. Neuron 20, 1211 – 1218. Salin, P.A., Scanziani, M., Malenka, R.C., Nicoll, R.A., 1996. Distinct short-term plasticity at two excitatory synapses in the hippocampus. Proc. Natl. Acad. Sci. USA 93, 13304–13309. Sawada, S., Higashima, M., Yamamoto, C., 1985. Inhibitors of high-affinity uptake augment depolarisations of hippocampal neurons induced by glutamate, kainate and related compounds. Exp. Brain Res. 60, 323 – 329. Sloviter, R.S., Damiano, B.P., 1981. On the relationship between kainic acid-induced epileptiform activity and hippocampal neuronal damage. Neuropharmacology 20, 1003–1011. Tarnawa, I., Engeberg, I., Flatman, J.A., 1990. GYKI 52466, an inhibitor of spinal reflexes is a potent quisqualate antagonist. In:
.
1277
Lubec, G., Rosenthal, G.A. (Eds.), Amino Acids, Chemistry, Biology and Medicine. ESCOM, Leiden, pp. 538 – 544. Vignes, M., Collingridge, G.L., 1997. The synaptic activation of kainate receptors. Nature 388, 179 – 182. Vignes, M., Mathiesen, C., Collingridge, G.L., 1996. Activation of a kainate-induced inward current in CA3 hippocampal neurons of immature rats in vitro. J. Physiol. 495.P, 44P. Vignes, M., Bleakman, D., Lodge, D., Collingridge, G.L., 1997. The synaptic activation of the GluR5 subtype of kainate receptor in area CA3 of the rat hippocampus. Neuropharmacology 36, 1477– 1481. Watkins, J.C., Evans, R.H., 1981. Excitatory amino acid transmitters. Annu. Rev. Pharmacol. Toxicol. 21, 165 – 204. Westbrook, G.L., Lothman, E.W., 1983. Cellular and synaptic basis of kainic acid-induced hippocampal epileptiform activity. Brain Res. 273, 97 – 109. Wilding, T.J., Huettner, J.E., 1995. Differential antagonism of alphaamino-3-hydroxy-4-isoxazolepropionic acid- and kainate-preferring glutamate receptors by 2,3-benzodiazepines. Mol. Pharmacol. 47, 582 – 587. Wisden, W., Seeburg, P.H., 1993. A complex mosaic of high affinity kainate receptors in rat brain. J. Neurosci. 13, 3582 – 3598. Zorumski, C.F., Yamada, K.A., Price, M.T., Olney, J.W., 1993. GYKI 52466, a 2,3-benzodiazepine, as a highly selective, noncompetitive antagonist of AMPA/kainate receptor responses. Neuron 10, 51 – 59.
The GluR5 subtype of kainate receptor regulates excitatory synaptic transmission in areas CA1 and CA3 of the rat hippocampus M. Vignes a,b, V.R.J. Clarke a, M.J. Parry a, D. Bleakman c,d, D. Lodge c,d, P.L. Ornstein d, G.L. Collingridge a,* b
a Department of Anatomy, Uni6ersity of Bristol, Uni6ersity Walk, Bristol BS8 1TD, UK Laboratoire ‘Plasticite´ Ce´re´brale’, EP 628 CNRS, Uni6ersite´ Montpellier II, Place Euge`ne Bataillon, 34095 Montpellier cedex 05, France c Lilly Research Centre Ltd., Erl Wood Manor, Windlesham GU20 6PH, UK d Eli Lilly and Company Ltd., Lilly Corporate Center, Indianapolis, IN 46285, USA
Accepted 12 August 1998
Abstract Activation of kainate receptors depresses excitatory synaptic transmission in the hippocampus. In the present study, we have utilised a GluR5 selective agonist, ATPA [(RS)-2-amino-3-(3-hydroxy-5-tert-butylisoxazol-4-yl)propanoic acid], and a GluR5 selective antagonist, LY294486 [(3SR,4aRS,6SR,8aRS)-6-({[(1H-tetrazol-5-yl)methyl]oxy}methyl)-1,2,3,4,4a,5,6,7,8,8a-decahydroisoquinoline-3-carboxylic acid], to determine whether GluR5 subunits are involved in this effect. ATPA mimicked the presynaptic depressant effects of kainate in the CA1 region of the hippocampus. It depressed reversibly AMPA (a-amino-3-hydroxy-5-methylisoxazole-4-propionic acid) receptor-mediated field excitatory postsynaptic potentials (field EPSPs) with an IC50 value of :0.60 mM. The dual-component excitatory postsynaptic current (EPSC) and the pharmacologically isolated NMDA (N-methyl-D-aspartate) receptor-mediated EPSC were depressed to a similar extent by 2 mM ATPA (61 9 7% and 58 96%, respectively). Depressions were associated with an increase in the paired-pulse facilitation ratio suggesting a presynaptic locus of action. LY294486 (20 mM) blocked the effects of 2 mM ATPA on NMDA receptor-mediated EPSCs in a reversible manner. In area CA3, 1 mM ATPA depressed reversibly mossy fibre-evoked synaptic transmission (by 82 9 10%). The effects of ATPA were not accompanied by any changes in the passive properties of CA1 or CA3 neurones. However, in experiments where K + , rather than Cs + , containing electrodes were used, a small outward current was observed. These results show that GluR5 subunits comprise or contribute to a kainate receptor that regulates excitatory synaptic transmission in both the CA1 and CA3 regions of the hippocampus. © 1998 Elsevier Science Ltd. All rights reserved. Keywords: ATPA; EPSP; Hippocampus; Kainic acid; LY294486; Mossy fibre; Presynaptic receptor; Synaptic transmission
1. Introduction It has been known for many years that kainate receptors can be distinguished from AMPA receptors (Davies et al., 1979, McLennan and Lodge, 1979, Watkins and Evans, 1981). However, the lack of selective kainate receptor agonists and antagonists has, until recently, greatly hindered the identification of the roles of kainate receptors in the vertebrate central nervous system. The use of the selective non-competitive AMPA receptor antagonists GYKI52466 (Tarnawa et al., 1990, * Corresponding author. Fax: +44 117 92916870.
Palmer and Lodge, 1993, Donevan and Rogawski, 1993, Zorumski et al., 1993) or GYKI53655 (Paternain et al., 1995, Wilding and Huettner, 1995, Partin and Mayer, 1996, Bleakman et al., 1996a) to prevent activation of AMPA receptors has allowed several effects to be ascribed to activation of kainate receptors. Most information has been obtained using the hippocampal slice preparation, in which it has been shown that kainate receptors mediate: (1) an inward current in CA1 and CA3 neurones (Chittajallu et al., 1996, Vignes et al., 1996, Castillo et al., 1997), (2) a synaptic current in CA3 neurones resulting from mossy fibre stimulation (Castillo et al., 1997, Vignes and Collingridge, 1997),
0028-3908/98/$ - see front matter © 1998 Elsevier Science Ltd. All rights reserved. PII: S0028-3908(98)00148-8
1270
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
(3) depression of g-aminobutyric acid (GABA)-mediated synaptic transmission (Clarke et al., 1997, Rodrı´guez-Moreno et al., 1997) and (4) depression of excitatory synaptic transmission (Chittajallu et al., 1996). The recent discovery of compounds (Bleakman et al., 1996b, Clarke et al., 1997) that are selective for the GluR5 kainate receptor subunit (Bettler et al., 1990) versus the other known subunits (GluR6, GluR7, KA-1 and KA-2; see Bettler and Mulle, 1995) led to the identification of a role for GluR5 subunits in the first three effects (Clarke et al., 1997, Vignes et al., 1997). In the present study, we have used the GluR5 selective agonist ATPA [(RS)-2-amino-3-(3-hydroxy-5-tertbutylisoxazol-4-yl)propanoic acid] and the GluR5 selective antagonist LY294486 to determine whether GluR5 subunits are involved in the presynaptic regulation of excitatory synaptic transmission in the hippocampus.
