HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Full Text (PDF) Supplemental Results and Figures All Versions of this Article: 98/5/2550    most recent 00376.2007v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Crepel, F. Articles by Daniel, H. Search for Related Content PubMed PubMed Citation Articles by Crepel, F. Articles by Daniel, H.

Developmental Changes in Agonist-Induced Retrograde Signaling at Parallel Fiber–Purkinje Cell Synapses: Role of Calcium-Induced Calcium Release

Pharmacologie de la Synapse, Institut de Biochimie et de Biophysique Moléculaire et Cellulaire, Université Paris-Sud, Orsay Cedex, France

Submitted 2 April 2007; accepted in final form 7 September 2007

	ABSTRACT

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
In cerebellar Purkinje cells (PCs), activation of postsynaptic mGluR1 receptors inhibits parallel fiber (PF) to PC synaptic transmission by retrograde signaling. However, results were conflicting with respect to whether endocannabinoids or glutamate (Glu) is the retrograde messenger involved. Experiments in cerebellar slices from 10- to 12-day-old rats and mice confirmed that suppression of PF-excitatory postsynaptic currents (EPSCs) by mGluR1 agonists was entirely blocked by cannabinoid receptor antagonists at this early developmental stage. In contrast, suppression of PF-EPSCs by mGluR1 agonists was only partly blocked by cannabinoid receptor antagonists in 18- to 22-day-old rats, and the remaining suppression was accompanied by an increase in paired-pulse facilitation. This endocannnabinoidindependent suppression of PF-EPSCs was potentiated by the Glu uptake inhibitor D-threo--benzyloxyaspartate (D-TBOA) and blocked by the desensitizing kainate (KA) receptors agonist SYM 2081, by nonsaturating concentrations of 6-cyano-7-nitroquinoxaline-2-3-dione (CNQX) [but not by GYKI 52466 hydrochloride (GYKI)] and by dialyzing PCs with guanosine 5'-[-thio]diphosphate (GDP-S). An endocannnabinoid-independent suppression of PF-EPSCs was also present in nearly mature wild-type mice but was absent in GluR6^–/– mice. The endocannnabinoid-independent suppression of PF-EPSCs induced by mGluR1 agonists and the KA-dependent component of depolarization-induced suppression of excitation (DSE) were blocked by ryanodine acting at a presynaptic level. We conclude that retrograde release of Glu by PCs participates in mGluR1 agonist-induced suppression of PF-EPSCs at nearly mature PF-PC synapses and that Glu operates through activation of presynaptic KA receptors located on PFs and prolonged release of calcium from presynaptic internal calcium stores.

	INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
In cerebellar Purkinje cells (PCs), activation of mGluR1 postsynaptic metabotropic glutamate receptors by selective agonists (Galante and Diana 2004; Levenes et al. 2001; Maejima et al. 2001) or by sustained parallel fiber (PF) stimulation (Brown et al. 2003; Maejima et al. 2001; Neale et al. 2001) inhibits both excitatory and inhibitory inputs to these neurons by retrograde signaling (Brown et al. 2003; Galante and Diana 2004; Levenes et al. 2001; Maejima et al. 2001; Marcaggi and Attwell 2005). Endocannabinoids have been favored as the retrograde messenger involved in this signaling (Galante and Diana 2004; Maejima et al. 2001), as was previously shown for depolarization-induced suppression of inhibition (DSI) (Diana et al. 2002; Glitsch et al. 1996, 2000; Kreitzer and Regehr 2001a; Llano et al. 1991a; Ohno-Shosaku et al. 2001; Pitler and Alger 1992, 1994; Vincent et al. 1992; Wang and Zucker 2001; Wilson and Nicoll 2001; Wilson et al. 2001) and for depolarization-induced suppression of excitation (DSE) (Brenowitz and Regehr 2003; Kreitzer and Regehr 2001b; Safo and Regehr 2005). However, the study by Levenes et al. (2001) contrasts with these results, suggesting instead that retrograde release of glutamate (Glu) by PCs was responsible for the observed agonist-dependent suppression of PF-excitatory postsynaptic currents (EPSCs), through activation of presynaptic ionotropic Glu receptors borne by PFs. Because most studies on the role of endocannabinoids in retrograde signaling at PF-PC synapses have been performed in juvenile rats and mice (but see Safo and Regehr 2005), whereas the study by Levenes et al. (2001) was performed in nearly mature rats, the apparent discrepancy on the nature of retrograde messengers involved in agonist-induced suppression of excitation at PF-PC synapses might be related to developmental differences. Indeed, we now know that DSE at PF-PC synapses is entirely mediated through retrograde release of endocannabinoids in juvenile rodents, whereas it also involves retrograde release of Glu in nearly mature animals (Crepel 2007).

Therefore these experiments were designed to determine whether such distinct mechanisms are also involved in suppression of PF-EPSCs by activation of postsynaptic mGluR1 in juvenile and nearly mature rats and mice. Because presynaptic kainate (KA) receptors are involved in DSE in nearly mature PF-PC synapses (Crepel 2007), emphasis was made on a possible role of these receptors in agonist-induced suppression of PF-EPSCs in nearly mature PF-PC synapses, as well as on a possible participation of presynaptic calcium-induced calcium release in this process.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Experimental procedures complied with guidelines of the French Animal Care Committee. They were performed on juvenile (10–12 days old) and on nearly mature (18–22 days old) male rats (Sprague-Dawley). Additional experiments were also performed on 22- to 24-day-old C56BL/6 and GluR6^–/– (on a hybrid 129Sv x C57BL/6 background) mice, as well as on juvenile (10–12 days old) C56BL/6 mice. In all cases, animals were stunned before decapitation, and parasagittal slices, 250 µm thick, were cut in ice-cold saline solution from the cerebellar vermis with a vibroslicer. Slices were incubated at room temperature in saline solution equilibrated with 95% O₂-5% CO₂ for 1 h. The recording chamber was perfused at a rate of 2 ml/min with oxygenated saline solution containing (in mM) 124 NaCl, 3 KCl, 24 NaHCO₃, 1.15 KH₂PO₄, 1.15 MgSO₄, 2 CaCl₂, 10 glucose, and the GABA_A antagonist bicuculline methochloride (10 µM, Sigma Aldrich, St. Quentin Fallavier, France), osmolarity 320 mOsm, final pH 7.35 at 27–28°C except when otherwise specified. PCs were directly visualized with Nomarski optics through a x40 water-immersion objective of an upright microscope (Zeiss).

Drugs were added to the superfusate. (S)-3,5-dihydroxyphenylglycine (DHPG), domoate, D-threo--benzyloxyaspartate (D-TBOA), 6-cyano7-nitroquinoxaline-2-3-dione (CNQX), GYKI 52466 hydrochloride (GYKI), D-2-amino-5-phosphopentanoic acid (D-APV), (2S,4R)-4methylglutamic acid (SYM 2081), 8-cyclopentyl-1,3-dipropylxanthine (DPCPX), CGP55845-A,N-(piperidin-1-yl)-5-(4-iodophenyl)-1(2,4dichlorophenyl)-4-methyl-1H-pyrazole-3-carboxamide (AM-251), and ryanodine were purchased from Tocris (Illkirch, France). The CB1 cannabinoid receptor antagonist SR141716-A [N-(piperidin-1-yl)-5- (-4chlorophenyl)-1-(2.4-dichlorophenyl)-4-methyl-1H-pyrazol-3-carboxamide hydrochloride] was provided by Sanofi-Recherche (Montpellier, France). N^G-nitro-L-argininemethyl ester (L-NAME) and guanosine 5'-[-this]diphosphate (GDP-S) was purchased from Sigma Aldrich. Stock solutions of drugs (dissolved in water or DMSO depending on manufacturer recommendations) were added to the oxygenated saline solution at the desired concentration.

Electrophysiology

Whole cell patch-clamp recordings were performed from PC somas, using an Axopatch-200A amplifier (Axon Instruments). Stimulating electrodes consisted in saline filled monopolar electrodes. PF stimulations were performed at 0.33 Hz, except for studies on unitary quantal events and on DSE, where stimulation rates were 0.1 and 0.5 Hz, respectively. Patch pipettes (2–4 M) were filled with a solution containing (in mM) 140 K-gluconate, 6 KCl, 10 HEPES, 0.75 EGTA, 1 MgCl₂, 4 Na2-ATP, and 0.4 Na-GTP, pH 7.35 with KOH; 300 mOsm. In experiments on DSE, patch pipettes (2–4 M) were filled with a solution containing (in mM) 140 Cs-gluconate, 6 KCl, 10 HEPES, 0.2 EGTA, 1 MgCl₂, 4 Na2-ATP, and 0.4 Na-GTP (pH and osmolarity adjusted accordingly). The components of internal solutions were purchased from Sigma Aldrich.

In the cells retained for analysis, access resistance (usually 5–10 M) was partially compensated (50–70%), according to the procedure described by Llano et al. (1991b). Cells were held at a membrane potential of –70 mV, and PF-EPSCs were subjected to 10-mV hyperpolarizing voltage steps that allowed monitoring of the passive electrical properties of the recorded cell throughout the experiment (Llano et al. 1991b).

For paired-pulse facilitation (PPF) experiments (Atluri and Regehr 1996; McNaughton 1982; Schultz et al. 1994), PF stimulations of the same intensity were applied to the cell with an interstimulus interval of 30 ms, and the ratio of the amplitude of the second PF-EPSC over the first one was calculated on-line. Mean PPF values were obtained by averaging PPFs in individual traces for each cell studied in a given condition. Although this conventional method can produce spurious results (Kim and Alger 2001), no use was made of the alternative method proposed by these authors because both methods produced very similar results in a recent study on DSE at PF-PC synapses (Crepel 2007). Because PPF increases associated to agonist-induced suppression of PF-EPSCs could be small in certain experimental conditions (see RESULTS) and obscured by variability of basal PPF across cells, mean PPFs were further normalized in these experiments by determining for each cell mean normalized PPF = 100 x (mean PPF/PPFi), where PPFi was the individual PPF value of the last trace preceding the depolarizing step. The contribution of presynaptic factors in the variation of synaptic responses was also examined by using the coefficient of variation (CV) (Kullmann 1994; Martin 1966), where CV is given by: CV² = (s/M)², in which s is the SD of the amplitude distribution of EPSCs corrected for the background noise, and M is the mean amplitude of EPSCs during the same epoch. In these experiments, CV were calculated on sets of 40–60 stable EPSCs according to the procedure previously described (Blond et al. 1997).

