CELLULAR MECHANISMS OF LONG-TERM DEPRESSION IN THE CEREBELLUM

Keywords

Abstract

Introduction Receptors to excitatory amino acids (EAAs) can be broadly divided in two groups: N-methyl-D-aspartate (NMDA) and non-NMDA receptor types (ref. in Mayer & Westbrook 1987). Until recently, the latter group was considered as constituted by 2 separate classes of receptors, i.e the Kainate (KA) and the Quisqualate (QA) receptors (Watkins, 1981). The situation is now more complex for several reasons. First, QA receptors can be split into 2 subclasses: the ionotropic QA receptors (Qi) coupled to a cationic ionic channel (also termed a-amino-3 hydroxy-5-methyl- isoxalone-4-propionate (AMPA) receptor), and the metabotropic QA receptors (mGlu) coupled to phospholipase C (PLC) (Recasens et al. 1987; Sladeczek et al. 1985; Sugiyama et al. 1987). Secondly, several subunits of the AMPA receptor have been cloned recently: these subunits named GluR1, 2, 3 and 4 (GluR1-4) exist in two versions (flip and flop) generated by an alternative splicing (Hollman et al. 1989; Boulter et al. 1990; Keinnen et al. 1990; Sommer et al. 1990) and are also characterized by a low affinity for KA (Nakanishi et al. 1990); a biochemical analysis of the AMPA receptor-channel purified from rat brain has shown that this receptor is a pentameric structure composed of combinations of the GluR1-4 subunits, the presence of additional subunits beeing very unlikely (Wenthold et al. 1992). Thirdly, one know now that high affinity KA receptors are also likely to be heteromeric structures since 5 subunits (GluR5-7 and KA1-2) have been recently cloned (ref. in Sommer & Seeburg 1992; see also Gasic & Hollman 1992). Finally, 5 subunits NMDA (NMDAR1 and NMDAR2A-D) and 6 mGlu receptors have also been recently identified (ref. in Nakanishi 1992), which makes the situation even more complex.

In most types of neurons studied so far, both NMDA and non-NMDA receptors to EAAs are well represented, as for instance in hippocampal and neocortical cells ( Artola & Singer 1987; 1990; Bindman et al. 1988; Collingridge et al. 1983; Hirsch & Crepel 1990). Moreover, they can be located on the same postsynaptic zones, as recently demonstrated in the hippocampus (Bekkers & Stevens 1989).

Another major reason for the current interest for receptors to EAAs is the discovery of their involvement in long-term changes in synaptic strength which is thougth to play a crucial role in learning and memory processes (Hebb 1949). In particular, it is now established that NMDA and / or non-NMDA receptors are responsible for the induction and the maintenance of long-term potentiation (LTP) of synaptic transmission in the hippocampus respectively (ref. in Bliss & Lynch 1988; see also Collingridge, Kehl & McLennan 1983).

The cerebellum of mammals is another interesting region of the brain to study the involvement EAA receptors in synaptic plasticity. First, the 2 main excitatory afferents to Purkinje cells (PCs), i.e. parallel fibres (PFs) and climbing fibres (CFs) (Eccles et al. 1967) are likely to use Glutamate (Glu) as neurotransmitters (Cuenod et al. 1989; Zhang et al. 1990; Herdon & Coyle, 1978; Hudson et al. 1976). Secondly, and in marked contrast with most other neuronal cell types, mature PCs only bear non-NMDA receptors (Crepel, Dhanjal & Sear 1982; see also ref. in Crepel & Audinat 1991), which make them an interesting model to study synaptic plasticity in the absence of this class of receptors. Finally, the participation of the cerebellum in motor learning was postulated by Brindley as early as 1964 (ref. in Ito 1984). This hypothesis was formalized by Marr (1969), who proposed the so-called "external teacher" theory derived from the Perceptron, a cybernetic machine invented by Rosenblatt in 1962. According to the theory, CFs are able to modify the gain of synaptic transmission between PFs and CPs during motor learning, to adjust the cerebellar output to the desired motor command. In this scheme, conjonctive activation of PCs by CFs and PFs lead to a LTP of synaptic transmission at PF-PC synapses. Later on, Albus revisited the theory and proposed instead that a long-term depression (LTD) of synaptic transmission occurs at PF- PC synapses following their co-activation with CFs (Albus 1971).

The present review will deal with main experimental data gathered over the last ten years on the participation of EAAs receptors of PCs in synaptic plasticity as a possible cellular basis of motor learning in this structure.

Discovery of Long-term depression. In keeping with the Marr-Albus theory of motor learning in the cerebellum, Ito and coworkers have been the first to demonstrate in in vivo experiments ( Ito et al. 1982; see also ref. in Ito, 1987; 1989) that, in rabbit cerebellum, conjunctive stimulation of PFs and CFs leads to LTD of synaptic transmission at PF-PC synapses. The fact that only those PF-PC synapses activated in conjunction with CFs are affected (ref. in Ito 1984) indicates that the changes in synaptic strength are restricted to the activated synapses. In the same paper (Ito et al. 1982), the authors also already showed that coactivation of PCs by CFs and by direct application of Glu in their dendritic fields by ionophoretic electrodes leads to a persistent decrease of the responsiveness of PCs to this agonist, an observation which was reproduced and extended later (see below), and which suggests that part of the phenomenon occurs postsynaptically.

The demonstration by Kano and Kato (1987) that pairing CF input with Glu or QA application on PC dendrites also induces subsequent LTD of synaptic transmission between PFs and PCs, whereas KA, Asp and NMDA are ineffective in this respect is of importance since it strongly suggests that AMPA receptors of PCs are indeed involved in LTD.