2. Materials and methods Experiments were performed on transverse hippocampal slices (400 mm) obtained from 12 – 18-dayold Wistar rats using a Vibroslice. Slices were collected and maintained in medium comprising (mM) NaCl (124), KCl (3), NaHCO3 (26), NaH2PO4 (1.25), CaCl2 (2), MgSO4 (1), D-glucose (10) (bubbled with O2/CO2 95:5). After at least 1 h of equilibration and recovery time, the slices were transferred to a recording chamber and perfused with the same medium (at a rate of approximately 2 ml min − 1). Data were collected and analysed on-line using LTP software (Anderson and Collingridge, 1997). Evoked field potentials were recorded in area CA1, at approximately 30°C, using glass microelectrodes containing 4 M NaCl following stimulation of the Schaffer collateral-commissural fibres every 30 s. Whole-cell patch-clamp recordings were obtained blind, at room temperature, using glass microelectrodes (5–7 MV; seal resistance approximately 10 GV) filled with a solution which comprised (mM) CsMeSO3 (or KMeS03) (120), NaCl (1), MgCl2 (1), 1,2-bis(o - aminophenoxy)ethane - N,N,N%,N% - tetraacetic acid (BAPTA) (10), N-(2,6-dimethylphenylcarbamoylmethyl)triethylammonium bromide (QX-314) (5), HEPES (5) (adjusted to pH 7.3) and, usually, Mg-ATP (4). Whole-cell patch-clamp recordings were obtained in the CA1 or CA3 region in the presence of 50 mM picrotoxin. Excitatory postsynaptic currents (EPSCs) were evoked by low-frequency stimulation of the Schaffer collateral-commissural and mossy fibre pathways, respectively. Access resistance was monitored continually and neurones discarded if this parameter changed by more than 20%. Unless otherwise stated, neurones were voltage-clamped at −60 mV. Compounds were applied by addition to the perfusing medium. Data are presented as mean9S.E.M and statistical significance
assessed using Student’s t-test (PB 0.05 considered significant and indicated in the figures by *). GYKI53655 [LY300168; 1-(4-aminophenyl)-3methylcarbamyl-4-methyl-7,8-methylenedioxy-3,4-dihydro-5H-2,3-benzodiazepine] and LY294486 [(3SR,4aRS,6SR,8aRS) - 6 - ({[(1H - tetrazol - 5 - yl)methyl]oxy} methyl) - 1,2,3,4,4a,5,6,7,8,8a - decahydroisoquinoline - 3carboxylic acid] were synthesised in house. Other compounds were obtained from Tocris Cookson.
3. Results Data were obtained from field and whole-cell patchclamp recordings made in the CA1 and CA3 areas of the hippocampus of 12–18-day-old Wistar rats.
3.1. ATPA depresses excitatory synaptic transmission The ability of ATPA to depress field excitatory postsynaptic potentials (EPSPs), evoked in area CA1 by stimulation of Schaffer collateral-commissural fibres, was investigated by applying a single concentration of ATPA for 10 min. A range of concentrations (from 30 nM to 10 mM) was tested, each application to a slice obtained from a different rat. The threshold concentration was approximately 100 nM, the IC50 was 0.60 mM and the maximum depression (at 10 mM) was approximately 70%. Depression of the EPSP was not associated with any change in the presynaptic fibre volley. Note that, at a concentration of 2 mM ATPA, a depression of 59% is predicted from the dose–response curve (Fig. 1).
3.2. E6idence for a presynaptic locus To address the locus of action of ATPA, additional experiments were performed under voltage-clamp conditions (at − 60 mV). The ability of ATPA to depress AMPA and NMDA receptor-mediated synaptic transmission was compared by studying either dual-component EPSCs (the peak of which at − 60 mV is mediated mainly via activation of AMPA receptors) or NMDA receptor-mediated EPSCs, with AMPA receptor activation eliminated using 50 mM GYKI53655. ATPA (2 mM) depressed AMPA and NMDA receptor-mediated synaptic transmission by 619 7% (n= 4) and 5896% (n= 4), respectively. The time-course of the ATPA-induced depression is shown in Fig. 2A. The depression of synaptic transmission was not associated with any effect on the holding current in either set of experiments (Fig. 2B). However, in the absence of GYKI53655, concentrations of ATPA greater than 2 mM resulted in inward currents, presumably due to
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
1271
Fig. 1. ATPA depresses field EPSPs in area CA1. (A) Graphs of pooled experiments, with traces from representative experiments at the time-points indicated (a –c), for three separate concentrations of ATPA. Each graph plots the mean 9 S.E.M (where larger than the symbol) for three slices. The data were normalised with respect to the 20-min baseline immediately preceding application of ATPA. (B) Dose – response curve plotting peak synaptic depression versus ATPA concentration. Each point plots the mean 9S.E.M. of three slices except dose at 3 mM (n=9).
activation of AMPA receptors (Lauridsen et al., 1985, Clarke et al., 1997). The observation that ATPA depresses both AMPA and NMDA receptor-mediated synaptic responses to a similar extent is most simply explained by a
presynaptic action to inhibit the amount of Lglutamate released. Consistent with this conclusion, depression of the EPSCs was associated with an increase in the paired-pulse facilitation ratio (Fig. 2C).
