| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
NEUROSCIENCE |
1 Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, Albert-Ludwigs-Universität, D-79104 Freiburg im Breisgau Germany
2 Endocannabinoid Research Group, Istituto di Chimica Biomolecolare, CNR, I-80078 Pozzuoli, Italy
 |
Abstract |
|---|
 Top Abstract IntroductionMethodsResultsDiscussionReferences |
|---|
(Received 16 August 2006;
accepted after revision 5 September 2006;
first published online 14 September 2006)
Corresponding author B. Szabo: Institut für Experimentelle und Klinische Pharmakologie und Toxikologie, Albert-Ludwigs-Universität; Albertstrasse 25, D-79104 Freiburg im Breisgau, Germany. Email: szabo{at}pharmakol.uni-freiburg.de
|
Introduction |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
i/o protein-coupled cannabinoid CB1 receptor is probably the most abundant G protein-coupled receptor in the central nervous system. It is the primary neuronal target of the phytocannabinoid 
9-tetrahydrocannabinol and the endogenous cannabinoids (endocannabinoids) (Howlett et al. 2002; Pertwee, 2005). Activation of CB1 receptors leads to presynaptic inhibition of synaptic transmission in many regions of the central and peripheral nervous system (Freund et al. 2003; Szabo & Schlicker, 2005). It has recently been discovered that endocannabinoids and CB1 receptors play a physiological role in both short- and long-term regulation of the strength of synaptic transmission (for review see Alger, 2002; Wilson & Nicoll, 2002; Freund et al. 2003; Gerdeman & Lovinger, 2003; Diana & Marty, 2004; Chevaleyre et al. 2006).
One form of short-term synaptic depression is triggered by depolarization of postsynaptic neurons. Endocannabinoids mediate depolarization-induced suppression of inhibitory synapses (depolarization-induced suppression of inhibition, DSI) (Llano et al. 1991; Ohno-Shosaku et al. 2001; Varma et al. 2001; Wilson & Nicoll, 2001; Diana et al. 2002) and depolarization-induced suppression of excitatory synapses (depolarization-induced suppression of excitation, DSE) (Kreitzer & Regehr, 2001; Ohno-Shosaku et al. 2002). DSI and DSE are thought to be due to retrograde synaptic signalling involving the following steps: depolarization of postsynaptic neurons elicits an increase in intracellular calcium concentration; the elevated calcium levels trigger endocannabinoid synthesis; the released endocannabinoids diffuse to presynaptic axon terminals where they inhibit GABA (DSI) or glutamate (DSE) release by acting on presynaptic CB1 receptors. Another form of endocannabinoid-mediated short-term retrograde synaptic signalling is triggered by activation of certain Gq/11 protein-coupled receptors on postsynaptic neurons (Maejima et al. 2001, 2005; Kim et al. 2002; Brown et al. 2003; Galante & Diana, 2004; Marcaggi & Attwell, 2005). Retrogradely diffusing endocannabinoids are also involved in long-term synaptic depression evoked by moderate- to high-frequency stimulation of presynaptic axons (for example, Gerdeman et al. 2002; Robbe et al. 2002; Chevaleyre & Castillo, 2003).
The two best-characterized endocannabinoids are anandamide (Devane et al. 1992; Di Marzo et al. 1994) and 2-arachidonoylglycerol (2-AG) (Mechoulam et al. 1995; Stella et al. 1997). The significance of the more recently discovered endocannabinoids noladin ether, virodhamine and N-arachidonoyl-dopamine is not yet clear (for review on endocannabinoids see Mechoulam et al. 1998; Piomelli, 2003; Di Marzo, 2005).
Although the role of endocannabinoids in retrograde synaptic signalling is well established, the knowledge on the chemical identity of the endocannabinoid involved and the chain of events leading to enhanced endocannabinoid release is limited. Thus, the endocannabinoid mediating DSI and DSE has been determined only in the hippocampus (Kim & Alger, 2004; Makara et al. 2005; Straiker & Mackie, 2005). The aim of the present study was to determine which of the two major endocannabinoids, anandamide or 2-AG, is involved in DSI at interneuron–Purkinje cell synapses in the cerebellar cortex. To this end, we studied the effects of inhibitors of endocannabinoid formation and degradation on DSI. In addition, involvement of intracellular messengers in the stimulation of endocannabinoid synthesis was also studied. Some of the findings have been published in abstract form (Urbanski et al. 2005; Szabo et al. 2005).
|
Methods |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Endocannabinoid production in N18TG2 neuroblastoma cells
Confluent N18TG2 cells (DSMG, Braunschweig, Germany) were incubated for 20 min at 37°C in Dulbecco's modified Eagle's medium supplemented with fetal bovine serum (10%) and 6-thioguanine (10–4 M), according to the manufacturer's instructions. Endocannabinoid production was stimulated by addition of the calcium ionophore ionomycin (3 x 10–6 M) to the incubation medium. After stimulation, cells plus media were extracted with chloroform/methanol (2/1; v/v). Extracts were purified by open bed chromatography and 2-AG and anandamide were quantified by isotope dilution liquid chromatography – atmospheric pressure chemical ionization – mass spectrometry (Bisogno et al. 2003).