Role of calcium in Long-term depression of synaptic transmission In in vivo experiments, involvement of Ca in induction of LTD was initially suggested by the observation that stellate cell inhibition prevents LTD from occurring (Ekerot & Kano 1985), probably by blocking Ca-dependent plateau potentials in PC dendrites following their activation by CFs (Ekerot & Oscarsson 1981). This observation led Ito to propose that the efficacy of PF-PC synapses is decreased as result of both the activation of Glu receptors of PCs and the Ca influx which occurs in these cells during their activation by CFs (ref. in Ito 1987; 1989).

Indeed, in more recent experiments in rat cerebellar slices, LTD of PF- mediated EPSPs is consistently induced by pairing these synaptic responses with Ca spikes directly induced in the postsynaptic cell by depolarizing current pulses (Fig. 1). In contrast, when only sodium spikes are induced in PCs during the pairing protocol, LTD is no longer observed and is replaced by LTP of PF-mediated EPSPs (Crepel & Jaillard 1991; see also below). Along the same line, in guinea pig cerebellar slices, LTD of PF-mediated EPSPs is not induced by their pairing with CF inputs when PCs are loaded with EGTA (Sakurai 1990). The involvement of Ca in LTD induction has been confirmed still more recently by 2 independent groups. First, in cultured PCs, Linden et al. (1991) have shown that LTD of AMPA-mediated currents is induced by conjunctive ionophoretic glutamate pulses and PC depolarization sufficient to produce Ca entry through voltage-gated Ca channels (Fig. 1). Secondly, by combining measurements of synaptic efficacy with fura-2 measurements of intracellular Ca concentration in single patch-clamped PCs in thin slices, Konnerth and coworkers (1992) have shown that in pairing experiments, a transient rise in internal Ca is sufficient to induce LTD (Fig. 1). It must be emphasized that the fact that Ca concentration goes back to normal level after LTD induction rules out the possibility that this form of synaptic plasticity is merely a pathological process. This is in keeping with the absence of LTD in pseudopairing experiments, i.e. when PCs are depolarized to activate voltage-dependent Ca channels in the absence of PF stimulation (Daniel et al. 1992), as well as with the fact that LTD was restricted to the activated synapses in the initial in vivo experiments of Ito and coworkers (see above).

Which glutamate receptors participate to LTD induction and expression? The role of AMPA receptors in induction and expression of LTD has been further investigated in whole cell clamped PCs in acute slices, by using the same type of pairing protocol between PF-EPSPS and Ca spike firing as before (see above). These experiments showed that when the pairing protocol was performed in the presence of a sufficient concentration of CNQX in the bath to block synaptic transmission, no LTD could be induced in any of the tested cells (Fig. 2). In contrast, when CNQX was bath applied after the induction of LTD, this did not prevent its maintenance after wash out of the compound (Crepel et al., unpublished data). Similarly, in cultured PCs, Linden and coworkers (1991) also showed that no LTD occurs unless AMPA receptors are activated during its induction phase (Fig. 2). Of course, these experiments bring further strong evidences in favor of an involvement of AMPA receptors of PCs in LTD induction.

Now, in experiments performed with the grease gap technique, Ito and Karachot (1990) have shown that co-activation of AMPA receptors of PCs by AMPA and of their mGlu receptors by trans-1-amino-cyclopentyl-1,3-dicarboxylate (trans- ACPD) is sufficient to induce a long lasting desensitization of AMPA receptors of these cells (Fig. 3). According to these authors, this suggests that activation of mGlu receptors of PCs also play an important role in LTD. This view was supported by more recent experiments by Linden and coworkers (1991) showing that in cultured PCs, induction of LTD by conjunctive depolarization and stimulation of Glu receptors requires both the activation of their AMPA and mGlu receptor subtypes.

We therefore decided to test this hypothesis differently in acute slices, specifically by co-activating AMPA and metabotropic receptors of intracellularly recorded PCs by PF stimulations and by bath application of trans-ACPD respectively. In all tested cells, trans-ACPD induced a marked decrease of PF- EPSPs . However, this effect was fully reversible after wash-out of the drug in all cells where no other effect of trans-ACPD was observed (see below), and this, even in cells where PFs were stimulated at 1 Hz during trans-ACPD application. In a fraction of the recorded cells, trans-ACPD also induced a transient depolarization of PCs and a bursting firing of Ca spikes. In this case only, a LTD of PF-mediated EPSPs was occasionally observed after wash-out of trans-ACPD. These results (Crepel et al. 1991) which have been recently confirmed by Glaum et al. (1992) do not support the hypothesis that mGlu receptors play a major role in LTD induction in acute slices, at least when voltage dependent Ca channels are not activated at the same time. In keeping with this view, Ross and coworkers (Miyakawa et al. 1992) have shown that Ca transients occuring during PF-mediated EPSPs are entirely blocked by CNQX, thus suggesting that they are not due to the activation of mGlu receptors, but rather due to the activation of voltage gated Ca channels.

In view of this apparent discrepancy with Ito and Karachot experiments, which can be attributed to differences in experimental protocols, we also looked for whether activation of mGlu receptors of PCs has at least a reinforcing effect on LTD of PF-mediated EPSPs induced by the same pairing protocol as described earlier, i.e coactivation of PCs by PFs and by direct depolarization of postsynaptic cells giving rise to Ca spike firing. These experiments showed that when the pairing protocol with Ca spikes was performed at the peak of the depressant effect of bath application of 50 M trans-ACPD on PF-mediated EPSPs, a robust LTD was now induced in most recorded cells (Fig. 3), i.e. significantly larger than LTD induced in the absence of trans-ACPD. Moreover and surprisingly enough, the same robust LTD was also observed in most cells where PFs were not stimulated during the period of Ca spike firing in the presence of trans-ACPD, or when CNQX was also added to the bath at the time of the pairing protocol (Fig. 3), which strongly suggest that this form of LTD does not involve activation of AMPA receptors of PCs for its induction (Daniel et al., 1992).