1272
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
Fig. 2. ATPA depresses EPSCs in area CA1. (A) A plot of peak EPSC amplitude versus time. Each point is the mean9 S.E.M. of eight slices (four dual-component EPSCs and four NMDA receptor-mediated EPSCs). (B) Plot of the effect on holding current for the same eight slices. (C) The graph plots the paired-pulse facilitation ratio (interpulse interval of 50 ms) before and during the peak of the ATPA-induced depression. Similar results were seen for five dual-component EPSC experiments and two isolated NMDA receptor-mediated EPSC experiments and so both sets of data have been pooled (n= 7). The traces show a representative paired-pulse facilitation experiment of dual-component EPSCs. The magnitude of the facilitation is more clearly evident when the depressed first EPSC is re-scaled to the peak of the control response.
3.3. Antagonism of the effect of ATPA
3.4. ATPA depresses EPSCs in area CA3
The activity of ATPA to depress excitatory synaptic transmission in the CA1 region of the hippocampus correlates with its potency as an agonist at the GluR5 subtype of kainate receptor (Clarke et al., 1997). Consistent with an action via GluR5, the GluR5-selective antagonist LY294486 (20 mM) reversibly antagonised the effects of ATPA (2 mM) on NMDA receptor-mediated EPSCs (Fig. 3). Thus, LY294486 had no effect on NMDA receptor-mediated EPSCs per se, or on the holding current, but reversed the depression induced by ATPA to 393% (n = 4). Collectively, therefore, these data suggest that GluR5-containing kainate receptors regulate excitatory synaptic transmission in the Schaffer collateral-commissural pathway, via a presynaptic mechanism.
Recent pharmacological experiments suggest that kainate receptors located postsynaptically on CA3 neurones contain GluR5 subunits, since their activation by either kainate or by the high frequency stimulation of mossy fibres is inhibited by GluR5 antagonists (Vignes et al., 1997). To determine whether GluR5-containing kainate receptors may also regulate the level of excitatory neurotransmission in the CA3 region of the hippocampus, we tested the effects of ATPA (1 mM, 5 min) on mossy fibre-evoked dual-component EPSCs. In each case the mossy fibre-evoked EPSC was depressed (by 829 10%; n= 4; Fig. 4A). As in CA1, the effects of ATPA were not associated with any change in the holding current (Fig. 4B). These data suggest that GluR5-containing kainate receptors also regulate exci-
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
1273
Fig. 3. Antagonism of ATPA-induced depressions in area CA1. (A) A plot of NMDA receptor-mediated EPSC amplitude versus time to illustrate the reversible block of ATPA-induced depressions by 20 mM LY294486. The traces are NMDA receptor-mediated EPSCs obtained at the times indicated (a – g). (B) Pooled data showing NMDA receptor-mediated EPSC amplitude in the presence of 1 mM ATPA (n = 4) and, for a different set of slices, in the presence of 2 mM ATPA and ATPA plus 20 mM LY294486 (n = 4).
tatory synaptic transmission in the mossy fibre pathway. The lack of any excitatory postsynaptic effect of ATPA is surprising. First, kainate is a potent excitant of CA3 neurones, an action that is antagonised by the selective GluR5 antagonist LY294486 (Vignes et al., 1997). Second, recordings using KMeSO4-filled sharp microelectrodes show ATPA hyperpolarised CA1 neurones of the adult hippocampus (Clarke et al., 1997). In five neurones we replaced Cs + with K + in the patch electrode and found that ATPA (1 mM) induced an outward current (of 1996 pA; Fig. 4C). In a further three neurones, ATPA generated an outward current in the presence of both 0.5 mM tetrodotoxin and 500 mM (S)-MCPG [(S)-a-methyl-4-carboxyphenylglycine; data not shown]. It is unlikely, therefore, that ATPA was inducing the release, in an action dependent manner, of a neurotransmitter to activate K + channels or that ATPA was activating mGlu receptors to elicit this response.