Brain slices
Ten- to 18-day-old NMRI mice were anaesthetized with isoflurane (> 3%) and decapitated. The brains were rapidly removed and placed in ice-cold artificial cerebrospinal fluid (ACSF) of the following composition (mM): NaCl 126, NaH2PO4 1.2, KCl 3, MgCl2 5, CaCl2 1, NaHCO3 26, glucose 20 and sodium lactate 4, pH 7.3–7.4 (after the solution was gassed with 95% O2–5% CO2). In most experiments, 250 µm thick sagittal slices of the cerebellar vermis were cut. In a few experiments, 300 µm thick oblique sagittal slices containing the caudate-putamen and the substantia nigra pars reticulata (SNR) or 300 µm thick coronal slices containing the hippocampus were prepared. Some experiments were carried out on 250 µm thick sagittal cerebellar slices prepared from 10- to 18-day-old Wistar rats. After cutting, the slices were stored in a Gibb chamber containing ACSF of the following composition (mM): NaCl 126, NaH2PO4 1.2, KCl 3, MgCl2 1, CaCl2 2.5, NaHCO3 26, glucose 10 and sodium lactate 4, pH 7.3–7.4. For patch clamping, brain slices were superfused at 20–24°C at a flow rate of 1.5 ml min–1 with ACSF of the following composition (mM): NaCl 126, NaH2PO4 1.2, KCl 3, MgCl2 1, CaCl2 2.5, NaHCO3 26 and glucose 10, pH 7.3–7.4.
Patch clamping
Neurons in slices were visualized with infrared video microscopy. Patch-clamp recordings were obtained with an EPC-9 amplifier under the control of TIDA software (HEKA Elektronik, Lambrecht, Germany). Series resistance compensation of 50% was usually applied. Series resistance was measured before and after recordings and experiments with major changes in series resistance (> 20%) were discarded. Inhibitory postsynaptic currents (IPSCs), spontaneous IPSCs (sIPSCs), miniature IPSCs (mIPSCs) and excitatory postsynaptic currents (EPSCs) were recorded at a holding potential of –70 mV with pipettes containing (mM): CsCl 147, MgCl2 1, Hepes 10, EGTA 1, ATP-Na2 4, GTP-Na 0.4 and N-ethyl-lidocaine chloride 2, pH 7.4. For IPSC and sIPSC recording, the superfusion ACSF contained the glutamate receptor antagonists 6,7-dinitroquinoxaline-2,3-dione (DNQX) (10–5 M) and DL-2-amino-5-phosphonopentanoic acid (AP5) (2.5 x 10–5 M). EPSCs were recorded in the presence of the GABAA receptor antagonist bicuculline (2 x 10–5 M) and the GABAB receptor antagonist CGP55845 (10–6 M). sIPSCs and mIPSCs were detected with the MiniAnalysis software (version 6.0.1; Synaptosoft, Decatur, GA, USA); the program allowed analysis of complex peaks consisting of several inhibitory currents. Amplitudes and recording times of sIPSCs and mIPSCs were transferred from MiniAnalysis to SigmaPlot (SPSS, Chicago, IL, USA), and a program written by us in Sigmaplot calculated cumulative amplitudes by summating amplitudes of all events within 10 s periods.
Fluorescence measurement of calcium concentrations in Purkinje cells
In addition to the intracellular solution used for recording postsynaptic currents, the patch pipette contained the low affinity calcium indicator Oregon green 488 BAPTA-5 N (Kd for calcium, 2 x 10–5
M; final concentration, 2 x 10–4
M). Fluorescence intensity in Purkinje cells was determined with an imaging system consisting of: Polychrome IV monochromatic light source, a cooled IMAGO VGA CCD camera and TILLvision imaging software (TILL Photonics, Gräfelfing, Germany). For measuring Oregon green fluorescence, the excitation wave length of the monochromatic light source was adjusted to 495 nm, and a 505DRLP dichroic filter and a 535AF45 bandpass emission filter were used (Omega Optical, Brattleboro, VT, USA). Fluorescence changes were evaluated in regions of interest (ROIs). Fluorescence values were corrected for background fluorescence. For further evaluation, ratios between stimulation-evoked fluorescence changes (F) and baseline fluorescence measured immediately before stimulation (F0) were calculated (F/F0 ratios).