The present results therefore suggests that LTD can be induced by 2 differents mechanisms: Both involve activation of voltage gated Ca channels of PCs, but one also primarily depends on the activation of ionotropic AMPA receptors (LTD AMPA), whereas the other does not involve this classe of receptors for its induction, but depends instead of a strong activation of metabotropic Glu receptors (LTD mGlu).

Interestingly enough, the induction of the LTD mGlu is blocked when slices are preincubated with 10 M Thapsigargin, whereas the LTD AMPA is not (Hemart et al., 1995). For the former, this suggests that Ca release from intracellular stores plays an important role in LTD induction. For the latter, i.e. LTD AMPA, this mechanism does not seem to operate. Now, what is the origin of Ca release from internal stores in LTD mGlu? Fluorometric measurements show little evidence for Ca release from intracellular stores in the presence of trans-ACPD (Llano et al., 1991; Staub et al., 1992; Vranesic et al., 1991). Therefore, another hypothesis is that activation of metabotropic receptors at PF-PC synapses by trans ACPD increases Ca currents flowing through voltage gated Ca channels, which in turn would be now sufficient to activate a Ca induced Ca release in the dendrites.

At this point, it must be emphasized that we have still no evidence that the AMPA receptor-dependent form of LTD in acute slices also depends on metabotropic receptors, in particular because no good antagonists of mGlu receptors is available actually.

Second messenger cascades involved in LTD One knows that the calcium dependent PKC I is very abundant in PCs (Hidaka et al. 1988; Nishizuka 1986). One also knows that mGlu receptors are located at PF-PC synapses (Martin et al. 1992), and are coupled with phospholipase C (PLC) (Nicoletti et al. 1986; Recasens et al. 1987; Sladeczek et al. 1985; Sugiyama et al. 1987). It was therefore temptating to postulate (Crepel & Krupa 1988) that the cascade of events leading to LTD involves a coactivation of PKC of PCs by Ca entry through voltage-gated ionic channels, and by diacylglycerol (DAG) produced by the activation of mGlu receptors by Glu released by PFs.

Indeed, a selective LTD of the responsiveness of PCs to Glu and QA (i.e. Asp- induced responses were unaffected) was obtained in about 40% of PCs in the presence of Phorbol esters known to activate PKC (Fig. 4), whereas inactive analogs were without any effect (Crepel & Krupa 1988; 1990). Moreover, LTD of PF-mediated EPSPs following their pairing with Ca spikes in rat cerebellar slices in vitro was nearly totally prevented by bath application of polymixin B, a potent blocker of PKC and, to a lesser extent of calmodulin (CAM)-dependent kinase (Crepel & Jaillard 1990), and the same results were obtained when the PKC inhibitor peptide 19-36 was directly injected in the recorded cells ( Hemart et al., unpublished data). Finally, these results were recently confirmed and extended in cultured PCs (Fig. 4; Linden & Connor 1991). It must be emphasized that they give an indirect support to the view that mGlu receptors are involved in LTD induction.

There are now experimental evidences that LTD also depends on another cascade of events beyond Ca entry into PCs. Indeed, it is known that Ca can induce the formation of NO from arginine by activating a CAM-dependent NO- synthase (ref. in Garthwaite, Charles & Chess-William 1988; Garthwaite, Southam & Anderton 1989; see also Ross, Bredt & Snyder 1990). Nitric oxide is highly diffusible and thus activates soluble guanylate cyclase (ref. in Tremblay, Gerzer & Hamet 1988) in cells where it is produced, as well as in surrounding cellular elements (ref. in Garthwaite, Southam & Anderton 1989 and in Ross, Bredt & Snyder 1990). This cascade of events can therefore activate cGMP-dependent protein kinases in PCs where this enzyme is particularly abundant (Lohmann et al. 1981) as well as in PFs and glial cells.

It is therefore of prime importance that, in intracellularly recorded PCs with sharp electrodes in acute slices, N-monomethylarginine (L-NMMA), a potent inhibitor of NO synthesis (Knowles et al. 1989), nearly totally prevented LTD of PF- mediated EPSPs following their pairing with Ca spikes, and that the same effect was obtained by methylene blue which acts by blocking activation of soluble guanylate cyclase by NO (Crepel & Jaillard 1990). At the same time, Ito & Karachot (1990) also established, with the grease-gap method, that desensitization of AMPA receptors of PCs is likely to require the production of cyclic GMP via the production of NO. Later on, Shibuki & Okada (1991) showed that protocols known to induce LTD indeed lead to the production of NO in cerebellar slices. However, either these studies on the role of NO in LTD induction were not performed at a cellular level (Ito & Karakot 1990; Shibuki & Okada 1991), or they only used blockers of NO to test its role in LTD (Crpel & Jaillard 1990). Furthermore, none of these studies dealed precisely with the site(s) of production and of action of NO. Finally, it has been recently claimed that LTD of glutamate currents in cultured PCs does not require NO signaling (Linden & Connor, 1992). The role of NO in the induction of LTD at PF-PC synapses was therefore re-investigated in whole-cell patch clamped PCs in thin slices in vitro (Daniel et al. 1993). It was thus confirmed that bath application of L-NMMA consistently partially prevents the induction of LTD of PF-mediated EPSPs induced by their pairing with Ca2+ spikes (Fig. 5), and that the effect of L- NMMA can be reversed by an excess of arginine. It was also shown that bath application of NO donnors and of 8-bromoguanosine 3': 5' cyclic monophosphate (8-bromo-c-GMP) are able to reproduce a LTD-like phenomenon. Finally, LTD of PF-mediated EPSPs was also induced when NO donnors or guanosine 3': 5' cyclic monophosphate (c-GMP) were directly dialyzed into PCs and these LTD-like effects partially occluded LTD induced by pairing protocols (Fig. 5). Therefore, these results show that NO plays indeed a role in LTD induction, and demonstrate for the first time that its site of action is probably the soluble guanylate cyclase of PCs. According to Ito and Karachot (1992), cGMP would in turn activate a cGMP dependent protein kinase (PKG), thereby allowing phosphorylation of its specific substrate G-substrate, a potent inhibitor of phosphatases. In this scheme, LTD would involve both phosphorylation of AMPA receptors of PCs by PKC, and inhibition of their de-phosphorylation by the NO route (fig. 6).