Fluorometry

Parallel fibers in coronal slices from nearly mature and juvenile rats (see RESULTS) were loaded by focal application of 100 µM of the low-affinity calcium sensitive dye Fluo-4FF-AM (Molecular Probes) as previously described (Levenes et al. 2001). After loading for 45 min, fluorescent signals from labeled PFs were recorded in a 20 x 50-µm window placed above the molecular layer, 500–800 µm away from the loading site and 100 µm above PC layer. The epifluorescence excitation light at 485 ± 22 nm was gated with an electromechanical shutter (Uniblitz, Rochester, NY), and the emitted light was collected by a photometer through a barrier filter at 530 ± 30 nm. The rather selective and weakly desensitizing KA receptor agonist domoate (Lerma et al. 1993) was applied for 1 min and at a concentration of 20 µM by local superperfusion through a theta-tube (rate of 0.5 ml/min) placed 50 µm above the surface of the slice, at the level of the molecular layer and parallel to labeled PFs. With such a protocol that minimized possible movements of the slice when switching superfusion through the theta-tube from standard saline solution to that with domoate included, this compound was unlikely to diffuse in significant concentration to nearby granule cells, whereas it was probably still at near saturing concentrations for presynaptic KA receptors (Renard et al. 1995), and this, even after partial dilution during superfusion of the tissue. In these experiments, the 1-min superfusion duration was chosen because, in pilot experiments, it was the shortest time that gave rise to reliable presynaptic calcium signals. Fluorescence signals corrected for dye bleaching and background fluorescence were expressed as relative fluorescence changes F/F, where F was the baseline fluorescence intensity and F was the change induced by domoate.

In sagittal slices, epifluorescence microscopy was also used to detect variations in intracellular free calcium concentration changes from an area of 90 x 90 µm centered on dendrites of recorded PCs, as previously described (Crepel 2007). In these experiments, the low affinity and impermeant calcium indicator Fluo-4FF (100 µM, Molecular Probes) was added to the same Cs-gluconate–based solution as used in experiments on DSE. The recording session started 30–45 min after whole cell break in, to allow diffusion of the dye to the dendrites. Fluorescence data corrected for dye bleaching and background fluorescence were again expressed as changes in F/F, where F was the baseline fluorescence intensity, and F was the fluorescence change induced by depolarization of PC soma from –70 to 0 mV for 1 s.

In all cases, pre- and postsynaptic calcium signals were recorded while superfusing the slices with a cocktail of GABA_A, GABA_B, and adenosine A1 receptor antagonists, i.e., bicuculline methochloride (10 µM), CGP55845 (300 nM), and DPCPX (100 nM), respectively. Fluorometric measurements were analyzed on- and off-line using the Acquis1 computer program (Biologic).

Statistical significance was assessed by paired or unpaired t-test, as appropriate, with P < 0.05 (2-tailed) considered significant. All error values given are mean ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Agonist-induced suppression of PF-EPSCs depends entirely on retrograde release of endocannabinoids in juvenile rodents but only partly in nearly mature ones

In 10- to 12-day-old mice, as shown in the previous study by Maejima et al. (2001), 5-min bath application of the selective mGluR1 and mGluR5 agonist DHPG, at an apparent saturing concentration of 100 µM (Canepari et al. 2004; Pin and Duvoisin 1995; Schoepp et al. 1994), induced a large and significant (P < 0.001) decrease in the mean amplitude of PF-EPSCs to 54.17 ± 5.12% of control (n = 6; Fig. 1, A and B). This suppression was accompanied by a large and significant (P < 0.002) increase in PPF, from 1.35 ± 0.06% in control to 1.86 ± 0.19% at the peak of the DHPG effect. This is in keeping with a presynaptic action of DHPG at PF-PC synapses through retrograde signaling in juvenile mice (Maejima et al. 2001). PF-EPSCs recovered their initial amplitude within <2 min of DHPG application and thereafter were potentiated for several minutes (123.48 ± 8.64% of control on average; Fig. 1, A and B). Very similar results were obtained in 10- to 12-day-old rats because bath application of 100 µM DHPG also induced a reversible decrease in the mean amplitude of PF-EPSCs to 68.22 ± 4.37% of control that was also followed by a transient potentiation of PF-EPSCs (n = 6; Fig. 1C). However, this potentiation was shorter than in 10- to 12-day-old mice and was later followed by a transient, albeit not significant, depression of PF-EPSCs that amounted to 8.98 ± 8.95% on average (Fig. 1C). In both mice and rats, the transient potentiation that followed the initial suppression of PF-EPSCs did not give rise to any significant variation in PPF (data not shown). It was therefore reminiscent of transient potentiations observed in older animals after blockade of endocannabinoid dependent- and endocannabinoid-independent components of DHPG-induced suppression of PF-EPSCs. Accordingly, these potentiating effects are likely to be induced at a postsynaptic level in immature and in nearly mature PCs (see DISCUSSION).

View larger version (27K): [in this window] [in a new window]   FIG. 1. Effects of (S)-3,5-dihydroxyphenylglycine (DHPG) on parallel fiber (PF)-excitatory postsynaptic currents (EPSCs) in control medium and in the presence of CB1 receptor antagonists in juvenile mice (A and B) and rats (C). A: superimposed sweeps of PF-EPSCs elicited in 1 Purkinje cell (PC) by 2 successive PF stimulations (interstimulus interval = 30 ms) in control solution (1), in the presence of 100 µM DHPG (2), and after washout (3) in an 11-day-old mouse. B: plot of mean (±SE; same in all figures except otherwise specified) normalized amplitudes of PF-EPSCs recorded from PCs over time in 10- to 12-day-old mice in control medium (black squares) and in the presence of 2 µM AM-251 (white squares). DHPG (100 µM) was added to the bath for 5 min, as indicated by corresponding horizontal bar. Note that 1, 2, and 3 correspond to numbers indicated in A. C: same plots as in B in 10- to 12-day-old rats in control solution (black squares) and in the presence of 1 µM SR141716-A (white squares).

In nearly mature (18–22 days old) rats (n = 10) and for all cells tested, 5-min bath application of 100 µM DHPG induced a transient and significant (P < 0.001) decrease in the amplitude of PF-EPSCs that amounted to 43.23 ± 5.08% of control (Fig. 2, A and B1). Most importantly, this transient suppression of PF-EPSCs was always accompanied by a significant increase in PPF (P < 0.001) from 1.32 ± 0.08 in control conditions to 1.95 ± 0.22 at the peak of the DHPG effect (Fig. 2B2). This is in keeping with a presynaptic site of action of DHPG at PF-PC synapses through retrograde signaling in nearly mature rats (Levenes et al. 2001). However, in this earlier study, suppression of PF-EPSCs of similar amplitude as those reported here were obtained with (S)-DHPG concentrations of only 50 µM (compare Fig. 2B1 of this study with Fig. 2B in Levenes et al. 2001). In pilot experiments, suppression of PF-EPSCs induced by 50 µM (S)-DHPG applications was on average 1.5 times smaller than that achieved with the same concentration in this earlier study (n = 5; Supplementary Fig. S1).¹ With 25 µM DHPG, the late (endocannabinoid-independent) component of suppression of PF-EPSCs was nearly totally absent and the initial component was still further reduced in amplitude (n = 5; Supplementary Fig. S1). This suggests that, taking into account the present recording chamber's exchange time, this nominal concentration was too low to fully saturate mGluR1 receptors during 5-min bath applications. Therefore concentrations of 100 µM DHPG that gave rise to more robust suppressions of PF-EPSCs were used throughout this study in nearly mature animals. This was also the case in experiments on juvenile rats and mice to compare results with mature ones in the same experimental conditions.

View larger version (34K): [in this window] [in a new window]   FIG. 2. Effects of DHPG on PF-EPSCs in control medium and in the presence of CB1 receptor antagonist in nearly mature rats. A: superimposed sweeps of PF-EPSCs elicited in 1 PC by 2 successive PF stimulations (interstimulus interval = 30 ms) in control solution (1), in the presence of 100 µM DHPG (2), and after washout (3) in an 18-day-old rat. B1: plot of mean normalized amplitudes of PF-EPSCs recorded from PCs over time in 18- to 22-day-old rats in control medium (black squares) and in the presence of 1 µM SR141716-A (white squares). Horizontal bar: duration of DHPG application as in Fig. 1. Note that 1 and 2 correspond to numbers indicated in A. B2: superimposed plots of mean paired-pulse facilitation (PPF) over control, 100 µM DHPG, and washing periods in control medium (black lozenges) and in the presence of 1 µM SR141716-A (white lozenges) for the same PCs as in B1.

Finally, and most importantly, in the presence of CB1 receptor antagonists, the remaining suppression of PF-EPSCs was still accompanied by a significant (P < 0.01) increase in PPF that amounted to 17%, i.e., from 1.29 ± 0.05 in control conditions to 1.51 ± 0.08 at the peak of the DHPG effect (Fig. 2B2). These latter results suggest that suppression of PF-EPSCs by DHPG in nearly mature rats not only involves activation of presynaptic CB1 receptors like in juvenile animals (Maejima et al. 2001), but also involves at least one other presynaptic mechanism. In contrast, kinetics of PF-EPSCs were not significantly affected. Thus mean values of the 10–90% rise time were 2.06 ± 0.21 and 1.90 ± 0.20 ms in control conditions and at the peak of DHPG-induced suppression of PF-EPSCs, respectively, and mean time constants of decay had values of 11.39 ± 0.75 and 10.90 ± 0.79 ms in the same conditions. Although kinetics are certainly severely biased in nearly mature PCs by dendritic filtering of synaptic currents, these data do not suggest that postsynaptic AMPA receptors at PF-PC synapses are sizably affected during the endocannabinoid-independent component of DHPG-induced suppression of PF-EPSCs (see DISCUSSION).

Sensitivity of the CB1 receptor–independent component of agonist-induced suppression of PF-EPSCs to a Glu uptake blocker and to a desensitizing KA receptor agonist in nearly mature rats

In nearly mature rats, retrograde release of Glu by PCs participates in DSE at PF-PC synapses through activation of presynaptic KA receptors that include GluR6 receptor subunits (Crepel 2007). In keeping with this recent finding, the previous study by Levenes et al. (2001) already suggested that activation of postsynaptic mGluR1 decreases PF-EPSCs through retrograde release of Glu and activation of presynaptic ionotropic Glu receptors borne by PFs. Accordingly, in 18- to 22-day-old rats, the endocannabinoid-independent component of agonist-induced suppression of PF-EPSCs should be enhanced in the presence of the Glu uptake inhibitor D-TBOA and suppressed by pharmacological blockade of presynaptic KA receptors.