Protocols and statistics
Recordings started 15 (electrophysiological recordings) or 40 min (calcium imaging) after establishment of the whole-cell configuration. DSI or DSE were usually elicited twice in a neuron, and the results of both trials were included in the evaluation. DSI and DSE were quantified by expressing cumulative sIPSC amplitudes as percentages of the initial reference value (PRE) (see Fig. 1). Amplitudes of electrical stimulation-evoked eIPSCs and eEPSCs were also expressed as percentages of the PRE value. Effects of drugs on PRE were systematically evaluated but reported only if they were significant. Solvent and drug superfusion is indicated in the figures; drug superfusion started at least 20 min before the respective DSI recordings. Drug-treated groups were compared with control groups generated during the same time period and receiving the appropriate solvent. Means ± S.E.M. are given throughout. Non-parametric statistical tests were used to identify significant differences. The two-tailed Mann–Whitney U test was used for comparisons between groups; significant differences are indicated by asterisks. The two-tailed Wilcoxon signed rank test was used for comparisons within groups (versus PRE); significant differences are indicated by filled symbols. P < 0.05 was taken as the limit of statistical significance, and only this level is indicated, even if P was < 0.01 or < 0.001. For multiple comparisons, the P levels were corrected according to Bonferroni.
|
Drugs were obtained from the following sources. Calbiochem (Darmstadt, Germany): 1,6-bis(cyclohexy-loximinocarbonylamino) hexane (RHC80267), autocamtide-2 related inhibitory peptide myristoylated, fluphenazine-mustard dihydrochloride and 2-[N-(2-hyd-roxyethyl)]-N-(4-methoxybenzenesulphonyl)]-amino-N-(4-chlorocinnamyl)-N-methylbenzylamine (KN-93); Cayman Chemicals (Ann Arbor, MI, USA): anandamide, 2-arachidonoyl glycerol, 3'carbamoyl-biphenyl-3-ylcyclohexylcarbamate (URB597) and arachidonoyl trifluoromethylketone (ATFMK); Endocannabinoid Research Group, Institute of Biomolecular Chemistry (Pozzuoli, Italy): arachidonoyl 5-hydroxytryptamine (AA-5-HT); F. Hoffman-La Roche (Basel, Switzerland): orlistat (also called tetrahydrolipstatin); Pharmacia & Upjohn (Kalamazoo, MI, USA): 4-[5-(4-chlorophenyl)-3-(trifluoromethyl)-1H-pyrazol-1-yl]benzenesulphonamide (SC-58236); Sanofi-Aventis (Chilly Mazarin, France): rimonabant (previously called SR141716A); Sigma (Milan, Italy): ionomycin.
Anandamide, 2-AG, ATFMK, fluphenazine-mustard, orlistat, RHC80267, rimonabant, SC-58236 and URB597 were dissolved in dimethylsulphoxide (DMSO), and stock solutions were stored at –32°C. Further dilutions were made with superfusion buffer; the final concentration of DMSO in the superfusion fluid was 
1 ml l–1. Control solutions (‘solvent’ in the figures) always contained the appropriate concentrations of DMSO. AA-5-HT was dissolved in ethanol; the final concentration of ethanol in the superfusion buffer was 1 ml l–1 (this was also the concentration of ethanol in the appropriate ‘solvent’ control group).
|
Results |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
2-AG is produced by cleavage of diacylglycerol by the recently cloned sn-1-diacylglycerol lipase that exists in two isoforms, and ß (Bisogno et al. 2003). This enzyme, as well as other diacylglycerol lipases, is inhibited by RHC80267 (Stella et al. 1997; Moriyama et al. 1999). A more potent and selective inhibitor of diacylglycerol lipase, orlistat, was identified by Bisogno et al. (2003); see also Bisogno et al. 2006. In the first series of experiments, we characterized the effects of orlistat on enzymes involved in endocannabinoid formation and degradation and on CB1 and CB2 cannabinoid receptors. Orlistat inhibited diacylglycerol lipase with an IC50 value of 6 x 10–8
M (Table 1). Monoglyceride lipase (the enzyme hydrolysing 2-AG), N-acylphosphatidylethanolamine-selective phospholipase D (the enzyme producing anandamide), fatty acid amide hydrolase (the enzyme hydrolysing anandamide) and triglyceride lipase were inhibited only at much higher concentrations (Table 1). Moreover, orlistat possessed only low affinity for CB1 and CB2 cannabinoid receptors (Table 1). Thus, orlistat selectively inhibits diacylglycerol lipase, and not other components of the endocannabinoid system. RHC80267, the compound which is traditionally used to inhibit diacylglycerol lipase, inhibited diacylglycerol lipase with an IC50 value of 6.5 ± 0.12 x 10–5
M (n
= 3). Thus, it is clearly less potent than orlistat.