However, in the cerebellum, NO-synthase has not been identified in PCs, but only in neighbouring elements (Bredt, Hwang & Snyder 1990; Southam, Morris & Garthwaite 1992). Thus, one hypothesis is that NO might be produced by the NO- synthase located in PFs and/or in basket cells when these neurons are activated by both PF stimulations (Eccles, Ito & Szentagothai 1967) and by the large efflux of potassium which follows the entry of Ca in PCs during pairing experiments. These possible paracrine interactions are illustrated in figure 6. However, it is still possible that the NO-synthase involved in the induction of LTD is located in PCs themselves, but is different from the already known forms of the enzyme, thus explaining why it has not been visualized so far.

On the other hand, in these experiments on the role of NO in LTD, the fact that the decrease in synaptic efficacy induced by a pairing protocol with Ca spikes was not completely occluded by that induced by cGMP supports the view that, in acute slices, LTD involves 2 different routes, via PKC and via NO respectively (Fig. 6). Accordingly, we have shown recently that the induction of LTD is totally prevented when cells are bathed with L-NMMA and dialysed at the same time through the patch pipette with the PKC inhibitor peptide 19-36 (Crepel et al., unpublished data), whereas, as mentioned before, blockade of LTD induction with only one of the inhibitors is generally only partial. The fact that LTD may involve 2 different routes might also explain why no effect of NO blockers was seen in experiments on LTD performed in dissociated cell cultures since, in particular, putative NO donnors as PFs and basket cells were possibly scarcely represented. It is also conceivable that cultured PCs only develop the PKC route and are thus insensitive to bath application of NO donnors (Linden & Connor, 1992).

Finally, an interesting observation of Ito and Karachot (1992) is that induction of LTD in slices by phorbol esters or by 8-bromo-cGMP requires concommitent activation of AMPA receptors. As we will see below, this fit well with recent experimental evidences showing that LTD induction requires, as an initial step, an agonist-dependent desensitization of AMPA receptors of PCs.

Does LTD involve desensitization of AMPA receptors of PCs ? The fact that coactivation of PCs by CFs and by direct application of Glu in their dendritic fields leads to a persistent decrease of their responsiveness to this agonist (Ito et al. 1982), led Ito to propose that induction of LTD might ultimately lead to a long-term desensitization of ionotropic Glu receptors of PCs, and thus to the observed decrease in synaptic efficacy. Accordingly, it has been shown later on that pairing ionophoretic application of Glu and Ca spike firing of PCs induces LTD of their responsiveness to this agonist, both in acute slices (Crepel & Krupa 1988) and in dissociated cultures (Linden & Connor 1991). However, the observed decrease in efficacy of Glu in activating PCs might be due to other causes than a true desensitization of Glu receptors.

Recently, the nootropic compound Aniracetam has been shown to markedly reduce desensitization of AMPA receptors and/or to decreases the closing rate constant for ion channel gating (Isaacson & Nicoll 1991; Ito et al. 1990; Tang et al. 1991; Vyklicky et al. 1991). It was therefore an important finding to show that in whole cell clamped PCs in acute slices, Aniracetam has a larger potentiating effect on PF-mediated EPSCs during expression of LTD than normally and that this compound also significantly blocks the induction of LTD (Hemart et al. 1994). Indeed, these data strongly supports the view that this change in synaptic efficacy involves a genuine change in the functional characteristics of these receptors. Furthermore, and in keeping with previous observations of Ito and Karachot (1992), these results also suggest that induction of LTD requires an initial agonist- dependent desensitization of AMPA receptors of PCs.

In contrast, the comparison of current-voltage curves of PF-mediated EPSCs in patch-clamped PCs in thin slices prior to and after LTD induction shows that this change in synaptic efficacy do not involve a change in reversal potential of the synaptic responses (Crepel et al., unpublished data).

Molecular composition of AMPA receptors of PCs Until recently, the subunit composition of native AMPA receptors of PCs was unknown. In order to further characterize the AMPA receptors involved in the LTD, a new method was developed, which combines whole-cell recordings and a molecular analysis, based on the polymerase chain reaction (PCR), of the messenger RNAs harvested into the patch pipette at the end of each recording. We found (Lambolez et al. 1992) that each single Purkinje cell recorded in cerebellar cultures or in olivo-cerebellar co-cultures expressed the messenger RNAs encoding the five following subunits of the AMPA receptor : the flip and flop versions of GluR1 and GluR2 as well as GluR3flip, GluR2 being the most abundant. In addition, GLuR3flop and GluR4flip were scarcely expressed in half of these neurones and GluR4flop was never detected. These results strongly suggest that the AMPA receptors of Purkinje cells are heterogenous with respect to their subunit composition. Whether this heterogeneity of the AMPA receptors within a single neurone could be the basis of a functional heterogeneity between synapses is still an open question.