In the presence of 1 µM SR141716-A and as expected for a Glu uptake inhibitor, application of 100 µM D-TBOA increased the amplitude of PF-EPSCs (Fig. 3A1). This increase ranged between 8 and 76% depending on cells, with a significant mean increase of 34.90 ± 9.79% (n = 9; P < 0.01). As seen in a previous study (Crepel 2007), the rather large variability of potentiating effects of D-TBOA on PF-EPSCs might result from the patterned expression of PC Glu transporters EAAT4 in rat cerebellar cortex (Wadiche and Jahr 2005). In all cells, this effect was accompanied by marked changes in EPSC kinetics (Fig. 3A1), probably because of slower clearance of Glu from synaptic cleft. At the plateau of the D-TBOA effect, application of 100 µM DHPG decreased the mean amplitude of PF-EPSCs that amounted to 39.78 ± 4.63% of control amplitude before DHPG application (Fig. 3, A2 and B1). This decrease in amplitude was significantly larger (P < 0.01) than that observed in the presence of SR141716-A alone (Fig. 3B1) and was accompanied by a significantly larger increase in PPF (P < 0.02) because PPF increased from 1.24 ± 0.17 in SR141716-A + D-TBOA containing medium to 1.77 ± 0.15 at the peak of the DHPG effect (Fig. 3B2). Although slower Glu clearance in the presence of D-TBOA may complicate the interpretation of this difference, the fact that D-TBOA alone did not induce any significant change in PPF (Crepel 2007) suggests that D-TBOA did not affect the presynaptic machinery to such an extent as to sizably affect the level of suppression of PF-EPSCs by DHPG. Therefore these results are consistent with the assumption that retrograde release of Glu participates, together with retrograde release of endocannabinoids, to the depressant effect of DHPG on PF-EPSCs in nearly mature rats. However, it should be noted that group 1 mGluRs are not restricted to PCs but are also found, for instance, on glial cells (Angulo et al. 2004; Karakossian and Otis 2004; Parpura et al. 1994).

View larger version (46K): [in this window] [in a new window]   FIG. 3. Effect of D-threo--benzyloxyaspartate (D-TBOA) on the CB1 receptor–independent component of agonist-induced suppression of PF-EPSCs in nearly mature rats. A1: superimposed sweeps of PF-EPSCs elicited by 2 successive PF stimulations in the presence of 1 µM SR141716-A and in the presence of 1 µM SR141716-A + 100 µM D-TBOA. A2: as in A1 when 100 µM DHPG was added to the bath and after washout of DHPG. Note the marked effect of D-TBOA on EPSC kinetics. B1: plots of mean normalized amplitudes of PF-EPSCs recorded from PCs over time in the presence of 1 µM SR141716-A alone (white squares) or of 1 µM SR141716-A + 100 µM D-TBOA (gray squares). Horizontal bar: duration of DHPG application as in Fig. 1. PF-EPSC amplitudes in the presence of D-TBOA were normalized with respect to values immediately before DHPG application. B2: plots of mean PPF of PF-EPSCs recorded in the same conditions and from the same PCs as in B1. White lozenges: 1 µM SR141716-A alone; gray lozenges: 1 µM SR141716-A + 100 µM D-TBOA.

View larger version (52K): [in this window] [in a new window]   FIG. 4. Sensitivity to (2S,4R)-4-methylglutamic acid (SYM 2081) of the CB1 receptor–independent component of agonist-induced suppression of PF-EPSCs in nearly mature rats. A: superimposed plots of mean normalized amplitudes of PF-EPSCs recorded from PCs over time in the presence of 1 µM SR141716-A alone (white squares), when 10 µM SYM 2081 was also present in the bath (gray squares), and when 300 nM CGP55845-A + 200 µM L-NAME were further added to the bath (black squares). Horizontal bar: duration of DHPG application as in Fig. 1. B: superimposed plots of mean normalized PPF of PF-EPSCs recorded in the same conditions and from the same PCs as in A. White lozenges: 1 µM SR141716-A alone; gray lozenges: 1 µM SR141716-A +10 µM SYM 2081; black lozenges: 1 µM SR141716-A + 10 µM SYM 2081 + 300 nM CGP55845-A + 200 µM L-NAME.

Sensitivity of the CB1 receptor–independent component of suppression of PF-EPSCs by DHPG to nonsaturating concentrations of CNQX and GYKI in nearly mature rats

Because SYM 2081 is a desensitizing KA receptor agonist rather than a genuine KA receptor antagonist, it was important to confirm the involvement of presynaptic KA receptors in the late phase of agonist-induced suppression of PF-EPSCs. We reasoned that the concentration of Glu achieved at the level of presynaptic KA receptors involved in the CB1 receptor–independent component of agonist-induced suppression of PF-EPSCs is likely to be much lower than that seen by postsynaptic AMPA receptors during PF-PC EPSCs. If so, it is possible that a nonsaturating concentration of the competitive AMPA/KA antagonist CNQX (Honoré et al. 1988) that only partly block PF-EPSCs is sufficient to fully antagonize presynaptic KA receptors and thus inhibit agonist-induced suppression of PF-EPSCs.

Indeed, bath application of 1 µM CNQX in the presence 1 µM SR141716-A reduced the amplitude of PF-EPSCs to 36.33 ± 3.76% of the control value (n = 11). Unexpectedly, this effect was accompanied by a large and highly significant (P < 0.001) PPF increase, from 1.30 ± 0.08 to 1.75 ± 0.14 (Fig. 5A1) that was reminiscent of that observed for basal PPF in the presence of SYM 2081 in nearly mature rats and in GluR6^–/– mice (Crepel 2007). In keeping with this latter result that suggests a presynaptic site of action of CNQX in addition to its well-established postsynaptic effect, mean CV of PF-EPSCs also significantly (P < 0.001) increased from to 0.045 ± 0.006 in control to 0.081 ± 0.012 at the steady state of the depressant effect of CNQX (n = 9; Fig. 5B).

View larger version (36K): [in this window] [in a new window]   FIG. 5. Sensitivity to 6-cyano-7-nitroquinoxaline-2-3-dione (CNQX) and GYKI 52466 hydrochloride (GYKI) of the CB1 receptor–independent component of agonist-induced suppression of PF-EPSCs in nearly mature rats. A1: superimposed plots of mean PPF of PF-EPSCs recorded from PCs over time in the presence of 1 µM SR141716-A alone, when 1 µM CNQX or 20 µM GYKI (black and white lozenges, respectively) were added to the superfusing medium, and when 100 µM DHPG was further added to the bath for 5 min as indicated by corresponding horizontal bars. A2: superimposed plots of mean normalized amplitude of PF-EPSCs recorded over time from the same PCs as in A1 in the presence of 1 µM SR141716-A + 1 µM CNQX (black squares) or in the presence of 1 µM SR141716-A + 20 µM GYKI (white and gray symbols) see below, and when 100 µM DHPG was further added to the bath for 5 min as indicated by the corresponding horizontal bar. Plots with gray and white squares correspond to the 3 and 8 cells where GYKI inhibited or not suppression of PF-EPSCs by DHPG, respectively, see RESULTS. B: histogram of mean CV (±SE) of PF-EPSC in PPF experiments. Bars labeled Con-1, CNQX-1, and GYKI-1 represent mean CV of first EPSCs within pairs recorded in the presence of 1 µM SR141716-A alone (white bar), of 1 µM SR141716-A + 1 µM CNQX (black bar), and of 1 µM SR141716-A + 20 µM GYKI (gray bar).

To preclude that the strong reduction in PF-EPSC amplitude by CNQX was not solely responsible for the lack of suppression of PF-EPSCs by DHPG in the experiments reported above, we also studied the effect of nonsaturating concentrations of GYKI, a noncompetitive and more selective AMPA receptor antagonist (Bureau et al. 1999; Renard et al. 1995; Wilding and Huettner 1995). On average, bath application of 20 µM GYKI reduced the amplitude of PF-EPSCs to 42.63 ± 3.92% of control values (n = 11), a decrease nonsignificantly different from that obtained with 1 µM CNQX. Like for experiments with CNQX, this decrease in EPSC amplitude was also accompanied by a significant (P < 0.01) PPF increase, although about only one half of that obtained in the presence of CNQX (Fig. 5A1). However, and in marked contrast with results obtained with CNQX, mean CV of PF-EPSCs did not significantly increase during the depressant effect of GYKI on PF-EPSCs, because mean CV values during the control period and at the steady state of the effect of GYKI were 0.046 ± 0.005 and 0.051 ± 0.006, respectively (Fig. 5B).

In 8 of the 11 tested cells, GYKI did not significantly inhibit the CB1 receptor–independent suppression of PF-EPSCs induced by bath application of 100 µM DHPG, because the mean decrease in peak amplitude of PF-EPSCs was 28.48 ± 7.07% compared with 22.11 ± 4.30% (n = 14) in the presence of SR141716-A alone (Fig. 5A2). Moreover, the time-course of PF-EPSC suppression was unchanged (cf. Fig. 2B1 and 5A2). In the remaining three cells, GYKI totally abolished the CB1 receptor–independent suppression of PF-EPSCs induced by bath application of 100 µM DHPG, which unmasked a short-term potentiation of PF-EPSCs (Fig. 5A2). Because there was no apparent difference between these two groups of cells with respect to the effect of GYKI on PF-EPSC amplitude and on basal PPF, DHPG results were pooled, leading to a mean suppressing effect of this compound on peak amplitude of PF-EPSCs that amounted to 20.83 ± 5.76%. This value was very close to that obtained in the presence of SR141716-A alone. Accordingly, DHPG suppression of PF-EPSCs was still accompanied by a significant (P < 0.05) and near 20% increase in mean PPF, from 1.54 ± 0.08 in the presence of SR141716-A + GYKI to 1.82 ± 0.18 at the peak of DHPG effect (Fig. 5A1).

Taken together, CNQX and GYKI results fully confirm that presynaptic KA receptors are likely to be involved in CB1 receptor–independent suppression of PF-EPSCs by mGluR1 agonists (see DISCUSSION).

Finally, because PF-EPSC amplitude did not reach a steady state during DHPG application (Figs. 2B1 and 5A2), no attempt was made to apply the CV method here. In contrast, such a near steady state was achieved during the endocannabinoid-independent component of DSE (Crepel 2007) and was accompanied by a significant increase in CV (see Supplementary Results and Supplementary Fig. S2), thus confirming involvement of presynaptic mechanisms in the CB1 receptor–independent component of DSE (Crepel 2007).