|
Characterization of DSI
Cerebellar cortical Purkinje cells were patch clamped, and spontaneous inhibitory postsynaptic currents (sIPSCs) were recorded at a membrane potential of –70 mV. Cumulative amplitudes of sIPSCs were calculated for 10 s periods and expressed as percentages of the initial reference value PRE (see Fig. 1A). Four kinds of depolarization protocols were applied after PRE (Fig. 1A). Either a series of one, three or nine depolarizing pulses (from –70 to 0 mV for 100 ms) were applied at 1 Hz, or the neuron was depolarized once (from –70 to 0 mV for 5 s). Already a single 100 ms depolarizing step led to significant suppression of sIPSCs (i.e. DSI occurred). The suppressions following three and nine 100 ms depolarizing pulses and the 5 s depolarization were greater and lasted longer. We decided to carry out all following experiments in Purkinje cells using nine depolarizing pulses – a stimulus causing maximal DSI.
Depolarization by nine pulses in the presence of solvent suppressed sIPSCs to 56% of PRE and the suppression lasted about 120 s (Fig. 1B and C). No DSI was observed in the presence of the CB1 cannabinoid receptor antagonist rimonabant (10–6 M; Fig. 1B and D). This observation points to involvement of endocannabinoids and CB1 receptors in DSI. Altogether, the properties of DSI in our preparation were similar to those observed previously in Purkinje cells (e.g. Diana et al. 2002; Brenowitz & Regehr, 2003; Szabo et al. 2004).
Interference with 2-AG production
Effects of diacylglycerol lipase inhibitors on DSI. In order to identify the endocannabinoid involved in DSI, at first we employed inhibitors of 2-AG production. Our hypothesis was that if DSI is mediated by 2-AG, then diacylglycerol lipase inhibitors should attenuate DSI. When orlistat (10–7–10–5 M) was superfused, it blocked DSI in a concentration-dependent manner (Fig. 2A and B), with full blockade occurring at the concentration of 10–5 M (Fig. 2B). Superfusion of another inhibitor of diacylglycerol lipase, RHC80267 (10–4 M), also significantly attenuated DSI (Fig. 2C). The less than full blockade by RHC80267 is probably due to the low potency of this inhibitor, as described above.
|
As described above, orlistat (10–5 M) slightly enhanced the cumulative amplitude of sIPSCs before DSI (from 12123 ± 716 to 13705 ± 951 pA (10 s)–1; an increase of 13%). It is theoretically possible – although not likely – that these orlistat-evoked sIPSCs cannot be suppressed by depolarization. In such a case, orlistat would apparently decrease the calculated DSI even without true interference with diacylglycerol lipase. According to the data shown in Fig. 2B, DSI was 44% in the solvent-treated group. Assuming the above mechanism, orlistat (10–5 M) could decrease DSI to 39%, even without true interference with diacylglycerol lipase. However, the complete abolishment of DSI by orlistat (10–5 M; Fig. 2B) obviously cannot be explained by this mechanism.
It must be noted that although relatively high concentrations of orlistat and RHC80267 were necessary to inhibit diacylglycerol lipase in brain slices, this does not mean that these drugs lost their selectivity for this enzyme. In the case of the highly lipophilic cannabinoids (e.g. WIN55212-2, and 9-tetrahydrocannabinol), the concentrations eliciting effects in brain slices are always several orders of magnitude higher than the affinities determined on cell monolayers or membrane preparations. This is probably due to poor diffusion into the brain slice (see Brown et al. 2004). Very probably, the same holds true for orlistat and RHC80267 (octanol/buffer partition coefficients, > 1000 and 
25 000, respectively).
Depolarization-induced suppression of sIPSCs in the cerebellar cortex is due to endocannabinoid-mediated inhibition of the firing of interneurons, inhibition of voltage-dependent calcium channels in the GABAergic axon terminals and inhibition of the vesicle release machinery in the axon terminals (Kreitzer et al. 2002; Diana & Marty, 2003). For obtaining more information on the involvement of diacylglycerol lipase in these processes, we tested the effect of orlistat on depolarization-induced suppression of electrically evoked IPSCs (eIPSCs) and miniature IPSCs (mIPSCs) (Fig. 3). eIPSCs in Purkinje cells were evoked by electrical stimulation in the outer molecular layer. Three seconds before each eIPSC, Purkinje cells were depolarized from –70 to 0 mV for 1 s. This depolarization led to suppression of eIPSCs (eIPSC-DSI) in the presence of solvent (Fig. 3Aa and Ab). Rimonabant (10–6 M) and orlistat prevented eIPSC-DSI (Fig. 3Aa, Ac and Ad). mIPSCs were isolated in the presence of tetrodotoxin (3 x 10–7 M). Nine depolarizing pulses (from –70 to 0 mV for 100 ms) at 1 Hz suppressed the following mIPSCs (i.e. mIPSC-DSI occurred) (Fig. 3B). No mIPSC-DSI occurred in the presence of orlistat (10–5 M) (Fig. 3B). Blockade of eIPSC-DSI and mIPSC-DSI by orlistat indicates that the endocannabinoid inhibiting GABA release from the axon terminals was produced by diacylglycerol lipase. Blockade of eIPSC-DSI and mIPSC-DSI by orlistat also excludes the possibility that orlistat attenuated sIPSC-DSI (Fig. 2A and B) by enhancing the number of sIPSCs which cannot be suppressed by depolarization.