4. Discussion The present results suggest that the presynaptic kainate receptor which regulates excitatory synaptic transmission in area CA1 of the hippocampus (Chittajallu et al., 1996) contains GluR5 subunits. In addition, a similar type of receptor regulates excitatory synaptic transmission in the mossy fibre pathway in area CA3 of the hippocampus.
4.1. A presynaptic GluR5 -containing kainate receptor in area CA1 The findings that a submaximal concentration of ATPA depressed AMPA and NMDA receptor-mediated synaptic transmission to the same extent, with no associated change in holding current, and that the paired-pulse facilitation ratio was increased, strongly suggest that the locus of action of ATPA is presynaptic. These data, therefore, confirm the existence of presynaptic kainate receptors on excitatory terminals, sug-
1274
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
Fig. 4. ATPA depresses EPSCs in area CA3. (A) The graph plots peak EPSC amplitude evoked by single shock stimulation of the mossy fibre pathway versus time. The frequency of stimulation was increased from 0.033 to 0.2 Hz to facilitate the mossy fibre response (180 9 20% of control; n=4; Salin et al., 1996). Representative traces were obtained at the time-points indicated (a – c). (B) Mean data for four neurones showing that ATPA has no effect on the holding current. (C) Mean data for five neurones showing that ATPA induces an outward current when K + is the major cation in the patch pipette.
gested on the basis of the actions of kainate and domoate in the CA1 region of the hippocampus (Chittajallu et al., 1996). Based on the effectiveness of ATPA and LY294486, it seems likely that this kainate receptor comprises or contains GluR5 subunits. Thus, the potency of ATPA at depressing EPSPs in the present experiments (IC50 0.60 mM) is similar to its potency at activating human homomeric GluR5 expressed in HEK293 cells (EC50 2.1 mM; Clarke et al., 1997). Based on binding experiments, ATPA is at least three orders of magnitude less potent at other kainate receptor subunits (GluR6, GluR7, KA2, GluR6+ KA2) and, in the present study, was highly effective at a concentration subthreshold for activating AMPA receptors (Clarke et al., 1997). LY294486 is also inactive at displacing binding to these human recombinant kainate receptors and its activity as an AMPA receptor antagonist can be excluded since all of the experiments were performed in the presence of sufficient GYKI53655 to eliminate the activation of AMPA receptors (Clarke et al., 1997). Of course, the possibility that over the same concentration ranges ATPA and
LY294486 respectively activate and inhibit some unknown receptor, or untested heteromeric assembly of kainate receptor subunits, cannot be discounted.
4.2. A presynaptic GluR5 -containing kainate receptor in area CA3 In view of the high concentration of kainate receptors in the CA3 of the hippocampus, especially within the mossy fibre termination zone (Foster et al., 1981, Monaghan and Cotman, 1982, Monaghan et al., 1985, Represa et al., 1987), and given the exquisite sensitivity of this region to the excitatory, epileptogenic and neurotoxic effects of kainate (Nadler, 1978, Robinson and Deadwyler, 1981, Sloviter and Damiano, 1981, Westbrook and Lothman, 1983; Kehl et al., 1984), we extended our analysis to the CA3 region. It is likely that we were studying the monosynaptic connection between mossy fibres and CA3 pyramidal neurones, based on the placement of the electrodes (Vignes and Collingridge, 1997) and since the pathway stimulated showed rapid and pronounced frequency facilitation, a characteristic of
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
1275
Fig. 5. A schematic representation of kainate receptor subunit composition at the mossy fibre synapse. GluR5 subunits are shaded black, other subunits (e.g. GluR6) are shaded white. One GluR5 subunit is located postsynaptically to explain the lack of activation by ATPA. The presynaptic (autoreceptor) is shown as a homomeric assembly but could be a heteromer with two or more GluR5 subunits. This hypothetical scheme assumes that more than one agonist molecule needs to bind to effectively open kainate receptors.