In conclusion, LTD induction can be tentatively explained in the following way, taking into account all experimental evidences mentioned above, as well as previously proposed schemes (Ito and Karachot, 1990; 1992). When AMPA receptors of PCs are activated by Glu released by PFs, their agonist-dependent desensitization leads to a conformational change which expose a phosphorylation site to PKC. As this kinase has been activated by Ca entry through voltage-gated Ca channels and by DAG produced by PLC when mGlu receptors at PF-PC synapses are also activated by Glu released by PFs, the AMPA receptors of PCs can now be phosphorylated. On the other hand, the NO route ultimately inhibits phosphatases, thus allowing the phosphorylated state of AMPA receptors to be maintained. As a consequence of this phosphorylation, kinetics of opening and closing of AMPA receptor coupled channels are likely to be affected, and/or a larger fraction of Glu receptors than normally is stabilized in a desensitized state at rest, thus explaining the maintenance of LTD (Fig. 6), at least on a relatively short time scale (see below). In this scheme, the LTD mGlu might be explained if one assume that in the absence of AMPA receptor activation by Glu, a phosphorylation of these receptors by PKC is still possible, but requires a much larger activation of the kinase, which is the case with bath application of trans-ACPD. Finally, it remains now to establish whether LTD is indeed involved in motor learning as suggested by recent in vivo experiments (Nagao & Ito 1991).

REFERENCES

Albus, J.S. (1971) A theory of cerebellar function. Mathematical Biosciences 10:25-61.

Artola, A. & Singer, W. (1987) Long-term potentiation and NMDA-receptors in rat visual cortex. Nature 330:649-652.

Artola, A. & Singer, W. (1990) The involvement of N-methyl-D-aspartate receptors in induction and maintenance of long-term potentiation in rat visual cortex. The European Journal of Neuroscience 2:254-269

Bekkers, J. M. & Stevens, C. F. (1989) NMDA and non-NMDA receptors are co-localized at individual excitatory synapses in cultured rat hippocampus. Nature 341:230-233.

Bindman, L. J., Murphy, K. P. S. J. & Pockett, S. (1988) Postsynaptic control of the induction of long-term changes in efficacy of transmission at neocortical synapses in slices of rat brain. Journal of Neurophysiology 60:1053-1065.

Bliss, T.V.P. & Lynch, M.A. (1988) Long-term potentiation of synaptic transmission in the hippocampus: properties and mechanisms. In: Long-term potentiation, from Biophysics to Behavior, ed. Landfield, P.W. and Deadwyler, S.A., pp.3-72. Alan R. Liss, Inc, New-York.

Boulter, J., Hollmann, M., O'Shea-Greenfield, A., Hartley, M., Deneris, E., Maron, C. & Heinemann, S. (1990) Molecular cloning and functional expression of glutamate receptor subunit genes. Science 249:1033-1037. Bredt, S., Hwang, P.M. & Snyder, S.H. (1990) Localization of nitric oxide synthase indicating a neural role for nitric oxide. Nature 347:768-770.

Collingridge, G.L., Kelh, S.J. & McLennan, H. (1983) Excitatory amino acids in synaptic transmission in the Schaffer collateral-commissural pathway of the rat hippocampus. Journal of Physiology 334:33-46.

Crepel, F. & Audinat, E. (1991) Excitatory amino acid receptors of cerebellar Purkinje cells: developement and plasticity. Progress in Biophysic and Molecular Biology 55:31-46.

Crepel, F., Daniel, H., Hemart, N. & Jaillard, D. (1991) Effects of ACPD and AP3 on parallel-fibre-mediated EPSPs of Purkinje cells in cerebellar slices in vitro. Experimental Brain Research 86:402-406.

Crepel, F., Dhanjal, S.S. & Sears, T.A. (1982) Effect of glutamate, aspartate and related derivatives on cerebellar Purkinje cell dendrites in the rat: an in vitro study. Journal of Physiology 329:297-317.

Crepel, F., & Jaillard, D. (1990) Protein kinases, nitric oxide and long-term depression of synapses in the cerebellum. NeuroReport 1:133-136.

Crepel, F., & Jaillard, D. (1991) Pairing of pre- and postsynaptic activities in cerebellar Purkinje cells induces long-term changes in synaptic efficacy. An in vitro study. Jounal of Physiology (Lond.) 432:123-141.

Crepel, F. & Krupa, M. (1988) Activation of protein kinase C induces a long term depression of glutamate sensitivity of cerebellar Purkinje cells. An in vitro study. Brain Research 458:397-401. Crepel, F. & Krupa, M. (1990) Modulation of the responsiveness of cerebellar Purkinje cells to excitatory amino acids. In: Excitatory amino acids and Neuronal plasticity, ed. Ben-Ari, Y.,New-York: Plenum press pp 323-329..

Cuenod, M., Do, K. Q., Vollenweider, F., Zollinger, M., Klein, A. & Streit, P. (1989) The puzzle of the transmitters in the climbing fibers. In: The olivocerebellar system in motor control, ed. Strata, P. Experimental Brain Research Series 17. pp. 161-176. Sringer-Verlag Berlin-Heidelberg.

Daniel, H., Hemart, N., Jaillard, D. & Crepel, F. (1992) Coactivation of metabotropic glutamate receptors and of voltage-gated calcium channels induces long-term depression in cerebellar Purkinje cells in vitro. Experimental Brain Research 90:327-331.

Daniel, H., Hemart, N., Jaillard, D. & Crepel, F. (1993) Long-term depression requires nitric oxide and guanosine 3':5' cyclic monophosphate production in rat cerebellar Purkinje cells. European Journal of Neuroscience 5:1079-1082.

Eccles, J. C., Ito, M. & Szentagothai, J. (1967) The cerebellum as a neuronal machine. Springer-Verlag Berlin.

Ekerot C.F. & Kano, M. (1985) Long-term depression of parallel fibre synapses following stimulation of climbing fibres. Brain Research 342:357-360.

Ekerot, C.F. & Oscarsson, O. (1981) Prolonged depolarization elicited in Purkinje cell dendrites by climbing fibre impulses in the cat. Journal of Physiology (Lond.) 318:207-221.

Garthwaite, J., Charles, S.L., & Chess-Williams, R. (1988) Endothelium- derived relaxing factor release on activation of NMDA receptors suggests role as intercellular messenger in the brain. Nature 336:385-388.

Garthwaite, J., Southam, E., & Anderton, M. (1989) A kainate receptor linked to nitric oxide synthesis from arginine. Journal of Neurochemistry 53:1952-1954.