|
|
|
|
As the next procedure in the identification of the endocannabinoid involved in DSI we used drugs that interfere with the degradation of endocannabinoids. Anandamide is uniquely metabolized in the brain by fatty acid amide hydrolase; inhibition of this enzyme always leads to major increases in brain anandamide levels (Di Marzo et al. 1994; Cravatt et al. 2001; Cravatt & Lichtman, 2003). By contrast, it is thought that fatty acid amide hydrolase plays only a minor role in 2-AG degradation; accordingly, inhibition of this enzyme does not result in increases in brain 2-AG levels (Lichtman et al. 2002; but see also Maione et al. 2006). Fatty acid amide hydrolase can be inhibited by URB597 and arachidonoyl-5-hydroxytryptamine (Bisogno et al. 1998; Kathuria et al. 2003). Our hypothesis was that if DSI is mediated by anandamide, then inhibitors of fatty acid amide hydrolase should potentiate DSI.
At first we tested the interaction of URB597 with exogenous anandamide and 2-AG. Anandamide was superfused in the presence of solvent or URB597 (Fig. 7A). In solvent-treated slices, anandamide caused a slowly developing and weak inhibition of sIPCSs. In the presence of URB597 (5 x 10–7 M), the inhibition by anandamide developed faster and it was also more pronounced. Similar experiments were carried out with 2-AG (Fig. 7B). 2-AG superfused in the presence of solvent caused a small inhibition of sIPSCs. This inhibition was not changed in the presence of URB597 (5 x 10–7 M). These observations indicate that under our experimental conditions anandamide degradation, but not 2-AG degradation, is inhibited by URB597 (5 x 10–7 M). Concerning the main question of our study, URB597 did not change DSI (Fig. 7D). The other fatty acid amide hydrolase inhibitor, arachidonoyl-5-hydroxytryptamine, also did not affect DSI (Fig. 7C).
|
Cyclooxygenase-2 can transform 2-AG to glyceryl prostaglandins (Kozak et al. 2000). Based on the observation that cyclooxygenase-2 inhibitors prolonged DSI in the hippocampus, it was suggested that cyclooxygenase-2 plays an essential role in the degradation of 2-AG released during DSI (Kim & Alger, 2004). We carried out a series of experiments to determine whether cyclooxygenase-2 is involved in the degradation of the endocannabinoid mediating DSI in the cerebellum. The cyclooxygenase-2 inhibitor SC-58236 (10–6 and 10–5 M) did not significantly augment or prolong DSI in the cerebellum (Fig. 8A). We reasoned that the role of cyclooxygenase-2 in endocannabinoid metabolism would be more pronounced in the absence of other endocannabinoid-metabolizing pathways. Accordingly, we tested the effects of SC-58236 in brain slices in which fatty acid amide hydrolase and monoglyceride lipase were blocked by URB597 (5 x 10–7 M) and ATFMK (10–5 M) (Fig. 8B). SC-58236 (10–6 M) did not affect DSI also under this condition (Fig. 8B). This suggests that cyclooxygenase-2 does not play a role in the degradation of the endocannabinoid mediating DSI in the cerebellum.
|
The above observations point to 2-AG as the endocannabinoid mediating DSI at the GABAergic synapses between cerebellar cortical interneurons and Purkinje cells in the mouse. We carried out four final series of experiments to find out whether 2-AG plays a role in DSI and DSE at other synapses.
First, we studied the GABAergic synapses between axons originating in the caudate-putamen (striato-nigral pathway) and substantia nigra pars reticulata neurons in the mouse (Fig. 9A). The striato-nigral pathway was stimulated in the caudate-putamen every 2 s, and eIPSCs were recorded in substantia nigra pars reticulata neurons. After the initial reference period PRE, substantia nigra pars reticulata neurons were depolarized for 5 s. In the presence of solvent, this depolarization suppressed the amplitude of eIPSCs to 75% of PRE (Fig. 9Aa and Ab), similarly as in our previous study (Wallmichrath & Szabo, 2002). The cannabinoid antagonist rimonabant (10–6 M; Fig. 9Aa and Ac) as well as the diacylglycerol lipase inhibitor orlistat (10–5 M; Fig. 9Aa and Ad) prevented the DSI. This set of observations points to involvement of 2-AG and CB1 receptors also at this synapse.
|
Third, we studied the GABAergic synapses between interneurons and Purkinje cells in the cerebellar cortex of the rat (Fig. 9C). The protocol was identical to the protocol used for studying DSI in the mouse cerebellar cortex. In the presence of solvent, depolarization of Purkinje cells by a series of nine pulses led to suppression of the cumulative amplitude of sIPSCs to 50% of PRE (Fig. 9C). In the presence of rimonabant (10–6 M) or orlistat (10–5 M), DSI no longer occurred (Fig. 9C).