DHPG sensitivity of PF-EPSCs in wild-type and GluR6^–/– mice

Presynaptic KA receptors located on PFs are likely to be heteromeric constructions that include GluR6 and KA2 receptor subunits (Lerma et al. 2001; Petralia et al. 1994). Their activation up- or down-regulates Glu release depending on agonist concentration (Delaney and Jahr 2002). Agonist-induced suppression of PF-EPSCs was studied in nearly mature GluR6^–/– mice and compared with that of wild-type mice of the same strain (see METHODS), with the assumption that invalidating GluR6 receptor subunits renders presynaptic KA receptors nonfunctional (Ruiz et al. 2005). All experiments in GluR6^–/– mice were performed in the presence of 50 µM D-APV to minimize possible developmental compensations by presynaptic NMDA receptors and in the presence of 1 µM SR141716-A to focus on the endocannabinoid-independent component of agonist-induced suppression of PF-EPSCs. For comparison, all experiments in wild-type mice were also performed in the presence of 50 µM D-APV and of 1 µM SR141716-A.

In 22- to 24-day-old wild-type mice, 5-min bath application of 100 µM DHPG induced an endocannabinoid-independent suppression of PF-EPSCs of similar amplitude and duration as in nearly mature rats, with a mean peak amplitude decrease of 17.62 ± 2.76% (n = 6; Fig. 6A). This suppression was accompanied by a significant (P < 0.001) increase in mean normalized PPF, from 100.39 ± 3.66% in control conditions to 116.27 ± 3.24% at the peak of the DHPG effect (Fig. 6B).

View larger version (39K): [in this window] [in a new window]   FIG. 6. Sensitivity of PF-EPSCs to DHPG in wild-type and GluR6–/– mice. A: superimposed plots of mean normalized amplitudes of PF-EPSCs recorded from PCs in the presence of 1 µM SR141716-A and when 100 µM DHPG was further added to the bath for 5 min (horizontal bar) in 22- to 24-day-old wild-type (white squares) and in 22- to 24-day-old GluR6–/– (black squares) mice. B: superimposed plots of mean normalized PPF of PF-EPSCs recorded in the same conditions and from the same PCs as in A. White lozenges: wild-type mice; black lozenges: GluR6–/– mice.

GDP-S sensitivity of the CB1 receptor–independent component of agonist-induced suppression of PF-EPSCs in nearly mature rats

As mentioned before, group 1 mGluRs are not restricted to PCs but are also found on various cell types including granule cells, molecular layer interneurons, and glial cells (Angulo et al. 2004; Baude et al. 1993; Parpura et al. 1994). As such, glutamate release from these cells might participate in DHPG-induced suppression of PF-EPSCs. In an attempt to exclude this possibility, we selectively blocked G protein activity in the postsynaptic compartment by dialyzing PCs through the patch pipette for 30 min after break-in with a conventional K-gluconate internal solution with 4 mM GDP-S added (Galante and Diana 2004); GDP-S is a nonhydrolyzable GTP analog that inhibits G protein activity. Experiments were performed in the presence of 1 µM SR141716-A to focus on the endocannabinoid-independent suppression of PF-EPSCs. In five of the six cells tested, the late phase of the endocannabinoid-independent suppression of PF-EPSCs induced by bath application of 100 µM DHPG was abolished and replaced by a short-term potentiation of PF-EPSCs, whereas its initial phase was much less affected (Fig. 7A). In these five cells, the late phase of mean normalized PPF increase associated with the endocannabinoid-independent suppression of PF-EPSCs was also inhibited and replaced by a nonsignificant 7.72 ± 2.83% decrease of the mean normalized PPF (Fig. 7B). In the remaining PC, the late phase of the endocannabinoid-independent suppression of PF-EPSCs was only partly abolished (data not shown). In contrast, in five other PCs dialyzed during the same period of time after break-in with a conventional K-gluconate internal solution without GDP-S, a clear-cut endocannabinoid-independent suppression of PF-EPSCs and associated PPF increase were still induced by bath application of 100 µM DHPG (Fig. 7, A and B).

View larger version (37K): [in this window] [in a new window]   FIG. 7. Sensitivity to guanosine 5'-[-thio]diphosphate (GDP-S) of the CB1 receptor–independent component of agonist-induced suppression of PF-EPSCs in nearly mature rats. A: superimposed plots of mean normalized amplitudes of PF-EPSCs recorded from PCs with a conventional K-gluconate internal solution (white squares) or with the same solution with 4 mM GDP-S added (black squares). In both cases, after 30 min dialysis of PCs, 100 µM DHPG was added to the bath at time 0 for 5 min as indicated by horizontal bar. In all experiments, 1 µM SR141716-A was present in the bathing medium throughout the recording period. B: superimposed plots of mean normalized PPF of PF-EPSCs recorded in the same conditions and from the same PCs as in A. White lozenges: conventional K-gluconate internal solution; black lozenges: same solution with 4 mM GDP-S added.

Ryanodine sensitivity of the CB1 receptor–independent component of agonist-induced suppression of PF-EPSCs and of DSE in nearly mature rats

Depolarization-induced potentiation of inhibition (DPI) also operates through Glu release from depolarized PCs. In this case, Glu activates presynaptic NMDA receptors, resulting in a slow build-up and decay (over several minutes) of calcium release from presynaptic ryanodine-sensitive calcium stores (Duguid and Smart 2004). As for DPI, the KA-dependent components of agonist-induced suppression of PF-EPSCs and of DSE at PF-PC synapses might therefore involve such a mechanism in nearly mature rats, as suggested previously (Crepel 2007; but see also Carter et al. 2002). This hypothesis was tested by the following experiments.

In the presence of 1 µM SR141716-A and of 100 µM ryanodine, the CB1 receptor–independent suppression of PF-EPSCs by bath application of 100 µM DHPG was strongly inhibited and replaced by a transient potentiation of PF-EPSCs, 112.84 ± 9.43% on average (n = 6; Fig. 8A1). In these cells, PPF increase associated with this CB1 receptor–independent suppression of PF-EPSCs was also strongly inhibited (Fig. 8A2). Interestingly, ryanodine also markedly reduced basal PPF because its mean value was only 1.09 ± 0.06 (n = 6; Fig. 8A2), thus suggesting that presynaptic ryanodine-sensitive calcium stores also contribute to basal PPF at PF-PC synapses in nearly mature rats.

View larger version (61K): [in this window] [in a new window]   FIG. 8. Sensitivity to ryanodine of agonist-induced suppression of PF-EPSCs and of depolarization-induced suppression of excitation (DSE) in nearly mature rats. A1: superimposed plots of mean normalized amplitudes of PF-EPSCs recorded from PCs over time in the presence of 1 µM SR141716-A alone (white squares) or of 1 µM SR141716-A + 100 µM ryanodine (black squares) before, during, and after 100 µM DHPG was further added to the bath, as indicated by the corresponding horizontal bar. A2: superimposed plots of mean PPF of PF-EPSCs recorded in the same conditions and from the same PCs as in A1. White lozenges: 1 µM SR141716-A alone; black lozenges: 1 µM SR141716-A +100 µM ryanodine. B1: superimposed plots of mean normalized amplitudes of PF-EPSCs recorded from PCs over time in the presence of 1 µM SR141716-A alone (white squares; data taken from Crepel 2007) or of 1 µM SR141716-A + 100 µM ryanodine before during and after a depolarizing pulse from –70 to 0 mV for 1 s (black squares), applied at time 0 (arrow). Plot with gray squares: same as plot with black squares, except that depolarizing pulse duration was 2 s. B2: superimposed plots of mean normalized PPF of PF-EPSCs recorded in the same conditions and from the same PCs as in B1. White lozenges: 1 µM SR141716-A alone; black lozenges: 1 µM SR141716-A +100 µM ryanodine.

Altogether, the marked effects of ryanodine on agonist-induced suppression of PF-EPSCs and on DSE at PF-PC synapses suggests that, as for DPI, their prolonged duration involves long-lasting calcium release from presynaptic ryanodine-sensitive calcium stores, after an initial and more short-lived activation of presynaptic KA receptors. Moreover, the fact that ryanodine also partly inhibited the early phase of the CB1 receptor–independent suppression of PF-EPSCs by DHPG, i.e., the portion of the response that was resistant to bath application of SYM 2081 (cf. Figs. 4A and 8B1), suggests in turn that ryanodine also inhibits the presynaptic GABA_B- and NO-dependent components of agonist-induced suppression of PF-EPSCs.

In these experiments, inhibition of the Glu-dependent component of the agonist-induced suppression of PF-EPSCs by ryanodine might also result from an inhibition of calcium release from postsynaptic ryanodine-sensitive calcium stores (Carter et al. 2002; Isokawa and Alger 2006) that would in turn inhibit retrograde release of Glu from PCs. To compensate, at least partly, for these postsynaptic effects, we used 2-s rather than 1-s depolarizing pulses from –70 to 0 mV in another series of DSE experiments in the presence of ryanodine. Indeed, this duration was likely to give rise to postsynaptic calcium transients of similar amplitude to those observed with 1-s depolarizing pulses in the absence of ryanodine (Fig. 9B1). As shown in Fig. 8B1, the Glu-dependent component of DSE (Crepel 2007) was still significantly inhibited (n = 9; P < 0.01), whereas the early residual DSE was not significantly different from that induced in the same conditions with 1-s depolarizing pulses (Fig. 8B1).

View larger version (44K): [in this window] [in a new window]   FIG. 9. In 18- to 22-day-old rats, ryanodine inhibited cytosolic calcium transients elicited in PFs by local superperfusion of domoate (A) and in PC dendrites by depolarizing voltage steps applied to the soma (B). A1: superimposed mean F/F plots of the initial phase (5 min) of presynaptic calcium signals induced in PFs by focal application of domoate for 1 min (arrow at time 0) in control condition and in the presence of 100 µM ryanodine in the bath (black and white lozenges, respectively). A2: same as in A1 over the entire recording period of 30 min. B1: superimposed mean F/F plots of the initial phase (1.5 s) of presynaptic calcium signals induced in PC dendrites by depolarizing voltage steps of 1 s from –70 to 0 mV (bottom trace) in control condition and in the presence of 100 µM ryanodine in the bath. B2: same as in B1 over the entire recording period of 15 s. Same symbols as in A1 and A2.