Gasic, G.P. & Hollman M. (1992) Molecular neurobiology of glutamate receptors. Annual review of Physiology 54:507-536.

Glaum, S.R., Slater, T., Rossi, D.J., & Miller, R.J. (1992) Role of metabotropic glutamate (ACPD) receptors at parallel fiber-Purkinje cell synapse. Journal of Neurophysiology 68:1453-1462.

Hebb, D.O. (1949). The organization of behavior. New-York: J. Wiley and Sons.

Hemart, N., Daniel, H., Jaillard, D. & Crepel, F. (1994) Properties of glutamate receptors are modified during long-term depression in cerebellar Purkinje cells. Neuroscience Research 19:213-221.

Hemart, N., Daniel, H., Jaillard, D. & Crepel, F. (1995) Receptors and second messengers involved in Long-term depression in rat cerebellar slices in vitro: a reappraisal. European Journal of Neuroscience 7:45-53.

Herdnon, R.M., & Coyle, J. T. (1978). Glutaminergic innervation, kainic acid and selective vulnerability in the cerebellum. In: Kainic acid as a tool in neurobiology, ed. McGeer, G., Olney, J.W. and McGeer, P.L., pp 189-200. New-York: Raven Press.

Hidaka, H., Tanaka, T., Onoda, K., Hagiwara, M., Watanabe, M., Ohta, H., Ito, Y., Tsurudome, M., & Yoshida, T. (1988) Cell-specific expression of protein kinase C isozymes in the rabbit cerebellum. Journal of Biological Chemistry 263:4523-4526

Hirsch, J.C. & Crepel, F. (1990) Use-dependent changes in synaptic eficacy in rat prefrontal neurons in vitro. Journal of Physiology 427:31-49.

Hollman, M., O'Shea-Greenfield, A., Rogers, S.W. & Heinemann, S. (1989) Cloning by functional expression of a member of the glutamate receptor family. Nature 342:643-648.

Hudson, B.D., Valcana, T., Bean, G. & Timiras, P.S. (1976) Glutamic acid: a strong candidate as the neurotransmitter of cerebellar granule cell. Neurochemical Research 1:73-82.

Isaacson, J.S. & Nicoll, R.A. (1991) Aniracetam reduces glutamate receptor desensitization and slows the decay of fast excitatory synaptic currents in the hippocampus. Proceedings of the National Academy of Sciences of the USA 88:10936-10940.

Ito, I., Tanabe, S., Kohda, A. & Sugiyama, H. (1990) Allosteric potentiation of quisqualate receptors by a nootropic drug aniracetam. Journal of Physiology (Lond.) 424:533-543.

Ito, M. (1984) The cerebellum and neural control. New-York: Raven Press.

Ito, M. (1987) Characterization of synaptic plasticity in cerebellar and cerebral neocortex. In: The neural and molecular bases of learning, ed. Changeux, J. P. and Nonishi, M., pp. 276-279. John Wiley & Sons, LTD.

Ito, M. (1989) Long term depression. Annual review in Neuroscience 12:85-102.

Ito, M. & Karachot, L. (1990) Messengers mediating long-term desensitization in cerebellar Purkinje cells. Neuroreport 1:129-132.

Ito, M. & Karachot, L. (1992) Protein kinases and phosphatase inhibitors mediating long-term desensitization of glutamate receptors in cerebellar Purkinje cells. Neuroscience Research, 14: 27-38.

Ito, M., Sakurai, M. & Tongroach, P. (1982) Climbing fibre induced depression of both mossy fibre responsiveness and glutamate sensitivity of cerebellar Purkinje cells. Journal of Physiology, (Lond.) 324:113-134.

Kano, M. & Kato, M. (1987) Quisqualate receptors are specifically involved in cerebellar synaptic plasticity. Nature 325:276-279.

Kasai, H. & Petersen, O.H. (1994) Spatial dynamics of second messengers: IP3 and cAMP as long-range and associative messengers. Trends in Neurosciences 17:95-101.

Keinnen, K., Wisden, W., Sommer, B., Werner, P., Herb, A., Verdoorn, T.A., Sakmann, B. & Seeburg, P.H. (1990) A family of AMPA-selective glutamate receptors. Science 249:556-560.

Klann, E., Chen, S.J. & Sweatt, J.D. (1993) Mechanism of protein kinase C activation during the induction and maintenance of long-term potentiation probed using a selective peptide substrate. Proceedings of the National Academy of Sciences of the USA 90: 8337-8341.

Knowles, R.G., Palacios, M., Palmer, R.M. & Moncada, S. (1989) Formation of nitric oxide from L-arginine in the central nervous system: a transduction mechanism for stimulation of the soluble guanylate cyclase. Proceedings of the National Academy of Sciences of the USA 86:5159-5162.

Konnerth, A., Dreesen, J., & Augustine, G.J. (1992) Brief dendritic calcium signals initiate long-lasting synaptic depression in cerebellar Purkinje cells. Proceedings of the National Academy of Sciences of the USA 89:7051-7055.

Lambolez, B., Audinat, E., Bochet, P., Crepel, F., & Rossier, J. (1992) AMPA receptor subunits expressed by single Purkinje cells. Neuron 9:247-258.

Linden, D.J., Dickinson, M.H., Smeyne, M. & Connor, J.A. (1991) A long term depression of AMPA currents in cultured cerebellar Purkinje neurons. Neuron 7:81-89.

Linden, D. J. & Connor, J.A. (1991). Participation of postsynaptic PKC in cerebellar long-term depression in culture. Science 254:1656-1659.

Linden, D. & Connor, J.A. (1992) Long-term depression of glutamate currents in cultured cerebellar Purkinje neurons does not require nitric oxide signalling. The European Journal of Neuroscience 4:10-15.