Fourth, we studied the glutamatergic synapses between parallel fibres and Purkinje cells in the cerebellar cortex of the mouse (Fig. 10). Parallel fibres were stimulated every 2 s, and eEPSCs were recorded in Purkinje cells. After the initial reference period PRE, Purkinje cells were depolarized by a series of nine pulses. In the presence of solvent, this depolarization suppressed the amplitude of eEPSCs to 30%–40% of PRE (Fig. 10Aa and 10Ab) (i.e. DSE occurred). In the presence of the cannabinoid antagonist rimonabant (10–6 M), DSE was abolished (Fig. 10Aa and 1Ac), thus verifying involvement of endocannabinoids and CB1 receptors. It is important to note that inhibition of diacylglycerol lipase by orlistat (10–5 M) fully prevented the DSE (Fig. 10Ba and Bc).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Inhibition of the enzyme responsible for degradation of 2-AG, monoglyceride lipase, led to potentiation of DSI in our experiments, pointing to a role of 2-AG. In contrast, a role for anandamide was ruled out by the observation that inhibition of the enzyme responsible for anandamide degradation, fatty acid amide hydrolase, did not enhance DSI. Similar observations led to the identification of 2-AG as the endocannabinoid mediating DSI and DSE in the hippocampus. Thus, DSI in hippocampal slices and DSE in cultured hippocampal neurons was not changed when fatty acid amide hydrolase was inihibited (Kim & Alger, 2004; Makara et al. 2005; Straiker & Mackie, 2005), but was prolonged when monoglyceride lipase was inhibited (Makara et al. 2005). In addition, a role of 2-AG in DSE in cultured hippocampal neurons is supported by the finding that only the kinetics of action of exogenous 2-AG, but not of anandamide and noladin ether, were compatible with the kinetics of DSE (Straiker & Mackie, 2005).
Inhibition of the enzyme responsible for 2-AG production, the sn-1-selective diacylglycerol lipase, by two inhibitors belonging to different chemical classes (RHC80267 and orlistat), led to attenuation and abolishment of DSI. This is the first convincing demonstration of a role of diacylglycerol lipase in DSI. We know only of one study in which a weak inhibition of DSE by RHC80267 was shown, in cultured hippocampal neurons (Straiker & Mackie, 2005). As shown by the data in Table 1, orlistat selectively inhibits diacylglycerol lipase: it does not interfere with 2-AG degradation, anandamide production and degradation and with cannabinoid receptors. In our brain slice experiments it did not activate or block CB1 receptors and it inhibited DSI at a step downstream of calcium entry into the postsynaptic Purkinje cells. Therefore, it is very likely that orlistat (and RHC80267) blocked DSI by inhibiting endocannabinoid production at the level of diacylglycerol lipase.
Consistent with a role of 2-AG in retrograde signalling, all necessary molecular components are present in the appropriate compartments at the synapses between interneurons and Purkinje cells. The enzyme producing 2-AG, diacylglycerol lipase, is highly expressed in the dendrites of Purkinje cells (Bisogno et al. 2003; Yoshida et al. 2006). The target receptors, CB1 receptors, are present in the presynaptic axon terminals of interneurons (Diana et al. 2002; Szabo et al. 2004; Kawamura et al. 2006). Finally, the enzyme degrading 2-AG, monoglyceride lipase, is present in axon terminals in the molecular layer (Dinh et al. 2002; Gulyas et al. 2004; Saario et al. 2004). The anandamide degrading enzyme fatty acid amide hydrolase is localized in Purkinje cell dendrites (Gulyas et al. 2004); this localization would be disadvantageous if retrograde signalling was to be mediated by anandamide. Consistent with the role of 2-AG in retrograde signalling is its high concentration in the brain (much higher than that of anandamide; Sugiura et al. 1995; Stella et al. 1997; Jung et al. 2005). Moreover, electrical stimulation of neurons leads to an increase in 2-AG biosynthesis, but not of anandamide biosynthesis (in the hippocampus; Stella et al. 1997).
The results of our experiments with orlistat suggest that DSE in the mouse cerebellum, DSI in the mouse substantia nigra pars reticulata and DSI in the rat cerebellum are also mediated by 2-AG. As shown by others, DSI and DSE in the hippocampus are also mediated by 2-AG (Kim & Alger, 2004; Makara et al. 2005; Straiker & Mackie, 2005).