this projection (Salin et al., 1996). However, the presence of extensive recurrent excitation within area CA3 precludes a definitive identification of the location of the ATPA-sensitive presynaptic receptors. The present finding that ATPA depresses excitatory synaptic transmission in area CA3 supports the existence of presynaptic kainate receptors in this region (Monaghan and Cotman, 1982, Sawada et al., 1985, Represa et al., 1987) and suggests that, as in area CA1, these receptors probably comprise or contain GluR5 subunits. It has been reported that kainate does not affect the frequency of miniature EPSCs in area CA3 (Castillo et al., 1997). However, interpretation of these data is complicated by the use of kainate, which will occupy AMPA receptors mediating the miniature EPSCs, and because certain presynaptic actions do not result in an alteration in the frequency of miniature EPSCs. We conclude that GluR5-containing kainate receptors are not only involved in postsynaptic events (Vignes and Collingridge, 1997) but also in presynaptic functions in the CA3 region of the hippocampus. Further experiments will be required to determine the mechanism by which activation of GluR5 receptors regulates excitatory synaptic transmission. One possibility is that ATPA causes a depolarisation block, analogous to the action of kainate on primary afferents (Agrawal and Evans, 1986). Alternatively it may involve a metabotropic action of GluR5 receptors, as suggested by the actions of kainate on inhibitory synap-
tic transmission in the hippocampus (Rodrı´guezMoreno and Lerma, 1998).
4.3. The ATPA paradox A perplexing observation is that ATPA had no effect on the holding current in Cs + -loaded cells. Under similar conditions, large inward currents are induced by nanomolar concentrations of kainate and these are reversibly antagonised by selective GluR5 antagonists, such as LY294486 (Vignes et al., 1997). The latter suggests that GluR5 subunits contribute to the kainate receptors which mediate postsynaptic excitation in the CA3 region of the hippocampus. Previously, ATPA has been shown to excite cells containing homomeric assemblies of GluR5 as well as DRG neurones (Clarke et al., 1997). Therefore, why did ATPA not induce inward currents in the present experiments? One radical explanation is that the postsynaptic receptor on CA3 neurones is a heteromer comprising GluR5 and other subunits in a stochiometric configuration. Activation of the GluR5 subunit in this heteromeric combination may be a necessary but not sufficient requirement for the induction of an inward current i.e. the complex needs more than ATPA for excitation to be observed. In this way, a GluR5 antagonist can prevent activation of the complex. Such a situation would result if, for example, two agonist molecules were required for full activation of the kainate receptor but GluR5 only
1276
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277
contributed one subunit to the complex. Such a scheme is consistent with the presence of GluR6 subunits in the postsynaptic kainate receptor (Mulle et al., 1998) and with the high levels of expression of this and other kainate receptor mRNAs in CA3 neurones (Wisden and Seeburg, 1993). A putative scheme is shown in Fig. 5. A second mystery is why ATPA hyperpolarises CA1 neurones (Clarke et al., 1997). The present finding that, when K + replaced Cs + as the main cation in the patch pipette, outward currents were observed suggests that the hyperpolarisation is due to activation of a K + conductance. Since the ATPA-induced hyperpolarisation is antagonised by LY294486 (Clarke et al., 1997) it is likely to be due to the action of ATPA on GluR5 subunits, rather than via some other class of receptor. This raises the possibility that GluR5 receptors can selectively activate a K + conductance under certain circumstances. One possibility is that the postsynaptic GluR5-containing receptor couples to its effector via a G-protein (see Henley, 1994, Rodrı´guez-Moreno and Lerma, 1998).
5. Concluding remarks The finding that GluR5 subunits are probably involved in a presynaptic kainate receptor on excitatory terminals in both the CA1 and CA3 regions of the hippocampus adds to the growing list of kainate receptors which contain this subunit — others being postsynaptic receptors on hippocampal CA3 neurones (Vignes et al., 1997) and presynaptic receptors on hippocampal GABAergic neurones (Clarke et al., 1997). Further studies are required to establish under what conditions these putative autoreceptors can be activated synaptically.
Acknowledgements Supported by the MRC and the Wellcome Trust. We would like to thank Marvin Hansen for the synthesis of LY300168 (GYKI53655).
References Agrawal, S.G., Evans, R.H., 1986. The primary afferent depolarizing action of kainate in the rat. Br. J. Pharmacol. 87, 345–355. Anderson, W.W., Collingridge, G.L., 1997. A data acquisition program for on-line analysis of long-term potentiation and long-term depression. Soc. Neurosci. Abstr. 23, 665. Bettler, B., Boulter, J., Hermans-Borgmeyer, I., O’Shea-Greenfield, A., Deneris, E.S., Moll, C., Borgmeyer, U., Hollmann, M., Heinemann, S., 1990. Cloning of a novel glutamate receptor subunit, GluR5: expression in the nervous system during development. Neuron 5, 583 – 595.