Llano, I., Dreesen, J., Kano, M., & Konnerth, A. (1991) Intradendritic release of calcium induced by glutamate in cerebellar Purkinje cells. Neuron 7: 577- 583.

Lohmann, S.M., Walter, U., Miller, P.E., Greengard, P. & Camilli, P.D. (1981) Immunohistochemical localization of cyclic GMP-dependent protein kinase in mammalian brain. Proceedings of the National Academy of Sciences of the USA 78:653-657.

Marr, D. (1969) A theory of cerebellar cortex. Journal of Physiology 202:437-470.

Martin, L.J., Blackstone, C.D., Huganir, R.L. & Price, D.L. (1992) Cellular localization of a metabotropic glutamate receptor in rat brain. Neuron 9:259- 270.

Mayer, M. L. & Westbrook, G. L. (1987) The physiology of excitatory amino acids in the vertebrate central nervous system. Progress in Neurobiology 28:197-276.

Miyakawa, H., Lev-Ram, V., Lasser-Ross, N. & Ross, W.N. (1992) Calcium transients evoked by climbing fiber and parallel fiber synaptic inputs in guinea pig cerebellar Purkinje neurons. Journal of Neurophysiology 4:1178- 1189.

Nagao, S. & Ito, M. (1991) Subdural application of hemoglobin to the cerebellum blocks vestibuloocular reflex adaptation. Neuroreport 2:193-196.

Nakanishi, N., Sneider, N.A. & Axel, R. (1990) A family of glutamate receptor genes: evidence for the formation of heteromultimeric receptors with distinct channel properties. Neuron 5:569-581.

Nakanishi, N. (1992) Molecular diversity of glutamate receptors and implications for brain function. Science 258:597-603.

Nicoletti, F., Meek, J.M., Iadora, M.J., Chuang, D.M., Roth, B.L. & Costa, E. (1986) Coupling of inositol phospholipid metabolism with excitatory amino acid recognition sites in rat hippocampus. Journal of Neurochemistry 46:40-46.

Nishizuka, Y. (1986) Studies and perspectives of protein kinase C. Science 233:305-311.

Recasens, M., Sassetti, I., Nourigat, A., Sladeczek, F. & Bockaert, J. (1987) Characterization of subtypes of excitatory amino acid receptors involved in the stimulation of inositol phosphate synthesis in rat-brain synaptoneurosomes. European Journal of Pharmacology 141:87-93.

Ross, C.A., Bredt, D. & Snyder, S.H. (1990) Messenger molecules in the cerebellum. Trends in Neurosciences 13:216-222.

Sacktor, T.C., Osten, P., Valsamis, H., Jiang, X., Naik, M.U. & Sublette, E. (1993) Persistent activation of the _ isoform of protein kinase C in the maintenance of long-term potentiation. Proceedings of the National Academy of Sciences of the U.S.A 90: 8342-8346.

Sakurai, M. (1990). Calcium is an intracellular mediator of the climbing fiber in induction of cerebellar long-term depression. Proceedings of the National Academy of Sciences of the U.S.A 87:8383-8385.

Shibuki, K. & Okada, D. (1991) Endogenous nitric oxide release required for long-term synaptic depression in the cerebellum. Nature 349:326-328.

Sladeczek, F., Pin, J.P., Recasens, M., Bockaert, J. & Weiss, S. (1985). Glutamate stimulates inositol phosphate formation in striatal neurones. Nature 314:717-719.

Sommer, B., Keinnen, K., Verdoorn, T.A., Wisden, W., Burnashev, N., Herb, A., Khler, M., Takagi, T., Sakmann, B. & Seeburg, P.H. (1990) Flip and Flop: A cell- specific functional switch in glutamate-operated channels of the CNS. Science 249:1580-1585.

Sommer, B. & Seeburg, P.H. (1992) Glutamate receptor channels: novel properties and new clones. Trends in Pharmacological Sciences 13:291-296.

Southam, E., Morris, R. & Garthwaite, J. (1992) Sources and targets of nitric oxide in rat cerebellum. Neuroscience Letters 137:241-244.

Staub, C., Vranesic, I., & Knpfel, T. (1992) Responses to metabotropic glutamate receptor activation in cerebellar Purkinje cells: induction of an inward current. European J. Neurosci., 4: 832-839

Sugiyama, H., Ito, I. & Hirono, C. (1987). A new type of glutamate receptor linked to inositol phospholipid metabolism. Nature 325:531-533.

Tang, C.M., Shi, Q.Y., Katchman, A. & Lynch, G. (1991) Modulation of the time course of fast EPSPs and glutamate channel kinetics by Aniracetam. Science 254:288-290.

Tremblay, J., Gerzer, R., & Hamet, P. (1988) Cyclic GMP in cell function. Advances in second messengers and phosphoprotein research. Vol. 22. Greengard, P. and Robison, G.A. Eds. pp 319-368. Raven Press: New York.

Vranesic, I., Batchelor, A., Ghwiler, B.H., Garthwaite, J., Staube, C., & Knpfel, T. (1991) Trans-ACPD-induced Ca2+ signals in cerebellar Purkinje cells. NeuroReport 2: 759-762.

Vyklicky, L., Patneau, D.K. & Mayer, M.L. (1991) Modulation of excitatory synaptic transmission by drugs that reduce desensitization at AMPA/kanate receptors. Neuron 7:971-984.

Watkins, J. C. (1981) Pharmacology of excitatory amino acid transmitters. In: Advances in Biochemical Psychopharmacology, Vol. 29. Amino Acid Neurotransmitters, ed. DeFeudis, F. W. and Mandel, P., pp 205-212. New York: Raven Press.

Wenthold, R.J., Yokotami, N., Doi, K & Wada, K. (1992) Immunochemical characterization of the non-NMDA receptor using subunit-specific antibodies. Journal of Biological chemistry 267:501-507.