Several observations indicate that 2-AG is the endocannabinoid released during short-term retrograde signalling triggered by stimulation of postsynaptic Gq/11 protein-coupled receptors. Thus, inhibition of GABAergic transmission elicited by stimulation of type 1 metabotropic glutamate receptors on cerebellar Purkinje cells was attenuated by the phospholipase C inhibitor U73122 and by RHC80267 (Galante & Diana, 2004). Activity-dependent short suppression of excitatory transmission in the cerebellar cortex and the ventral tegmental area was also inhibited by orlistat and RHC80267 (Melis et al. 2004; Safo & Regehr, 2005).
There are some recent observations on the endocannabinoid involved in long-term synaptic plasticity. Long-term depression in the cerebellum and hippocampus was abolished in the presence of RHC80267 and orlistat (Chevaleyre & Castillo, 2003; Safo & Regehr, 2005). These findings are compatible with a role of 2-AG. However, long-term depression in the amygdala was stronger in mice lacking fatty acid amide hydrolase and not changed by U73122 and RHC80267 – findings pointing to a role of anandamide (Azad et al. 2004).
Together, the experiments studying the chemical nature of the endocannabinoids involved in retrograde signalling suggest that 2-AG is involved in DSI and DSE, short-term retrograde signalling driven by activation of Gq/11 protein-coupled receptors and long-term synaptic plasticity at the majority of synapses. Yet, the role of anandamide should not be underestimated. For example, experiments in which the biosynthetic pathway of ananamide is manipulated may uncover hitherto unknown roles for anandamide in synaptic plasticity.
It is remarkable that pathways of 2-AG production may differ depending on the induction protocol. The operation of phospholipase C is necessary for long-term depression and for retrograde signalling elicited by stimulation of Gq/11 protein-coupled receptors, but not for DSI and DSE (Chevaleyre & Castillo, 2003; Hashimotodani et al. 2005; Maejima et al. 2005; Brenowitz et al. 2006; Edwards et al. 2006). The phospholipase C inhibitor U73122 did not change DSI in our experiments, showing that phospholipase C is also not necessary for DSI in the mouse cerebellar cortex. It has been shown that in neurons, diacylglycerol for 2-AG production can be generated not only by phospholipase C, but also via other pathways (Bisogno et al. 1999; Sugiura et al. 2002).
Less clear is the role of diacylglycerol lipase. In the hippocampus, retrograde signalling elicited by stimulation of muscarinic acetylcholine receptors and long-term depression were more sensitive than DSI to inhibitors of diacylglycerol lipase (Chevaleyre & Castillo, 2003; Edwards et al. 2006). Our own experiments verified the resistance of hippocampal DSI against the diacylglycerol lipase inhibitor orlistat. Similar observations have been made in the cerebellum by Safo & Regehr (2005). Thus, short-term retrograde signalling elicited by stimulation of parallel fibres and long-term depression elicited by combined stimulation of parallel fibres and climbing fibres was inhibited by the diacylglycerol lipase inhibitors RHC80267 and orlistat. In contrast, DSE and DSI were not blocked by RHC80267 and orlistat (Safo & Regehr, 2005; Brenowitz et al. 2006; the reason for the obvious discrepancy between these findings and our findings is not known). Thus, it seems that diacylglycerol lipase participates in the endo-cannabinoid production in retrograde signalling elicited by stimulation of Gq/11 protein-coupled receptors and in some forms of long-term depression. According to our own results, DSI and DSE can be mediated by 2-AG produced by diacylglycerol lipase. However, it seems that under certain conditions, DSI and DSE is mediated by 2-AG which is not produced by diacylglycerol lipase.
It is not known how elevated intracellular calcium levels lead to enhanced endocannabinoid production in the postsynaptic neuron during DSI and DSE. The role of calmodulin and Ca2+–calmodulin-dependent protein kinase II in long-term synaptic plasticity is well recognized (Kano et al. 1996; Fink & Meyer, 2002; Xia & Storm, 2005). Our results suggest for the first time that these two calcium-dependent messengers are essential for endocannabinoid production during short-term synaptic plasticity such as DSI. Calmodulin and Ca2+–calmodulin-dependent protein kinase II (a group of kinases) are present in cerebellar Purkinje cells and are activated by depolarization protocols similar to that used in the present study (see Kano et al. 1996). Theoretically, the activity of both phospholipase C and diacylglycerol lipase might be stimulated during calcium-induced endocannabinoid release. As phospholipase C is not necessary for DSI, it is likely that the target of calmodulin and Ca2+–calmodulin-dependent protein kinase II during DSI is diacylglycerol lipase. However, further experiments are necessary to characterize exactly the calcium-dependent signalling cascade operating in the postsynaptic cell during DSI and DSE.