Sensitivity to ryanodine of pre- and postsynaptic calcium signaling involved in the CB1 receptor–independent component of DSE in nearly mature rats

PFs loaded with Fluo-4FF-AM were subjected to focal superperfusion of 20 µM domoate for 1 min. This elicited a transient increase in the cytosolic calcium signal in these fibers that peaked shortly after the end of domoate application and relaxed slowly thereafter in <30 min (n = 6; Fig. 9, A1 and A2). Ryanodine (100 µM) present in the bath for 30 min markedly reduced the duration of domoate-induced calcium transients, whereas the peak calcium transient was only marginally affected (n = 5; Fig. 9, A1 and A2). Indeed, peak F/F in control conditions and in the presence of ryanodine were 3.58 ± 0.10 and 3.42 ± 0.94%, respectively; these values are not significantly different. To further quantify differences in calcium transients recorded in control conditions and in the presence of ryanodine, we determined for each cell = (F/F for each point of F/F plots) between the beginning of domoate application and 30 min thereafter. We averaged these values for all cells recorded in a given condition (see Crepel 2007). Mean was significantly lower in the presence of ryanodine, because it amounted to only 0.61 ± 0.14 compared with 1.52 ± 0.30 in its absence (P < 0.05). These results strongly suggest that prolonged calcium release from presynaptic ryanodine-sensitive stores occurs in PFs after brief activation of presynaptic KA receptors.

In control bathing medium, calcium signals induced in PC dendrites by the above described DSE protocol (depolarizing voltage steps from –70 to 0 mV for 1 s) peaked to F/F = 14.10 ± 2.85% at the end of the depolarizing steps (Fig. 9B1) and rapidly decayed over the course of 10–15 s (n = 6; Fig. 9B2). Superfusion of the slices for 30 min with 100 µM ryanodine significantly (P < 0.05) inhibited these depolarization-induced calcium transients because mean F/F was now only 7.77 ± 1.63% (n = 6; Fig. 9, B1 and B2). Differences between control and ryanodine results were further quantified by determining again, for each cell, mean values calculated from the start of the depolarizing step until 4 s after, i.e., the period of time during which F/F plots in control conditions and in the presence of ryanodine appeared clearly different (Fig. 9B2). The mean value was significantly lower (P < 0.05) with ryanodine (4.56 ± 1.17) than in its absence (8.52 ± 1.27). Here again and in keeping with previous results by Carter et al. (2002), this suggests that calcium release from intracellular ryanodine-sensitive stores participates in calcium transients induced in PC dendrites by short depolarizing voltage steps identical as those used to induce DSE (Crepel 2007). However, these calcium transients were much shorter than those induced in PFs by domoate application, and moreover, they were also much shorter than the duration of the KA-sensitive component of DSE in nearly mature rats (Crepel 2007). Therefore electrophysiological and fluorometric experiments with ryanodine clearly suggest that prolonged calcium release from presynaptic ryanodine-sensitive stores is responsible for the prolonged duration of DSE and of agonist-induced suppression of PF-EPSCs in nearly mature rats.

Origin of the lack of the Glu-dependent component of agonist-induced suppression of PF-EPSCs in immature rats

Depolarization-induced potentiation of inhibition that operates through Glu release from depolarized PCs is strictly dependent on the activity of surrounding Glu transporters (Duguid and Smart 2004). Therefore the lack of a Glu-dependent component of agonist-induced suppression of PF-EPSCs in juvenile animals might simply result from an unbalance between Glu release and Glu uptake in favor of the latter at early developmental stages. In 10- to 12-day-old rats and in the presence of 1 µM SR141716-A, bath application of the Glu uptake inhibitor D-TBOA (100 µM) did not unmask any sizeable Glu-dependent component of agonist-induced depression of PF-EPSCs (n = 5; Fig. 10A). However, the transient potentiation elicited by DHPG was significantly (P < 0.001) inhibited compared with that observed in the presence of SR141716-A alone (Fig. 10A), suggesting that it was counterbalanced by a DHPG-induced suppression of PF-EPSCs of similar amplitude and time-course. Thus and as shown for DSE (Crepel 2007), the absence of a Glu-dependent component of agonist-induced depression of PF-EPSCs in juvenile rats may be partly explained by the activity of surrounding Glu transporters. However, this absence might also be caused by other factors such as incomplete maturation of presynaptic KA receptors or of presynaptic calcium-induced calcium release, because D-TBOA failed to reveal any fully developed KA-dependent component of agonist-induced suppression of PF-EPSCs in these animals. Therefore fluorometric experiments were also performed to test these hypotheses.

View larger version (54K): [in this window] [in a new window]   FIG. 10. Origin of the lack of kainate (KA)-dependent component of agonist-induced suppression of PF-EPSCs in immature rats. A: plot of mean normalized amplitudes of PF-EPSCs recorded from PCs over time in 10- to 12-day-old rats in the presence of 1 µM SR141716-A (white squares) and in the presence of 1 µM SR141716-A + 100 µM D-TBOA in the bath (gray squares). DHPG (100 µM) was added to the bath for 5 min, as indicated by the corresponding horizontal bar. B1–B3: presynaptic calcium signals induced in PFs by focal application of domoate for 1 min at time 0 (horizontal bar) in 12-day-old rats. B1: superimposed mean F/F plots in the presence of 1 µM SR141716-A alone (white lozenges), in the presence of 1 µM SR141716-A + 100 µM D-TBOA (gray lozenges), and when 100 µM ryanodine was also present in the bath (black lozenges). B2: same as in B1 when ryanodine was replaced by 50 µM D-APV in the bath (black circles). B3: superimposed mean F/F plots in the presence of 1 µM SR141716-A alone (white lozenges), in the presence of 1 µM SR141716-A + 100 µM D-TBOA + 50 µM D-APV (black circles), and when 1 µM TTX was also present in the bath (gray circles).

In contrast, when domoate application was performed in the presence of 100 µM D-TBOA, cytosolic calcium signals were significantly (P < 0.001) larger (F/F = 4.74 ± 0.53%) and now decayed over 15 min (n = 5; Fig. 10B1). Moreover, this decay was markedly shortened when 100 µM ryanodine was also present in the bath (n = 5; Fig. 10B1), so that the mean value was significantly lower (P < 0.05) in the presence of ryanodine (0.96 ± 0.30) than in its absence (2.65 ± 0.47). These results therefore suggest that at least some calcium-induced calcium release may be triggered in immature PFs.

However, the prominent effect of D-TBOA on cytosolic calcium signals was puzzling. Indeed, and to the best of our knowledge, significant uptake of this agonist is unlikely. However, domoate is known to induce release of excitatory amino acids from cultured cerebellar granule cells through reversal of the Glu transporter (Berman and Murray 1997). Therefore the possibility exists that, in these experiments, domoate not only acts through activation of presynaptic KA receptors, but also activates other presynaptic Glu receptors after release of Glu from the superfused molecular layer. We therefore tested this possibility by studying the effect of bath-applied D-APV (50 µM) on calcium transients elicited in PF by focal superfusion of the slices for 1 min with 20 µM domoate in the presence of D-TBOA. We performed these experiments because NMDA receptors are characterized by their high affinity for Glu and are probably present and functional on PFs (Casado et al. 2000). However, this conclusion has been challenged recently (Shin and Linden 2005) and, to date, it still has not been established by electron microscopic analysis that NMDA receptors exist on PFs. As shown in Fig. 10B2, D-APV shortened calcium transients induced by domoate in the presence of D-TBOA (n = 6) to nearly the same extent as that observed with ryanodine because mean values were not significantly different, i.e., 0.96 ± 0.18 and 0.96 ± 0.30, respectively. This latter result indicates that activation of presynaptic NMDA receptors probably contributes to calcium transients induced by domoate in the presence of Glu uptake blockers, thus corroborating the aforementioned hypothesis, and also supports the previous finding that NMDA receptors are present on PFs (Casado et al. 2000). Furthermore, these results also strongly suggest that activation of NMDA receptors is necessary to trigger calcium-induced calcium release in immature PFs. In contrast, D-APV did not significantly affect calcium transients induced by domoate in control bathing medium in 20- to 22-day-old rats (n = 4; data not shown), suggesting that presynaptic NMDA receptors are not recruited in this case.

In 12-day-old rats, the mean values of calcium transients elicited in the presence of D-TBOA and D-APV or in the presence of D-TBOA and ryanodine were still significantly (P < 0.05) higher (0.96 ± 0.18 and 0.96 ± 0.30, respectively) than in control medium (0.10 ± 0.02; Fig. 10B2). It is therefore likely that an additional mechanism participates to calcium transients induced by domoate in PFs in the presence of D-TBOA. One hypothesis is that Glu uptake blockers lead to a progressive accumulation of Glu within slices that tonically activates granule cells. Such activity might therefore increase basal calcium concentration within PFs and facilitate induction of calcium-induced calcium release on activation of presynaptic KA receptors. If this were so, prolonged calcium transients induced by domoate in the presence of D-TBOA and D-APV should be no longer observed in the presence of TTX. However, in six experiments performed in the presence of 1 µM TTX, calcium transients induced by focal application of 20 µM domoate in the presence of both 100 µM D-TBOA and 50 µM D-APV was not significantly altered compared with that induced when TTX was omitted (Fig. 10B3). At the moment, we have no other plausible explanation to explain that calcium transients elicited in the presence of D-TBOA and D-APV or in the presence of D-TBOA and ryanodine were still significantly larger and of longer duration than in control medium.

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
In keeping with the study by Levenes et al. (2001) and Crepel (2007), these data strongly suggest that, in nearly mature rats, retrograde release of Glu is involved in the suppression of synaptic transmission at PF-PC synapses after activation of postsynaptic mGluR1. Moreover, results with GDP-S are consistent with the view that Glu mainly originates from the recorded PC themselves, with only minor (if any) contribution of spillover arising from neighboring cells. As for the KA-dependent component of DSE at PF-PC synapses in nearly mature rats (Crepel 2007), Glu is likely to operate through activation of presynaptic KA receptors located on PFs and prolonged calcium release from presynaptic ryanodine-sensitive calcium stores. Finally, results in nearly mature wild-type and GluR6^–/– mice further suggest that, as for DSE (Crepel 2007), presynaptic KA receptors involved in agonist-induced suppression of PF-EPSCs include GluR6 receptor subunits.

These results also show that retrograde release of endocannabinoids is another major component in agonist-induced suppression of excitation at these nearly mature PF-PC synapses. In contrast, in juvenile rats and mice, suppression of synaptic transmission at PF-PC synapses by activation of postsynaptic mGluR1 is entirely mediated through retrograde release of endocannabinoids by PCs and activation of presynaptic CB1 receptors. Our results are in agreement with those previously obtained in juvenile animals for agonist-induced suppression of PF-EPSCs (Maejima et al. 2001) and for DSE (Brenowitz and Regehr 2003; Crepel 2007; Kreitzer and Regehr 2001b).