Zhang, N., Walberg, F., Laake, J.H., Meldrum, B.S. & Ottersen, O.P. (1990) Aspartate-like and glutamate-like immunoreactivities in the inferior olive and climbing system: a light microscopic and semi-quantitative electron microscopic study in rat and baboon (Papio anubis). Neuroscience 38:61-80.

FIGURE LEGENDS

[Note: Figures avaiable only in hard copy version]

FIGURE 1: Role of calcium in long term depression of PF-transmission (diagrams of experimental arrangements are shown on the left of each panel). A: Example of LTD in a cerebellar slice, induced by pairing (P) PF-mediated EPSCs with depolarization of PCs giving rise to Ca spike (duration=1min). Plot of EPSC amplitudes against time before and after the pairing. Insets display control EPSC (1) and EPSC 15 min after the end of the pairing period (2) (Crepel et al., unpublished) . B: Conjunctive depolarization of cultured PC to -20 mV (from t=0 to 4 min , P) and glutamate application (at 0.05 Hz) produces LTD of glutamate currents. Insets display corresponding glutamate current traces before and after LTD induction. (adapted fron Linden, D.J. et al. (1991), Neuron 7: 81-89 ). C: Time course of changes of PF-EPSC amplitudes and [Ca]i ( measured with fura- 2 fluorescence signal ) in a PC of cerebellar slice. At time=0 min, a series of eight depolarizing pulses ( 10 ms duration, from -60 to 0 mV) were paired with the PF stimulation. (adapted from Konnerth, A. et al. (1992), PNAS 89: 7051-7055) .

FIGURE 2: Involvement of AMPA receptor in LTD induction. A: In cerebellar slices, pairing PF-mediated EPSCs with Ca spikes does not induce LTD in the presence of 4 m CNQX in the bath. Plot of EPSC amplitudes against time before and after the pairing protocol (P). Insets: averaged EPSCs at the indicated times (Crepel et al., unpublished ). B: Conjunctive application of quisqualate with depolarization of cultured PC (P) in the presence of 20 m CNQX failed to induce LTD of quisqualate currents after washout of CNQX. A second conjunctive stimulation (applied at t=30 min) in the absence of CNQX induced LTD. (adapted from Linden, D.J. et al. (1991), Neuron 7: 81-89).

FIGURE 3: Involvement of metabotropic receptor in LTD induction. A: Effects of conjunctive application of AMPA (10 M) and t-ACPD (0.3 mM). Records A-F were taken at 10 min intervals and show AMPA-induced potentials. Upward arrow indicates that C lies between B and D. (experimental arrangement: w, slice in a wedgeform; arrows indicate flow of perfusate; e and f are recording electrodes connected to d.c. amplifier; g, ground). (adapted from Ito, M. & Karachot, L. (1990) Neuroreport 1:129-132 ) . B: Effects of pairing of PF-mediated EPSPs with Ca spikes performed in the presence of t-ACPD in the bath. Plot of EPSP amplitudes against time. Insets display control EPSP (1), EPSP under application of t-ACPD (2) and EPSP 15 min after the pairing protocol performed in the presence of trans-ACPD (3). C: Plot of EPSPs amplitudes against time in another Purkinje cell before and after a pseudo-pairing protocol performed in the presence of trans-ACPD. Insets: EPSPs at the indicated time. ( In this protocol, PF stimulation was omitted during the 1 min period of Ca spike firing ) (B and C adapted from Daniel et al., (1992). Experimental Brain Research 90: 327-331) .

FIGURE 4: Involvement of PKC pathway in LTD. A: Effect of phorbol esters on excitatory amino acid induced responses. Post- stimulus time histograms of the responses of an extracellularly recorded PC to Glu, Asp and Quis, before (A1) and after (A2) bath application of 400 m of phorbol 12 13 -dibutyrate (PDBu) (adapted from Crepel and Krupa (1990) , Excitatory amino acids and Neural plasticity, Plenum press, pp 323-329 ) . B: Conjunctive depolarization of cultured PC and glutamate application (P) produces LTD of glutamate currents with the following internal perfusates in the pipette: vehicle ( D ) and vehicle plus [glu 27] PKC(19-36) ( o ,10 m), but no LTD occured with vehicle plus BAPTA ( D , 20 m) and vehicle plus PKC (19-36) (o , 10 m). Insets: EPSCs at the indicated times. Scale bars = 100 pA, 2 s. (adapted from Linden, D.J. & Connor, J.A. (1991) Science 254: 1656-1659 ) .

FIGURE 5: Involvement of NO pathway in LTD A: Effects of pairing on PF-mediated EPSPs in standard bathing medium. Averaged EPSP evoked in a Purkinje cell during the control period (A1), then 1 min (A2) and 15 min (A3) after the end of the pairing period with Ca spikes (A2). The EPSPs in A1 and A4 are superimposed (A5). B: Plot of EPSP amplitudes against time for the cell illustrated in A, before and after the pairing protocol (P). C to E: Plot of EPSP amplitudes against time in 3 others Purkinje cells before and after a pairing protocol (P) with respectively, 30 m L-NMMA in the bath (C), 3 mM SIN-1 added in the recording pipette (D) and 0.5 mM cGMP added in the recording pipette(E). Insets: EPSPs at the indicated times. (adapted from Daniel et al. (1993 ) The European Journal of Neuroscience 5: 1079-1082 ) .

FIGURE 6: Schematic diagram of the signal transduction processes involving the PKC and NO-pathway, that are presumed to underlie LTD. In this scheme, we have illustrated the hypothesis according to which NO-synthase involved in LTD is located outside PCs.

G: Glutamate m-GLUR: Metabotropic receptor AMPA-R: AMPA receptor G: G protein PLC: Phospholipase C PKC: Protein kinase C NO: Nitric oxide cGMP: Cyclic guanosine monophosphate PKG: Protein kinase G Pi: Inorganic phosphate P: Phosphorylation