A corollary observation of the present study is that cyclooxygenase-2 does not participate in 2-AG degradation in the cerebellum. We employed SC-58236, a potent and selective inhibitor of cyclooxygenase-2 (IC50 for inhibition of cyclooxygenase-2, 10–8 M; IC50 for inhibition of cyclooxygenase-1, 1.8 x 10–5 M; Penning et al. 1997). It is very likely that at concentrations 100- to 1000-fold higher than its IC50 value for inhibition of cyclooxygenase-2 in cell free systems, SC-58236 indeed inhibited cyclooxygenase-2 in our cerebellar slices. Yet, the magnitude and duration of DSI were not changed, even when other 2-AG degrading pathways were inhibited. Thus, it is probable that in the cerebellum, 2-AG is solely metabolized by monoglyceride lipase, as suggested also by previous biochemical experiments (Saario et al. 2004). Notably, the recent study of Straiker & Mackie (2005) also showed that cyclooxygenase-2 does not play a role in the metabolism of 2-AG during hippocampal DSE.
Activity-dependent alteration of the strength of synaptic transmission – synaptic plasticity – is a widely occurring phenomenon and is thought to be the basis of memory formation and learning. Endocannabinoids mediate short- and long-term depression of synaptic strength in many brain regions (Alger, 2002; Wilson & Nicoll, 2002; Freund et al. 2003; Gerdeman & Lovinger, 2003; Diana & Marty, 2004; Chevaleyre et al. 2006). Importantly, endocannabinoid signalling can be activated by physiological and pathophysiological stimuli (e.g. Robbe et al. 2002; Brown et al. 2003; Chen et al. 2003; Melis et al. 2004; Brenowitz & Regehr, 2005; Maejima et al. 2005; Brenowitz et al. 2006). The identification of the endocannabinoid, and of the steps leading to its synthesis, involved in one form of short-term depression is an important advance in the understanding of these endocannabinoid-mediated processes. Developing drugs that interfere with the synthesis and degradation of 2-AG will make it possible to influence synaptic plasticity and its behavioural correlates.
|
Footnotes |
|---|
|
References |
|---|
|
TopAbstractIntroductionMethodsResultsDiscussionReferences |
|---|
Appendino G, de Petrocellis L, Trevisani M, Minassi A, Daddario N, Moriello AS et al. (2005). Development of the first ultra-potent ‘capsaicinoid’ agonist at transient receptor potential vanilloid type 1 (TRPV1) channels and its therapeutic potential. J Pharmacol Exp Ther 312, 561–570.
Azad SC, Monory K, Marsicano G, Cravatt BF, Lutz B, Zieglgänsberger W & Rammes G (2004). Circuitry for associative plasticity in the amygdala involves endocannabinoid signaling. J Neurosci 24, 9953–9961.
Bisogno T, Cascio MG, Saha B, Mahadevan A, Urbani P, Minassi A et al. (2006). Development of the first potent and specific inhibitors of endocannabinoid biosynthesis. Biochim Biophys Acta 1761, 205–212.[Medline]
Bisogno T, Howell F, Williams G, Minassi A, Cascio MG, Ligresti A et al. (2003). Cloning of the first sn1-DAG lipases points to the spatial and temporal regulation of endocannabinoid signaling in the brain. J Cell Biol 163, 463–468.
Bisogno T, Melck D, De Petrocellis L, Bobrov MY, Gretskaya NM, Bezuglov VV, Sitachitta N, Gerwick WH & Di Marzo V (1998). Arachidonoylsrotonin and other novel inhibitors of fatty acid amide hydrolase. Biochem Biophys Res Commun 248, 515–522.[CrossRef][Medline]
Bisogno T, Melck D, De Petrocellis L & Di Marzo V (1999). Phosphatidic acid as the biosynthetic precursor of the endocannabinoid 2-arachidonoylglycerol in intact mouse neuroblastoma cells stimulated with ionomycin. J Neurochem 72, 2113–2119.[CrossRef][Medline]
Brenowitz SD, Best AR & Regehr WG (2006). Sustained elevation of dendritic calcium evokes widespread endocannabinoid release and suppression of synapses onto cerebellar Purkinje cells. J Neurosci 26, 6841–6850.
Brenowitz SD & Regehr WG (2003). Calcium dependence of retrograde inhibition by endocannabinoids at synapses onto Purkinje cells. J Neurosci 23, 6373–6384.
Brenowitz SD & Regehr WG (2005). Associative short-term synaptic plasticity mediated by endocannabinoids. Neuron 45, 419–431.[CrossRef][Medline]
Brown SP, Brenowitz SD & Regehr WG (2003). Brief presynaptic bursts evoke synapse-specific retrograde inhibition mediated by endogenous cannabinoids. Nat Neurosci 6, 1048–1057.[CrossRef][Medline]
Brown SP, Safo PK & Regehr WG (2004). Endocannabinoids inhib