Suppression of PF-EPSCs by DHPG in juvenile rats and mice depends entirely on retrograde release of endocannabinoids

In juvenile rats and mice, DHPG-induced suppression of PF-EPSCs exclusively involved retrograde release of endocannabinoids by PCs and activation of presynaptic CB1 receptors. This is in complete agreement with previously published results in juvenile mice by Maejima et al. (2001). The lack of any detectable Glu-, GABA_B-, or NO-dependent components of suppression of PF-EPSCs by DHPG in juvenile animals suggests in turn that these pathways are likely to mature later than endocannabinoid signaling. However, the present results (see above) also strongly suggest that the absence of endocannabinoid-independent suppression of PF-EPSCs by DHPG in juvenile rats is partly caused by an unbalance between Glu release and Glu uptake in favor of the latter at early developmental stages, coupled with either immature presynaptic KA receptors and/or not fully developed calcium release from ryanodine sensitive internal stores.

In this study and in that of Maejima et al. (2001) in juvenile animals, a transient potentiating effect of DHPG on PF-EPSC was observed, in particular in the presence of CB1 receptor antagonists. This was also true in nearly mature synapses when all pathways of the DHPG suppressive effect on PF-EPSCs were blocked (see RESULTS). This potentiation might be of postsynaptic origin because it was not accompanied by any significant change in PPF, in keeping with the potentiating effect of mGluR1 agonists on PC responsiveness to ionophoretic application of Glu in their dendritic field (Levenes et al. 2001). Further studies will have to unravel underlying mechanisms of this transient potentiation of PF-EPSCs.

Therefore sequential development of mechanisms underlying short-term plasticity at PF-PC synapses may at least partly explain apparent contradictions between data by Levenes et al. (2001) and by Maejima et al. (2001).

DHPG suppression of PF-EPSCs in nearly mature rats depends partly on retrograde release of glutamate and activation of presynaptic KA receptors

In nearly mature rats, suppression of PF-EPSCs by DHPG was only partly blocked by CB1 receptor antagonists. Moreover, the remaining component was potentiated by Glu uptake inhibitors and markedly inhibited by the desensitizing KA receptor agonist SYM 2081 and by nonsaturating concentrations of the competitive AMPA/KA receptor antagonist CNQX (Honoré et al. 1988). However, this inhibitory effect was not reproduced by nonsaturating concentrations of the more selective AMPA receptor antagonist GYKI (Bureau et al. 1999; Renard et al. 1995; Wilding and Huettner 1995). These results strongly suggest that the depressant effect of DHPG on PF-EPSCs not only involves retrograde release of endocannabinoids as in immature rats and mice, but also involves retrograde release of Glu by PCs and subsequent activation of KA receptors. Moreover, and as mentioned above, results in wild-type and GluR6^–/– mice further suggest that these KA receptors include GluR6 subunits. Finally, the nearly complete blockade of endocannabinoid-independent suppression of PF-EPSCs with GDP-S makes unlikely that Glu responsible for this suppression might be spilling over in significant concentrations from neighboring cellular elements such as glial cells (Angulo et al. 2004; Parpura et al. 1994).

The fact that the endocannabinoid-independent suppression of PF-EPSCs by mGluR1 agonists was still accompanied by an increase in PPF that itself was potentiated by Glu uptake inhibitors and markedly inhibited by SYM 2081 as well as by nonsaturating concentrations of CNQX further suggests that, like for DSE in nearly mature rats (Crepel 2007), KA receptors involved in agonist-induced suppression of PF-EPSCs are located on PFs. In contrast, the absence of significant modification of PF-EPSC kinetics during the endocannabinoid-independent suppression of PF-EPSCs (see RESULTS) suggests that released Glu does not reach postsynaptic AMPA receptors in significant concentrations, probably because of their lower sensitivity to glutamate than KA receptors (Lerma et al. 2001) and retarded transmitter diffusion around synaptic spines (Barbour et al. 1994).

Finally, DHPG suppression of PF-EPSCs in nearly mature rats is also likely to involve activation of presynaptic GABA_B receptors and release of NO from molecular layer inhibitory interneurons (see RESULTS). Such a complexity of effects pertaining to DHPG may well have led to oversight of the CB1 receptor–dependent component of the PF-EPSC suppression induced by mGluR1 agonists in the previous study by Levenes et al. (2001).

Effect of nonsaturating concentrations of CNQX and of GYKI on PPF and CV

In a previous study, Delaney and Jahr (2002) showed that weak activation of presynaptic KA receptors borne by PFs up-regulate Glu release by these fibers. This fits well with these results showing that nonsaturating concentrations of CNQX increase basal PPF and CV. Indeed, if one assumes that the basal extracellular Glu concentration is sufficient to induce weak tonic activation of presynaptic KA receptors that in turn increase release probability at PF-PC synapses, reversal of such an effect by CNQX will effectively lead to the observed increases in basal PPF and CV. Similarly, Delaney and Jahr (2002) showed that larger activation of presynaptic KA receptors borne by PFs down-regulates Glu release by these fibers. As such, the larger Glu concentration achieved through retrograde Glu release during DSE or during agonist-induced suppression of PF-EPSCs is now likely to down-regulate Glu release at these synapses, perhaps through depletion of the readily releasable pool of synaptic vesicles (Crepel 2007; Levenes et al. 2001), explaining the associated PPF increase.

In this study, the increase in basal PPF induced by nonsaturating concentrations of GYKI raises a question, in particular because it was not accompanied by a corresponding increase in mean CV, at least when considering the first EPSCs within PF-EPSC pairs in PPF experiments. Accordingly, and even though this PPF increase was only one half of that induced by nonsaturating concentrations of CNQX causing similar reduction in PF-EPSC amplitude (see RESULTS), it remains difficult to interpret these results solely on the basis of a partial antagonist effect of GYKI on presynaptic KA receptors that contain GluR6 subunits (Bureau et al. 1999). Indeed, this PPF increase was observed in all tested cells, whereas the strong inhibition by GYKI of the effect of DHPG on PF-EPSCs only concerned less than a third of them (see RESULTS). Among other possible explanations for such a puzzling effect of GYKI, one can point to potassium efflux that occurs through postsynaptic AMPA receptors during PF-EPSCs. In PPF experiments, if one assumes that clearance of potassium released during pairs of EPSCs is incomplete when the next pair is elicited (PF stimulations at 0.33 Hz), such a residual increase in extracellular potassium concentration might contribute to steady depolarization of active zones, thereby leading to a decrease of basal PPF. Reversal of this effect by strong blockade of postsynaptic AMPA receptors by GYKI would therefore reduce release probability of Glu during EPSCs and thus explain the observed 16% basal PPF increase in these experiments. In keeping with this hypothesis, we nearly always noticed a progressive increase in PF-EPSC amplitude and a correlative progressive decrease of associated PPF during the first minutes of PF stimulations, before reaching a quasi steady state (see Fig. 5A). Along the same line, increasing the frequency of PF stimulation from 0.1 to 0.33 Hz decreased within a few minutes basal PPF from 1.94 ± 0.18 to 1.42 ± 0.20 in the four tested cells, and this effect was accompanied by a correlative increase in amplitude of PF-EPSCs. Noteworthy, CV seemed more reliable than PPF in establishing that GYKI suppresses PF-EPSCs by selectively blocking postsynaptic AMPA receptors in these experiments because this depressant effect was not accompanied by any significant correlative increase in CV (see RESULTS). Following this interpretation, only about one half of the basal PPF increase accompanying inhibition of PF-EPSCs by nonsaturating concentrations of CNQX would be genuinely caused by its blocking effect on presynaptic KA receptors, in agreement with the correlative increase in CV.

Finally, it must be emphasized that the ratio r = mean basal PPF increase/mean EPSC amplitude decrease during superfusion of the slices with GYKI was only 0.29 compared with 0.97 and 0.76, respectively, for the ratio R = mean PPF increase/mean EPSC amplitude decrease during DHPG application in the presence of SR141716-A + GYKI and in the presence of SR141716-A alone (see RESULTS). Therefore the mechanism proposed above to explain the unexpected effects on PPF of blocking postsynaptic AMPA receptors by GYKI can hardly explain the bulk of the PPF increase accompanying suppression of PF-EPSCs by mGluR1 agonists. The latter remains therefore most likely caused by activation of presynaptic KA receptors by Glu released by PCs. Concerning the KA-dependent component of DSE, the ratio R was close to 0.5 on average (Crepel 2007), so that it was no longer possible, by simply comparing r and R values, to exclude that a significant contribution of prolonged clearance of potassium after PF-EPSCs participates in the observed PPF increase. However, the KA-dependent component of DSE was also accompanied by a significant increase in CV and, moreover, its prolonged time-course closely matched that of presynaptic ryanodine sensitive calcium transients induced by short domoate application to PFs (see RESULTS). Therefore here again, PPF and CV increases accompanying DSE of PF-EPSCs are most likely primarily caused by activation of presynaptic KA receptors by Glu released by PCs. On the whole, these data are also consistent with earlier reports on dendritic release of Glu from neocortical neurons (Ali et al. 2001; Harkany et al. 2004).

Contribution of pre- and postsynaptic ryanodine-sensitive calcium stores to the KA-dependent components of DSE and to agonist-induced suppression of PF-EPSCs in nearly mature rats

Results of fluorometric experiments and those showing a nearly complete inhibition of the KA-dependent components of DSE and of agonist-induced suppression of PF-EPSCs by ryanodine suggest that their long duration is caused by prolonged calcium release from presynaptic ryanodine-sensitive calcium stores, after initial activation of presynaptic KA receptors by Glu released by PCs. In particular, no prolonged ryanodine-sensitive calcium transient could be observed in PC dendrites after the same depolarizing pulses applied to PC soma as those used to induced DSE, whereas such prolonged ryanodine-sensitive calcium transients were readily elicited within PFs by focal application of low concentrations of the selective KA agonist domoate (see RESULTS). Because of technical constraints (see METHODS), we used focal applications of domoate lasting 60 s instead of only 4 s as in experiments by Duguid and Smart (2004). This may well explain why the duration of presynaptic ryanodine-sensitive calcium transients in these experiments was about twice that of inhibitory synaptic noise elicited in PCs by brief activation of presynaptic NMDA receptors (Duguid and Smart 2004) and also nearly 3 times longer than duration of the KA-dependent component of DSE (see Fig. 8B1) (Crepel 2007).

In addition, these results also suggest that presynaptic ryanodine-sensitive calcium stores participate in the residual calcium increase responsible for basal PPF at PF-PC synapses because ryanodine also markedly reduced basal PPF values (see RESULTS). Finally, one cannot totally exclude that calcium release from postsynaptic ryanodine-sensitive calcium stores (Carter et al. 2002; Isokawa and Alger 2006) also participates to retrograde release of Glu from PCs. However, this contribution is unlikely to be important because, in DSE experiments in the presence of ryanodine, compensation for inhibition of postsynaptic calcium-induced calcium release by increasing postsynaptic stimulation failed to restore normal DSE (see RESULTS).