Bettler, B., Mulle, C., 1995. Neurotransmitter receptors II. AMPA and kainate receptors. Neuropharmacology 34, 123 – 139. Bleakman, D., Ballyk, B.A., Schoepp, D.D., Palmer, A.J., Bath, C.P., Sharpe, E.F., Woolley, M.L., Bufton, H.R., Kamboj, R.K., Tarnawa, I., Lodge, D., 1996a. Activity of 2,3-benzodiazepines at native rat and recombinant human glutamate receptors in vitro: stereospecificity and selectivity profiles. Neuropharmacology 35, 1689 – 1702. Bleakman, D., Schoepp, D.D., Ballyk, B., Bufton, H., Sharpe, E.F., Thomas, K., Ornstein, P.L., Kamboj, R.K., 1996b. Pharmacological discrimination of GluR5 and GluR6 kainate receptor subtypes by (3S,4aR,6R,8aR)-6-[2-(1(2)H-tetrazole-5-yl)ethyl]decahydroisoquinoline-3 carboxylic acid. Mol. Pharmacol. 49, 581– 585. Castillo, P.E., Malenka, R.C., Nicoll, R.A., 1997. Kainate receptors mediate a slow postsynaptic current in hippocampal CA3 neurons. Nature 388, 182 – 188. Chittajallu, R., Vignes, M., Dev, K.K., Barnes, J.M., Collingridge, G.L., Henley, J.M., 1996. Regulation of glutamate release by presynaptic kainate receptors in the hippocampus. Nature 379, 78 – 81. Clarke, V.R.J., Ballyk, B.A., Hoo, K.H., Mandelzys, A., Pellizzari, A., Bath, C.P., Thomas, J., Sharpe, E.F., Davies, C.H., Ornstein, P.L., Schoepp, D.D., Kamboj, R.K., Collingridge, G.L., Lodge, D., Bleakman, D., 1997. A hippocampal GluR5 kainate receptor regulating inhibitory synaptic transmission. Nature 389, 599–603. Davies, J., Evans, R.H., Francis, A.A., Watkins, J.C., 1979. Excitatory amino acid receptors and synaptic excitation in the mammalian central nervous system. J. Physiol. Paris 75, 641–654. Donevan, S.D., Rogawski, M.A., 1993. GYKI 52466, a 2,3-benzodiazepine, is a highly selective, noncompetitive antagonist of AMPA/kainate receptor responses. Neuron 10, 51 – 59. Foster, A.C., Mena, E.E., Monaghan, D.T., Cotman, C.W., 1981. Synaptic localization of kainic acid binding sites. Nature 289, 73 – 75. Henley, J.M., 1994. Kainate-binding proteins: phylogeny, structures and possible functions. Trends Pharmacol. Sci. 15, 182 –190. Kehl, S.J., McLennan, H., Collingridge, G.L., 1984. Effects of folic and kainic acids on synaptic responses of hippocampal neurones. Neuroscience 11, 111 – 124. Lauridsen, J., Honore´, T., Krogsgaard-Larson, P., 1985. Ibotenic acid analogues. Synthesis, molecular flexibility, and in vitro activity of agonists and antagonists at central glutamic acid receptors. J. Med. Chem. 28, 668 – 672. McLennan, H., Lodge, D., 1979. The antagonism of amino acid-induced excitation of spinal neurones in the cat. Brain Res. 169, 83 – 90. Monaghan, D.T., Cotman, C.W., 1982. The distribution of [3H]kainic acid binding sites in rat CNS as determined by autoradiography. Brain Res. 252, 91 – 100. Monaghan, D.T., Yao, D., Cotman, C.W. L, 1985. L-[3H]Glutamate binding to kainate-, NMDA-, and AMPA-sensitive binding sites: an autoradiographic analysis. Brain Res. 340, 378 – 383. Mulle, C., Sailer, A., Pe´rez-Otan˜o, I., Dickinson-Anson, H., Castillo, P.E., Bureau, I., Maron, C., Gage, F.H., Mann, J.R., Bettler, B., Heinemann, S.F., 1998. Altered synaptic physiology and reduced susceptibility to kainate-induced seizures in GluR6-deficient mice. Nature 392, 601 – 605. Nadler, J., 1978. Intraventricular kainic acid preferentially destroys hippocampal pyramidal cells. Nature 271, 676 – 677. Palmer, A.J., Lodge, D., 1993. Cyclothiazide reverses AMPA receptor antagonism of the 2,3-benzodiazepine, GYKI 53655. Eur. J. Pharmacol. 244, 193 – 194. Partin, K.M., Mayer, M.A., 1996. Negative allosteric modulation of wild-type and mutant AMPA receptors by GYKI 53655. Mol. Pharmacol. 49, 142 – 148.