It must be acknowledged that these results seem to contradict previous observations by Carter et al. (2002), who concluded that presynaptic ryanodine-sensitive calcium stores do not contribute to synaptic transmission at PF-PC synapses in 10- to 22-day-old rats. However, and interestingly enough, these authors suggested that internal calcium stores may well play a role in presynaptic function at later developmental stages as is the case for granule cells and their associated parallel fibers in the avian cerebellum where ryanodine receptors are only prominent in mature animals (Ouyang et al. 1997). Therefore the possibility remains that subtle age differences of animals under study explain the apparent contradiction between the two set of results because, in particular, ages of rats in these experiments correspond to the upper limit of those included in that of Carter et al. (2002).

Finally, the mechanisms described here could take place in physiological conditions and have functional consequences. Indeed, one knows that postsynaptic mGluR1 are activated by high-frequency activity of PFs (Batchelor et al. 1994) and that the prolonged discharge of excitatory quantal events that follows short PF tetanus is sensitive to the mGluR1 antagonist AIDA and to the KA receptor blocker SYM 2081 (Crepel 2007; Levenes et al. 2001). Thus if one assumes that this Glu-dependent discharge of excitatory quantal events is sufficient to trigger action potentials along these fibers as previously suggested (Levenes et al. 2001), this would introduce a form of local communication among PCs sharing the same PF input. However, in these experiments, this enhanced discharge was terminated a few seconds after PF tetanus, whereas in this study, the duration of the KA-sensitive component of suppression of PF-EPSCs by DHPG outlasted its washout by several minutes (see RESULTS). Therefore quantal events after short PF tetanus were likely to be caused by the initial activation of presynaptic KA receptors by Glu released by PCs, i.e., the initial phase of the KA-dependent suppression of PF-EPSCs, rather than due to the subsequent prolonged ryanodine-sensitive calcium-induced calcium release seen here. One plausible hypothesis is that the initial activation of presynaptic KA receptors after short PF tetanus was insufficient to trigger such calcium-induced calcium release. Further studies will be required to determine physiological conditions in which fully developed endocannabinoid-independent suppression of PF-EPSCs can be induced.

	ACKNOWLEDGMENTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
We thank A. Marty for advice and comments during these experiments and on the manuscript and C. Mulle for kindly providing GluR6^–/– mice and for helpful discussions. We also thank G. Sadoc for the Acquis1 software.

	FOOTNOTES

 
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

¹ The online version of this article contains supplemental data.

Address for reprint requests and other correspondence: F. Crepel, Pharmacologie de la Synapse, IBBMC, Bât. 430, Univ. Paris-Sud, 91405 Orsay Cedex, France (E-mail: francis.crepel{at}u-psud.fr)

	REFERENCES

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION ACKNOWLEDGMENTS REFERENCES

 
Ali AB, Rossier J, Staiger JF, Audinat E. Kainate receptors regulate unitary IPSCs elicited in pyramidal cells by fast-spiking interneurons in the neocortex. J Neurosci 21: 2992–2999, 2001.[Abstract/Free Full Text]

Angulo MC, Kozlov AS, Charpak S, Audinat E. Glutamate released from glial cells synchronizes neuronal activity in the hippocampus. J Neurosci 24: 6920–6927, 2004.[Abstract/Free Full Text]

Atluri P, Regehr WG. Determinants of the time course of facilitation at the Granule cell to Purkinje cell synapse. J Neurosci 16: 5661–5671, 1996.[Abstract/Free Full Text]

Barbour B, Keller BU, Llano I, Marty A. Prolonged presence of glutamate during excitatory synaptic transmission to cerebellar Purkinje cells. Neuron 12: 1331–1343, 1994.[CrossRef][Web of Science][Medline]

Batchelor AM, Madge DJ, Garthwaite J. Synaptic activation of metabotropic glutamate receptors in the parallel fibre-Purkinje cell pathway in rat cerebellar slices. Neurosci Lett 63: 911–915, 1994.

Baude A, Nusser Z, Roberts JDB, Mulvihill E, McIlhinney RAJ, Somogyi P. The metabotropic glutamate receptor (mGluR1a) is concentrated at perisynaptic membrane of neuronal subpopulations as detected by immunogold reaction. Neuron 11: 771–787, 1993.[CrossRef][Web of Science][Medline]

Berman FW, Murray TF. Domoic acid neurotoxicity in cultured cerebellar granule neurons is mediated predominantly by NMDA receptors that are activated as a consequence of excitatory amino acid release. J Neurochem 69: 693–703, 1997.[Web of Science][Medline]

Blond O, Daniel H, Jaillard D, Crepel F. Pre and postsynaptic effects of nitric oxide at synapses between parallel fibers and Purkinje cells; involvement in cerebellar long-term suppression. Neuroscience 77: 945–954, 1997.[CrossRef][Web of Science][Medline]

Bredt S, Hwang PM, Snyder SH. Localization of nitric oxide synthase indicating a neural role for nitric oxide. Nature 347: 768–770, 1990.[CrossRef][Medline]

Brenowitz SD, Regehr WG. Calcium dependence of retrograde inhibition by endocannabinoids at synapses onto Purkinje cells. J Neurosci 23: 6373–6384, 2003.[Abstract/Free Full Text]

Brown SP, Brenowitz SD, Regehr WG. Brief pre-synaptic bursts evoke synapse-specific retrograde inhibition mediated by endogenous cannabinoids. Nat Neurosci 6: 1048–1057, 2003.[CrossRef][Web of Science][Medline]

Bureau I, Bischoff S, Heinemann SF, Mulle C. Kainate receptor-mediated responses in the CA1 field of wild-type and GluR6-deficient mice. J Neurosci 19: 653–663, 1999.[Abstract/Free Full Text]

Canepari M, Auger C, Ogden D. Calcium ion permeability and single-channel properties of the metabotropic slow EPSC of rat Purkinje neurons. J Neurosci 24: 3563–3573, 2004.[Abstract/Free Full Text]

Carter AG, Vogt KE, Foster KA, Regehr WG. Assessing the role of calcium-induced calcium release in short-term presynaptic plasticity at excitatory central synapses. J Neurosci 22: 21–28, 2002.[Abstract/Free Full Text]

Casado M, Dieudonne S, Ascher P. Pre-synaptic N-methyl-D-aspartate receptors at the parallel fiber-purkinje cell synapse. Proc Natl Acad Sci USA 97: 11593–11597, 2000.[Abstract/Free Full Text]

Cho K, Francis JC, Hirbec H, Dev K, Brown MW, Henley JM, Bashir ZI. Regulation of kainate receptors by protein kinase C and metabotropic glutamate receptors. J Physiol 548: 723–730, 2003.[Abstract/Free Full Text]

Cossart R, Epsztein J, Tyzio R, Becq H, Hirsch J, Ben-Ari Y, Crepel V. Quantal release of glutamate generates pure kainate and mixed AMPA/kainate EPSCs in hippocampal neurons. Neuron 35: 147–159, 2002.[CrossRef][Web of Science][Medline]

Crepel F. Developmental changes in retrograde messengers involved in depolarization-induced suppression of excitation at parallel fiber-Purkinje cell synapses in rodents. J Neurophysiol 97: 824–836, 2007.[Abstract/Free Full Text]

Delaney AJ, Jahr CE. Kainate receptors differentially regulate release at two parallel fiber synapses. Neuron 36: 476–482, 2002.

Devane W, Dysarz F, Johnson M, Melvin L, Howlett A. Determination and characterisation of a cannabinoid receptor in rat brain. Mol Pharmacol 34: 605–613, 1988.[Abstract]

DeVries SH. Bipolar cells use kainate and AMPA receptors to filter visual information into separate channels. Neuron 28: 847–856, 2000.[CrossRef][Web of Science][Medline]

Diana MA, Levenes C, Mackie K, Marty A. Short-term retrograde inhibition of GABAergic synaptic currents in rat Purkinje cells is mediated by endogenous cannabinoids. J Neurosci 22: 200–208, 2002.[Abstract/Free Full Text]

Dittman JS, Regehr WG. Mechanism and kinetics of heterosynaptic depression at a cerebellar synapse. J Neurosci 17: 9048–9059, 1997.[Abstract/Free Full Text]

Duguid IC, Smart TG. Retrograde activation of pre-synaptic NMDA receptors enhances GABA release at cerebellar interneuron-Purkinje cell synapses. Nat Neurosci 7: 525–533, 2004.[CrossRef][Web of Science][Medline]

Epsztein J, Represa A, Jorquera I, Ben-Ari Y, Crepel V. Recurrent mossy fibers establish aberrant kainate receptor-operated synapses on granule cells from epileptic rats. J Neurosci 25: 8229–8239, 2005.[Abstract/Free Full Text]

Galante M, Diana MA. Group I metabotropic glutamate receptors inhibit GABA release at interneuron-Purkinje cell synapses through endocannabinoid production. J Neurosci 24: 4865–4874, 2004.[Abstract/Free Full Text]

Gatley SJ, Gifford AN, Volkow ND, Lan R, Makriyannis A. 123I-labeled AM251: a radioiodinated ligand which binds in vivo to mouse brain cannabinoid CB1 receptors. Eur J Pharmacol 307: 331–338, 1996.[CrossRef][Web of Science][Medline]

Glitsch M, Llano I, Marty A. Glutamate as a candidate retrograde messenger at interneurone-Purkinje cell synapses of rat cerebellum. J Physiol 497: 531–537, 1996.[Abstract/Free Full Text]

Glitsch M, Parra P, Llano I. The retrograde inhibition of IPSCs in rat cerebellar purkinje cells is highly sensitive to intracellular calcium. Eur J Neurosci 12: 987–993, 2000.[CrossRef][Web of Science][Medline]

Harkany T, Holmgren C, Hartig W, Qureshi T, Chaudhry FA, Storm-Mathisen J, Dobszay MB, Berghuis P, Schulte G, Sousa KM, Fremeau RTJ, Edwards RH, Mackie K, Ernfors P, Zilberter Y. Endocannabinoid-independent retrograde signaling at inhibitory synapses in layer 2/3 of neocortex: involvement of vesicular glutamate transporter 3. J Neurosci 24: 4978–4988, 2004.[Abstract/Free Full Text]

Honoré T, Davies SN, Drejer J, Fletcher EJ, Jacobsen P, Lodge D, Nielsen FE. Quinoxalinediones: potent competitive non-NMDA glutamate receptor antagonists. Science 241: 701–703, 1988.[Abstract/Free Full Text]

Isokawa M, Alger BE. Ryanodine receptor regulates endogenous cannabinoid mobilization in the hippocampus. J Neurophysiol 96: 3001–3011, 2006.

Karakossian MH, Otis TS. Excitation of cerebellar interneurons by group I metabotropic glutamate receptors. J Neurophysiol 92: 1558–1565, 2004.[Abstract/Free Full Text]

Knowles RG, Palacios M, Palmer RMJ, Moncada S. Formation of nitric oxide from L-arginine in the central nervous system: a transduction mechanism for stimulation of the soluble guanylate cyclase. Proc Natl Acad Sci USA 86: 5159–5162, 1989.[Abstract/Free Full Text]

Kim J, Alger BE. Random response fluctuations lead to spurious paired-pulse facilitation. J Neurosci 21: 9608–9618, 2001.[Abstract/Free Full Text]

Kreitzer AC, Regehr WG. Cerebellar depolarization-induced suppression of inhibition is mediated by endogenous cannabinoids. J Neurosci 21: 1–5, 2001a.[Web of Science][Medline]

Kreitzer AC, Regehr WG. Retrograde inhibition of pre-synaptic calcium influx by endogenous cannabinoids at excitatory synapses onto Purkinje cells. Neuron 29: 717–727, 2001b.[CrossRef][Web of Science][Medline]

Kullmann DM. Amplitude fluctuations of dual-component EPSCs in hippocampal pyramidal cells : implications for long-term potentiation. Neuron 12: 1111–1120, 1994.[CrossRef][Web of Science][Medline]

Lerma J, Paternain AV, Naranja JR, Mellstrom B. Functional kainate selective glutamate receptors in cultured hippocampal neurons. Proc Natl Acad Sci USA 90: 11688–11692, 1993.[Abstract/Free Full Text]

Lerma J, Paternain AV, Rodriguez-Moreno A, Lopez-Garcia JC. Molecular physiology of kainate receptors. Physiol Rev 81: 971–998, 2001.[Abstract/Free Full Text]

Levenes C, Daniel H, Crepel F. Retrograde modulation of transmitter release by post-synaptic subtype 1 metabotropic glutamate receptors in the rat cerebellum. J Physiol 537: 125–140, 2001.[Abstract/Free Full Text]

Levenes C, Daniel H, Soubrie P, Crepel F. Cannabinoids decrease excitatory synaptic transmission and impair long- term depression in rat cerebellar Purkinje cells. J Physiol 510: 867–879, 1998.[Abstract/Free Full Text]

Li P, Wilding TJ, Kim SJ, Calejesan AA, Huettner JE, Zhuo M. Kainate-receptor-mediated sensory synaptic transmission in mammalian spinal cord. Nature 397: 161–164, 1999.[CrossRef][Medline]

Llano I, Leresche N, Marty A. Calcium entry increases the sensitivity of cerebellar Purkinje cells to applied GABA and decreases inhibitory synaptic currents. Neuron 6: 565–574, 1991a.[CrossRef][Web of Science][Medline]

Llano I, Marty A, Armstrong C, Konnerth A. Synaptic- and agonist-induced excitatory currents of Purkinje cells in rat cerebellar slices. J Physiol 434: 183–213, 1991b.[Abstract/Free Full Text]

Maejima T, Hashimoto K, Yoshida T, Aiba A, Kano M. Pre-synaptic inhibition caused by retrograde signal from metabotropic glutamate to cannabinoid receptors. Neuron 31: 463–476, 2001.[CrossRef][Web of Science][Medline]

Marcaggi P, Attwell D. Endocannabinoid signaling depends on the spatial pattern of synapse activation. Nat Neurosci 8: 776–781, 2005.[CrossRef][Web of Science][Medline]

Martin AR. Quantal nature of synaptic transmission. Physiol Rev 46: 51–66, 1966.[Web of Science]

McNaughton B. Long-term synaptic enhancement and short-term potentiation in rat fascia dentata act through different mechanisms. J Physiol 324: 249–262, 1982.[Abstract/Free Full Text]

Neale SE, Garthwaite J, Batchelor AM. mGlu1 receptors mediate a post-tetanic suppression at parallel fibre-Purkinje cell synapses in rat cerebellum. Eur J Neurosci 14: 1313–1319, 2001.[CrossRef][Web of Science][Medline]

Ohno-Shosaku T, Maejima T, Kano M. Endogenous cannabinoids mediate retrograde signals from depolarized post-synaptic neurons to pre-synaptic terminals. Neuron 29: 729–738, 2001.[CrossRef][Web of Science][Medline]

Ouyang Y, Martone ME, Deerinck TJ, Airey JA, Sutko JL, Ellisman MH. Differential distribution and subcellular localization of ryanodine receptor isoforms in the chicken cerebellum during development. Brain Res 775: 52–62, 1997.[CrossRef][Web of Science][Medline]

Parpura V, Basarsky TA, Liu F, Jeftinija K, Jeftinija S, Haydon PG. Glutamate-mediated astrocyte-neuron signalling. Nature 369: 744–747, 1994.[CrossRef][Medline]

Petitet F, Marin L, Doble A. Biochemical and pharmacological characterization of cannabinoid binding sites using [3H]SR141716A. Neuroreport 7: 789–792, 1996.[Web of Science][Medline]

Petralia RS, Wang YX, Wenthold RJ. Histological and ultrastructural localisation of the kainate receptor subunits KA2 and GluR6/7, in the rat nervous system using selective antipeptide antibodies. J Comp Neurol 349: 85–110, 1994.[CrossRef][Web of Science][Medline]

Pin JP, Duvoisin R. The metabotropic glutamate receptors: structure and functions. Neuropharmacology 34: 1–26, 1995.[CrossRef][Web of Science][Medline]

Pitler TA, Alger BE. Post-synaptic spike firing reduces synaptic GABAA responses in hippocampal pyramidal cells. J Neurosci 12: 4122–4132, 1992.[Abstract]

Pitler TA, Alger BE. Depolarization-induced suppression of GABAergic inhibition in rat hippocampal pyramidal cells: G protein involvement in a pre-synaptic mechanism. Neuron 13: 1447–1455, 1994.[CrossRef][Web of Science][Medline]

Renard A, Crepel F, Audinat E. Evidence for two types of non-NMDA receptors in rat cerebellar Purkinje cells maintained in slice cultures. Neuropharmacology 34: 335–346, 1995.[CrossRef][Web of Science][Medline]

Rinaldi-Carmona M, Barth F, Héaulme M, Shire D, Calandra B, Congy C, Martinez S, Maruani J, Néliat G, Caput D, Ferrara P, Soubrié P, Brelière J, Le Fur G. SR 141716A, a potent and selective antagonist of the brain cannabinoid receptor. FEBS Lett 350: 240–244, 1994.[CrossRef][Web of Science][Medline]

Ruiz A, Sachidhanandam S, Utvik JK, Coussen F, Mulle C. Distinct subunits in heteromeric kainate receptors mediate ionotropic and metabotropic function at hippocampal mossy fiber synapses. J Neurosci 25: 11710–11718, 2005.[Abstract/Free Full Text]

Safo PK, Regehr WG. Endocannabinoids control the induction of cerebellar LTD. Neuron 48: 647–659, 2005.[CrossRef][Web of Science][Medline]

Schoepp DD, Goldsworthy J, Johnson BG, Salhoff CR, Baker SR. 3,5-dihydroxyphenylglycine is a highly selective agonist for phosphoinositide-linked metabotropic glutamate receptors in the rat hippocampus. J Neurochem 63: 769–772, 1994.[Web of Science][Medline]

Schultz P, Cook E, Johnston D. Changes in paired-pulse facilitation suggest pre-synaptic involvement in long-term potentiation. J Neurosci 14: 5325–5337, 1994.[Abstract]

Shin JH, Linden DJ. An NMDA receptor/nitric oxide cascade is involved in cerebellar LTD but is not localized to the parallel fiber terminal. J Neurophysiol 94: 4281–4289, 2005.[Abstract/Free Full Text]

Vincent P, Armstrong CM, Marty A. Inhibitory synaptic currents in rat cerebellar Purkinje cells: modulation by post-synaptic depolarization. J Physiol 456: 453–471, 1992.[Abstract/Free Full Text]

Vincent SR, Kimura H. Histochemical mapping of nitric oxide synthase in the rat brain. Neuroscience 46: 755–784, 1992.[CrossRef][Web of Science][Medline]

Wadiche JI, Jahr CE. Patterned expression of Purkinje cell glutamate transporters control synaptic plasticity. Nat Neurosci 8: 1329–1334, 2005.[CrossRef][Web of Science][Medline]

Wang J, Zucker RS. Photolysis-induced suppression of inhibition in rat hippocampal CA1 pyramidal neurons. J Physiol 533: 757–763, 2001.[Abstract/Free Full Text]

Wilding TJ, Huettner JE. Differential antagonism of alpha-amino-3hydroxy-5-methyl-4-isoxazolepropionic acid-preferring and kainate-preferring receptors by 2,3-benzodiazepines. Mol Pharmacol 47: 582–587, 1995.[Abstract]

Wilson RI, Kunos G, Nicoll RA. Pre-synaptic specificity of endocannabinoid signaling in the hippocampus. Neuron 31: 453–462, 2001.[CrossRef][Web of Science][Medline]

Wilson RI, Nicoll RA. Endogenous cannabinoids mediate retrograde signalling at hippocampal synapses. Nature 410: 588–592, 2001.[CrossRef][Medline]

Zhou LM, Gu ZQ, Costa AM, Yamada KA, Mansson PE, Giordano T, Skolnick P, Jones KA. (2S,4R)-4-methylglutamic acid (SYM 2081): a selective, high-affinity ligand for kainate receptors. Pharmacology 280: 422–427, 1997.

This article has been cited by other articles:

				Y. S. Kim, J. H. Shin, F. S. Hall, and D. J. Linden Dopamine Signaling Is Required for Depolarization-Induced Slow Current in Cerebellar Purkinje Cells J. Neurosci., July 1, 2009; 29(26): 8530 - 8538. [Abstract] [Full Text] [PDF]

				M. Kano, T. Ohno-Shosaku, Y. Hashimotodani, M. Uchigashima, and M. Watanabe Endocannabinoid-Mediated Control of Synaptic Transmission Physiol Rev, January 1, 2009; 89(1): 309 - 380. [Abstract] [Full Text] [PDF]

This Article Abstract Full Text (PDF) Supplemental Results and Figures All Versions of this Article: 98/5/2550    most recent 00376.2007v1 Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Crepel, F. Articles by Daniel, H. Search for Related Content PubMed PubMed Citation Articles by Crepel, F. Articles by Daniel, H.