M. Vignes et al. / Neuropharmacology 37 (1998) 1269–1277 Paternain, A.V., Morales, M., Lerma, J., 1995. Selective antagonism of AMPA receptors unmasks kainate receptor-mediated responses in hippocampal neurons. Neuron 14, 185–189. Represa, A., Tremblay, E., Ben-Ari, Y., 1987. Kainate binding sites in hippocampal mossy fibres: localisation and plasticity. Neuroscience 20, 739 – 748. Robinson, J.H., Deadwyler, S.A., 1981. Kainic acid produces depolarization of CA3 pyramidal cells in the in vitro hippocampal slice. Brain Res. 221, 117–127. Rodrı´guez-Moreno, A., Herreras, O., Lerma, J., 1997. Kainate receptors presynaptically downregulate GABAergic inhibition in the rat hippocampus. Neuron 19, 893–901. Rodrı´guez-Moreno, A., Lerma, J., 1998. Kainate receptor modulation of GABA release involves a metabotropic function. Neuron 20, 1211 – 1218. Salin, P.A., Scanziani, M., Malenka, R.C., Nicoll, R.A., 1996. Distinct short-term plasticity at two excitatory synapses in the hippocampus. Proc. Natl. Acad. Sci. USA 93, 13304–13309. Sawada, S., Higashima, M., Yamamoto, C., 1985. Inhibitors of high-affinity uptake augment depolarisations of hippocampal neurons induced by glutamate, kainate and related compounds. Exp. Brain Res. 60, 323 – 329. Sloviter, R.S., Damiano, B.P., 1981. On the relationship between kainic acid-induced epileptiform activity and hippocampal neuronal damage. Neuropharmacology 20, 1003–1011. Tarnawa, I., Engeberg, I., Flatman, J.A., 1990. GYKI 52466, an inhibitor of spinal reflexes is a potent quisqualate antagonist. In:
.
1277
Lubec, G., Rosenthal, G.A. (Eds.), Amino Acids, Chemistry, Biology and Medicine. ESCOM, Leiden, pp. 538 – 544. Vignes, M., Collingridge, G.L., 1997. The synaptic activation of kainate receptors. Nature 388, 179 – 182. Vignes, M., Mathiesen, C., Collingridge, G.L., 1996. Activation of a kainate-induced inward current in CA3 hippocampal neurons of immature rats in vitro. J. Physiol. 495.P, 44P. Vignes, M., Bleakman, D., Lodge, D., Collingridge, G.L., 1997. The synaptic activation of the GluR5 subtype of kainate receptor in area CA3 of the rat hippocampus. Neuropharmacology 36, 1477– 1481. Watkins, J.C., Evans, R.H., 1981. Excitatory amino acid transmitters. Annu. Rev. Pharmacol. Toxicol. 21, 165 – 204. Westbrook, G.L., Lothman, E.W., 1983. Cellular and synaptic basis of kainic acid-induced hippocampal epileptiform activity. Brain Res. 273, 97 – 109. Wilding, T.J., Huettner, J.E., 1995. Differential antagonism of alphaamino-3-hydroxy-4-isoxazolepropionic acid- and kainate-preferring glutamate receptors by 2,3-benzodiazepines. Mol. Pharmacol. 47, 582 – 587. Wisden, W., Seeburg, P.H., 1993. A complex mosaic of high affinity kainate receptors in rat brain. J. Neurosci. 13, 3582 – 3598. Zorumski, C.F., Yamada, K.A., Price, M.T., Olney, J.W., 1993. GYKI 52466, a 2,3-benzodiazepine, as a highly selective, noncompetitive antagonist of AMPA/kainate receptor responses. Neuron 10, 51 – 59.